New technology

Surg Endosc (1997) 11: 864867

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Pathophysiological and clinical aspects of the CO2 pneumoperitoneum (CO2-PP)

H. Gebhardt,1 A. Bautz,1 M. Ross,1 D. Loose,2 H. Wulf,2 H. Schaube1
1 2

Department of General and Thoracic Surgery, University of Kiel, Arnold-Heller-Str. 7, 24105 Kiel, Germany Department of Anaesthesiology, University of Kiel, Arnold-Heller-Str. 7, 24105 Kiel, Germany

Received: 13 December 1996/Accepted: 8 January 1997

Abstract. Experimental studies demonstrated a severe cardiac load of the CO2 pneumoperitoneum caused by an accelerated after- and a decreased preload. Patients displaying cardiovascular risks are therefore often rejected from laparoscopic surgery. Hence, the pathophysiological changes and the intraoperative risk of the CO2 pneumoperitoneum in high-risk cardiopulmonary patients (NYHA IIIII, n 15) undergoing laparoscopic cholecystectomy are described. The changes in cardiac after- and preload seem to be due to the elevated intraabdominal pressure rather than transperitoneally resorbed CO2 and are reversible by desufflation. In one patient conversion to open operation had to be performed because of a severe drop in cardiac output and right ventricle ejection fraction. Mixed oxygen saturation was predicting intraoperative worsening in this case. The described pathophysiological changes may seem to be well tolerated even in high-risk cardiac patients. Monitoring of hemodynamics should include an arterial catheter line and blood gas analyses. Pharmacologic interventions or pressureless laparoscopic procedures might not be necessary as long as laparoscopic cholecystectomy is performed. Key words: Laparoscopic surgery CO2 pneumoperitoneum Pathophysiology Intraoperative risk Monitoring

The reduction in pain and pulmonary dysfunction achieved via laparoscopic surgery decreases the potential for postoperative (p.op.) complications and prolonged recovery. Patients displaying cardiovascular risks or patients undergoing lengthy operations should profit most from these benefits, since preexisting diseases are not exacerbated as after laparotomy. However, the scientific evaluation of the pathophysiological changes caused by CO2 pneumoperitoneum
Correspondence to: H. Gebhardt

(CO2-PP) and elevated intraabdominal pressure (IAP) has lagged behind the clinical use of this technique. The intraoperative stress may be greater than under laparotomy since experimental studies have demonstrated severe changes in heart rate, systemic vascular resistance, cardiac output, and pulmonary function [5, 6, 1618]. The increased intraabdominal pressure led to a reduction of venous reflux from the periphery and squeezed the venous reservoir within the abdominal cavity. Cardiac afterload parameters, such as the systemic vascular resistance, increased significantly. Establishing the pneumoperitoneum caused a 40% reduction in cardiac output, raising concerns about the clinical use of this method [1, 5]. However, only few clinical studies have been undertaken to evaluate the pathophysiological changes since an invasive monitoring is required to investigate the hemodynamic alterations observed in experimental studies. A reason for this might be that, so far, the majority of laparoscopic operations have been performed in young and healthy patients not allowing a substantial, invasive hemodynamic monitoring because of ethical and moral problems [1, 8, 9, 14]. Only a few studies have been done to evaluate the indications and perioperative risks of laparoscopic surgery in high-risk cardiac patients. Despite the significant benefits, patients with cardiopulmonary risk are therefore often rejected from laparoscopic surgery. For further evaluation of the pathophysiology and the safety of this method we investigated in a prospective, clinical study the pathophysiological changes caused by CO2 pneumoperitoneum that occur in patients with underlying heart and/or pulmonary diseases who undergo laparoscopic cholecystectomy using an intensive, invasive hemodynamic monitoring throughout the operative procedures.

Materials and methods

Fifteen patients with cardiopulmonary insufficiency stage II (n 9) and stage III (n 6), according to the classification of the New York Heart

865 Table. 1. Patient characteristicsa Operation time (min) 47 71 88 56 91 57 64 74 58 69 74 60 61 68 88

Age 47 53 67 56 75 81 83 46 73 84 51 66 78 63 78
a

Sex F F M F F F M F F F F M M F M

Cardiac disease Aortic stenosis, P.myoc.inf. I.H.D., C.C.P. Mitral stenosis, heart block, artrial fibrillation Idio.h.suba.st., left ventricular hypertrophy I.H.D., stable angina P.myoc.inf., artrial fibrillation P.myoc.inf., C.C.P. Tricuspidal insuff., P.myoc.inf. I.H.D., P.myoc.inf. I.H.D., exertional angina Cor pulmonale, stable angina ACVB, heart block I.H.D., C.C.P. Cor pulmonale, I.H.D. P.myoc.inf., I.H.D.

NYHA II III III II II II II III II II II III III III II

Others Diabetes mellitus II Peripheral vascular disease, diabetes mellitus II Hypertension

Hypertension, diabetes mellitus II Hypertension COLD Hypertension COLD, diabetes mellitus II

P.myoc.inf., postmyocardial infarction; I.H.D., ischemic heart disease, C.C.P., congestive cardiomyopathy; Idio.h.suba.st., idiopathic hypertrophic subaortic stenosis.

Association, were scheduled for laparoscopic cholecystectomy and considered as candidates for the study. Four of these patients suffered from a valvular heart disease, eight an ischemic heart disease, and two from a cor pulmonale as a result of pulmonary hypertension. Informed consent was obtained from all participants. The characteristics of the patients and the duration of the operation are presented in Table 1. The mean duration of the CO2-PP was 68 12 min at an intraabdominal pressure (IAP) of 14 mmHg, the mean age of the patients 66.7 14.3 years. General anesthesia was induced with etomidate (0.2 mg/kg body weight i.v.), fentanyl (0.1 mg/kg body weight i.v.), succinylcholine (11.5 mg/kg body weight i.v.) and maintained with isoflurane (0.41 vol.%), N2O/O2 2:1, sufentanyl and vecuronium bromide (0.1 mg/kg body weight). Mechanical ventilation (Draeger Cicero, Draeger Co., FRG) was performed with the respiratory rate adjusted to achieve normal arterial (3035 mmHg) and end-tidal pCO2 values. The invasive hemodynamic monitoring included a pulmonary artery Swan-Ganz Thermodilution catheter (Model 744 H-7.5 F, Baxter Int. Coop., USA) and a radial arterial catheter line (1.0, Argyle Mediport 2, Sherwood Int. Co., USA) for measurement of centralvenous pressure (CVP), pulmonary arterial wedge pressure (PAWP), cardiac output (CO), right ventricle ejection fraction (RVEF), and mean arterial blood pressure (MAP) using on-line hemodynamic monitoring (Sirecust 401-1 and 961, Siemens Corp., FRG). For an indirect measurement of the intrathoracal pressure changes, a catheter line was placed in the esophagus (IEP) at the level of the right ventricle (Sirecust 401-1, Siemens Corp., FRG). Calculated values like the systemic vascular resistance (SVR) were obtained from CVP, CO, and MAP. By subtracting the IEP from the CVP the transmural right atrial pressure (TMP) was obtained. After induction of anesthesia hemodynamic and ventilatory parameters were normalized to achieve SVR of less than 1,800 dyn s/cm5, CO of more than 3.9 l/min, and PCWP between 8 and 12 mmHg. The abdominal insufflation of the CO2-PP at an intraabdominal pressure of 14 mmHg was then performed. Directly before and 2 min after establishing CO2-PP a set of data was obtained. Additional values under the pneumoperitoneum and after desufflation of it were obtained every 5 min. A final set of data were collected 90 min after desufflation of the abdomen. In addition the end-tidal carbon dioxide concentrations (VEXPCO2) were measured and the positive end-expiratory pressure (PEEP), peak inspiratory pressure (PEAK), as well as the inspiratory plateau phase (PLAT) and tidal volume (Vt) were protocolled by the respiratory system (Draeger Cicero, Draeger Co., FRG). Arterial blood samples were analyzed with a radiometer ABL2 blood-gas analyser which was calibrated every hour for measurement of arterial pCO2 (PaCO2) and O2 (PaO2). Statistical analysis was performed with a two-way analysis of variance for repeated measures, and p values < 0.05 were regarded as significant.

Results Significant hemodynamic but not ventilatory alterations were observed in all patients. Already at 2 min after establishing the CO2 pneumoperitoneum we noticed a significant increase in the CVP from 15.5 2.9 mmHg to 22.4 3.4 mmHg. This was followed by a decrease to 18.4 mmHg 2.4 mmHg during the elevated intraabdominal pressure caused by the CO2-PP. However, immediately after induction of the elevated IAP the intraesophageal (IEP) and the intrathoracal pressure increased even more strongly. As a result the TMP (Fig. 1), a calculated value (CVP-IEP) giving the actual information about precardial load, showed a constant decrease throughout CO2-PP from 15.2 3.4 mmMHg to 3 2.8 mmHg (45 min CO2-PP). Only after release of the pneumoperitoneum did these parameters of cardiac preload returned to baseline, preinsufflation values (15 min after desufflation). The MAP increased initially (5 min CO2-PP) after establishing the CO2-PP from 86 12 mmHg to 111 18 mmHg. PAP also increased from 22.1 4 to 32.2 5 mmHg after 5 min. The MAP- and PAP-baseline values were reached before desufflation (45 min). Other afterload parameters such as SVR (Fig. 2) rose from 1,770 224 to 2,415 221 dyn s/cm5 immediately after induction of CO2-PP raised the elevated IAP, before any reasonable amount of carbon dioxide was resorbed transperitoneally. These afterload parameters returned to almost normal values of 1,615 241 dyn s/cm5 after about 60 min of CO2-PP. As a consequence of these changes in cardiac after- and preload, CO (Fig. 3) dropped after induction of CO2-PP from 3.7 0.6 to 2.8 0.4 l/min. This was followed by an increase of up to 4.2 0.6 (30 min CO2-PP). After desufflation a peak increase of 5.04 0.78 l/min occurred. The increase in CO was accompanied by an increase in heart rate. Cardiac stroke volume decreased constantly through-

866

Fig. 1. Transmural right atrial pressure (TMP) in high-risk cardiac patients (NYHA IIIII, n 15) before (10 min, 0 min), during (060 min), and after (60120 min) CO2-pneumoperitoneum at an intraabdominal pressure of 14 mmHg.

Fig. 3. Cardiac output in high-risk cardiac patients (NYHA IIIII, n 15) before (10 min, 0 min), during (060 min), and after (60120 min) CO2 pneumoperitoneum at an intraabdominal pressure of 14 mmHg.

Fig. 2. Systemic vascular resistance in high-risk cardiac patients (NYHA IIIII, n 15) before (10 min, 0 min), during (060 min), and after (60120 min) CO2 pneumoperitoneum at an intraabdominal pressure of 14 mmHg.

Fig. 4. Right ventricle ejection fraction in high-risk cardiac patients (NYHA IIIII, n 15) before (10 min, 0 min), during (060 min), and after (60120 min) CO2 pneumoperitoneum at an intraabdominal pressure of 14 mmHg.

out the CO2-PP. The decrease in cardiac pre- and the increase in afterload led to a constant drop of the right ventricle ejection fraction (RVEF) from 45 3.4% to 33 2.1% after 60 min of CO2-PP (Fig. 4). In one patient the ejection fraction dropped to 23% after 32 min of CO2-PP, demanding conversion to open operation. This patient began the operation with a low CO of 2.98 l/min and a high SVR of 2,254 dyn s/cm5. During the operation the mixed oxygen saturation decreased significantly, reflecting inadequate pulmonary perfusion. After desufflation these parameters returned to preinsufflation values within 4 min and the patient did well during the postoperative course. VexpCO2 increased from preinsufflation values of 238 14 ml/min to 298 28 ml/min (30 min after insufflation) since minute ventilation was adapted to maintain normal PaCO2 between 30 and 35 mmHg. PEEP values did not show any significant changes. After establishing the CO2PP PLAT values increased significantly from 20 3.4 mbar to 27.5 2.7 mbar and PEAK values from 26.8 4.5 to 33.7 2.4 mbar. The changes in PEAK values were not regarded as significant.

Discussion Experimental pull-through manometries via V. femoralis up to V. cava superior under CO2-PP in the sheep have already demonstrated a reduction zone in the elevated IAP begin-

ning at the diaphragma along the basal parts of the thorax up to the level of the right atrium [5]. Only at that level the intravasal pressure within V. cava superior was like the IEP changes plus the CVP. The compression of the basal parts of the thorax might be responsible for the increase in PLAT and PEAK. This compression and the diaphragmatic elevation decrease pulmonar functional residual capacity, anatomical deadspace, and size of airways [5, 14]. They might also be responsible for the augmented pulmonar resistance under CO2-PP. However, there seem to be no significant changes in PEAK, PEEP, PaO2, and PaCO2 during laparoscopic cholecystectomy as long as an adapted, controlled mechanical ventilation is performed. The results of the pull-through manometries via V. femoralis superior to V. cava superior and the alterations in TMP (Fig. 2) indicate that CVP alone is not an appropriate indicator of cardiac preload, since the observed increase is caused by the elevated IAP carrying off into the basal parts of the thorax, altering CVP. As shown by the TMP, there is actually a constant decrease in cardiac preload during the pneumoperitoneum. This has to be kept in mind during laparoscopic procedures under CO2-PP. Other groups described a 40% reduction of the flow within intraabdominal V. cava and V. mesenterica as well as a stasis and dilatation within common V. femoralis [2, 8, 12]. These findings and the changes in TMP indicate a shift of blood volume from the center to the periphery with the elevated IAP reducing blood flow within V. cava abdominalis. This augments the decrease of cardiac preload and

867

might be, in contrast to the findings of Safran and Ho et al., one of the main causes of the observed changes in cardiac output [6, 16]. The venous stasis in the periphery might also promote venous thrombosis of the lower extremities. Intermittent sequential pneumatic compression, as suggested by Millard et al., reversed venous stasis within common femoral vein, returning peak velocity to normal [12]. This might be of special interest in preventing thrombosis in high-risk patients or in the case of longstanding laparoscopic operations. The parameters of cardiac afterload, as there is the SVR and PAR, increased up to 35% also immediately after induction of the CO2-PP before any reasonable amount of CO2 was resorbed. During the time investigated, these parameters returned to baseline values 60 min after insufflation. The initial onset of hemodynamic alterations, the reverse effects after desufflation, and the results of the pullthrough manometry as well as the TMP, indicate that the observed hemodynamic changes are caused by direct effects of the elevated IAP rather than systemic effects of transperitoneally resorbed CO2 as suggested by Ho et al. [6]. Kashtan et al. and Leighton et al. described similar changes after hydroperitoneum, helium or nitrous oxide pneumoperitoneum in the dog, further suggesting that the elevated IAP is the main cause [7, 10]. The CO2-PP, and the elevated IAP, also lead to an initial release of epinephrine, norepinephrine, and vasopressin as soon as 5 min after insufflation. The elevated catecholamine levels returned to baseline values only after desufflation [11, 13]. The observed changes in cardiac afterload, which are also described by Koksoy et al. [9], are therefore not only due to direct effects of the elevated IAP on the vascular system but also to increased catecholamine levels augmenting systemic vascular resistance. The changes in cardiac pre- and afterload caused the observed alterations in CO (Fig. 3) and RVEF (Fig. 4). These might be critical for the organism, since cardiac wall tension and oxygen demand increase. Simultaneously, coronary perfusion is reduced because of a reduced stroke volume and shortened diastolic filling time during tachycardia. Hypovolemia, aggravating these hemodynamic changes, should therefore be avoided as long as CO2-PP is used for laparoscopic procedures. However, these data show that these changes are tolerated, and most of the patients with myocardial insufficiency respond well to CO2-PP. Only in one patient did conversion to open operation have to be performed because of a severe drop in CO and RVEF. In this case a reduced mixed oxygen saturation and an increase in SVR was predicting intraoperative worsening. The effects were easily reversible by taking away the elevated IAP by removing the CO2-PP. Despite the described pathophysiological alterations of the CO2-PP, laparoscopic surgery might be safely performed even in high-risk cardiac patients as long as careful perioperative monitoring and controlled mechanical ventilation are performed. The intraoperative monitoring should include an arterial catheter line, in-line capnography for VEXPCO2, and a close look at the mixed oxygen saturation predicting intraoperative worsening. Safran et al. recom-

mended the use of a pulmonary arterial catheter line [16]. We do not regard this as mandatory any more since the mixed oxygen saturation reflected inadequate perfusion and the increase in systemic vascular resistance as well as the drop in CO and RVEF. Extraperitoneal carbon dioxide insufflation, pharmacologic interventions, or so-called pressureless laparoscopic procedures, as described by Feigh and Chin et al., Rademaker et al. as well as Paolucci and Gutt, might not be necessary as long as laparoscopic cholecystectomy is performed [3, 4, 15, 18]. References
1. Bannenberg JJ, Rademaker BM, Grundeman PF, Kalkman CJ, Meijer DW, Klopper PJ (1995) Hemodynamics during laparoscopy in the supine or prone position. An experimental study. Surg Endosc 2: 125 127 2. Beebe DS, McNavin MP, Boyle M (1991) Evidence of venous stasis after abdominal insufflation for laparoscopic surgery. Anaesthesiol 77(3A): 144148 3. Chin AK, Eaton J, Tsoi EKM, Smith RS, Fry WR, Henderson J, McColl MB, Moll FH (1994) Gasless laparoscopy using a planar lifting technique. J Am Coll Surg 178: 401403 4. Feig BW, Berger DH, Dougherty TB, Dupuin JF, Bartholomew HSI, Hickey RC, Ota DM (1994) Pharmacologic intervention can reestablish baseline hemodynamic parameters during laparoscopy. Surgery 116: 733741 5. Gebhardt JH, Fa ndrich F, Ross M, Schaube H, Loose D (1996) Intraoperative risk and hemodynamic effects of the CO2 -pneumoperitoneum in laparoscopic surgery. An experimental study. Min Invas Ther Allied Technol 5: 207210 6. Ho HSH, Saunders CJ, Corso FA, Wolfe BM (1993) The effects of CO2-pneumoperitoneum on hemodynamics in hemorrhaged animals. Surgery 114(2): 381388 7. Kashtan J, Gren JF, Parson EQ, Holcroft JW (1981) Hemodynamic effects of increased intraabdominal pressure. J Surg Res 30: 249256 8. Kazuhiro I, Tetsuto T, Hirotoshi W, Yasuhiro T (1994) Intraabdominal venous pressure during laparoscopic cholecystectomy. HBP Surg 8: 1317 9. Koksoy C, Kuzu MA, Kurt I, Kurt N, Yerdel MA, Tezcan C, Aras N (1995) Hemodynamic effects of pneumoperitoneum during laparoscopic cholecystectomy: a prospective comparative study using bioimpedance cardiography. Br J Surg 82(7): 972974 10. Leighton TA, Liu SY, Bongard FS (1993) Comparative cardiopulmonary effects of carbon dioxide versus helium pneumoperitoneum. Surgery 113: 527531 11. Mikami O, Kawakita S, Fujise K, Shingu K, Takahashi H, Matsuda T (1996) Catecholamine release caused by carbon dioxide insufflation during laparoscopic surgery. J Urol 155(4): 13681371 12. Millard JA, Hill BB, Cook PS (1993) Intermittent sequential pneumatic compression in prevention of venous stasis associated with pneumoperitoneum during laparoscopic cholecystectomy. Arch Surg 128: 914919 13. Punnonen R, Viinama ki O (1982) Vasopressin release during laparoscopy: role of increased intraabdominal pressure. Lancet 16: 175176 14. Puri GD, Singh H (1992) Ventilatory effects of laparoscopy under general anaesthesia. Br J Anaesth 68: 211213 15. Rademaker BM, Meyer DW, Bannenberg JJ, Klopper PJ, Kalkman CJ (1995) Laparoscopy without pneumoperitoneum. Effects of abdominal wall retraction vs. carbon-dioxide insufflation on hemodynamics and gas exchange in pigs. Surg Endosc 9(7): 797801 16. Safran DB, Orlando R III (1994) Physiological effects of the pneumoperitoneum. Am J Surg 167: 281286 17. Westerband A, van de Water JM, Amzallag M (1992) Cardiovascular changes during laparoscopic cholecystectomy. Surg Gynecol Obstet 175: 535538 18. Wright DM, Serpell MG, Baxter JN, ODwyer PJ (1995) Effect of extraperitoneal carbon dioxide insufflation on intraoperative blood gas and hemodynamic changes. Surg Endosc 9(11): 11691172

Surg Endosc (1997) 11: 802804

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Immediately recognizable benefits and drawbacks after laparoscopic colon resection for benign disease
R. Bergamaschi, J.-P. Arnaud
Department of Visceral Surgery, Angers University Hospital, 4 rue Larrey, 49033 Angers, France Received: 15 October 1996/Accepted: 13 December 1996

Abstract Background: A prospective assessment of the impact of laparoscopic colon resection (LCR) was carried out in order to quantify immediately recognizable benefits and limitations of this approach. Methods: Elective LCR was attempted in 95 selected patients (mean age 64 years, range 3981 years) presenting with benign disease of the colon. A completely intracorporeal approach was adopted. Results were compared with a control group of 90 patients who had previously undergone open colectomy (OC) by the same surgeons at the same institution. Results: There were no perioperative deaths. Intraoperative complications included difficult extraction of accidentally detached anvil (n 1), air leak at colonoscopy (n 2), and conversion to OC (n 1). Operating time was significantly longer after LCR compared with OC (180 10.3 vs 116 97, p < 0.001). Passage of flatus (3.5 1.2 days vs 4.4 1.4, p < 0.5) and morbidity (4 vs 3, p 0.48) were not significantly different in the two groups. Hospital stay was significantly shorter after LCR (5.2 1.3 days vs 12.2 1.9 days, p < 0.001). Theater and ward costs were, respectively, significantly increased ($ 2,829.6 340 vs $ 1,422 318, p < 0.001) and decreased ($ 2,600 366 vs $ 6,022 916, p < 0.001) in LCR patients compared with the OC group. There was no significant difference in total hospital costs ($ 10,929 369 vs $ 9,944 1,014). Conclusions: LCR does not appear to offer any immediately recognizable advantages. Key words: Laparoscopy Colectomy Complications

achieval of a certain number of immediately recognizable benefits for the patients. Claims on less postoperative pain, immediate patient mobilizing, shorter postoperative ileus, earlier oral solid intake, reduced hospital stay, more costeffectiveness, quick return to normal activities, and improved cosmetic results have characterized the literature of the early 1990s [4, 6, 10]. A minimally invasive approach may, in theory, not be devoid of advantages provided that colon resection is performed in accordance to the standards of surgical technique and that indications for surgery are not influenced by the change of access [8]. This prospective evaluation of laparoscopic colon resection (LCR) in a selected group of patients with benign disease of the colon was carried out in order to identify and quantify immediately recognizable benefits as well as limitations of this approach.

Materials and methods

Between January 1992 and November 1994, all patients presenting with benign disease of the colon necessitating elective surgery were considered as candidates for laparoscopic colon resection (LCR). Exclusion criteria included previous extensive abdominal surgery, previous and/or ongoing treatment for malignant disease, liver cirrhosis with portal hypertension, severe coagulopathy, intraabdominal abscess, and emergency surgery. Data regarding operating time, intra- and postoperative complications, passage of flatus, duration of hospital stay, and operating room, ward, and total hospital costs were recorded prospectively. Patients who underwent LCR were compared with a control group of 90 patients who underwent elective open colectomy (OC) by the same surgeons at the same institution from January 1990 to December 1991. A mechanical bowel preparation was achieved using 2 l of polyethylene glycol ingested orally during 2 days before surgery. Broad-spectrum intravenous antibiotics were given 1 h preoperatively. Whenever indicated and feasable, preoperative colonic tattooing was made to aid lesion identification at LCR. Patients were given perioperative epidural analgesia and general endotracheal anesthesia and underwent placement of nasogastric tube and urinary catheter. Pneumoperitoneum was induced using carbon dioxide insufflated to a pressure of 12 mmHg by placement of a trocar in the infraumbilical skin using a cut-down technique. The approach to right LCR differed from a previously reported technique [18] with regard to the side-to-side ileocolic anastomosis which was fashioned intracorporeally with Endo-GIA 60 stapler (USSC, Norwalk, CT). Left LCR was carried out

The main rationale for choosing a laparoscopic approach to the surgical treatment of benign diseases of the colon is the
Presented at the 4th International Congress of The European Association for Endoscopic Surgery (EAES), Trondheim, Norway, 2326 June 1996 Correspondence to: R. Bergamaschi, Institute of Surgery, University of Bergen, Diakonissehjemmets University Hospital, N-5009 Bergen, Norway

803 Table 1. Demographics of the patientsa LCR (n 95) Age (years) Weight (kg) ASA grade, I:II Resection type, right: left Indications for surgery Diverticular disease Crohns disease Villous adenoma Volvulus Ischemic colitis Previous surgeryb
a

Table 2. Morbidity after Troidl classification [21] OC (n 90) 65 9.2 79 13.6 66:24 39:51 35 (38.5) 38 (42.2) 9 (10) 8 (8.8) (0) 10 (11.1) p NS NS NS NS NS NS NS NS NS NS Stage II III IV Air leaks Anvil detached Pleural effusion Deep vein thrombosis Urinary retention Wound hematoma Wound abscess LCR (n 94) 2 1 1 1 1 1 OC (n 90)

64 9.8 80 15.2 70:25 41:54 48 (50) 39 (41) 4 (4.2) 3 (3.1) 1 (1) 12 (12.6)

1 2

Mean values the standard error of the mean (SEM); values in parentheses are percentages; NS, not significant b Previous cholecystectomy and/or appendectomy

according to a lately described technique [1]. All ileocolic and colorectal anastomoses were checked for leaks with air pressure, respectively, via a colonoscope or submerging the pelvis in saline and insufflating via a 50-ml syringe. Fascial closure of all trocar sites was performed regardless of the cannula size. All specimens were sent for histology. Morbidity was defined after the Troidl Classification [22]. Criteria for discharge were the same after OC and LCR. Patients were dismissed tolerating oral solid food intake, after the passage of one bowel movement, and with no evidence of sepsis. Data were expressed as mean standard error of mean and calculated by a statistical software program (Harvard Graphics 1.0Software Publishing Corp., 1991). Students t-test, Fishers exact test, and chi-squared test were used where appropriate. Statistical significance was set at p < 0.05.

tients (4.2%) (Table 2). A patient with a superficial abscess at the infraumbilical site was readmitted for incision and drainage. Morbidity rates (4 vs 3, p 0.48) were not significantly different in the LCR and OC groups. Table 3 compares the two patient groups with respect to operating time, first flatus, length of stay, and operating room, ward, and total hospital costs. Discussion A note of caution about the risk of replacing traditional colectomy with two operations has been expressed [19]. However, in spite of some criteria that have been given to define laparoscopy-assisted colectomy [11, 23], one can hardly see how laparoscopy-assisted procedures actually differ from electively converted operations. Vascular and/or bowel division and/or anastomosis fashioning are often performed extracorporeally through a minilaparotomy. However, speaking beyond definitions, it is recommendable to keep conversion rates to a minimum. In fact, converted procedures appear to be associated with high morbidity rates [20]. Thus, a policy of attempting all colon resections laparoscopically should be discouraged. Reported conversion rates vary from 3% to 48% depending on definition, patient selection, which phase of the learning curve, and the bowel segment to be resected [20]. Zucker et al. [24] reported a 3% conversion rate in patients selected based on their ability to understand the rationale for the celioscopic approach, provided the absence of morbid obesity and previous extensive abdominal surgery. Milson et al. [12] achieved a 9.3% conversion rate operating on healthy nonobese patients requiring surgery limited to one colon segment. Reissman et al. [17] had a 7% conversion rate in unselected patients. Low conversion rates should be accomplished via careful preoperative patient selection backed by extensive experience with open colorectal surgery, and of course not by excesses of zeal in trying to postpone a necessary conversion. Claims of shorter postoperative ileus, earlier oral solid intake, and reduced hospital stay after LCR [4, 6, 10] have been supported by a few subsequent controlled studies [5, 13, 15]. However, data from a recent prospective study [9] could not fully confirm that LCR leads to shorter transient postoperative gastrointestinal hypomotility. Moreover, it has been shown that early oral intake is possible after open colorectal surgery [3]. Therefore, it might be very difficult to provide evidence of significant differences in timing for discharge after open and laparoscopic surgery. Shorter hos-

Results The LCR and OC patient groups were not significantly different with regard to age, weight, ASA grading, type of resection, previous minor abdominal surgery (cholecystectomy and/or appendectomy), and indications for surgery (Table 1). Conversion rate was 2% (1/54) in patients undergoing left LCR. Laparoscopy was deliberately abrupted in a 53-year-old obese (body mass index [BMX] 30) man presenting with stenosing diverticular disease of the sigmoid because operating time had reached 4 h. The thickness of the mesentery and the presence of small bowel in the pelvis accounted for the duration of the resection. This patient had postoperative ileus during 5 days and delayed wound healing which recovered without further surgery. There were no deaths. Intraoperative complications occurred in three patients (3%). A 63-year-old man presenting with stenosing diverticular disease of the sigmoid had an end-to-side colorectal anastomosis performed with a 28-mm circular stapler. During per anum extraction of the circular stapler, the anvil was accidentally detached and subsequently removed with forceps. Residual disease left behind at the transection site of the sigmoid rectum probably made it easier to staple the anastomosis on the anterior rectal wall. A barium enema carried out 6 months after surgery revealed an asymptomatic anastomotic stenosis. Two of 41 patients (5%) with ileocolostomy had intraoperative air leaks at colonoscopy with air pressure. Each of two anastomoses was reinforced with sutures. Postoperative complications occurred in four of 94 pa-

804 Table 3. Variables compareda: laparoscopic colon resection (LCR) vs open colectomy (OC) LCR (n 94) Operating time (min) Flatus (days) Hospital stay (days) Operating room costs ($) Ward costs ($) Total hospital costs ($)
a

OC (n 90) 116 9.7 4.4 1.4 12.2 1.9 1,422 318 6,022 916 9,944 1,014

long-term benefits (reduced adhesion formation and decreased incidence of small-bowel obstruction) [21] might magnify the cost-effectiveness of the celioscopic approach. References
1. Bergamaschi R, Arnaud JP (1997) Intracorporeal colorectal anastomosis following laparoscopic left colon resection. Surg Endosc 11:800 801 2. Bessler M, Whelan RL, Halverson A. Treat MR, Nowygrod R (1994) Is immune function better preserved after laparoscopic versus open colon resection? Surg Endosc 8: 881883 3. Binderow SR, Cohen SM, Wexner SD, Nogueras JJ (1994) Must early postoperative oral intake be limited to laparoscopy? Dis Colon Rectum 37: 584589 4. Corbitt JD (1992) Preliminary experience with laparoscopic-guided colectomy. Surg Laparosc Endosc 2: 7981 5. Falk PM, Beart RW Jr, Wexner SD, Thorson AG, Jagelman DG, Lavery IC, Johansen OB, Fitzgibbons RJ Jr (1993) Laparoscopic colectomy. A critical appraisal. Dis Colon Rectum 36: 2834 6. Fowler DL, White SA (1991) Laparoscopic assisted sigmoid resection. Surg Laparosc Endosc 1: 183188 7. Franklin ME, Rosenthal D, Norem RF (1995) Prospective evaluation of laparoscopic colon resection versus open colon resection for adenocarcinoma. Surg Endosc 9: 811816 8. Herfarth C, Schumpelick V, Siewert JR (1994) Pitfalls of minimally invasive surgery. Surg Endosc 8: 847 9. Hotokezaka M, Dix J, Mentis EP, Minasi JS, Schirmer BD (1996) Gastrointestinal recovery following laparoscopic vs open colon surgery. Surg Endosc 10: 485489 10. Jacobs M, Verdeja JC, Goldstein HS (1991) Minimally invasive colon resection (laparoscopic colectomy). Surg Laparosc Endosc 1: 144150 11. Lumley JW, Fielding GA, Rhodes M, Nathanson LK, Siu S, Stitz RW (1996) Laparoscopic-assisted colorectal surgery. Lessons learned from 240 consecutive patients. Dis Colon Rectum 39: 155159 12. Milsom JW, Lavery IC, Church JM, Stolfi VM, Fazio VW (1994) Use of laparoscopic techniques in colorectal surgery. Dis Colon Rectum 37: 215218 13. Musser DJ, Boorse RC, Madera F. Reed JF (1994) Laparoscopic colectomy: at what cost? Surg Laparosc Endosc 4: 15 14. Ortega A, Beart R, Anthone G, Schlinker R (1994) Laparoscopic bowel resection and consecutive series (abstract). Dis Colon Rectum 37: 22 15. Peters WR, Bartels TL (1993) Minimally invasive colectomy: are the potential benefits realized? Dis Colon Rectum 36: 751756 16. Pfeifer J, Wexner SD, Reissman P, Bernstein M, Nogueras JJ, Singh S, Weiss E. (1995) Laparoscopic vs open colon surgery. Cost and outcome. Surg Endosc 9: 13221326 17. Reissman P, Cohen S, Weiss EG, Wexner SD (1996) Laparoscopic colorectal surgery: ascending the learning curve. World J Surg 20: 277282 18. Schlinkert RT (1991) Laparoscopic-assisted right hemicolectomy. Dis Colon Rectum 34: 10301031 19. Scott HJ, Spencer J (1995) Colectomy: the role of laparoscopy. Surg Laparosc Endosc 5: 382386 20. Slim K, Pezet D, Riff Y, Clark E, Chipponi J (1995) High morbidity rate after converted laparoscopic colorectal surgery. Br J Surg 82: 14061408 21. Thompson, JN, Whawell SA (1995) Pathogenesis and prevention of adhesion formation. Br J Surg 82: 35 22. Troidl H, Spangenberger W, Dietrich A, Neugebauer E (1991) Laparoskopische Cholecystectomie. Erste Erfahrungen and Ergebnisse bei 300 Operationen: eine prospektive Beobachtungsstudie. Chirurg 62: 257265 23. Wishner JD, Baker JW Jr, Hoffman GC, Hubbard GWII, Gould RJ, Wohlgemuth SD, Ruffin WR, Melick CF (1995) Laparoscopicassisted colectomy. The learning curve. Surg Endosc 9: 11791183 24. Zucker KA, Pitcher DE, Martin DT, Ford RS (1994) Laparoscopicassisted colon resection. Surg Endosc 8: 1218

p <0.001 NS <0.001 <0.001 <0.001 NS

180 10.3 3.5 1.2 5.2 1.3 2,829.6 340 2,600 366 10,929 369

Mean values the standard error of the mean (SEM)

pital stay after LCR represents scanty evidence when it is compared with a length of stay of 11 days [14] or 12.2 days [7] following traditional colectomy. As stated elsewhere [16], better cosmesis is, for the time being, the only proven, but often irrelevant, benefit of LCR. Nevertheless, it must be acknowledged that laparoscopic surgery has contributed to making us review our traditional routines in open surgery. Immediately recognizable limitations of LCR include a learning phase, prolonged operating time, and new complications. It has been estimated that 3550 procedures are necessary before the learning curve flattens [23]. Furthermore, it has been shown that the learning curve is in most cases steep [5], although it may depend on the type of bowel resection [17]. A prolonged operating time at any phase of the learning period may in fact point out the limits of current instruments. Decreasing operating time must be achieved without an increase in complication rates. Data from a large prospective study [11] have shown that overall morbidity following LCR is not increased when compared with open surgery. However, the definition of morbidity may vary [17] and the use of one classification is certainly desirable [22]. New complications such as resection of the wrong colon segment and port site hernias [17] may be overcome by a more widespread use of intraoperative colonoscopy and routine surgical closure of port wounds regardless of their size. Early claims on increased cost-effectiveness of LCR [13] have been disconfirmed by convincing evidence showing that increased operating-room costs often offset possible gains from shorter hospital stay [5, 16]. Direct costs of a completely intracorporeal approach may differ markedly among institutions and should not be analyzed as isolated data [19]. A judicious use of reusable laparoscopic equipment may help contain these costs. Early return to work is unlikely to lead to economic benefit since most patients with colorectal diseases are retired [19]. Although it might be very difficult to provide evidence of increased costeffectiveness of LCR, a cost-utility analysis should be most appropriate because it would measure quality of life following LCR. No immediately recognizable advantages following LCR can be surmised from the present study. Still-unproven favorable physiologic effects (decreased cell-mediated immunosuppression and blood loss) [2] and still-unknown

News and notices

Surg Endosc (1997) 11: 880882

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

New Address for the European Association for Endoscopic Surgery (E.A.E.S.)
Effective January 1, 1997, the new correspondence, telephone, and fax numbers of the E.A.E.S. office are: E.A.E.S. Office, c/o Mrs. Ria Palmen Luchthavenweg 81 Unit 1.42 5657 EA Eindhoven The Netherlands or: P.O. Box 335 5500 AH Veldhoven The Netherlands Tel: +31 40 2525288 Fax: +31 40 2523102

Fellowships in Minimally Invasive Surgery The University of Pittsburgh Medical Center Pittsburgh, PA, USA
One year fellowships in advanced minimally invasive surgery in both general and thoracic surgery are being offered at the University of Pittsburgh Medical Center beginning on July 1, 1997. Requirements include completion of residency training programs in the desired area. The fellowships include a competitive salary and travel allowance. Interested candidates should send a letter of inquiry with curriculum vitae to: Philip R. Schauer, MD (General Surgery) or James Luketich, MD (Thoracic Surgery) The University of Pittsburgh Medical Center 3471 Fifth Avenue Suite 300 Pittsburgh, PA 15213-3221

Volunteer Surgeons Needed Northwestern Nicaragua Laparoscopic Surgery Teaching Program, Leon, Nicaragua
Volunteer surgeons are needed to tutor laparoscopic cholecystectomy for this non-profit collaboration between the Nicaraguan Ministry of Health, the National Autonomous University of Nicaragua, and Medical Training Worldwide. The program consists of tutoring general surgeons who have already undergone a basic laparoscopic cholecystectomy course. Medical Training Worldwide will provide donated equipment and supplies when needed. For further information, please contact: Medical Training Worldwide Ramon Berguer, MD, Chairman Tel: 707-423-5192 Fax: 707-423-7578 e-mail: berguer.r@martinez.va.gov

Fellowships in Laparoscopic Surgery Staten Island University Hospital Staten Island, NY USA
A one year fellowship, to start July 1, 1997, in advanced laparoscopic surgery is being offered at Staten Island University Hospital. The selected fellow will be exposed to many advanced general laparoscopic surgeries including: hiatal hernia repair, splenectomy, adrenalectomy, bowel resection, and others. Participation in research projects will be encouraged. For further information, please contact: Barbara Coleman Coordinator, Surgical residency program Tel: 718-226-9508

Fellowship in Minimally Invasive Surgery George Washington Medical Center Washington, DC USA
A one-year fellowship is being offered at the George Washington University Medical Center. Interested candidates will be exposed to a broad range of endosurgical Education and Research Center. Active participation in clinical and basic science research projects is also encouraged. For further information, please contact: Carole Smith 202-994-8425 or, send curriculum vitae to: Dr. Jonathan M. Sackier Director of Endosurgical Education and Research George Washington University Medical Center Department of Surgery 2150 Pennsylvania Avenue, N.W. 6B-417 Washington, DC 20037, USA

Essentials of Laparoscopic Surgery Surgical Skills Unit University of Dundee Scotland, UK

Under the direction of Professor A. Cuschieri the Surgical Skills Unit is offering a three-day practical course designed for surgeons who wish to undertake the procedures such as laparoscopic cholecystectomy. This intensely practical program develops the necessary operating skills, emphasizes safe practice, and highlights the common pitfalls and difficulties encountered when starting out. Each workshop has a maximum of 18 participants who will learn both camera and instrument-manipulation skills in a purpose-built skills laboratory. During the course there is a live demonstration of a laparoscopic cholecystectomy. The unit has a large library of operative videos edited by Professor Cuschieri, and the latest books on endoscopic surgery are on display in our Resource area. Course fee including lunch and course materials is $860.

881 For further details and a brochure please contact: Julie Struthers, Unit Co-ordinator Surgical Skills Unit Ninewells Hospital and Medical School Dundee DD1 9SY Tel: +44 382 645857 Fax: +44 382 646042 recreational pursuits which can be arranged by the facility to suit your personal agenda. For further details please contact: Carole Smith: Department of Surgery 2150 Pennsylvania Avenue NW 6B Washington, DC 20037, USA Tel: (202) 994-8425

Advanced Endoscopic Skills Surgical Skills Unit University of Dundee Scotland, UK

Each month Professor Cuschieri Surgical Skills Unit offers a 412 day course in Advanced Endoscopic Skills. The course is intensely practical with hands on experience on a range of simulated models. The program is designed for experienced endoscopic surgeons and covers advanced dissection techniques, extracorporeal knotting techniques, needle control, suturing, internal tying technique, stapling, and anastomotic technique. Individual workstations and a maximum course number of 10 participants allows for personal tuition. The unit offers an extensive collection of surgical videos and the latest books and publications on endoscopic surgery. In addition, participating surgeons will have the opportunity to see live advanced laparoscopic and/or thoracoscopic procedures conducted by Professor Cuschieri and his team. The course is endorsed by SAGES. Course fee including lunch and course materials is $1850. For further details and a brochure please contact: Julie Struthers, Unit Co-ordinator Surgical Skills Unit Ninewells Hospital and Medical School Dundee DD1 9SY Tel: +44 382 645857 Fax: +44 382 646042

Courses at the Royal Adelaide Centre for Endoscopic Surgery

Basic and Advanced Laparoscopic Skills Courses are conducted by the Royal Adelaide Centre for Endoscopic Surgery on a regular basis. The courses are limited to six places to maximize skill development and tuition. Basic courses are conducted over two days for trainees and surgeons seeking an introduction to laparoscopic cholecystectomy. Animal viscera in simulators is used to develop practical skills. Advanced courses are conducted over four days for surgeons already experienced in laparoscopic cholecystectomy who wish to undertake more advanced procedures. A wide range of procedures are included, although practical sessions can be tailored to one or two procedures at the participants request. Practical skills are developed using training simulators and anaesthetised pigs. Course fees: $A300 ($US225) for the basic course and $A1,600 ($US1,200) for the advanced course. For further details and brochure, please contact: Dr. D. I. Watson or Professor G. G. Jamieson The Royal Adelaide Centre for Endoscopic Surgery Department of Surgery Royal Adelaide Hospital Adelaide SA 5000 Australia Tel: +61 8 224 5516 Fax: +61 8 232 3471

The Practical Aspects of Laparoscopic Fundoplication Surgical Skills Unit University of Dundee Scotland, UK
A three-day course, led by Professor Cuschieri, designed for experienced laparoscopists wishing to include fundoplication in their practice. The course covers the technical details of total and partial fundoplication using small group format and personal tuition on detailed simulated models. There will be an opportunity to observe one of these procedures live during the course. Maximum course number is six. Course fee including lunch is $1850. For further details and a brochure please contact: Julie Struthers, Unit Co-ordinator Surgical Skills Unit Ninewells Hospital and Medical School Dundee DD1 9SY Tel: +44 382 645857 Fax: +44 382 646042

Advanced Laparoscopic Suturing and Surgical Skills Courses MOET Institute San Francisco, CA, USA
Courses are offered year-round by individual arrangement. The MOET Institute is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians and designates these CME activities for 2040 credit hours in Category 1 of the Physicians Recognition Award of the American Medical Association. These programs are also endorsed by the Society of Gastrointestinal Endoscopic Surgeons (SAGES). For further information, please contact: Wanda Toy, Program Administrator Microsurgery & Operative Endoscopy Training (MOET) Institute 153 States Street San Francisco, CA 94114, USA Tel: (415) 626-3400 Fax: (415) 626-3444

Courses at George Washington University Endosurgical Educational and Research Center

George Washington University Endosurgical Educational and Research Center is proud to offer a wide range of surgical endoscopy courses. These courses include advanced laparoscopic skills such as Nissen fundoplication, colon resection, common bile duct exploration, suturing, as well as subspecialty courses. Individual surgeons needs can be met with private tuition. The Washington D.C. area is a marvelous destination to visit for

Courses at WISE Washington Institute for Surgical Endoscopy Washington, DC, USA
The Washington Institute of Surgical Endoscopy is pleased to offer the following courses:

882 Laparoscopic antireflux and hiatal hernia surgery (July 1415, 1997); Laparoscopic management of the common bile duct and difficult cholecystectomy (May 1516, August 1112, November 1011, 1997); Laparoscopic colon and rectal surgery (June 2021, September 1516, December 45, 1997). Also, courses for operating room nurses and technicians will be run on a monthly basis and personal instruction and preceptorship is available. For further information, please call: Carole Smith Washington Institute of Surgical Endoscopy 2150 Pennsylvania Avenue, N.W. Washington, DC 20037 Tel: 202-994-9425

9th International Meeting Society for Minimally Invasive Therapy July 1416, 1997 Kyoto, Japan
Scientific program to include: Plenary, Parallel, Poster, and Video sessions. Host Chairman: Professor Osamu Yoshida, Department of Urology, Kyoto University, 54 Shogoin Kawahara-sho, Sakyo, Kyoto 606, Japan. Phone: +81 75 751-3328, Fax: +81 75 751-3740. This meeting coincides with the Gion Festival in Kyoto, one of the greatest festivals in Japan. For further information, please contact: Secretariat of SMIT 9th Annual International Meeting c/o Academic Conference Planning 383 Murakami-cho Fushimika, Kyoto 612 Japan Tel: +81 75 611-2008 Fax: +81 75 603-3816

Call for Abstracts Society of American Gastrointestinal Endoscopic Surgeons (SAGES) 1998 Annual Meeting April 14, 1998 Seattle, WA, USA
Abstract deadlines: Oral and Poster abstracts: September 12, 1997 Video Submissions: September 18, 1997 For further information, or to obtain an abstract form, please contact: SAGES Program Committee Society of American Gastrointestinal Endoscopic Surgeons Suite #3000 2716 Ocean Park Boulevard Los Angeles, CA 90405 Tel: (310) 314-2404 Fax: (310) 314-2585 e-mail: SAGESMail@AOL.com

Colorectal Disease in 1998 February 1921, 1998 Fort Lauderdale, FL, USA Symposium Director: Steven D. Wexner, MD
Cleveland Clinic Florida presents its ninth annual postgraduate course. Provides an intensive, in-depth, analytical review of all aspects of colorectal disease, including laparoscopy; colorectal carcinoma screening and genetics, inflammatory bowel disease; and pouch surgery. There will be a review of both basic and advanced principles of diagnosis and management of disease. Video techniques will be shown as well. The faculty is internationally represented and includes leading experts in the field. Simultaneous Spanish and Italian translation is available. For more information, please contact: Cleveland Clinic Florida Department of Education 2950 West Cypress Creek Road Fort Lauderdale, FL 33309-1743 Tel: 800-359-6101, ext. 6066 Fax: 954-978-5539

European Course on Laparoscopic Surgery (English language) November 1821, 1997 Brussels, Belgium
Course director: G.B. Cadiere For further information, please contact: Administrative Secretariat Conference Services s.a. Avenue de lObservatoire, 3 bte 17 B-1180 Bruxelles Tel: (32 2) 375 16 48 Fax: (32 2) 375 32 99

Courses Offered at the University of Minnesota Minneapolis, Minnesota, USA September 17, 1997: Fourth Annual Conference, Molecular Biology of Colorectal Cancer September 17, 1997: Sixth Annual Conference, Endorectal Ultrasonography September 1820, 1997: Sixtieth Annual Conference, Principles of Colon and Rectal Surgery
For further information, please contact: Continuing Medical Education University of Minnesota 615 Washington Avenue SE, Suite 107 Minneapolis, MN 55414 Tel: 800-776-8636 Fax: (612) 626-7766

6th World Congress of Endoscopy Surgery Roma 98 6th International Congress of European Association for Endoscopic Surgery June 36, 1998 Rome, Italy
The program will include: the latest, original high quality research; symposia; plenary lectures; abstract presentations (video, oral, and posters); EAES and SAGES postgraduate courses, OMED postgraduate course on therapeutic endoscopy; working team reports; educational center and learning corner; meeting of the International Society of Nurses and Associates; original and non original scientific reports; and a world expo of new technology in surgery. For further information, please contact: Congress Secretariat: Studio EGA Viale Tiziano, 19 00196 Rome, Italy Tel: +39 6 322-1806 Fax: +39 6 324-0143

Letters to the editor

Surg Endosc (1997) 11: 875876

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic colectomy
We read with great interest the papers of the SAGES Postgraduate Course published in the December 1995 issue of Surgical Endoscopy, and we find that we cannot agree with some statements made by D.M. Ota in his work, Laparoscopic colon resection for cancer. The author describes the basic principles of colorectal cancer management in open and laparoscopic surgery. We do agree with Ota that the laparoscopic procedures should reproduce the principles of open surgery. Nevertheless, Ota says that it is not possible to accomplish a complete staging of the disease through the laparoscopic approach and that, therefore, retroperitoneal and periportal adenopathy can be missed. Ota also states that early ligation of regional blood vessels is not feasible during laparoscopic colectomy and that anterior resection is also not feasible because the laparoscopic intestinal stapler is straight and the stapled suture line can only be 1012 cm from the anal verge. We do not want to discuss whether radical hemicolectomy is preferable to segmental colectomy for cancer treatment since this is an open question and the assertions made by the author in this respect are not arguable. But we would like to assert some technical principles which totally differ from those affirmed by Ota. 1. The identification of metastatic disease is certainly time consuming but at the same time extremely precise by laparoscopy. As far as the periportal nodes are concerned, during laparoscopic hepatectomy and laparoscopic D2 gastrectomy it has been demonstrated that it is possible to visualize and dissect the element of the hepatoduodenal ligament, carrying out the dissection of the group of nodes #12 according to the JSCC classification [3]. As far as lumbar and aortic node dissection is concerned, it has been demonstrated that this is feasible in colorectal cancer, cancer of the testis, and cervical, endometrial, and ovarian cancer [2]. In our experience (165 laparoscopic colorectal resections out of 250 laparoscopic colectomies, 101 of which for cancer), in 12 cases an extended lymphadenectomy, including infraaortic node dissection, iliac, hypogastric, and obturatory node dissection have been performed. Intraoperative complications and postoperative morbidity and mortality were similar to those of open operations. In one case, a lesion of the aorta occurred during the disection due to the detachment of the left ovarian artery at its origin. The lesion was immediately repaired without any need of conversion. 2. The early ligature of the inferior mesenteric vessels is a basic step and the key element of laparoscopic colorectal procedures. Such a maneuver is not only doable (Fig. 1) but makes even easier the accomplishment of the operation since the preparation of the inferior mesenteric artery at its origin is carried out along an avascular plane and the dissection of the mesentery without previous colonic mobilization allows a clear visualization of the gonadic vessels, the ureter, and the genitofemoral nerve [1]. In this way the procedure is safer, quicker, and oncologically correct: Only two ligatures are required! The high ligature (the ligature and division of the artery are performed close to its origin from the aorta) of the mesenteric vessels is performed by us as the first step of a colorectal resection; thus, a no-touch procedure is accomplished. In our opinion this does not increase patient survival, but it does allow a precise staging of the disease, which is essential for postoperative adjuvant chemotherapy. 3. Very low anastomoses (Fig. 2), even coloanal or ileoanal anastomoses, are performed using one of the following techniques: a. Double linear endostapler application with instrument loaded with the 30-mm- or 35-mm-long cartridges and inserted through the cannula placed in the right inferior quadrant, while the rectum is pulled toward the left and a pressure is applied on the perineum to push upward the elevatormuscles plane. b. Application of a standard roticulator stapler through a suprapubic minilaparotomy, with the operation continued with a gasless technique.

Correspondence to: M. M. Lirici

Fig. 1. High division of inferior mesenteric artery is achieved after either ligating or clipping (two large clips 12 mm long, each side) the vessel at its origin from the aorta.

876

Fig. 2. Multiple applications of linear endostapler with the described technique allow a very low division of the rectum (the line of transection, below the peritoneal reflection, is brought down to the inferior third of the rectum) and performance of very low colorectal anastomoses and coloanal or ileoanal anastomoses according to the double stapling procedure.

In both cases the anastomosis is then accomplished according to the double stapling technique. In 20 cases of very low anastomoses performed through the laparoscopic approach, we found no technical difference compared to open surgery. On the contrary, the dissection of mesorectum seems to be more accurate by the laparoscopic route. Laparoscopic colorectal procedures can really reproduce the techniques performed in open surgery: They should be considered a surgical option for that reasonnot because radical operations have not shown real advantages compared to segmental colonic resections.

lymphadenectomy in gynecologic malignancies. Obstet Gynecol 82: 741747 3. Huscher C, Chiodini S, Recher A, Battiston C, Tarantini M, Soccio M (1995) Laparoscopy-assisted gastrectomy for cancer: initial experience. Proceedings International Gastric Cancer Congress, Kyoto, pp 1215 1218

C. S. G. Hu scher1 M. M. Lirici2 L. Angelini2

1

References
1. Buess G, Manncke K, Mehran J, Lirici MM (1993) State of the art of laparoscopic colorectal surgery. End Surg 1: 312 2. Childers JM, Hatch KD, Surwit EA (1993) Laparoscopic paraortic

Department of General Surgery Ospedale Vallecamonica 25040 Esine (BS) Italy 2 4th Department of Surgery Policlinico Umberto I University Hospital Viale del Policlinico 00161 Rome Italy

Surg Endosc (1997) 11: 852855

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Telementoring
A practical option in surgical training
J. C. Rosser,1 M. Wood,2 J. H. Payne,3 T. M. Fullum,4 G. B. Lisehora,5 L. E. Rosser,1 P. J. Barcia,5 R. S. Savalgi1
1 2

Department of Surgery, Yale University School of Medicine, 333 Cedar Street, New Haven, CT, USA Riverview Hospital, Detroit, MI, USA 3 Kaiser Foundation Hospital, Honolulu, HI, USA 4 Providence Hospital, Washington DC, USA 5 Tripler Army Medical Center, Honolulu, HI, USA Received: 17 May 1996/Accepted: 19 August 1996

Abstract Background: Telemedicine offers significant advantages in bringing consulting support to distant colleagues. There is a shortage of surgeons trained in performing advanced laparoscopic operations. Aim: Our aim was to evaluate the role of telementoring in the training of advanced laparoscopic surgical procedures. Methods: Student surgeons received a uniform training format to enhance their laparoscopic skills and intracorporeal suturing techniques and specific procedural training in laparoscopic colonic resections and Nissen fundoplication. Subsequently, operating rooms were equipped with three cameras. Telestrator (teleguidance device), instant replay (to critique errors), and CD-ROM programs (to provide information of reference) were used as intraoperative educational assistance tools. In phase I, four colonic resections were performed with the mentor in the operating room (group A) and four colonic resections were performed with the mentor on the hospital grounds, but not in the operating room (group B). The voice and video signals were received at the mentors location, using coaxial cable. In phase II, two Nissen fundoplications were performed with the mentors in the operating room (group C) and two Nissen fundoplications were performed with the mentors positioned five miles away from the operating room (group D), using currently existing land lines at the T-1 level. Results: There were no differences in the performances of the surgeons and outcome of the operations between groups A & B and C & D. It was possible to tackle the intraoperative problems effectively. Conclusions: The telementoring concept is potentially a safe and cost-effective option for advanced training in lap-

aroscopic operations. Further investigation is necessary before routine transcontinental patient applications are attempted. Key words: Telemedicine Telementoring Surgical training Laparoscopic surgery

Telemedicine has previously been defined as live two-way interactive video communications between a physician and a patient and/or another physician, where all participants are able to see and hear one another much like a face to face encounter. This basically constitutes the remote practice of medicine. This concept has gained recent notoriety because of the great advances in telecommunications and the increasing cost-effectiveness associated with its utilization. If the subject is more closely reviewed, it will be revealed that telemedicine is not a recent concept. Telemedicine began in the 1950s during the early days of television. Forty years later, we now have a resurgence of the telemedicine initiative. Most of the current applications are based on teleconferencing, utilizing interactive vocal communication, high-quality clinical still photographs, and video images which are not of full-motion quality. Telementoring is a telemedicine technique that involves the remote guidance of a treatment or investigational procedure (Fig. 1), where the student has no or limited experience with the featured technique. The multiple educational and technological requirements place telementoring in a much more sophisticated and higher-risk category than standard telemedicine applications. These factors dictate that a standard training protocol be formulated so that quality assurance can be maintained. Materials
The technology that makes telementoring possible may seem to be futuristic and to require Star Wars technology, but in fact the technology is

Correspondence to: J. C. Rosser, Jr., Yale University School of Medicine, Department of Surgery, 40 Temple Street, Suite 3A, New Haven, CT 06510, USA

853

Fig. 1. This picture shows operating room setup.

obtainable today as a cost-effective system of transferring skill by guiding actions. The heart of this technology is the coder/decoder (CODEC), an image-processing computer with the ability to digitize analog signals and compress video, audio, and other data for transmission through a digital network. The CODEC (CLI Rembrandt, Computer Labs, Inc., San Jose, CA) manages enormous amounts of information in real time, which requires a high level of compression technology. Two of these units are needed. One is placed at the control site and the other in the operating theater. The ultimate test of the development of any telementoring technique is the ability to complete a mission with the use of this technology at great distances, due to the inherent data transmission characteristics that must be compensated. Tactical information deployment (TID) of clinical reference material was achieved by using a Power Macintosh 7200/90 Macintosh (Apple Computers, Inc., Cupertino, CA) Power PC with quadruple-speed CDROM drive, 16 MB of RAM, and a 500-MB hard drive. A CD-ROM on laparoscopic colon resection and Nissen fundoplication (Yale Laparoscopic Series Procedure Specific Interactive Multimedia CD-ROMNissen Fundoplication and Colon Resection, New Haven, CT) can provide the less-experienced surgeon with fast and easy access to reference information about a specific laparoscopic procedure. The standard utilization of a VCR (Sharp VHS VC-A534U, Sharp, Inc., Mahwah, NJ) to record the operative procedure is advantageous to clarify any inaccuracies by using instant replay, thus avoiding them in the future. Along with this apparatus, the mentor must be able to guide the novice surgeon by utilizing a surgistrator, that is, a telestrator (ACECAT II, Boecker Instruments, Tucson, AZ), a necessary component which allows the mentor to direct the learning surgeon by placing marks on the screen and pinpointing an exact location for dissection or avoidance (Fig. 2). Finally, wireless headphones and microphones (Clear-com CS 222, Intercom Systems, Berkeley, CA) must be worn by the operative team and the mentor to complete the two-way communication. No cameras were placed at the control site (Fig. 3), but there was a tripod-mounted unit (Sony CCD-V5000, Sony, Corp., Tokyo, Japan) in the operating room to survey the outside of the abdomen and the teams movements. The external camera was mounted to view the abdomen for assistance on trocar placement and view the hand movements of the student surgeon.

Fig. 2. This picture displays the two 30-inch monitors used in the telementoring application. The monitor on the right demonstrates the use of teleguidance instrumentation (a telestrator), seen as white Xs on the screen. The monitor on the left shows a view of the operating room. Fig. 3. A typical command center for telementoring at Yale University School of Medicine.

Methods
The effectiveness of this technology should not rely solely on electronics. It should be dictated by the human resources commanding it. To apply this technology [1] competently to a clinical situation, we must proceed within a structured and well-thought-out strategy, with a complete educational agenda that covers all aspects of the operation and skill required. A preprocedural assessment and enhancement of laparoscopic surgical skills to ensure patient safety is also very important. A three-day laparoscopic skills and suturing program was conducted for all participating surgeons. The program sent the surgeons through a series of drills and exercises. A large database demonstrates results as correlating closely with the ability to

suture intracorporeally. This provides a yardstick to evaluate each participants skill capability and generate a checkpoint to document progress [46]. At the end of the course, each participant could competently suture. This greatly increased the surgeons confidence that intraoperative complications could be effectively handled. It is mandatory to establish a standardized and tactical approach to the procedure to promote a stable and anticipated operative sequence with both mentor and student sharing common ground. For this purpose a step-by-step protocol was developed and all the participants were trained to follow the same steps. A telementoring simulation laboratory was developed [7] to introduce participants to the experience of being instructed from a remote site. This was utilized during the animate laboratory exercise where the students conducted the different procedures on porcine specimens. A telecommunication system similar to our standard setup was utilized with the exception of the CODEC. To better assist in the guidance of the student surgeon and in order to establish a standardized database for each surgeon that is mentored, the concept of Tactical Information Deployment (TID) was utilized. This was made possible by the development of procedure-specific CD-ROM multimedia interactive computer programs that can rapidly deploy reference information to the video monitor in the operating suite.

Results Phase I was conducted at Riverview Hospital in Detroit, MI (Fig. 4). A closed television circuit without compression technology was utilized. The control center was situated in a truck, approximately 500 yards away from the hospital. There were a total of eight laparoscopic colon resections

854

review of anomalous anatomical presentations (digitized illustrations); reviewing the correct method to achieve an operative goal in the operation (digitized video clips); and reviewing suturing technique needed to achieve intracorporeal suturing in certain areas of the abdomen.

Discussion Telemedicine is not a recent innovation. It began in the United States at the University of Nebraska in the late 1950s. In 1959, the University of Nebraska implemented a telemedicine network to support clinical continuing medical education and training, and also research collaboration applications in geographically remote areas of Nebraska. The majority of the early experiments in telemedicine eventually were closed down. At the same time during the investigation of the remote delivery of health care, there was an increase in efforts to expand the health-care delivery system in the US and telemedicine entered a period of stagnation. The potential advantages of telemedicine were revisited as a result of the health-care delivery crisis, ever-tightening budgetary concerns, and the impact of managed care. Telecommunications and other technological breakthroughs also fueled renewed enthusiasm. Credible, financially sound programs have now been established. In 1986, the Mayo Clinic implemented a two-way satellite program between their Rochester, Minnesota, campus and remote clinics in Scottsdale, Arizona, and Jacksonville, Florida. In 1991, the Medical College of Georgia established a telemedicine link between its Augusta, Georgia, campus and Dodge County Hospital in Eastman, Georgia, to provide consultation and continuing education to the 87-bed hospital [8]. Telementoring is an advanced application of telemedicine. It adds new dimensions to current educational and clinical practices. It involves the remote guidance of a procedure where the student has no or limited experience. This article is a review of the results of phase I and phase II of our telementoring project. Others have also begun investigating this new application. Go et al. used videoconferencing applications to evaluate early technology [1]. Ranshaw et al. [3] teleproctored a Georgian rural surgeon in more than 24 cases. All cases were successfully completed laparoscopically without any complications. They utilized one-half and full T1 lines and demonstrated that they can be used for telementoring. Moore et al. [2] developed an in-house telementoring system without the use of computer compression technology. These studies represent preliminary investigations. There is a need for further critical evaluation of this technology in clinical settings. There are several impressions that have come from our initial experience. At this time, with current CODEC computer compression algorithms, one-half T-1 bandwidth is the minimum required for maximal motion display capability with minimal delay. The two-way audio link is best handled with individual headsets rather than relying on one speaker and microphone for all parties in the remote OR. Well-established military and aviation communication speech patterns are crucial for prompt execution of commands. In addition to an annotator to target important landmarks, the use of instant replay with a VCR allows for

Fig. 4. Pictorial depiction of phase I and phase II of telementoring.

performed. Four of the cases were performed via the standard mentoring process (group A). In this group the mentor was in the OR with the student surgeon. The other four laparoscopic colon cases were performed utilizing telementoring techniques (group B). In group A (Table 1) the average OR time (mean SEM) was 231.0 36.4 min. The average blood loss was 155.0 48.4 ml. The return to consumption of liquid and solids was an average of 2.6 0.3 days and 4.0 0.4 days, respectively. The average length of stay in the hospital was 6.6 1.0 days and patients returned to normal activity in 25.0 6.2 days. In comparison with group A, the OR time in group B was slightly longer (318 79 min). All cases were successfully completed laparoscopically without mentor intervention. In group B the blood loss was 263.0 122.5 ml and the return to consumption of liquid and solids was 2.3 0.3 days and 4.3 0.6 days, respectively. The average length of stay in the hospital was 6.3 1.2 days, and the return to normal activity was 16.0 2.5 days. There were no complications in either group. Statistical parameters were very similar in both the groups and therefore the next phase of the mission was begun. Phase II was held in Honolulu, Hawaii (Fig. 4). Full CODEC (coder/decoder)-mediated computer compression technology was utilized. There were a total of four cases; two were performed with the mentor (group C) in the OR and the other two cases were performed in the control center, located approximately 5 miles from the hospital (Table 2). The OR times for groups C and D were 278.0 10 and 280.0 10 min, respectively. The blood loss for both group C (90.0 10 ml) and D (110.0 14.1 ml) was minimal. For group C the consumption of liquid and solids began in 1 day and 2 days, in comparison to group D, in 1 day and 4 days, respectively. In both groups all the patients stayed in the hospital for 3 days. For group C and group D the mean time required before returning to normal activity was 10.0 2.0 days and 12.0 1.0 days, respectively. The results between both groups showed no significant difference. TID was found useful in many instances. These included

855 Table 1. Results from phase I involving both mentored (group A) and telementored (group B) groups featuring laparoscopic colon resection (CODEC not utilized)a Group A (n 4) B (n 4)
a

OR time (minutes) 231.0 36.4 318 79.7

Blood loss (ml) 155.0 48.4 263.0 122.5

PO liquid (days) 2.6 0.3 2.3 0.3

PO solid (days) 4.0 0.4 4.33 0.6

Admn (days) 6.6 1.0 6.3 1.2

RTNA (days) 25.0 6.2 16.0 2.5

A, Colon resection with mentor in OR; B, Colon resection with mentor not in OR; mean SEM

Table 2. Results from phase II involving both mentored (group C) and telementored (group D) groups featuring laparoscopic Nissen fundoplication (CODEC utilized)a Group C (n 2) D (n 2)
a

OR time (minutes) 278.0 10.0 280.0 10.0

Blood loss (ml) 90.0 10.0 110.0 14.1

PO liquid (days) 1.0 0.0 1.0 0.0

PO solid (days) 2.0 0.0 2.0 0.0

Admn (days) 3.0 0.0 3.0 0.0

RTNA (days) 10.0 2.0 12.0 1.0

C, Nissen fundoplication with mentor in OR; D, Nissen fundoplication with mentor not in OR; mean SEM

the rapid review and critique of unsafe or unwanted operative techniques. The use of a concept called Tactical Information Deployment (TID) provides the surgeon with rapid access to reference information in the operating suite. Multimedia Interactive CD-ROMs with digitized movie clips, illustrations, sound bits, and the latest academic review of the literature arm the surgeon with a database that establishes an unprecedented clinical adaptive capability. The laparoscopic and intracorporeal suturing skills of those to be mentored must be assessed (and thereby established) to maintain the safety of the patient. Adherence to a uniform, step-by-step tactical operative protocol that is part of the training regimen provides common ground for instructor and student. This must be established before the mission is undertaken. An animal laboratory practical [6] that is conducted under telementoring guidance conditions is essential to prepare the students to utilize all that they have learned while interfacing with technology. The results of our initial experience are very encouraging. We feel that this concept, if properly established, can not only impact the delivery of health care for areas that are

underserved but also accelerate the safe deployment of advanced procedural capability worldwide. References
1. Go PMNYH, Payne JH Jr, Satava RM, Rosser JC (1996) Teleconferencing bridges two oceans and shrinks the surgical world. Surg Endosc 10: 105106 2. Moore RG, Adams JB, Partin AW, Docimo SG, Kavoussi LR (1996) Telementoring of laparoscopic procedures. Surg Endosc 10: 107110 3. Ranshaw B, Tucker J, Duncan T, Mason E, Lucas G (1996) Laparoscopic herniorrhaphy: a review of 900 cases. Surg Endosc 10: 255 4. Rosser J, Rosser L, Savalgi R (1996) Computer based training in laparoscopic surgery. Surg Endosc 10: 257 5. Rosser J, Rosser L, Savalgi R (1996) Objective evaluation in laparoscopic surgical training and credentialling. Surg Endosc 10: 257 6. Rosser J, Rosser L, Savalgi R (1996) Incorporation of telementoring in the surgical training of advanced laparoscopic operations. Program and Abstracts, 4th International Congress of The European Association for Endoscopic Surgeons, A49 7. Stix G (1995) Boot camp for surgeons. Sci Am 273: 24 8. Whitehead R (1995) The evolution of telemedicine: then, now, and what will be. Teleconference 14: 911

Original articles
Surg Endosc (1997) 11: 800801

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Intracorporeal colorectal anastomosis following laparoscopic left colon resection

R. Bergamaschi, J.-P. Arnaud
Department of Visceral Surgery, Angers University Hospital, 4 rue Larrey, 49033 Angers, France Received: 26 March 1996/Accepted: 9 September 1996

Abstract Background: The aim of this study was to assess the impact of an intracorporeal double-stapled colorectal anastomosis upon the outcome of laparoscopic left colon resection. Methods: Fifty-four selected patients underwent elective laparoscopic left colon resection for benign disease. Once resection was completed, a 33-mm suprapubic port allowed insertion of the anvil of a circular stapler into the colon, which was closed by a handsewn purse-string suture using the T-needle technique. The circular stapler was passed transanally to perform a double-stapled anastomosis. Specimens were delivered in a plastic bag via the suprapubic port. Results: There were no deaths. Minor intraoperative and postoperative complications occurred in 3.7% and 9.2% of the patients, respectively. Median operating time was 125 min (range 80210 min). Complete proximal and distal doughnuts were obtained in all patients and anastomoses were all methylene blue tight. Median hospital stay was 4 (range 37) days. Conclusions: Fashioning double-stapled colorectal anastomoses intracorporeally is feasable and safe. Key words: Laparoscopy Colectomy Surgical technique

bowel. Since resecting and anastomosing segments of colon totally within the abdomen may become the procedure of choice, we describe here an approach to intracorporeal colorectal anastomosis that uses the double-stapling technique [3]. Patients
Between January 1992 and November 1994, 54 selected patients (31 men and 23 women; mean age 51, range 3981 years) underwent elective laparoscopic left colon resection for diverticulitis (n 40), polyps (n 10), ischemic colitis (n 3), and endometriosis (n 1). Patients with previous extensive abdominal surgery and previous and/or ongoing treatment for malignant disease were excluded.

Technique
A mechanical bowel preparation is achieved using 2 l of polyethylene glycol ingested orally during 2 days before surgery. Broad-spectrum intravenous antibiotics are given 1 h preoperatively. Patients are given perioperative epidural analgesia and general endotracheal anesthesia and undergo placement of nasogastric tube and urinary catheter. Patients are placed in the lithotomy position with the thighs unflexed. The surgeon, cameraman, and scrub nurse stand on the patients right side. The cameraman and the scrub nurse stand on the surgeons left and right side, respectively. Pneumoperitoneum is induced using carbon dioxide insufflated to a pressure of 10 mmHg by placement of a trocar in the infraumbilical skin using a cut-down technique. A 0 forward-viewing telescope is employed. A 10-mm port and a 12-mm port are placed lateral to the rectus muscle sheath in the right hypocondrium and iliac fossa, respectively. Patients are turned into a steep Trendelenburg tilt and right lateral decubitus after trocar introduction. The sigmoid colon is retracted with a Babcock clamp and dissected free from the retroperitoneum using curved scissors with monopolar cautery. The left ureter is identified. Further proximal mobilization of the bowel and of the splenic flexure is performed in a similar way according to the extent of the resection in order to ensure a tension-free anastomosis. The visceral peritoneum of the mesentery of the intended specimen is incised. The proximal and distal margins of the specimen are divided with intestinal cartridges of an endoscopic 30-mm linear stapler. The mesentery is divided using diathermy, clips, and a linear stapler with vascular cartridges. An additional 33-mm trocar (Endopath, Ethicon, Somerville, NJ) is inserted suprapubically once the surgeon has judged that conversion to an open procedure is unlike to occur. When necessary, the large bowel may be grasped with a Babcock clamp introduced through the 33-mm cannula and

Despite the lack of convincing statistical evidence in support of laparoscopic colectomy as an approach which results in immediately recognizable benefits to the patient and costeffectiveness, it is a fact that an increasing number of surgeons are attempting it to treat benign diseases of the large
Presented at the 6th joint meeting of the International Gastro-Surgical Club (IGSC), Bangkok, Thailand, 36 December 1995, and at the 5th World Congress of the Society of American Gastrointestinal Endoscopic Surgeons (SAGES), Philadelphia, Pennsylvania, USA, 1317 March 1996 Correspondence to: R. Bergamaschi, Institute of Surgery, University of Bergen, Diakonissehjemmets University Hospital, N-5009 Bergen, Norway

801

difficult extraction of circular stapler (n 1) and end-toside anastomosis (n 1). Postoperative complications (9.2%) included wound hematoma (n 1), wound infection (n 2), pleural effusion (n 1), and urinary retention (n 1). Median operating-room cost was $3,040 (range $2,7843,424). Conclusion The procedure described here is not by any means a new surgical technique. In fact, only access to the doublestapling technique [3] has been changed. A variety of methods have been described for fashioning intracorporeal colorectal anastomoses [1, 2, 4, 5]. All these procedures share the potential disadvantages of operating with an open rectal stump. Besides, a triple-stapled anastomosis [2, 4] appears to unnecessarily complicate surgical technique. With regard to specimen delivery, a transanal route has been proposed either in association with transanal endoscopic microsurgery [4] or using a snare placed through a colonoscope [1]. Contradicting Darzi et al. [2], transanal specimen removal seems to us inappropiate regardless of the size of the lesion and particularly when operating on cancer. A 33-mm suprapubic port allows safe specimen retrieval and construction of a colorectal anastomosis with a closed rectal stump. It has been claimed [7] that an intracorporeal anastomosis would entail a hardly justifiable increase in operating time and costs when compared to open surgery. The present data, which in part include the authors learning curve, do not appear to support these claims. In fact, a judicious use of reusable laparoscopic equipment may help contain costs. No data are available in the literature comparing extracorporeal and intracorporeal anastomoses in terms of immediately recognizable benefits to the patient and costeffectiveness. In this respect, a costutility analysis would be most appropiate because it measures quality of life.
Acknowledgment. The authors are grateful to Hans Knapp, South Tyrol, for having designed Figure 1.

Fig. 1. Fashioning hand-sewn purse strings intracorporeally.

held by the scrub nurse, thus allowing the surgeon to use a two-handed technique during dissection of the mesentery. A plastic bag introduced via the 33-mm port allows temporary intraabdominal storage of the specimen. The anvil of a 31-mm or 33-mm-diameter circular stapler is then inserted through the suprapubic trocar and held into the proximal colon after complete excision of the staple line. A noncrushing bowel clamp passed via the trocar in the right hypocondrium and held by the cameraman may be placed over the proximal colon to avoid fecal spillage. The bowel lumen is generally not widened by dilators prior to anvil introduction, while intravenous glucagon is given when necessary [6]. During the earlier part of our experience, hand-sutured purse strings were made with a ski-shaped needle and extracorporeal knotting. Following the development of the T-needle technique (Endo-Stitch, USSC, Norwalk, CT) purse strings are made and tied intracorporeally around the purse-string notch of the anvil (Fig. 1). The rectal stump is irrigated by luminal wash-out in order to verify that the linear staple row is tight before introduction of the circular stapler. A double-stapled colorectal anastomosis is performed according to a previously described technique [3]. The circular stapler is removed and both tissue doughnuts are inspected for completeness and thickness. The integrity of the anastomosis is checked by irrigation with methylene blue after the application of a noncrushing intestinal clamp just proximal to the circular staple line. In case of intraoperative leak, the anastomosis is reinforced with sutures. The specimen in a plastic bag is delivered through the 33-mm suprapubic port.

References
1. Bleday R, Babineau T, Forse RA (1993) Laparoscopic surgery for colon and rectal cancer. Semin Surg Oncol 9: 5964 2. Darzi A, Super P, Guillou PJ, Monson JRT (1994) Laparoscopic sigmoid colectomy: total laparoscopic approach. Dis Colon Rectum 37: 268271 3. Knight CD, Griffen FD (1980) An improved technique for low anterior resection of the rectum using the EEA stapler. Surgery 88: 710714 4. Ko kerling F, Gastinger I, Gall CW, Schneider B, Krause W, Gall FP (1992) Laparoskopische kolorektale Chirurgie: Kolon- und Rektumanastomosen in triple-stapling-technique. Minimal Invasive Chir 1: 4450 5. Mentges B, Bue G, Scha fer D, Manncke K, Becker HD (1995) Transanal endoscopic microsurgery and combined operations. In: MorenoGonza lez E, Escart n P, Lygidakis NJ, Balibrea JL, Pajares JM, Hidalgo Pascual M (eds) Joint meeting of surgery, gastroenterology and endoscopy. Jarpyo, Madrid, pp 339345 6. Moseson MD, Hoexter B, Labow SB (1980) Glucagon, a useful adjunct in anastomosis with a stapling device. Dis Colon Rectum 23: 25 7. Scott HJ, Spencer J (1995) Colectomy: the role of laparoscopy. Surg Laparosc Endosc 5: 382386

Results No perioperative deaths occurred. Median operating time was 125 (range 80210) min. Median length of ileus was 3.7 (range 25) days. Median hospital stay was 4 (range 37) days. Intraoperative complications (3.7%) included

Surg Endosc (1997) 11: 878

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Cost comparison: disposable vs reusable instruments

I read with interest the article by Demoulin et al. in the May 1996 issue of Surgical Endoscopy. [1] We have been doing this kind of investigation since 1992, studying more than 6,000 cycles of sterile processing of surgical laparoscopies (see our publication) in our institution. [2] We agree with the article, although our calculation base was different. We did not assume the life span for an instrument (50 cycles in 3 years), but did calculate expenses for buying and repair, beginning in 1990, until the present. Calculation then was done for 1 year, adding all costs (instruments, repair, staff, processing, location, installation, devices, energy, water, waste disposal). We developed a documentation sheet to analyze the suitability of instruments, devices, detergents, and processing parameters. Suitability in our terms means handling, function (failure), reusability. Meanwhile, we have finished the first prospective analysis of instrument cycles with new instruments (three trays, 100 cycles each) which will be published next. Some 70100 surgical laparoscopies are performed each month (75% cholecystectomies). Sterile processing in our hospital is strictly separated from the operating theatre; thus, transportation is needed. We try to avoid manual cleaning, preferring automatic devices. Precleaning might be necessary, especially for working tips and inner surfaces of tubes, the main localization of remaining contamination (apart from insulation on tubes). The percentage of laparoscopies in operations was 25%, which is important with regard to costs for processing. We calculated the costs for comparable laparoscopic instruments and their repair with USD 40 (including bipolar forceps for coagulation, but neglecting clips). Processing costs were calculated with USD 30 or less depending on the skills of the personnel and the quality of automatic devices. The comment on biocompatibility of the different materials used for disposables and reusables seems important to us. During operation the instrument is implanted into biological tissue. It is appropriate to ask for information concerning composition of the plastics and/or steel compounds, a question of good manufacturing practice (GMP). Finally, we agree that there are no hygienic problems in the sense of nosocomial infection. In fact, there are very few data available concerning clinical evidence of remaining contamination, although it has been proven that contamination may remain after cleaning, but then it is supposed to be inactivated by steam sterilization (own data, 1993). [3] Therefore we are now studying cleaning modalities and remaining contamination on all surgical instruments to acquire clinical data for hygienic in vitro testing (multicenter study 199798). Any test soil used must be correlated to clinical data of remaining contamination (quality and quantity of detritus). In conclusion, we recommend reusables. Disposables will be necessary, as well, especially for even smaller instruments than needed in laparoscopy or arthroscopy. The finer the instrument the finer it must be treated and the fewer times it can be used.

Th. W. Fengler
Kranoldstr. 24 D-12051 Berlin Germany

References
1. Demoulin L, Kesteloot K, Penninckx F (1996) A cost comparison of disposable vs reusable instruments in laparoscopic cholecystectomy. Surg Endosc 10: 520525 2. Fengler ThW, Pahlke H, Kraas E (1995) Five years of experience with laparoscopic instruments and accessories. Repair index and cleaning modalities after 3000 laparoscopic cholecystectomies. Minimal invasive Medizin 6(4): 153158 3. Fengler ThW, Pahlke H, Kraas E (1996) Minimal invasive instrumentemaximal expandierende Kosten? Erfahrung aus 5000 Arbeitszyklen am Krh. Moabit, Berlin, Geburtsh. u. Frauenheilk. 56(11).

Surg Endosc (1997) 11: 830833

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopically assisted gastric surgery using Dexterity Pneumo Sleeve

T. Naitoh, M. Gagner
Department of General Surgery/A-80, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, OH 44195, USA Received: 18 September 1996/Accepted: 26 December 1996

Abstract Background: Laparoscopic surgery has been successfully applied to several gastrointestinal procedures. Although the totally laparoscopic gastrectomy is feasible, tactile sensation and manipulation of the organ as well as the lesion are decreased when compared to open surgery. The Dexterity Pneumo Sleeve is a new device which allows the surgeon to insert a hand into the abdominal cavity while preserving the pneumoperitoneum. This device was used for patients who underwent laparoscopic gastric surgery. Methods: The first patient presented with a non-Hodgkins lymphoma of the stomach. A laparoscopically assisted distal gastrectomy was performed with Roux-en-Y reconstruction. The second patient had a 5-cm leiomyoma involving the greater curve of the stomach, and this device was used for manipulation of the tumor. The last patient suffered from morbid obesity with its associated medical complications and a ventral hernia. The Sleeve was applied at the hernia site and a laparoscopically assisted gastric bypass was performed. Results: The Pneumo Sleeve was useful in these cases for tactile localization of the tumor and for retraction and manipulation of the stomach and surrounding upper abdominal organs. Conclusions: The utilization of this device resulted in a more easily performed dissection, resection, and anastomosis and was felt to decrease operation time. Key words: Gastric surgery Lymphoma Leiomyoma Morbid obesity Laparoscopic surgery Surgical device

exception. Goh et al. reported two cases of totally intraabdominal laparoscopic Billroth II gastrectomy in 1992 [5]. Watson et al. described the technique of totally laparoscopic Billroth II gastrectomy for early gastric cancer in 1995 [14]. Although the totally laparoscopic gastrectomy for carcinoma is feasible, there are still some concerns about detection of other lesions and surgical margins [3, 8]. A persistent problem with laparoscopic surgery is the inability of the surgeon to palpate the abdominal contents during the operation. This lack of tactile sensation can lead to poor general abdominal exploration, difficulty in extraction of organs, or a relatively long operation time compared to conventional open procedures. The Dexterity Pneumo Sleeve (Dexterity, Research Triangle Park, NC) is a new device which allows the surgeon to insert a hand into the abdominal cavity through a relatively small incision while preserving the ability to work under pneumoperitoneum [15]. The use of this product provides the possibility of hand assistance during laparoscopic surgery and tactile sensation of the lesion which might not be apparent with the use of instrumentation alone. This product was recently approved for clinical use in the United States. During our early experience in this project, we had three patients who underwent laparoscopically assisted gastric surgery using the Dexterity Pneumo Sleeve. Description of Dexterity Pneumo Sleeve The device has two main components: the Protector Retractor and the Pneumo Sleeve (Fig. 1). The Protector Retractor is an open-ended cylinder with a flexible ring at each end. One ring is inserted through the incision into the peritoneal cavity while the other remains outside of the incision. The retractor lines the wound and acts to keep the incision open and to protect against wound contamination. The Pneumo Sleeve is approximately 34 inches long and 912 inches wide and is made of a biocompatible, sealable polyolefin-type material. The distal side of the sleeve has a fenestration and an adhesive flange for attaching to the pa-

Laparoscopic surgery has been recently applied to several gastrointestinal procedures to reduce pain and effect a faster recovery and better cosmetic result. Gastric surgery is no
Correspondence to: M. Gagner

831

Fig. 2. Stomach was retracted with the inserted hand and transected with endoscopic linear staplers.

Fig. 1. The Dexterity Pneumo Sleeve set, which allows the surgeon to insert a hand into the abdominal cavity while preserving an ability to work under pneumoperitoneum.

tients abdomen. The proximal end of the sleeve has a cuff (constructed of Velcro or molded plastic) which is designed to secure the sleeve around the surgeons arm and prevent leakage of pneumoperitoneum gas. Generally, the length of the skin incision (in centimeters) required to insert the surgeons hand is almost equal to the surgeons glove size. Surgical technique
The patients who underwent laparoscopic-assisted gastric surgery using the Dexterity Pneumo Sleeve signed a specific informed consent which was approved by the Institutional Review Board of the Cleveland Clinic Foundation.

proximately 20 to 25 cm from the ligament of Treitz and the distal loop was brought up over the transverse colon. After putting a stay suture between the stomach and jejunum, a small incision was made in the anterior wall of the stomach and antimesenteric side of the jejunum, respectively. The endoscopic linear stapler was inserted through the hole and fired twice to create a 6-cm gastrojejunostomy. The remaining enterotomy was closed with 3-0 running suture using the hand-assisted intracorporeal knot-tying technique. Then a loop of more distal jejunum was selected and pulled through the skin incision of the Pneumo Sleeve; the jejunojejunostomy was created extracorporeally. The abdominal cavity was irrigated and two drains were left next to the anastomosis. Trocars were all removed, and all fascial wounds were closed with 0 Polyglactin 910 sutures and 4-0 sutures for subcutaneous. Approximate blood loss was 150 ml and duration of operation was 360 min. The patient started normal diet on 5th postoperative day and was discharged on postoperative day 8 without complications.

Case 2
A 78-year-old male who was otherwise in good health presented with a history of gastrointestinal bleeding. The examination and subsequent workup revealed a 5-cm submucosal tumor of the stomach which was involving the anterior wall of the gastric body. Biopsies returned probable leiomyoma and computerized tomography demonstrated the lesion without any signs of malignancy or metastasis. A 1.5-cm gallstone was also seen in the gallbladder. After usual laparoscopic cholecystectomy, an 8-cm skin incision was made in the infraumbilical area. The Dexterity Pneumo Sleeve was applied to the skin, and the left hand of the operator was inserted inside of the abdominal cavity. After insufflation of carbon dioxide gas up to 15 mmHg, a 10-mm trocar was inserted in the left subcostal area, and a 12-mm trocar for the insertion of the endoscopic linear stapler was inserted in the left paramedian area. Using 10-mm 30 angle laparoscope, the lesion was taken by the inserted hand and lifted up to the anterior abdomen. Confirming the adequate margins with manual palpation, a wedge resection of the stomach was performed with four cartridges of the endoscopic linear stapler, 12 mm in diameter, 30 mm in length (Fig. 3). The specimen was retrieved through the Pneumo Sleeve and sent to pathology for immediate examination. It was confirmed to be leiomyoma with clear margins. One stitch of 2-0 silk was applied with a laparoscopic needle holder, using intracorporeal knot tying technique, to the superior staple line to control bleeding in that area. The Pneumo Sleeve and trocars were all removed and wounds were closed with 0 Polyglactin 910 sutures for fascia and 4-0 for the skin. Duration of the procedures was 140 min. Estimated blood loss was about 20 ml. A postoperative course was uncomplicated. The patient started normal diet on the 4th postoperative day and was discharged on the 7th postoperative day.

Case 1
The patient is 66-year-old female who weighed 119 kg and was 170 cm tall. She presented with upper gastrointestinal bleeding due to ulceration of a non-Hodgkins lymphoma of the stomach. The lymphoma was on the lesser curvature of the gastric body starting at approximately 5 cm distal to the gastroesophageal (GE) junction; the ulcer was approximately 7 cm in diameter. She was scheduled to undergo laparoscopically assisted distal gastrectomy with Roux-en-Y gastroenterostomy. Surgery was performed with the patient supine with legs in the split position. The surgeon stood between the legs and an assistant was on each side. An 8-cm skin incision was made in the lower midline and the Dexterity set was applied to the skin surrounding the wound. A 10-mm trocar was inserted in the left flank and pneumoperitoneum was created. A 10mm and a 12-mm trocar were then inserted in the left subcostal region as well as a 10-mm trocar in the right subcostal area. The left hand of the surgeon was inserted into the abdominal cavity through the Pneumo Sleeve, and the tumor was easily palpable, as were its margins. Using the Harmonic Scalpel (UltraCision Inc., Smithfield, RI), the gastrocolic ligament was taken down to mobilize the greater curvature of the stomach. The right gastroepiploic artery was transected after applying clips. Then the duodenum bulb was dissected and the first portion of the duodenum was transected with an endoscopic linear stapler. While retracting the stomach inferiorly with the left hand, the lesser curvature was dissected with the Harmonic Scalpel until a point just distal to the GE junction. The short gastric arteries were taken in a similar fashion. Clips were utilized to control the larger vessels. Then, using the endoscopic linear stapler with several cartridges, the distal stomach was divided (Fig. 2). This specimen was taken out through the Sleeve and sent to pathology. Next, the jejunum was divided with an endoscopic linear stapler ap-

Case 3
The patient was a morbidly obese 44-year-old male who presented with a ventral hernia. He weighed 172 kg and was 170 cm tall. He had required coronary artery bypass grafting in 1992 and subsequently developed a

832

Fig. 3. The tumor was taken by the inserted hand, and a wedge resection was performed with staplers.

ventral hernia in the epigastric region. The hernia was repaired in 1995 but recurred shortly thereafter. The patient was felt to be a proper surgical candidate for both his morbid obesity as well as the ventral hernia. The patient was scheduled for a laparoscopically assisted gastric bypass with gastric division [10], Roux-en-Y gastrojejunostomy (Fig. 4), tube gastrostomy for feeding, and ventral hernia repair. Under general anesthesia an 8-cm upper midline incision was made in the hernia site, and all adhesions were lysed. The Dexterity Pneumo Sleeve was then applied on the skin. The left hand of the surgeon was inserted and a left paramedian trocar, 10 mm in diameter, was inserted. After insufflation of carbon dioxide gas, two more 10-mm trocars were inserted in the left lower paramedian area. The greater curvature of the stomach was mobilized to the GE junction using the Harmonic Scalpel. The tissue surrounding the distal esophagus and proximal stomach was dissected and the stomach was divided into two pouches 1 cm distal of the GE junction using the endoscopic linear stapler. The jejunum was divided 20 cm distal to the ligament of Treitz, and the distal limb of jejunum was brought up over the transverse colon. The gastrojejunostomy was performed end to side with interrupted 2-0 silk sutures using a laparoscopic needle holder. The nasogastric tube was passed through the anastomosis and into the jejunal limb prior to completion of the anterior row of sutures. The Pneumo Sleeve was removed and the Roux-en-Y jejunojejunostomy was made extracorporeally through the upper abdominal incision. A 20-Fr gastrostomy tube was inserted into the anterior upper gastric pouch, and a feeding jejunostomy tube was inserted 40 cm distal to the Roux-en-Y anastomosis. A drain was left near the GE junction and the abdominal cavity was irrigated with saline. The hernia was repaired with mesh and nonabsorbable 0 sutures. The skin was closed with a 4-0 subcuticular stitch. Estimated blood loss was about 500 ml. Duration of procedure was 390 min. The patients hospital course was complicated by pneumonia and he was discharged on the 31st postoperative day tolerating a soft diet.

Fig. 4. Stomach was transected using endoscopic linear staplers. Then gastrojejunostomy was created with intracorporeal suturing method. Finally the Roux-en-Y jejunojejunostomy was made extracorporeally through the upper abdominal incision.

Discussion The fervor over laparoscopic cholecystectomy cultivated tremendous interest in expanding this type of minimally invasive surgery into other areas. The combinations of improved instrumentation, better operating room staff familiarity, and advanced surgical skills, along with patient education in this subject, have all been kindling factors. Several articles have reported the feasibility of laparoscopic gastric surgery. For example, Goh et al. reported two patients who underwent complete laparoscopic Billroth II gastrectomy for a gastric ulcer in 1992 [5]. Lointier et al. also reported a case with Billroth II gastrectomy for a gastric ulcer in 1993 [9]. For malignancy, Ablassmaier et al. described a laparoscopic distal gastrectomy for malignant lymphoma in 1994

[1], and Watson and associates presented a case of laparoscopic distal gastrectomy for early gastric cancer in 1995 [14]. Based on our experiences in advanced laparoscopic surgery, we believe that performing totally laparoscopic gastric surgery, including resections, is complicated but feasible. Cases involving morbid obesity or gastric tumors, however, add an extra degree of difficulty. Especially when removing a mass in the stomach, the lack of tactile sensation may lead to misdiagnosis, lack of localization of the tumor, and inadequate assessment of lymphnode or liver metastasis. This is, indeed, one of the reasons why laparoscopic surgery has not been widely accepted in the oncologic setting [3]. Llorente reported a case with laparoscopic gastric resection due to gastric leiomyoma and explained that the tumor was identified by grasping with a Babcock clamp, but the center and perimeter of the lesion could not be ascertained [8]. In addition, laparoscopic gastrectomy requires extension of the skin incision to retrieve the specimen. Based on several case reports, range of the wound size for specimen retrieval is from 25 mm to 50 mm [1, 2, 4, 69, 1114]. The Dexterity Pneumo Sleeve, in our laboratory and clinical investigation, appears to fill at least part of the void described above. The hand inserted into the abdominal cavity is an excellent instrument for retraction, suturing, knot tying, and gentle manipulation of the tumor. Even intracorporeal suturing, which many surgeons hesitate to do, can be done very easily and quickly. Using the Pneumo Sleeve, the size of the skin incision for the application of this device was 8 cm. Considering the benefit of using this device, the size of the skin incision is still acceptable. A 3- or 4-cm extension of the wound yields many advantages to laparoscopic surgeries such as tactile sensation, finger dissection, and retraction. These abilities improve the quality of the exploration of the abdominal cavity and decrease the opera-

833

tion time. In cases 1 and 3, operation times may not appear to be decreased significantly; however, considering the obesity of the patients, the severe adhesions, and difficulty of these procedures, we believe that the time saved in each case would have been measured in hours. Finally, the use of this device is not advanced for standard laparoscopic operations, such as cholecystectomy or hernioplasty, but it may be quite useful and suitable for advanced laparoscopic surgical cases. References
1. Ablassmaier B, Steinhilper U, Bandl WD, Ziehen T, Munster W, Fockersperger H (1994) 100 Jahre nach Billroth . . . Laparoskopische distale Magenresektion (Billroth I und Billroth II). Der Chirurg 65: 367372 2. Anvari M, Park A (1994) Laparoscopic-assisted vagotomy and distal gastrectomy. Surg Endosc 8: 13121315 3. Craven JL (1995) Laparoscopic Billroth II gastrectomy for early gastric cancer [letter; comment]. Br J Surg 82: 17001701 4. Goh P, Kum CK (1993) Laparoscopic Billroth II gastrectomy: a review. Surg Oncol 2(Suppl)1: 1318 5. Goh P, Tekant Y, Isaac J, Kum CK, Ngoi SS (1992) The technique of laparoscopic Billroth II gastrectomy. Surg Laparosc Endosc 2: 258260

6. Johanet H, Cossa JP, Hamdan M, Marmuse JP, Le Goff JY, Benhamou G (1994) Laparoscopic gastrectomy for obstructing duodenal ulcer. J Laparoendosc Surg 4: 447450 7. Kitano S, Iso Y, Moriyama M, Sugimachi K (1994) Laparoscopyassisted Billroth I gastrectomy. Surg Laparosc Endosc 4: 146148 8. Llorente J (1994) Laparoscopic gastric resection for gastric leiomyoma. Surg Endosc 8: 887889 9. Lointier P, Leroux S, Ferrier C, Dapoigny M (1993) A technique of laparoscopic gastrectomy and Billroth II gastrojejunostomy. J Laparoendosc Surg 3: 353364 10. Lonroth H, Dalenback J, Haglind E, Lundell L (1996) Laparoscopic gastric bypass. Surg Endosc 10: 636638 11. Nagai Y, Tanimura H, Takifuji K, Kashiwagi H, Yamoto H, Nakatani Y (1995) Laparoscope-assisted Billroth I gastrectomy. Surg Laparosc Endosc 5: 281287 12. Uyama I, Ogiwara H, Takahara T, Kato Y, Kikuchi K, Iida S (1995) Laparoscopic Billroth I gastrectomy for gastric ulcer: technique and case report. Surg Laparosc Endosc 5: 209213 13. VanHouden CE (1994) Laparoscopic bilateral truncal vagotomy, antrectomy, and Billroth I anastomosis for prepyloric ulcer. Surg Laparosc Endosc 4: 457460 14. Watson DI, Devitt PG, Game PA (1995) Laparoscopic Billroth II gastrectomy for early gastric cancer. Br J Surg 82: 661662 15. OReilly MJ, Saye WB, Mullins SG, et al (1996) Technique of handassisted laparoscopic surgery. J Laparoendosc Surg 6(4): 239244

Surg Endosc (1997) 11: 815818

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopy without trocars

A. M. Ternamian
Division of Endoscopic Surgery, Department of Obstetrics and Gynecology, St. Josephs Health Centre, 30 The Queensway, Toronto, Ontario, M6R 1B5, Canada Received: 1 July 1996/Accepted: 11 November 1996

Abstract Background: Laparoscopic access devices have consisted of a cannula and sharp trocar. A trocarless endoscopic access cannula was designed to improve safety, eliminate blind entry, save time, and decrease cost. It consists of a threaded hollow cannula with a blunt penetrating distal end. Methods: After pneumoperitoneum is established with the Veress needle, the laparoscope is inserted into the cannula, and using the blunt tip at the distal end of the thread, the anterior rectus fascia is engaged through a 57-mm horizontal fascial incision. Rotation advances the blunt cannula tip past the anterior rectus fascia, muscle, and posterior rectus fascia and through the peritoneum under continuous video control with no axial force. Results: The trocarless rotational access cannula (TRAC) was used successfully in 135 consecutive unselected patients with no complications. Conclusions: This access cannula and methods offer many advantages while improving safety and cost. Key words: Laparoscopy Cannula insertion Surgical technique Endoscopic complications The development of laparoscopic procedures is one of the major surgical advances of the past decade; laparoscopy is the second most common gynecologic procedure performed today (next to abortions) [21]. Despite considerable technological innovation, most primary port insertions remain a blind procedure, with potentially serious yet avoidable complications [27]. Pneumoperitoneum is induced by inserting the Veress needle at or near the umbilicus. Customarily, the primary port is then introduced blindly by applying axial force on the sharp trocar and cannula. Surgeons rely on experience to feel the different landmarks and layers of the abdominal wall. The incidence of serious complications is reported as about one per 1,000 patients for bowel and vascular injuries
Present address: 105 Glenayr Road, Toronto, Ontario M5P 3C1, Canada

[5]. However, most studies are retrospective, and the true incidence may be underreported [22]. Too-shallow insertion can produce preperitoneal emphysema with loss of the ability to enter the peritoneal cavity, necessitating repeated attempts at insertion [27]. Toodeep or sideways insertion may cause retroperitoneal emphysema, CO2 embolus, or bowel or vessel injury [20]. The axial force necessary to pass a sharp reusable trocar is twice as great as that needed for disposable ones (14.55 vs 7.14 lb) [6]. The less the force applied during insertion, the greater the control. Reduced force decreases risk of inadvertent injury, especially in thin or pediatric patients where the distance between umbilicus and major abdominal vessels is considerably less [11, 15]. Although smooth, conetipped trocars are less likely to lacerate blood vessels on entry, greater axial force is necessary to insert them compared to the sharp, bevelled trocars [12, 20]. Disposable shielded trocars have not eliminated bowel perforations or vascular injuries. The safety shield, although designed to spring once fascial resistance is lost, has a 20 50-ms lag time to shield deployment, so quick, uncontrolled, deep axial thrusts can injure bowel or vessels [5, 23]. Conversely, slow insertion can hold the safety shield at the fascial or subfascial level, while the exposed sharp trocar-tip penetrates the peritoneum unprotected. The safety shield can trigger before penetrating the peritoneum, necessitating repeated trocar passes. The shield is less important than the external force applied at the other end of the trocar [4]. As yet there are no scientific data to indicate that the safety shield provides additional protection [24]. However, there have been inadvertent visceral injuries attributed to the shielded models [1, 22]. Moreover, open laparoscopic entry using Hassons technique does not necessarily avoid or prevent bowel injuries [14]. Penfields survey found six bowel injuries in 10,840 open laparoscopies, an incidence similar to laparotomies; half of these injuries were not noticed during the initial surgery [18]. The need for placement of fascial sutures, difficulty maintaining an airtight seal, poor visualization,

816

Fig. 2. Transverse 57-mm fascial incision along fascial fibers (white arrow). Note the blunt cannula tip at 3 o-clock (black arrow) to engage slitlike fascial window.

Fig. 1. Trocarless visual access cannula developed, in 5- and 10-mm diameters. With scope inserted, blunt distal end engages fascial window; rotation will elevate fascia along the thread.

longer insertion time, and the bulk of the cannula make this method less practical [17]. Thus, a safer visual access method with a trocarless reusable cannula has been developed.

Materials and methods

The trocarless rotational access cannula (TRAC) consists of a single sturdy stainless-steel device with a proximal valve segment and a distal cannula segment. A thread winds along the cannulas outer surface and ends as a blunt tip (Fig. 1). Several 10- and 5-mm cannulas were developed and used on 135 consecutive unselected gynecologic patients at St. Josephs Health Centre, Toronto, from November 1994 to June 1996. Informed consent was obtained. The age; weight; height; pre- and postoperative diagnosis; operative procedure and findings; number, type, and size of cannulas; insertion time; incision length; and previous pelvic or laparoscopic surgery and complications were recorded at each operation on a standard form. Under general endotracheal anaesthesia, pulse oximetry, and end-tidal CO2 monitoring, the patient is placed in dorsal lithotomy position, prepped, draped, and catheterized by the surgeon. Using a small scalpel, a 15-mm vertical subumbilical skin incision is made; the subcutaneous tissue is teased laterally using dental pushers and small right-angle retractors; the white anterior rectus fascia is exposed; and CO2 is insufflated using the Veress needle. Then the Veress needle is removed and a 57-mm horizontal fascial incision is made along the anterior fascial fibres (Fig. 2). The warmed laparoscope with camera connected is inserted into the cannula and held by the surgeon. The blunt

cannula tip is then engaged into the slitlike anterior rectus fascial incision using the surgeons other hand, and the cannula is rotated clockwise around a fixed laparoscope. Without exerting any axial force, the fascia and muscle stretches radially and engages snugly the initial thread on the outer rim of the cannula (Fig. 3). Further rotation elevates the fascia which travels along the outer thread, while the blunt tip of the cannula advances radially and safely enters the peritoneal cavity, under visual control. Just before penetrating the thin, stretched, transparent peritoneal membrane, the CO2-distended peritoneal cavity appears dark blue-gray, and nonadherent peristalsing bowel may be seen. When adhesions are present, glistening white light reflections are seen, and an alternate site should be selected. For extraperitoneal procedures, the cannula is not advanced further, and the laparoscope is tunnelled visually to the desired space, e.g., the space of Retzius or inguinal region. At the end of the procedure, the cannula is rotated counterclockwise, and the different tissue layers are seen on the videomonitor to disengage off the thread and converge into apposition. The separated rectus muscle fibers meet vertically at midline and seal the peritoneal window, while the fascial halves shut horizontally under magnified vision. Any omentum or bowel lodged along the cannulas tract would be recognized immediately. The skin is then sutured with interrupted subcuticular 3-0 absorbable suture. No fascial sutures are necessary, as the window is an overlapping gridiron incision with intact muscle and fascial fibers.

Results A total of 177 TRAC ports were created on 135 patients, 10 mm (135 patients) and 5 mm (42 patients) in diameter, with no failed insertions. The mean weight of the patients was 65 kg (range 42 to 120), and the mean age was 38.2 years (range 18 to 76). None of the patients in this small group developed abdominal wall bleeding during insertion or intraoperative complications. We did not encounter CO2 leakage around the cannula. The mean insertion time was 2.4 min. No difficulty was encountered in accessing the heavier patients. Four patients required a laparotomy after successful insertion for unrelated reasons. The majority of patients were discharged within 6 h after surgery. Two patients were readmitted 24 h after dis-

817

Fig. 3. With video-monitor control, rotation of cannula lifts anterior fascia while maintaining airtight seal and advances blunt tip radially into peritoneum.

charge, both for paralytic ileus, which resolved spontaneously by the next day. There were seven superficial umbilical wound infections. Follow-up in 4 weeks found no complications, and the skin scars were comparable to conventional trocar-site scars. No port-site hernias have been detected thus far, but longer follow-up is necessary. Discussion Laparoscopy is the most commonly performed abdominal surgery in the developed world [13]. Despite the relative low rate of serious complications, their absolute numbers present a public health and medicolegal challenge. Deziels survey of 77,604 laparoscopies revealed an incidence of 0.14% for bowel and 0.25% for vascular injuries, where the mortality from bowel injuries was 4.6% and from vascular injuries 8.8% [7]. Yuzpes survey reported that more than one-fourth of Canadian laparoscopists had patients who experienced bowel, bladder, or vascular injury (27% Veress induced, 26% primary trocar induced, 15% secondary trocar induced) [27]. Kaali found that as much as one-third of the standard sharp trocar tip penetrates the peritoneal cavity before the anticipated loss of fascial resistance is perceived manually by the surgeon [13]. Furthermore, razor-sharp bevelled trocars are known to slice across muscle and fascial fibers during insertion and create a tripolar fascial rent; this interferes with the muscle shutter mechanism and predisposes to incisional hernias [3, 16, 25]. Unexpected injuries to viscera continue to occur despite surgical training, experience, and up-to-date equipment [19]. Patients expecting band-aid surgery and intact body image are particularly vulnerable to suggestions of malpractice when confronting complications. As an everexpanding menu of procedures are performed endoscopically, this increasing trend of litigation is worrisome. For each new indication, safety, cost, efficacy, risks and benefits, and extent and duration of surgery must be compared with those of the same procedure performed by laparotomy [9, 10, 26]. Most complications during laparoscopy are associated

with establishment of the pneumoperitoneum, introduction of the primary port, or insertion of secondary trocars. Sharp trocars, especially at the primary ports, cause the most serious injuries [28]. Thus, manufacturers have introduced costly disposable visual-entry trocars; however, considerable axial force is still necessary to access the peritoneal cavity. TRAC instead provides reusable visual access, without applying axial penetrating force; only the forearm muscles are used to rotate instead of the stronger upper arm and shoulder-girdle muscles. Continuous video control shows the surgeon the exact whereabouts of the cannula tip and enables the operator to stop advancing the cannula if complications are anticipated or white adhesions, instead of blue-gray peritoneum, are encountered. During the surgery, the snugly engaged thread stabilizes the cannula and prevents slippage or gas leakage. This simple, cognitively friendly cannula has no moving parts and no trocar; it is inexpensive, easy to maintain and sterilize, and provides controlled visual insertion and removal. It acts as a stabilizer with a tight seal, and probably decreases incidence of hernias. It can be used at primary and secondary port sites, and it can be adapted for minilaparoscopy and automated entry. TRAC is designed to improve safety and decrease cost and has several applications to laparoscopy at other sites (arthroscopy, thoracoscopy, urology, etc.).
Acknowledgment. The author gratefully acknowledges the invaluable guidance, encouragement, and technical assistance of Dr. Mervyn Deitel, Departments of Surgery and Nutritional Sciences, University of Toronto, and Dr. Garo Damla, Department of Obstetrics & Gynecology, St. Elizabeth Hospital, Boston, Massachusetts, in the preparation of this manuscript. Mr. Roger Harris, director of the Audio-Visual Department, St. Josephs Health Centre, Toronto, Ontario, deserves special thanks for the numerous hours spent documenting on video and preparing the artwork. The Trocarless Rotational Access Cannulae used in this study were kindly manufactured to our specifications by Richard Wolf, GMBH of Knittlingen, Germany, Karl Storz, GMBH & Co. of Tuttlingen, Germany, and Core Dynamics Inc. of Jacksonville, Florida.

References
1. Apelgren KN, Scheeres DE (1994) Aortic injury: a catastrophic complication of laparoscopic cholecystectomy. Surg Endosc 8: 689691

818 2. Berqavist D, Berqavist A (1987) Vascular injuries during gynecologic surgery. Acta Obstet Gynecol 66: 1923 3. Bhoyrul A, Mori T, Way LW (1996) Radially expanding dilatation: a superior method of laparoscopic trocar access. Surg Endosc 10: 775 778 4. Cooperman AM (1995) Complications of laparoscopic surgery. In: Arregui ME, Fitzgibbons RJ, Katkhouda N, McKernan JB, Reich H (eds) Principles of laparoscopic surgery, basic and advanced techniques. Springer-Verlag, New York, pp 7177 5. Corson SL (1994) Endoscopic complications: causes, prevention and litigation. In: Soderstrom RM, Hulka JF, Corson SL (eds) Syllabus postgraduate course I. American Association Gynecologic Laparoscopists 23rd Annual meeting, New York, pp 1721 6. Corson SL, Batzer FR, Gocial B, Maislin G (1989) Measurement of the force necessary for laparoscopic trocar entry. J Reprod Med 34: 282284 7. Deziel DJ, Millikan KW, Economou SG, Doolas A, Ko ST, Airan MC (1993) Complications of laparoscopic cholecystectomy: a national survey of 4,292 hospitals and an analysis of 77,604 cases. Am J Surg 165: 914 8. Gomel V, Taylor PJ (1995) Complications and their management. In: Gomel V, Taylor PJ (eds) Diagnostic & operative gynecologic laparoscopy. Mosby, St Louis, pp 299308 9. Grimes DA (1992) Frontiers of operative laparoscopy: a review and critique of the evidence. Am J Obstet Gynecol 166: 10621070 10. Hill DJ, Maher PJ, Wood CE, Lolatgis N, Lawrence A, Dowling B, Lawrence M (1994) Complications of laparoscopic hysterectomy. J Am Assoc Gynecol Laparosc 1: 159162 11. Hurd WW, Bude RO, Delancey JOL, Guavin JM, Aisen AM (1991) Abdominal wall characterization by MIR and CT imaging. J Reprod Med 36: 473476 12. Hurd WW, Pearl ML, DeLancey JOL, Qiunt EM, Garnett B, Bud RD (1993) Laparoscopic injury of abdominal wall blood vessels: a report of three cases. Obstet Gynecol 32: 673676 13. Kaali SG (1993) Introduction of the Opti-Trocar. J Am Assoc Gynecol Laparosc 1: 5053 14. Levy BS, Hulka JF, Peterson HB, Phillips JM (1993) Operative laparoscopy: American Association of Gynecologic Laparoscopists 1993 membership survey. J Am Assoc Gynecol Laparosc 1: 301305 McDougall EM, Figenshau RS, Clayman RV, Monk TG, Smith DS (1994) Laparoscopic pneumoperitoneum: impact of body habitus. J Laparosc Surg 4: 385391 McMurrick PJ, Polglase AL (1993) Early incisional hernia after use of the 12-mm port for laparoscopic surgery. Aust N Z J Surg 63: 574575 Ohl DA, Faerber GJ, Hurd WW (1994) Urologic laparoscopy with a new blunt-tipped trocar: safe, rapid access without the use of fascial sutures. Urology 43: 861864 Penfield AJ (1985) How to prevent complications of open laparoscopy. J Reprod Med 30: 660663 Peterson HB, Hulka JF, Phillips JM (1990) American Association of Gynecologic Laparoscopists 1988 membership survey on operative laparoscopy. J Reprod Med 35: 587589 Semm K, Friedrich ER (1987) Operative manual for endoscopic abdominal surgery. Year Book, Chicago, pp 150151 Soderstrom RM, Levy BS (1990) American association of gynecologic laparoscopists manual of endoscopy. Port City Press, Baltimore, pp 5355 Soderstrom RM (1993) Bowel injury litigation after laparoscopy. J Am Assoc Gynecol Laparosc 1: 7477 Soderstrom RM (1993) Basic operative techniques. In: Soderstrom RM (ed) Operative laparoscopy. The masters technique. Raven Press, New York, pp 2544 Soderstrom RM, Levinson C, Levy BS (1993) Complications of operative laparoscopy. In: Soderstrom RM (ed) Operative laparoscopy. The masters techniques. Raven Press, New York, pp 187197 Turner DJ (1996) A new, radially expanding access system for laparoscopic procedures versus conventional cannulas. J Am Assoc Gynecol Laparosc 3: 609615 Vasquez JM, Demarque AM, Diamond MP (1994) Vascular complications of laparoscopic surgery. J Am Assoc Gynecol Laparosc 1: 163167 Yuzpe AA (1990) Pneumoperitoneum and trocar injuries in laparoscopy: a survey on possible contributing factors and prevention. J Reprod Med 35: 485490

15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.

Surg Endosc (1997) 11: 838841

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic cholecystectomy and time-course changes in renal function

The effect of the retraction method on renal function
Y. Miki,1 K. Iwase,2 W. Kamiike,2 E. Taniguchi,2 K. Sakaguchi,3 J. Sumimura,1 H. Matsuda,2 I. Nagai1
1 2 3

Department of Surgery, Kinan General Hospital, Wakayama 646, Japan First Department of Surgery, Osaka University Medical School, 2-2 Yamadaoka, Suita, Osaka 565, Japan Department of Medicine, Sumitomo Hospital, Japan

Received: 26 March 1996/Accepted: 27 July 1996

Abstract Background: Recently, the retraction method has been used to reduce intraabdominal pressure (IAP) during laparoscopic surgery. The purpose of this study was to determine the serial changes in renal function during laparoscopic cholecystectomy (LC) using the retraction method. Methods: Urine output, effective renal plasma flow (ERPF), and glomerular filtration rate (GFR) were measured serially in seven patients who underwent LC with 12 mmHg pneumoperitoneum (High-IAP group) and five who underwent LC using the retraction method with 4 mmHg pneumoperitoneum (Low-IAP group). Results: Urine output, ERPF, and GFR were decreased during pneumoperitoneum in the High-IAP group, whereas no significant changes in any of these parameters were observed in the Low-IAP group. Conclusions: Our findings demonstrate that reduction of IAP to 4 mmHg using the retraction method prevents the transient renal dysfunction caused by prolonged 12 mmHg pneumoperitoneum during LC, suggesting that the retraction method reduces the risk of perioperative renal dysfunction during laparoscopic surgery. Key words: Laparoscopy Cholecystectomy Pneumoperitoneum Renal function Retraction method

have revealed the systemic hemodynamic changes that occur during laparoscopic cholecystectomy (LC) with prolonged pneumoperitoneum [5]. It has also been reported that continuous elevation of intraabdominal pressure (IAP) might cause critical changes in systemic hemodynamics, especially in patients with heart disease [3]. Therefore, the retraction method with minimal or no pneumoperitoneum has recently been used to reduce the risk of systemic hemodynamic changes in laparoscopic surgery [9]. Increased IAP may have effects on renal function as well as on systemic hemodynamics. Acute renal failure associated with increased IAP has been reported [11]. Prolonged continuous pneumoperitoneum over 60 min may cause transient renal dysfunction even when IAP is maintained at only 12 mmHg during LC [4]. Reducing IAP using the retraction method may reduce the risk of transient renal dysfunction as well as systemic hemodynamic changes during laparoscopic operations. However, the effect of the retraction method on renal function is still unknown. Precise determination of the serial changes in renal function during laparoscopic surgery using the retraction method may be useful for reducing the risks of postoperative renal dysfunction. The purpose of this study was therefore to examine the serial changes in renal function during laparoscopic surgery using the retraction method. Materials and methods

Recent advances in operative procedures and surgical instrumentation have enabled the application of laparoscopic surgery to various kinds of intraabdominal surgery [7, 10, 13]. Laparoscopic surgery is also used for patients with reduced renal function due to underlying diseases such as diabetes, hypertension, and atherosclerosis. Previous studies
Correspondence to: W. Kamiike

Patients
The subjects included seven patients who underwent LC with a conventional method at 12 mmHg IAP (High-IAP group) and five who underwent LC using the retraction method at 4 mmHg IAP (Low-IAP group). All patients were operated upon for chronic cholecystitis due to cholelithiasis. Patients with choledocholithiasis, heart disease, or renal dysfunction (blood urea nitrogen > 20 mg/dl, serum creatinine > 1.5 mg/dl, or creatinine clearance during 24 h < 50 ml/min) were excluded from the present study. There was no significant difference in mean age, distribution by sex, body

839 Table 1. Preoperative characteristics and operative results High-IAP group (n 7) Preoperative characteristics Sex Age (years) Body weight (kg) Blood urea nitrogen (mg/dl) Serum creatinine (mg/dl) Creatinine clearance (ml/min/m2) Anesthesia and operation Fentanyl citrate (mg) Droperidole (mg) Pancuronium bromide (mg) Operating time (min) Intraoperative infusion (ml) 3M/4F 37 11 54 10 14 4 0.9 0.4 76 20 0.4 0.1 9.5 1.0 8.2 0.7 85 12 2,108 382 Low-IAP group (n 7) 2M/3F 46 20 62 16 12 5 0.7 0.2 96 8.4 0.4 0.1 9.7 1.2 7.8 1.3 65 17 1,850 503 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

weight, or preoperative renal function parameters between the two groups (Table 1).

Fig. 1. Under direct visualization with videoscopic monitoring, the Ushaped steel needle (white arrow) is inserted in the epigastrium into the peritoneal cavity on the left side of the falciform ligament and pierced through the dorsal edge of the ligament from left to right (black arrow: trocar at the epigastrium, L: liver, F: flaciform ligament).

Anesthesia
Orotracheal intubation was performed after rapid induction of anesthesia with 5 mg/kg of thiopental sodium and 1.2 mg/kg of suxamethonium chloride. General anesthesia was induced with fentanyl citrate, droperidole, and nitrous oxide. Intraoperative mechanical ventilation was performed with a mixture of 67% nitrous oxide and 33% oxygen. There was no significant difference in the total amounts of fentanyl citrate, droperidole, or pancuronium bromide used between the two groups. There was no significant difference between the two groups in the time of operation or amount of intraoperative infusion (Table 1).

Operation
IAP during laparoscopic surgery was maintained continuously at 12 or 4 mmHg with an electric CO2 insufflator. In the Low-IAP group, with laparoscopic surgery performed using the retraction method, operative procedure was begun at 12 mmHg IAP. A laparoscope was inserted through a trocar in the lower portion of the umbilicus after establishment of pneumoperitoneum at 12 mmHg. Under direct visualization with videoscopic monitoring, a U-shaped steel needle was inserted at their epigastric portion into the peritoneal cavity on the left side of the falciform ligament, pierced through the dorsal edge of the ligament from left to right (Fig. 1), and pushed out of the abdominal wall through the right side of the ligament. The needle was pulled upward using a small winching device with an attached wire in order to elevate the abdominal wall along with the ligament, and then IAP was reduced to 4 mmHg (Fig. 2). Three additional trocars were then inserted along the right costal margin as 12 mmHg, in the usual fashion. This method makes it possible to lift up the falciform ligament, around ligament and liver, which yields a sufficient view of the surgical field even at 4 mmHg IAP.

Fig. 2. The U-shaped retractor needle is pulled upward using a small winching device with an attached wire (white arrow).

Measurement of hemodynamic parameters

The right radial artery was cannulated under general anesthesia. Systolic blood pressure (BP), heart rate (HR), and PaCO2 from gas analysis of systemic arterial blood were measured at six timepoints during the operation: (1) following induction of anesthesia, (2) following initiation of pneumoperitoneum, (3) (30 and 4) 60 min after the initiation of pneumoperitoneum, and (5) before and (6) 15 min after pneumoperitoneum. The ventilation rate was increased as required to maintain PaCO2 below 45 mmHg.

undertaken serially during four intervals derived from the six timepoints described above (12, 23, 34, and 56). Beginning 3 h before surgery, 0.9% saline was administered to all patients by continuous drip infusion at a rate of 80 ml/h. Following primary bolus injection of 20 mg/kg paminohippuric acid (PAH) and 50 mg/kg sodium thiosulfate (STS), continuous intravenous drip infusion of 17 mg/min PAH and 50 mg/min STS was begun 2 h prior to surgery, and it was continued until termination of the anesthesia to maintain blood concentrations of these agents at a constant level. The effective renal plasma flow (ERPF) was measured as the clearance of PAH (C[PAH]) [2], and the glomerular filtration rate (GFR) was measured as the clearance of STS (C[STS]) [8]. Values are expressed as means standard deviation. Statistical analysis was performed using one-way analysis of variance for repeated measurements and multiple comparisons, the unpaired t-test, or the 2 test; p values less than 0.05 were taken to indicate statistical significance.

Results

Operative results Determination of renal function

For each patient, after informed consent for participation in the study had been obtained, urine output was determined and renal functional tests were

There were no significant differences between the two groups in amounts of anesthetic drugs administered, operating time, or intraoperative infusion volume (Table 1).

840 Table 2. Serial changes in BP, HR, PaCO2, PAH, and STSa Timepoints High-IAP group Systolic BP (mmHg) HR (beats/min) PaCO2 (mmHg) PAH (mg/dl) STS (mg/dl) Systolic BP (mmHg) HR (beats/min) PaCO2 (mmHg) PAH (mg/dl) STS (mg/dl) 1 120 15 85 12 30 6 1.3 0.5 6.2 2.3 122 14 85 7 34 5 3.6 1.6 5.3 2.2 2 128 25 84 9 37 7 1.2 0.5 6.4 2.4 130 20 90 4 40 10 4.2 1.9 5.9 2.0 3 125 19 82 8 44 5 1.5 0.6 6.2 2.7 124 11 92 6 41 5 4.0 1.5 5.6 1.7 4 136 23 80 10 43 7 1.5 0.7 6.5 2.5 136 21 87 9 38 4 4.0 1.5 5.5 1.8 5 140 23 81 9 42 5 1.5 0.6 6.6 2.9 130 22 90 8 42 8 4.0 1.5 5.6 1.2 6 142 20 78 10 40 6 1.6 0.5 6.6 2.7 128 12 86 9 38 8 3.7 1.0 5.5 1.4

Low-IAP group

Timepoints: 1, initiation of anesthesia; 2, initiation of pneumoperitoneum; 3, 30 min after the initiation of pneumoperitoneum; 4, 60 min after the initiation of pneumoperitoneum; 5, just prior to depneumoperitoneum; 6, 15 min after depneumoperitoneum. Abbreviations as in text.

Serial changes in BP, HR, and PaCO2 There were no significant differences between the two groups in systolic BP, HR, or PaCO2 at any timepoint tested (Table 2). Serial changes in plasma concentrations of PAH and STS There were no significant differences between the two groups in plasma concentrations of PAH and STS at any timepoint tested. Plasma concentrations were maintained at a constant level in both groups (Table 2). These finding suggest that the clearances of PAH and STS revealed ERPF and GFR, respectively. Serial changes in urine output Urine outputs during the four intervals tested were 101 45, 42 13, 24 4, and 74 21 l/min/kg in the High-IAP group and 58 19, 102 39, 45 16, and 47 16 l/min/ kg in the Low-IAP group (Fig. 3). In the High-IAP group, the rate of urine output from 30 min to 60 min after initiation of pneumoperitoneum was significantly lower than that preceding pneumoperitoneum and that following depneumoperitoneum. Serial changes in C(PAH) C(PAH) values measured for the four intervals were 648 133, 493 57, 410 63, and 754 92 ml/min in the High-IAP group and 400 75, 402 63, 341 86, and 379 71 ml/min in the Low-IAP group (Fig. 4). In the High-IAP group, C(PAH) from 30 min to 60 min after initiation of pneumoperitoneum was significantly lower than that preceding pneumoperitoneum and that following depneumoperitoneum. Serial changes in C(STS) C(STS) values measured for the four intervals were 180 34, 137 17, 110 13, and 222 26 ml/min in the HighIAP group and 114 14, 123 16, 114 9, and 125 23 ml/min in the Low-IAP group (Fig. 5). In the High-IAP group, C(STS) from 30 min to 60 min after initiation of pneumoperitoneum was significantly lower than that preceding pneumoperitoneum and that following depneumoperitoneum.

Discussion Alterations in renal functions occurring during LC in the clinical setting might be affected not only by changes in IAP but also by general anesthesia, intraoperative infusions, and specific operative procedures used for cholecystectomy. In the present study, there was no significant difference in preoperative renal function test results, method of anesthesia, amounts of anesthetic drugs administered, volume of intraoperative infusion, or operative procedures between the High-IAP and Low-IAP groups. Elevation of PaCO2 might significantly affect changes over time in renal function. It has been reported that the sympathetic nervous system is stimulated by increased PaCO2 following the absorption of CO2 through the peritoneum and by the elevated intrathoracic pressure due to the elevation of the diaphragm induced by pneumoperitoneum [1, 12, 14]. In the present study, the rate of mechanical ventilation was increased as necessary to maintain PaCO2 below 45 mmHg. Consequently, there was no significant difference in PaCO2 between the two groups at any timepoint of measurement. These findings suggest that the difference in serial changes in renal function between the two groups was caused by differences in IAP. It has been reported that prolonged, continuous 12 mmHg pneumoperitoneum results in gradual decrease in urine output, ERPF, and GFR during LC [4]. These changes in renal function are not caused by systemic hemodynamic changes or changes in levels of humoral factors such as plasma renin activity, human atrial natriuretic peptide, aldosterone, angiotensin II, antidiuretic hormone, adrenaline, and noradrenaline [4]. In the present study, significant decreases in urine output, ERPF, and GFR were observed during the period 3060 min after initiation of pneumoperitoneum in High-IAP group. ERPF is primarily determined by the difference between renal arterial and venous pressures. It seems quite unlikely that the pneumoperitoneum induced with an IAP of only 12 mmHg resulted in stenosis of the renal artery. It has been reported that the pressure in the inferior vena cava is continuously maintained at almost the same level as IAP during pneumoperitoneum in the clinical setting with general anesthesia [6]. Those findings suggest that the decrease in ERPF during pneumoperitoneum resulted from an elevation of the renal venous pressure induced by the increase in IAP. Our findings demonstrated that reduction of IAP using the retraction method was able to prevent the gradual decrease in ERPF, GFR, and

841

Fig. 3. Serial changes in urine output. 1: Initiation of anesthesia, 2: initiation of pneumoperitoneum, 3: 30 min after the initiation of pneumoperitoneum, 4: 60 min after the initiation of pneumoperitoneum, 5: just prior to depneumoperitoneum, 6: 15 min after depneumoperitoneum. Values are means standard deviations.

urine output caused by prolonged, continuous 12 mmHg pneumoperitoneum. Continuous drip infusion of PAH and STS was used in this study to evaluate serial changes in ERPF and GFR. Clearance of PAH yields the most accurate value of ERPF when the plasma concentration of PAH is maintained within the range of 1 to 5 mg/dl [2], while clearance of STS yields the most accurate value of GFR when the plasma concentration of STS is maintained within the range of 10 to 50 mg/dl [8]. The plasma levels of PAH and STS were maintained within these ranges at all timepoints tested in the present study. Furthermore, the plasma levels of these two agents did not change significantly over time during operation in either of the two groups. Clearances of PAH and STS were, therefore, considered useful for measuring ERPF and GFR in the present study. In conclusion, reduction of IAP to 4 mmHg using the retraction method prevented the decreases in ERPF, GFR, and urine output caused by prolonged continuous 12 mmHg pneumoperitoneum during laparoscopic cholecystectomy. Our findings suggest that the retraction method can reduce the risk of perioperative renal dysfunction, especially in the patients with underlying renal dysfunction.
Acknowledgment. The authors are grateful to the late Hiroaki Takenaka, MD, for his kind advice concerning research design and operative procedures. We would also like to thank the nursing staff of the Operation Unit of Kinan General Hospital for their technical assistance.

References
1. Alexander GB, Brown EL (1969) Physiologic alterations during pelvic laparoscopy. Am J Obstet Gynecol 105: 10781081 2. Dworkin LD, Brenner BM (1991) Total renal blood flowmethods of measurement. In: Brenner BM, Rector FC Jr (eds) The kidney. 4th ed. Saunders, Philadelphia, pp 179782 3. Iwase K, Takenaka H, Yagura A (1992) Hemodynamic changes during laparoscopic cholecystectomy in patients with heart disease. Endoscopy 24:771773 4. Iwase K, Takenaka H, Ishizaka T, Ohata T, Oshima S, Sakaguchi (1993) Serial changes in renal function during laparoscopic cholecystectomy. Eur Surg Res 25:203212 5. Iwase K, Takenaka H, Yagura A (1993) Hemodynamic alterations during laparoscopic cholecystectomya comparison with minilaparotomy cholecystectomy. Digest Surg 10: 59 6. Iwase K, Takao T, Watanabe H (1994) Intra-abdominal venous pressure during laparoscopic cholecystectomy. HPB Surg 8:1317 7. Matsuda M, Nishiyama M, Hanai T, Saeki S, Watanabe T (1995) Laparoscopic omental patch repair for perforated peptic ulcer. Ann Surg 221: 236240 8. Newman EV, Gilman A, Philips Fs (1946) The renal clearance of thiosulfate in man. Bull Johns Hopkins Hosp 79: 229242 9. Nagai H, Kondo Y, Yasuda T, Kasahara K, Kanazawa K (1993) An abdominal wall-lift method of laparoscopic cholecystectomy without peritoneal insufflation. Surg Laparosc Endosc 3: 175179 10. Quattelebaum JK, Flanders HD, Usher CH (1993) Laparoscopically assisted colectomy. Surg Laparosc Endosc 3: 8187 11. Richards WO, Scovill W, Shin B, Reed W (1983) Acute renal failure associated with increased intra abdominal pressure. Ann Surg 197: 183187 12. Richardson DW, Wasserman AJ, Patterson JL Jr (1961) General and regional circulatory responses to change in blood pH and carbon dioxide tension. J Clin Invest 40: 3143 13. Taniguchi E, Kamiike W, Iwase K, Nishida T, Miyata M, Ohashi S, Okada T, Matsuda H (1995) Laparoscopic extramucosal myectomy with anterior fundoplication (Dor) for esophagial achalasia using intraoperative manometry. Surg Endosc 9: 817819 14. Wittgen CM, Andrus CH, Fitzgerald SD (1991) Analysis of the hemodynamic and ventilatory effects of laparoscopic cholecystectomy. Arch Surg 126:9971001

Fig. 4. Serial changes in clearance of P-aminohippuric acid (C[PAH]). 1: Initiation of anesthesia, 2: initiation of pneumoperitoneum, 3: 30 min after the initiation of pneumoperitoneum, 4: 60 min after the initiation of pneumoperitoneum, 5: just prior to depneumoperitoneum, 6: 15 min after depneumoperitoneum. Values are means standard deviations.

Fig. 5. Serial changes in clearance of sodium thiosulfate (C[STS]). 1: Initiation of anesthesia, 2: initiation of pneumoperitoneum, 3: 30 min after the initiation of pneumoperitoneum, 4: 60 min after the initiation of pneumoperitoneum, 5: just prior to depneumoperitoneum, 6: 15 min after depneumoperitoneum. Values are means standard deviations.

Surg Endosc (1997) 11: 812814

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Cost and benefit of the trained laparoscopic team

A comparative study of a designated nursing team vs a nontrained team
T. A. G. Kenyon,1 M. P. Lenker,1 T. W. Bax,2 L. L. Swanstrom1
1 2

Minimally Invasive Surgery Program, Legacy Portland Hospitals, 2801 North Graham Avenue, Suite 120, Portland, OR 97227, USA Department of Surgery, Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, Portland, OR 97201, USA

Received: 5 July 1996/Accepted: 9 January 1997

Abstract Background: In spite of the emergence of laparoscopic cholecystectomy as the gold standard for treatment of symptomatic gallstones, questions still remain regarding its overall cost effectiveness, especially at low-volume centers where operating room (OR) time and operative complications are higher. We hypothesize that the presence of a well-organized, dedicated laparoscopic OR team will improve surgical outcomes for this procedure. This study compares the operative results of an advanced and a basic laparoscopic surgeon using either a designated laparoscopic operating team or a nondesignated team. Methods: The hospital records for 71 elective laparoscopic cholecystectomies with cholangiograms were retrospectively reviewed and anesthesia times and conversion rates were analyzed. Procedures were performed either at a hospital with a dedicated laparoscopy team or a hospital with nondedicated OR personnel. All procedures were done by an advanced laparoscopic surgeon or a basic laparoscopic surgeon. Results: Case characteristics were evenly matched between sites and surgeons. The mean total anesthesia time at the dedicated site was 120.8 min, compared to 152.3 min at the nondedicated site with a mean difference of 31.5 min (p 0.001). A 12% conversion rate was documented at the nondedicated site. There were no conversions at the site with a dedicated laparoscopy team. No major complications were encountered in this series. Conclusion: This study demonstrates that having a designated laparoscopic trained team provides a time savings to both advanced and basic laparoscopic surgeons. Although no major complications were encountered, there was a significant conversion rate for the less experienced surgeon operating without the support of a trained team. The end result from having a dedicated team in endoscopic surgery

is decreased operative time, an improvement in patient care, and decreased costs to the patient and institution. Key words: Laparoscopy Team Cholecystectomy Nursing Cost

Laparoscopic cholecystectomy (LC) is now widely regarded as the gold standard for the treatment of symptomatic cholelithiasis [3, 5]. Endoscopic surgery is increasingly the procedure of choice for many other operations as well. As laparoscopic surgeries become more complex, the operating room (OR) staff has added responsibilities for dealing with the intricate technical aspects of these advanced surgeries. It is widely accepted that particular specialties need trained staff members to function at their optimal level [4, 6]. Surgical special teams such as the Heart Team, Trauma Team, Pediatric Team, and others benefit from a designated staff which is focused and proficient in these areas [2]. At our center we have established a Laparoscopy Team trained in all aspects of assisting and circulating for endoscopic surgeries. We hypothesized that having such a dedicated team resulted in decreased operative times and fewer operative complicationsmeasurable indicators of cost effectiveness and improved patient care.

Methods
To document the presumed benefits of such a team, we reviewed two surgeons uncomplicated laparoscopic cholecystectomy (LC) experiences. The LC procedure was chosen because it was established and familiar to both the OR staffs and surgeons. OR times, conversion rates, and complications were collected from hospital records dated 1990 through 1993. Data was collected only on LC with intraoperative cholangiograms from two sites. Operations that involved more than a LC with cholangiograms or that were listed as emergent were not evaluated in order to minimize patient and surgical practice variations. Also, procedures that converted to open were not included in the overall OR time results. Two operating

Correspondence to: L.L. Swanstrom

813 rooms (alpha and beta), at different institutions but part of the same hospital system, were chosen for the review. As part of the same health care system similar equipment and instruments were used at both sites. Site alpha had a designated, trained laparoscopic team. The team members included anesthesia, nursing, OR tech, and support staff. All members were specifically chosen for their interest in laparoscopic procedures. Training of all team members included a syllabus and a 2-h didactic session on the physiology of pneumoperitoneum and potential complications of laparoscopy. Nursing staff and OR technicians went through an orientation period totaling 40 h where they worked under the supervision of an established team member. In addition they participated in a 1-day training course covering the indications and techniques of laparoscopic procedures, instrumentation, equipment trouble shooting, and OR safety. This course concluded with a hands-on animate lab which allowed the participants to act as assistant, camera person, and surgeon in common laparoscopic procedures. The beta site was staffed at similar levels by random OR personnel all familiar with LC. These nurses and OR techs had been inserviced on laparoscopy as part of their orientation. The one surgeon was considered an Advanced Laparoscopic Surgeon (>200 LC cases and routinely performing other advanced procedures) and the other surgeon was considered a Basic Laparoscopic Surgeon (<50 LC cases and not performing advanced procedures). Both surgeons were on active staff at both institutions. All patients had a general anesthetic administered by anesthesiologists belonging to the same practice group. Seventy-one cases met the criteria for review. Statistical significance was determined using the standard t-test and Student-Newman-Keuls tests.

Fig. 1. Y Surgeon, basic laparoscopic surgeon. X Surgeon, advanced laparoscopic surgeon. Combined, X and Y surgeons anesthesia times for beta and alpha sites. Alpha Site, designated trained laparoscopic team. Beta Site, randomly assigned OR team.

Results OR times, conversion rates, and operative complications were the main outcome parameters reviewed. There were no major intraoperative complications. Four cases were converted to open for obscure anatomy (two) or bleeding (two). These cases were not included in the analysis of operative times. Of the 71 reviewed cases, 27 cholecystectomies were performed at the alpha site and 44 cholecystectomies were done at the beta site. Thirty-five cases (23 at site alpha, 12 at site beta) were reviewed for the experienced surgeon and 36 cases (11 at site alpha, 25 at site beta) for the basic laparoscopic surgeon. Patient age, sex, and ASA class were not significantly different between the two groups. The mean total anesthesia time at the alpha site was 120.8 (6.6) min, compared to 152.3 (4.8) min at the beta site with a mean difference of 31.5 min (p 0.001). The basic surgeons mean total anesthesia time at the alpha site was 144.2 (11.7) min vs 175.7 (5.7) min at the beta site, for a mean difference of 31.5 min (p < 0.05) The advanced surgeons mean total anesthesia time at the alpha site was 97.5 (6.3) min vs 128.9 (7.7) min at the beta site, for a mean difference of 31.4 min (p < 0.05) (Fig. 1). There were no cases by either surgeon converted to open at the alpha site. There was, however, a 9% rate of conversion (four cases) for cases done at the beta site. All conversions were cases done by the basic surgeon. This surgeons conversion rate was therefore 14% at site beta (Table 1). Because of the relatively small numbers involved, the conversion rates between the sites and the surgeons are not statistically different as determined by the Fisher exact test (p > 0.11). The fact, however, that all of the conversions occurred at one site and with one surgeon is probably not a chance occurrence. A randomization test demonstrates that p 1/64; therefore this finding trends toward significance and requires an explanation. Conclusion Eight years after its introduction in the United States, laparoscopic cholecystectomy is available as a surgical option

Table 1. Laparoscopic to open conversion rates between site alpha and beta, and surgeon X and surgeon Ya Site Alpha Beta Surgeon X Y Overall Surgeon X + site alpha Surgeon X + site beta Surgeon Y + site alpha Surgeon Y + site beta 0/27 cases 4/40 cases 0/35 cases 4/36 cases 0 0 0 4 (14%) (0%) (10%) (0%) (9%)

Alpha site, designated trained laparoscopic team; Beta site, randomly assigned OR team; Y surgeon, basic laparoscopic surgeon; X surgeon, advanced laparoscopic surgeon

to almost all patients. In many hospitals, however, it is still not well integrated into OR practices or performed with maximum efficiency. Laparoscopy is increasingly being applied to the full spectrum of surgical procedures. At specialty centers, surgeons are routinely performing laparoscopic operations on the foregut, colon, liver, spleen, kidneys, and pancreas. Investigational laparoscopic procedures are being used for vascular and spine surgery as well. These complex laparoscopic procedures can take several hours at best and efficiency in the operating room may improve outcomes and certainly will reduce cost; results that are being closely looked at by patients and third-party payers. This study demonstrated a significant time savings during a basic procedure, laparoscopic cholecystectomy, of 31 min per case. This translated to a 26% reduction in anesthesia time for the advanced laparoscopic surgeon and an 18% reduction for the basic laparoscopic surgeon. At our institution a 31-min reduction results in a $651 savings (based on standard operating room and anesthesia charges of $21 per minute). One would expect that this cost-effectiveness benefit would be magnified in longer, more complex, and unfamiliar procedures. In addition, the risk of converting a laparoscopic procedure to an open procedure was significantly higher when the procedure was performed in a nondesignated setting. This is especially true early in a surgeons experience.

814

The converted cases in this study were excluded from the OR time analysis and also not listed as complications. Other reports, however, have shown that converted cases are both more costly and more prone to have complications [1, 7]. Patients and surgeons, for the most part, would certainly prefer that these procedures be accomplished laparoscopically if it can be done safely and effectively. Although this study was not constructed to directly measure them, we have found that a dedicated laparoscopic team has other advantages as well. Laparoscopic procedures require technically advanced equipment; correct setup and handling of this equipment decreases potential downtime as well as accidental injuries to patients and staff. Intraoperative complications are frequently the result of poorly choreographed team efforts, and cognizant and interested assistants can help recognize potentially missed injuries. Additionally, knowledgeable and appropriate handling of equipment decreases damage and prolongs instrument life. Having a team center on the details of the procedure with the knowledge to anticipate and troubleshoot has allowed our faculty surgeons to focus on the details of the operation, which results in fewer delays and shorter operative times. Based on this study we recommend increased OR staff training in laparoscopy and wider acceptance of dedicated laparoscopic OR teams. Although this training takes time

and effort, we believe that the end result will be a team that is more cost-effective in providing efficient and safer care for the increasing numbers of patients undergoing minimally invasive procedures.
Acknowledgment. Thanks to Louis Homer, Ph.D. for data review and statistical analysis.

References
1. Buanes T, Mjaland O (1996) Complications in laparoscopic and open cholecystectomy: a prospective comparative trial. Surg Laparosc Endosc 4: 266272 2. Huffman M (1995) Competency based oriention for perioperative cardiovascular nurses. AORN J 61: 722729 3. National Institutes of Health: Consensus development conference statement (1993) Gallstones and laparoscopic cholecystectomy. JAMA 269: 10181024 4. Paz-Partlow M, Berci G, Sackier JM (1992) Adequate trainingthe basis of a proficient operating room. J Surg Endosc 6: 105 5. Soper NJ, Stockman PT, Dunnegan DL (1992) Laparoscopic cholecystectomy: the new Gold Standard? Arch Surg 127: 917921 6. Springle AD, Snell WE, Boissoneau R (1993) Specialty surgical teams: results of a study. AORN J 58: 11701180 7. Traverso LW, Hargrave K (1995) A prospective cost analysis of laparoscopic cholecystectomy. Am J Surg 169: 503506

Surg Endosc (1997) 11: 809811

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Duration of postlaparoscopic pneumoperitoneum

K. Draper,1 R. Jefson,1 R. Jongeward, Jr.,2 M. McLeod1
1 2

Department of Surgery, Kalamazoo Center for Medical Studies, Michigan State University, 1000 Oakland Drive, Kalamazoo, MI 49008-1284, USA Department of Radiology, Kalamazoo Center for Medical Studies, Michigan State University, 1000 Oakland Drive, Kalamazoo, MI 49008-1284, USA

Received: 22 March 1996/Accepted: 12 July 1996

Abstract Background: Patients who present with abdominal pain after recent laparoscopic surgery present a diagnostic dilemma when pneumoperitoneum is present. Previous studies do not define the duration of postlaparoscopic pneumoperitoneum. In this study, we attempted to define the duration of laparoscopic pneumoperitoneum and to identify factors which affect resolution time. Methods: We followed 57 patients who underwent laparoscopic cholecystectomy (34), inguinal herniorraphy (20), or appendectomy (three). Serial abdominal films were taken until all residual gas was resolved. Results: Thirty patients resolved their pneumoperitoneum within 24 h; 16 patients resolved between 24 h and 3 days; nine patients resolved between 3 and 7 days; two patients resolved between 7 and 9 days. Mean resolution time for all patients was 2.6 2.1 days. There was no apparent difference in resolution time between the three types of procedures; however, the sample size may be insufficient. Duration of the pneumoperitoneum did not correlate with gender, age, weight, initial volume of CO2 used, length of time for the procedure, or postoperative complications. Sixteen patients had bile spillage during cholecystectomy which significantly reduced the duration of postoperative pneumoperitoneum (p < 0.008), resulting in a mean resolution time of 1.3 0.9 days. While 14 patients reported postoperative shoulder pain, no correlation was found between the presence or duration of shoulder pain and the extent or duration of pneumoperitoneum. Conclusions: We conclude that the residual pneumoperitoneum following laparoscopic surgery resolves within 3 days in 81% of patients and within 7 days in 96% of patients. The resolution time was significantly less in patients sustaining intraoperative bile spillage during cholecystectomy. There was no correlation found between postoperative shoulder pain and the presence or duration of the pneumoperitoneum.

Key words: Postoperative Laparoscopic Pneumoperitoneum

The possibility of a perforated intraabdominal viscus must be considered in any patient who presents with abdominal pain and is found to have pneumoperitoneum on an abdominal imaging study. However, patients who present with abdominal pain, free intraperitoneal air, and a recent history of having had a laparoscopic surgical procedure often present the surgeon with a diagnostic dilemma. This is especially true when there are no other clear indications for surgical exploration. Many patients retain a pneumoperitoneum of varying size following laparoscopic procedures during which carbon dioxide is routinely introduced into the peritoneal cavity to create an operative pneumoperitoneum that facilitates the completion of the planned procedure [2, 3, 5]. The presence of right shoulder pain in the postoperative period is commonly attributed to the presence of the subdiaphragmatic component of the residual pneumoperitoneum [1, 4]. Therefore, we studied 57 patients who underwent laparoscopic cholecystectomy, appendectomy, or inguinal herniorrhaphy in a community hospital setting in order to clarify the expected rate of absorption of the induced pneumoperitoneum associated with a routine laparoscopic procedure. Materials and methods
Fifty-nine consecutive patients who underwent either laparoscopic cholecystectomy or appendectomy or inguinal herniorrhaphy and signed informed consent to participate, between July 1, 1994, and July 31, 1995, were recruited into this study. No patient denied consent to participate in the study during this period, and all procedures were performed by one of the authors (R.J.) in either the inpatient or outpatient facilities of both Borgess Medical Center and Bronson Methodist Hospital in Kalamazoo, Michigan. This study was approved by the Human Use Committees of both hospitals. Two patients did not complete the entire postoperative radiological series and had to be deleted from the study and final analysis of data.

Correspondence to: R. Jefson, 1631 Gull Road, Suite 207, Kalamazoo, MI 49001, USA

810 Laparoscopic cholecystectomy was performed using the standard American technique. Four trocars were introduced and the gallbladder was removed after identifying the cystic duct and cystic artery. These structures were clipped and transected and the gallbladder was then dissected free in a retrograde manner from its attachments to the liver. Laparoscopic herniorrhaphy was performed using the transabdominal preperitoneal (TAPP) procedure. Three trocars were introduced. The peritoneum covering the inguinal canal and its contents was then opened and the hernia sac was identified and reduced. The posterior wall of the inguinal canal was reinforced by the application and stapled fixation of mesh. Laparoscopic appendectomy was performed by the placement of three trocars, identification of the appendix, and transection of the mesoappendix and appendiceal base using a laparoscopic GIA stapler. Either a Snowden-Pencer or Wolff laparoscopic carbon dioxide (CO2) insufflator was used to induce and maintain the operative pneumoperitoneum at a pressure of approximately 15 mmHg. The volume of CO2 used to establish the initial pneumoperitoneum as well as the total volume of CO2 used throughout the entire procedure was recorded in 43 patients. Richard-Allan Reflex STR disposable trocars were used in all procedures. The supraumbilical port was vented to room air using the flapper valve to permit the passive evacuation of the induced operative pneumoperitoneum (CO2) prior to removing this port. A single left lateral decubitus abdominal X-ray was obtained 24 h postoperatively in all patients. All films were reviewed by the same radiologist for the presence of pneumoperitoneum. If residual pneumoperitoneum was noted 24 h postoperatively, the patient was reevaluated on an outpatient basis with a follow-up left lateral decubitus film at 48-h intervals until no residual pneumoperitoneum was detected. Patient demographics were recorded preoperatively and included age, sex, race, height, and weight. We looked for a possible correlation between the rate of absorption of the postoperative or retained pneumoperitoneum and multiple variables. These variables included the type of procedure performed (herniorrhaphy vs cholecystectomy or appendectomy), the presence of acute vs chronic cholecystitis, the level of preoperative hemoglobin, a previous history of abdominal surgery, the duration of the procedure, the amount of CO2 used for the procedure, and the occurrence of intraoperative events such as any bile or gallstone spillage during cholecystectomy. Patients undergoing cholecystectomy were classified as having acute cholecystitis if they presented preoperatively with an elevated leukocyte count (a WBC >11.9), fever (oral temperature >101.5F), and right upper quadrant abdominal pain, in addition to their symptoms of biliary colic (postprandial pain, nausea, vomiting). To evaluate the correlation between postoperative shoulder pain and the presence of residual pneumoperitoneum, all patients were asked whether they experienced any shoulder pain, and if so, for how long. Data was evaluated using Pearson r correlation coefficients to identify relationships between continuous variables. Significant relationships were defined as those in which the correlation coefficient r value was >0.7 or <0.7. The effect of discrete variables (for example, the patients gender) on the persistence of the retained pneumoperitoneum was evaluated using the Students t-test. Comparisons of the duration of the retained pneumoperitoneum among the different procedures were evaluated using one-way ANOVA. Differences were considered significant when p 0.05.

Results Two patients were deleted from the original 59 patients entered into the study as described in Materials and Methods. Among the remaining 57 patients, 24 (42%) were male and 33 (58%) were female. The mean age for the entire group was 44 16 years, ranging from 17 to 77 years. Thirty-four patients (61% of study group) underwent laparoscopic cholecystectomy (LAC), 28 (82%) of whom were female. Eleven patients (32%) had intraoperative cholangiograms performed. Twenty patients (33% of study group) underwent laparoscopic herniorrhaphy (LAH), 18 (90%) of whom were male. Three patients undergoing laparoscopic herniorrhaphy underwent bilateral herniorraphies. Three patients underwent laparoscopic appendectomy (LAA). Forty-three patients were treated at Borgess Medical Center, and 14 patients were treated at Bronson Methodist

Hospital. Among the 34 patients who underwent LAC, eight (24%) patients had acute cholecystitis, and 26 (76%) patients had chronic cholecystitis using the criteria described in Materials and Methods. Intraoperatively, 16 (47%) of the patients who underwent LAC had the spillage of either bile or gallstones into the peritoneal cavity. In this operative series, five (8.8%) patients experienced any postoperative complication. Postoperative complications included one wound infection, one subhepatic fluid collection, two hematomas, and one episode of congestive heart failure. The two postoperative hematomas occurred in patients undergoing LAH and were located one each in the retroperitoneum and the soft tissues of the groin. No patients experienced a perforated viscus at any time during the study. The mean volume of CO2 required to establish an adequate operative pneumoperitoneum was 5.4 1.8 l, and the mean total volume of CO2 used for the entire procedure was 60.4 37.4 l per procedure. The amount of CO2 used to establish the operative pneumoperitoneum was not significantly different between male and female patients (p 0.6) or between patients with and without prior intraabdominal operations (p 0.9). There was no correlation found in this study between the weight of a patient and the amount of CO2 used to establish the initial pneumoperitoneum (Pearson r coefficient 0.370). Thirty (53%) patients resolved their induced pneumoperitoneum within 24 h; 16 (28%) patients resolved their pneumoperitoneum between 24 h and 3 days; nine (16%) patients resolved their pneumoperitoneum between 3 and 7 days; and the remaining two (3%) patients had persistence of their postoperative pneumoperitoneum for 7 to 9 days. Comparisons of the continuous variables of age, height, weight, preoperative hemoglobin, history of abdominal surgery, and the amount of CO2 used to the duration of postoperative pneumoperitoneum revealed no significant relationships. Among the laparoscopic procedures performed in this study, there was no significant difference observed among the mean durations of postoperative pneumoperitoneum in that the mean duration of postoperative pneumoperitoneum was 2.3 2 days, 3.1 2.3 days, and 2.3 1.2 for LAC, LAH, and LAA, respectively (p 0.3). However, the mean duration of postoperative pneumoperitoneum was significantly shortened by the intraoperative occurrence of bile or gallstone spillage in that the mean duration of postoperative pneumoperitoneum without intraoperative spillage of either bile or gallstones was 3.1 2.4 days whereas the mean duration of postoperative pneumoperitoneum following the intraoperative spillage of either bile or gallstones was 1.3 0.9 days (p 0.008). Fourteen (25%) patients experienced postoperative shoulder pain. The presence of postoperative shoulder pain was not significantly related to the amount of CO2 used, either that used to initially induce an operative pneumoperitoneum (p 0.7) or the total amount of CO2 used throughout the procedure (p 0.3). Further, the duration of shoulder pain did not correlate with the duration of postoperative pneumoperitoneum (Pearson r coefficient 0.137). Six patients experienced postoperative shoulder pain for a longer time than their postoperative pneumoperitoneum lasted.

811

Discussion There are few available published data on the rate of absorption and disappearance of the residual carbon dioxide in the peritoneal cavity following laparoscopic procedures. McAllister et al. reported on 27 patients who were evaluated by computerized axial tomography of the abdomen within 24 h following laparoscopic cholecystectomy [2]. In this series, 19 (70.4%) patients demonstrated residual postoperative pneumoperitoneum at 24 h. In contrast, Smith et al., reported on 30 patients who were evaluated by plain abdominal roentgenograms for postoperative pneumoperitoneum 24 h following laparoscopic cholecystectomy [5]. In this second series, only three (10%) patients were found to have residual pneumoperitoneum persisting after 24 h. In both of the above studies, no further evaluations were attempted beyond 24 h. It is unclear from these studies what the expected rate of absorption of retained carbon dioxide should be following a routine laparoscopic procedure. Further, a more recent study reported on 50 patients who were evaluated by serial plain abdominal roentgenograms for 2 weeks following laparoscopic cholecystectomy. In this series, Millitz et al. observed that 27 (54%) patients absorbed all retained postoperative pneumoperitoneum by 6 h. All but one among the remaining 23 patients cleared their residual postoperative pneumoperitoneum within 1 week of their procedure [3]. In our study, 81% of our patients resorbed their residual postoperative pneumoperitoneum within 3 days, and 97% resorbed their residual postoperative pneumoperitoneum within 7 days following laparoscopic surgery. The remaining 3% of patients in our present study required up to 9 days to completely resorb their residual postoperative pneumoperitoneum. The type of laparoscopic procedure performed did not appear to significantly influence the rate of resolution of the postoperative residual pneumoperitoneum. The mean duration of the pneumoperitoneum following herniorrhaphy appears slightly increased at 3.1 2.3 days; however, the mean duration of the pneumoperitoneum in the cholecystectomy patients without bile or gallstone spillage is almost exactly the same at 3.1 2.4 days. The number of laparoscopic appendectomies in the study is too small (n 3) to be of statistical significance. It remains possible that a larger study with a more rigorous experimental design may establish a real difference in these rates. Millitz et al. observed an inverse correlation between the patients body weight and the duration of the residual postoperative pneumoperitoneum [3]. Our present study did not confirm this finding, nor did it identify any significant correlation between the amount of carbon dioxide needed to create an operative pneumoperitoneum, the amount used to

complete the procedure, or the rate of resorption of the postoperative pneumoperitoneum and the age, sex, height, race, preoperative hemoglobin level, or the occurrence of a prior intraabdominal surgical procedure. The rate of resorption of the postoperative pneumoperitoneum was significantly influenced by the spillage of bile alone or gallstones with the associated bile during laparoscopic cholecystectomy. Patients sustaining the spillage of bile or gallstones with bile during laparoscopic cholecystectomy resolved their postoperative pneumoperitoneum faster than those who did not experience bile or gallstone spillage. It is possible that bile has a potentiating or stimulating effect on carbon dioxide absorption by the peritoneal epithelium. However, the basic mechanism that underlies this finding remains to be elucidated. Finally, the presence of postoperative shoulder pain following a laparoscopic procedure has been commonly attributed to the presence of residual carbon dioxide [4, 5]. Our study failed to show any correlation between the subjective presence and duration of shoulder pain with the duration of postoperative pneumoperitoneum. Although the sample size was small, with only 14 (25%) of 57 patients reporting postoperative shoulder pain, six (43%) of 14 reported the persistance of shoulder pain after the resolution of their residual pneumoperitoneum. It remains possible that the sensitivity of the method used in this study to document pneumoperitoneum may have been inadequate to identify a small, obscure residual pneumoperitoneum capable of causing symptoms despite its size.
Acknowledgment. We thank the Richard-Allan Corporation, Richland, Michigan, for providing the laparoscopic trocars used in this study. This study was funded in part by the Bronson Clinical Investigational Unit Community Research Fund. We thank Marian Cowley for assistance with the preparation of the manuscript.

References
1. Fredman B, Jedeikin R, Olsfanger D, Flor P, Gruzman A (1994) Residual pneumoperitoneum: a cause of postoperative pain after laparoscopic cholecystectomy. Anesth Analg 79: 152154 2. McAllister JD, DAltorio RA, Snyder A (1991) CT findings after uncomplicated percutaneous laparoscopic cholecystectomy. J Comput Assist Tomogr 15(5): 770772 3. Millitz K, Moote DJ, Sparrow RK, Girotti MJ, Holliday RL, McLarty TD (1994) Pneumoperitoneum after laparoscopic cholecystectomy: frequency and duration as seen on upright chest radiographs. AJR Am J Roentgenol 163: 837839 4. Perry PC, Tombrello R (1993) Effect of fluid installation of postlaparoscopy pain. J Reprod Med 38(10): 768770 5. Smith R, Kolyn D, Pymar H, Sauerbrie E, Pace RI (1992) Ultrasonographic and radiological evaluation of patients after laparoscopic cholecystectomy. Can J Surg 35(1): 5558

Surg Endosc (1997) 11: 879

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The author replies

We appreciate the valuable comments of Dr. Th. W. Fengler. Indeed, we had to restrict ourselves to a retrospective cost analysis, as the study was performed in the framework of the Masters thesis of the first author (L. Demoulin), which meant substantial limitations on the available time and budget. The main advantage of a prospective cost analysis, as performed by Dr. Fengler, is that the instrument lifetime is not assumed and the actual processing and maintenance costs can be calculated. It is important to note, however, that the results go in the same direction. It would be interesting to have the precise results and costing data of Dr. Fenglers research, which would enable a full comparison. Furthermore, we fully agree that a prospective analysis of the safety (e.g., infection risk) and functionality (e.g., user friendliness) of the various types of instruments would add value to the results already available in the literature. We therefore look forward to the publication of the results of Dr. Fenglers new study with great interest. L. Demoulin K. Kesteloot F. Penninckx
Faculty of Medicine Centre for Health Services and Nursing Research Kapucijnenvoer 35 B-3000 Leuven Belgium

Case reports
Surg Endosc (1997) 11: 846847

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Umbilical sinus tract with gallstone extrusion after laparoscopic cholecystectomy

D. Pohl,1 P. J. Milman,2 H. Rothenberg3
1 2 3

Department of Surgery, Flushing Hospital Medical Center, 4500 Parsons Boulevard, Flushing, NY 11355, USA Department of Gastroenterology, Long Island Jewish Medical Center, 270-07 76th Avenue, New Hyde Park, NY 11040, USA Department of Surgery, Long Island Jewish Medical Center, 270-07 76th Avenue, New Hyde Park, NY 11040 USA

Received: 14 August 1995/Accepted: 20 October 1995

Abstract. There are an increasing number of reports concerning complications after laparoscopic cholecystectomy. We report a case of persistent elevation of serum alkaline phosphatase and spillage of several gallstones through an abdominocutaneous sinus tract to the umbilicus with spontaneous resolution. Key words: Cholecystectomy Laparoscopic Complications

Laparoscopic cholecystectomy has become the preferred surgical treatment for symptomatic cholelithiasis and cholecystitis. It has been shown to be safe, efficient, and costeffective [18]. As a relatively new procedure, all of the complications are not yet known. Gallbladder perforation and spillage of bile or stones are common secondary to manipulation of the gallbladder during dissection and extraction [14, 15, 21]. An increasing number of complications due to retained gallstones are reported in the literature. We report a case of an abdominocutaneous sinus tract to the umbilicus and expulsion of several gallstones.

Case report
An 82-year-old female with a past surgical history of appendectomy and oophorectomy presented with indigestion for several weeks and with acute onset of right upper quadrant (RUQ) and epigastric pain, nausea, and vomiting. Temperature on admission was 39.5C. She was jaundiced. The abdomen was tender in the RUQ. Pertinent lab values were: WBC 15,000/ mm3, serum bilirubin 2.5 mg/dl, alkaline phosphatase (ALP) 210 IU/l, AST 121 IU/l. The ultrasound showed cholelithiasis, and the biliary tree was not dilated. On hepatobiliary scan, the gallbladder and intestine were not visualized after 24 h. Endoscopic retrograde cholangiopancreatography (ERCP) demonstrated cholelithiasis (multiple), choledocholithiasis, and a

periampullary diverticulum. A sphincterotomy was performed, and one large stone was extracted from the common bile duct. The patient was started on ampicillin/sulbactam. She became afebrile and pain free. Laparoscopic cholecystectomy was performed 2 days later. The gallbladder appeared acutely and chronically inflamed with adhesions. No gross bile or stone spillage was noted. The gallbladder was extracted through the umbilical port. Closure of the fascia was done under direct vision. Pathology revealed acute and chronic cholecystitis, a 0.6-cm hole in the wall at the fundus, and multiple brown stones. The patient was discharged. One week later, she was readmitted for persistent abdominal pain, tenderness, and low-grade fever. The wounds were healed. Computed tomography (CT) showed retained gallstones under the anterior abdominal muscles, in the RUQ, and in the pelvis. She became asymptomatic with antibiotics and was discharged. Three weeks after surgery, on a follow-up visit, her lab values were WBC 21,000/mm3, bilirubin 2.1 mg/dl, ALP 240 IU/l. ERCP showed choledocholithiasis, and the common bile duct was not dilated. A pigmented stone was removed and a biliary stent was placed. The WBC and bilirubin returned to normal but the ALP and the GGT remained elevated. Repeat ERCP revealed that the stent had passed spontaneously. There were no residual common bile duct stones. Four months after surgery she reported drainage of blood, pus, and stones from her umbilicus. A repeat CT scan demonstrated the same retained gallstones under the abdominal wall in the area of the umbilicus, now with inflammatory reaction around them and thickening around the umbilicus probably as a result of a sinus tract. During the next 2 months the drainage of pigmented stones persisted. One large stone was removed in the office. The patient had no systemic symptoms and no other complaints. After 3 months, the drainage resolved spontaneously and the sinus tract closed. ALP and GGT remained elevated. Mitochondrial antibodies were 1:160 (normal < 1:20). A presumptive diagnosis of primary biliary cirrhosis was made. The patient refused liver biopsy and treatment. She remains asymptomatic.

Discussion This report of an abdominocutaneous sinus tract with spillage of pigmented gallstones demonstrates only one form of complication after intraabdominal spillage of gallstones. Most of the reports in the literature are anecdotal. There is only one prospective study [14]. The rate of bile and gallstone spillage is very low (<1%) in some studies [9, 10, 18] but high in other surveys (>30%) [1, 14]. Spillage occurs as a result of traction tears, coagulation injuries, insufficient

Correspondence to: D. Pohl

847

clip application to the cystic duct, or perforation during forceful extraction through the umbilicus [3, 7, 14]. In the United States, 70% of gallstones are made of cholesterol and 30% contain bilirubin pigment. According to most reports, cholesterol stones are sterile whereas pigment stones can contain bacteria [7, 13, 16]. In case of spillage, cultures should be taken and the patient should be kept on appropriate antibiotics postoperatively [3, 14, 15]. Most experiments in animals have shown that pigmented gallstones implanted in the abdomen or the abdominal wall cause adhesions but no infection [4, 5, 13, 20, 21]. Only Johnston found an increased number of abscesses [8]. Clinically, lost stones were incidentally the cause of small-bowel obstruction [6], subcutaneous abscesses [6, 17, 21, 22], intraabdominal abscesses [3, 11, 19, 23], and enteric and cutaneous fistulas [2, 7, 15]. Stones were found subdiaphragmatic, subhepatic, intrapelvic, and intrathoracic. In our patient, a leak was not observed intraoperatively and probably occurred during extraction. Several stones were detected by CT scan in the subhepatic space, in the area of the umbilicus and in the pelvis. Only the periumbilical stones caused a problem. The other stones remained silent. Choledocholithiasis and possibly primary biliary cirrhosis made this case more complex. However, they were not related to the complication, which could have occurred in simple cholecystitis. The recommendation is to observe closely for gallbladder perforation and close tears when possible with an endoclip, endoloop, or a suture and to remove the gallbladder in a plastic bag. If a retrieval bag is unavailable, the umbilical trocar site incision should be enlarged to facilitate the forceless removal of the gallbladder without risking a tear or spillage of gallstones. In case stones are lost, an attempt should be made to retrieve all gallstones by suctioning, grasping, or using a Dormia basket. The area should be irrigated extensively [3, 7, 8, 12, 14, 17, 20]. Conclusion Laparoscopic cholecystectomy is a safe procedure. With an increasing number of reports about complications secondary to lost gallstones, the surgeon must be very careful in handling the gallbladder during dissection and extraction. References
1. Bernard HR, Hartman TW (1993) Complications after laparoscopic cholecystectomy. Am J Surg 165: 533535

2. Cappucino H, Cargill S, Nguyen T (1994) Laparoscopic cholecystectomy: 563 cases at a community teaching hospital and a review of 12201 cases in the literature. Surg Laparosc Endosc 4: 213221 3. Catarci M, Zaraca F, Scaccia M, Carboni M (1993) Lost intraperitoneal stones after laparoscopic cholecystectomy: harmless sequela or reason for reoperation? Surg Laparosc Endosc 3: 318322 4. Cline RW, Poulos E, Clifford EJ (1994) An assessment of potential complications caused by intraperitoneal gallstones. Am Surg 60: 303 305 5. Cohen RV, Pereira PRB, Barros MV, Ferreira EAB, Tolosa EM (1994) Is the retrieval of lost peritoneal gallstones worthwhile? Surg Endosc 8: 1360 6. Dittrich K, Weiss H (1995) Duenndarmileus durch einen verlorenen Gallenstein. Chirurg 66: 443445 7. Golub R, Nwogu C, Cantu R, Stein H (1994) Gallstone shrapnel contamination during laparoscopic cholecystectomy. Surg Endosc 8: 898 900 8. Johnston S, OMalley K, McEntee G, Grace P, Smyth E, BouchierHayes D (1994) The need to retrieve the dropped stone during laparoscopic cholecystectomy. Am J Surg 167: 608610 9. Larson GM, Vitale GC, Casey J, Evans JS, Gilliam G, Heuser L, McGee G, Rao M, Scherm MJ, Voyles R (1992) Multipractice analysis of laparoscopic cholecystectomy in 1983 patients. Am J Surg 163: 221226 10. Lee VS, Chari RS, Cucchiaro G, Meyers WC (1993) Complications of laparoscopic cholecystectomy. Am J Surg 165: 527532 11. Leslie KA, Rankin RN, Duff JH (1994) Lost gallstones during laparoscopic cholecystectomy: are they really benign? CJS 37: 240242 12. SAGES (1994) Guidelines for the clinical application of laparoscopic biliary tract surgery. Surg Endosc 8: 14571458 13. Sax HC, Adams JT (1993) The fate of the spilled gallstone. Arch Surg 128: 469 14. Soper NJ, Dunnegan DL (1991) Does intraoperative gallbladder perforation influence the early outcome of laparoscopic cholecystectomy? Surg Laparosc Endosc 1: 156161 15. Steerman PH (1994) Delayed peritoneal-cutaneous sinus from unretrieved gallstones. Surg Laparosc Endosc 4: 452453 16. Stewart L, Smith AL, Pellegrini CA, Motson RW, Way LW (1987) Pigment gallstones form as a composite of bacterial microcolonies and pigment solids. Ann Surg 206: 242249 17. Targarona EM, Balague C, Cifuentes A, Mart nez J, Tr as M (1995) The spilled stone. Surg Endosc 9: 768773 18. The Southern Surgeons Club (1991) A prospective analysis of 1518 laparoscopic cholecystectomies. N Engl J Med 324: 10731078 19. Van Brunt PH, Lansafame RJ (1994) Subhepatic inflammatory mass after laparoscopic cholecystectomy. Arch Surg 129: 882883 20. Welch NT, Hinder RA, Ciurej T, Bacon N (1991) Laparoscopic capture of escaped gallstones. Surg Laparosc Endosc 1: 4244 21. Welch N, Hinder RA, Fitzgibbons RJ, Rouse JW (1991) Gallstones in the peritoneal cavity. A clinical and experimental study. Surg Laparosc Endosc 1: 246247 22. Wetscher G, Schwab F, Fend F, Glaser K, Ladurner D, Bodner E (1994) Subcutaneous abscess due to gallstones lost during laparoscopic cholecystectomy. Endoscopy 26: 324325 23. Wilton PB, Andy OJ, Peters JJ, Thomas CF, Patel VS, Scott-Conner CEH (1993) Laparoscopic cholecystectomy. Leave no stone unturned. Surg Endosc 7: 537538

Surg Endosc (1997) 11: 842845

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Mirizzis syndrome
Diagnostic and therapeutic controversies in the laparoscopic era
E. M. Targarona, E. Andrade, C. Balague , J. Ardid, M. Tr as
Service of General and Digestive Surgery, Hospital Clinic, University of Barcelona, Villarroel 170, 08036 Barcelona, Spain Received: 18 September 1996/Accepted: 3 December 1996

Abstract Background: Mirizzis syndrome (MS) is an unusual cause of obstructive jaundice. It can mimic bile duct cancer, and the role of laparoscopic surgery is not well defined. The aim of this paper is to report five cases and describe the pitfalls encountered in its diagnosis and treatment with a laparoscopic approach. Methods: From January 1992 to January 1996, five cases of MS out of 560 patients with gallstones prospectively treated and recorded were found (0.9%). Results: There were two men and three women, (mean age: 54 years [3093]). In one case diagnosis of bile duct carcinoma was established but surgery revealed MS. Four cases were approached by laparoscopy, but all of them were converted: in two, due to a distorted anatomy, in a third due in the difficulty of visualizing the distal end of the bile duct, and in the last case due to the impossibility of retrieving the stones. All were treated with a cholecochorrhaphy over a T tube, except one, in which a hepaticojejunostomy was performed. Morbidity and mortality were nil, and they remain asymptomatic after a mean follow-up of 19 months (336). Conclusions: MS constitutes an important laparoscopic challenge, both to clearance of duct stones and to the proper reconstruction of the biliary duct. A prudent policy is to perform a dissection trial and convert if local conditions are not clear for an experienced laparoscopic surgeon. Key words: Bile duct stones Laparoscopy Cholecystocholedochal fistula

dice. Its definition has been revised from the initial description by Mirizzi [5], who defined it as obstructive jaundice consecutive to existence of a gallstone that compresses the bile duct; it has been extended to cases in which there is an erosion of the gallbladder wall or the cystic duct so that it communicates with the bile duct, forming a cholecystocholedochal fistula [3, 6]. Endoscopic retrograde cholangiography (ERCP) is the most useful diagnostic tool, but it usually fails to resolve this situation; surgical therapy may be required, but can sometimes be a difficult procedure [1, 4, 12]. Surgical therapy of biliary lithiasis has been modified since the introduction of laparoscopic surgery, but the role of this technical option in Mirizzis syndrome is not well defined. The aim of this paper is to report five cases recently observed and comment on the diagnostic and therapeutic difficulties that can present. Materials and methods
From January 1992 to March 1996, following the introduction of laparoscopic surgery in our hospital, 560 patients with biliary lithiasis were treated consecutively and clinical data were prospectively recorded. Four hundred sixty were operated on by laparoscopy and 100 suspected of harboring bile duct stones were randomized to open surgery or endoscopic therapy with sphincterotomy in a prospective trial [11]. Five out of these 560 (0.9%) patients presented a clinicoradiological picture compatible with Mirizzis syndrome.

Results The diagnosis of Mirizzis syndrome was established after the demonstration of a compression or obstruction of the bile duct secondary to stones located in the gallbladder neck or cystic duct. There were two men and three women with a mean age of 54 years (3093). All patients developed jaundice (mean Bi: 9.5 g/l (8.710.5), with cholangitis in two cases and asthenia and weight loss in one. Ultrasonography showed an atrophic gallbladder with stones in four cases, and in one case the gallbladder was not identified, suggesting the existence of a bile duct tumor. In one case

Mirizzis syndrome is an unusual clinicopathological manifestation of biliary lithiasis and a cause of obstructive jaunPresented as a poster presentation in the annual meeting of the Society of American Gastrointestinal Endoscopic Surgeons (SAGES), San Diego, California, USA, 2122 March, 1997 Correspondence to: M. Tr as

843

Fig. 2. Csendes type III Mirizzis syndrome. The gallbladder has disappeared and communicates with the common duct, which is filled with stones.

Fig. 1. A large stone (6 cm) located below erodes the bile duct (Csendes type III) and compresses all the elements of the hepatic hilum, which induced a profuse hemmorrhage during surgery.

ultrasound (US) showed a 6 4 cm (Fig. 1) stone located in the right hypocondrium and an enlarged spleen with collateral circulation in the hepatic hilum. The portal vein was not visualized. Dilated bile duct was observed in all cases. Computed tomography (CT) scan was used in one case, in which the suspicion of a Klatskin tumor was established. ERCP was performed in all cases, and this exploration established the diagnosis of Mirizzis syndrome in all cases but one, which was thought to be a bile duct cancer. All patients diagnosed with Mirizzis syndrome were referred for surgery after ERCP. The patient diagnosed with bile duct cancer was considered unsuitable for surgery and he was proposed for percutaneous stenting. After a failed attempt of intubation the patient presented abdominal pain and fever and was reevaluated by the surgical team; elective surgery was proposed. During surgery a scleroatrophic gallbladder with a stone eroding the common hepatic duct was observed. After cholecystectomy, and due the wide wall defect, a hepaticojejunostomy was constructed. Perioperative and deferred histological study showed inflammatory changes without signs of malignancy. The other four patients were approached by laparoscopy. In one case, a normal liver was observed, but it was converted because of significant adhesions and profuse bleeding during dissection of the liver edge (Fig. 1). Dissection and identification of the gallbladder were complicated by adhesions and profuse hemmorrhage due to collateral circulation, which required transfusion of 4 units of

blood. The hemmorrhage stopped when the gallbladder was opened and the stone was retrieved. The other three cases were converted to open surgery, in two due the impossibility of retrieving all the stones located and impacted in the cholecystocholedochal fistula (Fig. 2), and in one case due to the impossibility of finding and locating properly the distal arm of the T tube on the distal bile duct. In all cases a subtotal cholecystectomy was performed and cholecystocholedochal fistula was repaired with a choledochorraphy over the T tube. All patients evolved satisfactorily without morbidity and were discharged 12 days after surgery (range 621). T-tube control cholangiography was performed later and the T tube was withdrawn within 3 months in all cases. The patients remain asymptomatic. Definitive histologic study showed severe chronic cholecystitis in all cases.

Discussion Mirizzis syndrome was first described by Mirizzi in 1984: In the initial description it was defined as obstruction of the proximal bile duct secondary to external compression by a large stone located in the Hartmann pouch or secondary to local inflammatory changes [5]. In recent years, extension of Mirizzis eponym has been used to define obstructive jaundice induced by different grades of compression and erosion of the bile duct wall by the Hartmann pouch, which can evolve to a complete cholecystocholedochal fistula, with compression and dilation of the proximal bile duct by the stone. Several authors have attempted to classify Mirizzis syndrome. McSherry et al. [6] proposed a type I in which a fistula is not present and type II when a communication between the gallbladder and the bile duct is present. Csendes et al. [3], after the analysis of a series of 219 cases,

844 Table 1. Laparoscopic treatment of Mirizzis syndrome; literature review Author Rust [10] Paul [8] Binnie [2] Meng [7] Posta [9] Targarona [11] Year 1991 1992 1992 1995 1995 1996 N 1 1 1 1 1 4 Mirizzi type McSherry I, Csendes I McSherry I, Csendes I McSherry II, Csendes II McSherry I, Csendes II McSherry II, Csendes II McSherry II, Csendes III Treatment Conversion Transcystic choledocoscopic extraction of stone and lap cholecystectomy Endoscopic stenting + lap cholecystectomy + T tube Peroperative sonographic identification of cystic stone, lap cholecystectomy and T tube Conversion Conversion and choledochorraphy over T tube

defined four types or stages of the same disease. In Csendes classification type I involves external compression alone; type II involves erosion of one-third of the circumference of the common bile duct; in type III, up to two-thirds of the common bile duct is eroded; and type IV is the complete destruction of the bile duct. The incidence of this clinicopathology is low, under 1% in most series of cholecystectomies. These five cases presented with jaundice and were diagnosed by ERCP. Exact diagnosis of obstructive lesions in the bile duct bifurcation may be difficult, and it is not exceptional for benign lesions to mimic malignant conditions or vice versa. Exhaustive workup, including surgical exploration, should be performed before the lesion is considered malignant and unresectable, especially when stones are visualized in the tumor [13]. In selected cases, therapeutic flexible endoscopy may succeed in retrieving the stones [1, 11, 12], and the use of electrohydraulic lithotripsy and stone retrieval with the aid of a baby scope through the cystic duct has been proposed. In most cases it is not possible to resolve this situation definitively due to the size of the stone and impaction in the bile duct, but palliative intubation can be accomplished in order to decompress the obstructed biliary tree. One case in this series developed profuse intraoperative hemorrhage as a consequence of collateral circulation, probably due to prehapatic portal hypertension as a result of compression of a large stone. Although ultrasonography showed signs of portal hypertension, it was not considered during the workup of the patient. After extraction of the stone the hemorrhage ceased and postoperative ultrasonography showed a patent portal vein. Extrahepatic compression of the portal vein in adults is a rare etiology of portal hypertension, but the size of the stone and its anatomic position could explain vein compression. It was an associated difficulty during the operation, which should be considered in cases in which a large stone is located across the bile duct. Finally, the role of laparoscopic surgery in treatment of this pathology is not well defined and is controversial [2, 710] (Table 1). Four of the five cases presented in this series were approached by laparoscopy, but all of them were converted: in one case due to distorted anatomy and local venous hypertension, in a second due to difficulty in visualizing the distal end of the bile duct for introduction of the distal arm of the T tube, and in the other two cases, due to

impossibility of retrieving all the impacted stones in the cholecystocholedochal fistula. Five recent papers [2, 710] described finding Mirizzis syndrome during laparoscopic cholecystectomy. Two of them [9, 10] considered it a formal contraindication for laparoscopic surgery and those were converted, but Paul et al. [8] successfully treated one patient with a McSherry and Csendes type I Mirizzis syndrome. In the case of Paul et al. [8], there was no biliobiliary fistula, and the stone located in the cystic duct that compressed the bile duct was retrieved proximally through the cystic duct with a basket catheter. Meng et al. [7] also resolved a McSherrys type I case, aided by intraoperative ultrasound that facilitated the identification of the stone in order to avoid bile duct injury. Binnie et al. [2] reported the successful treatment of a Mirizzi type II (McSherry), associated to preoperative endoscopic stenting. Visualization of the stent facilitated the identification of the lumen of the bile duct and the placement of a T tube. We do not think that Mirizzis syndrome should be considered a formal contraindication for the laparoscopic approach, but it constitutes a great technical challenge, both for a complete clearance of duct stones and because a high degree of expertise is required to assure that the biliary duct is properly reconstructed. A prudent policy is to perform a dissection trial and convert if local conditions are not clear for an experienced laparoscopic surgeon.

References
1. Binmoeller KF, Thonke F, Soehendra N (1993) Endoscopic treatment of Mirizzis syndrome. Gastroint Endosc 39: 532536 2. Binnie NR, Nixon SJ, Palmer KR (1992) Mirizzi syndrome managed by endoscopic stenting and laparoscopic cholecystectomy. Br J Surg 79: 647 3. Csendes A, Carlos J, Burdiles P, Maluenda F, Nava O (1989) Mirizzi syndrome and cholecystobiliary fistula: a unifying classification. Br J Surg 76: 11391143 4. Delcenseries R, Joly JP, Dupas JL (1992) Endoscopic diagnosis and treatment of Mirizzis syndrome. J Clin Gastroenterol 15: 343346 5. Mirizzi PL (1948) S ndrome del conducto hepa tico. J Int Chir 8: 731.0 6. McSherry CK, Ferstenberg H, Virshup M (1982) The Mirizzi syndrome: suggested classification and surgical therapy. Surg Gastroenterol 1: 219225 7. Meng WCS, Kwok SPY, Kely SB, Lau WY, Li AKC (1995) Management of Mirizzi syndrome by laparoscopic cholecystectomy and laparoscopic ultrasonography. Br J Surg 82: 396 8. Paul MG, Burris DG, McGuire AM, Thorfinnson HD, Schonekas H

845 (1992) Laparoscopic surgery in the treatment of Mirizzis syndrome. J Laparoendosc Surg 2: 157163 9. Posta ZG (1995) Unexpected Mirizzi anatomy. A major hazard to the common bile duct during laparoscopic cholecystectomy. Surg Endosc 5: 412414 10. Rust KR, Clancy TV, Warren, Merstendorf J, Maxwell JG Mirizzis syndrome: a contraindication to coeliscopic cholecystectomy. J Laparoendosc Surg 1: 133137 11. Targarona EM, Perez Ayuso RM, Bordas JM, Pros I, Martinez J, Ros E, et al. (1996) Randomised trial of endoscopic sphincterotomy with gallbladder in situ versus open surgery for common bile duct calculi in high-risk patient. Lancet 347: 926929 12. Walker JM, Karim BF (1982) Carcinoma of the cystic duct mimicking the Mirizzi syndrome. Am J Gastroenterol 77: 936938 13. Weiss SL, Pupols AZ, Starling JR, Gould HR (1986) Mirizzi syndrome simulating a tumor by ERC. Dig Dis Sci 31: 100102

Editorial
Surg Endosc (1997) 11: 797799

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic surgery for cure of colorectal cancer

One good reason for considering a change of access to radical surgery for colorectal cancer should be the proven superiority of the new approach to the treatment of benign diseases of the same viscera. For the time being this does not seem to be the case for laparoscopic colorectal surgery [8]. In fact, there is so far no convincing evidence showing that minimal invasive colorectal surgery is advantageous over its traditional counterpart with regard to the immediately recognizable outcome [22]. On the other hand, the immediately recognizable drawbacks of laparoscopic colorectal surgery (learning curve, prolonged operating time, high morbidity rates following conversion to open surgery, and the new complications) are well documented [24, 25]. In spite of this, an increasing number of surgeons are today attempting to perform laparoscopic surgery for cure of colorectal cancer outside randomized controlled trials. It is therefore appropriate to review the cancer issues that may make laparoscopic colorectal surgery for cure of malignancy an unsafe procedure. colon tumors. A restricted ability to pack away the small bowel can interfere with this requirement. Laparoscopy-induced disease progression? The laparoscopic approach is suspected of being responsible for previously unknown tumor spreading. Anecdotal reports support an increased rate of abdominal wound metastases. The overall incidence cannot be determined given that the total number of patients undergoing such procedures is, regrettably, unknown. The American Society of Colon and Rectal Surgeons Laparoscopic Registry has reported a 1.1% wound recurrence rate at 1-year follow-up for 480 patients [26]. As many operations have been performed during the last 2 years, there is reason to fear that the true incidence might be over 4%. Most recurrences (80%) have been reported to occur within 1 year (range 126 months) at port sites which had not been used for specimen retrieval even when the specimen was removed in a plastic bag [27]. Only 63% of these patients had advanced disease at the time of laparoscopy. Four patients (9%) were reported as Dukes A adenocarcinomas, which is hardly comprehensible unless understaging or perforation occurred. More than 50% of abdominal wound metastases were associated with peritoneal carcinomatosis. Over 80% of these patients with peritoneal tumor dissemination had Dukes C cancers at the time of laparoscopy and died within 1 year of the date of recurrence diagnosis. Two patients (17%) with Dukes B colon tumors developed abdominal wound metastases with peritoneal carcinomatosis (9 and 10 months after laparoscopy), which were treated by cytoreductive surgery and intraperitoneal chemotherapy [15]. Experimental data [19] support a significant increase in abdominal wound metastasis rates after laparoscopic resection compared with open excision. It is still unclear whether metastatic mechanisms such as viable cancer cells directly implanted by increased contact with tumor-laden instruments, cannulas, stability threads [2], specimen and/or vehiculated by carbon dioxide within vapor particles that condense [10] and/or transported by cell-laden fluid at the time desufflation [28] account for all wound metastases. Other experimental data [4, 13, 14] are controversial as to whether pneumoperitoneum enhances spread, implantation, and growth of free, viable cancer cells on ischemic and traumatized sites of visceral and/or parietal peritoneum. Interestingly, some of these data [4] have pointed out that the topography of carcinomatosis depends on port placement.

Disease localization and demarcation View magnification and the loss of the palpating hand have considerable consequences for disease localization and demarcation. Colon cancer can be intraluminal and/or intramural and preoperative tatooing or metal-clip application is mandatory for intraoperative lesion identification (with fluoroscopy in the case of clips). In case of failure, laparoscopic ultrasound can facilitate the localization of early disease and the evaluation of adequate resection margins. Perioperative colonoscopy may, however, be necessary in case of extremely mobile large villous adenomas for exact tumor location. All intraoperative diagnostic methods share the disadvantage of a further prolonged operating time. Failure to identify the lesion before resection leads to increased manipulation of the tumor-bearing bowel segment (prior to proximal vessel division) or resection of a sound colon segment. Disease demarcation is a critical issue, too, as T4 cancers require open en bloc surgery, as do transverse

Correspondence to: R. Bergamaschi, Institute of Surgery, University of Bergen, Diakonissehjemmets University Hospital, N-5009 Bergen, Norway

798

Patient selection If the risk of pneumoperitoneum-enhanced tumor shedding, implantation, and take cannot be ruled out, laparoscopyinduced disease progression could occur in all adenocarcinomas extending to the serosal layer of the bowel wall. Endoscopic ultrasonography (EUS) allows fairly accurate assessment of the depth of colonic wall invasion, whereas CT-scan findings poorly correlate with pathology [11]. Preoperative knowledge of the T-factor adds to selection of colon cancer patients unsuitable for celioscopic resection. In fact, laparoscopy-induced peritoneal tumor dissemination could occur not only in patients with T3 resectable cancers but also in patients with unresectable disease undergoing laparoscopy for palliation. Although the view in the low pelvis may be even better than in open surgery [20], laparoscopy should not be attempted in patients with cancer of the middle third of the rectum. This is particularly true for lesions 56 cm from the anal margin where there is doubt that a sphincter-saving procedure can be performed. Not doing so may in turn increase the relative number of abdominoperineal resections [23]. Moreover, it has been easier to prove the feasibility of a truly laparoscopic abdominoperineal excision [6] since there is no need for distal suture line or for laparotomy. (The bulky specimen with its attendant mesentery is delivered perineally.) The technical limit of the distal suture line may in part be overcome using bilaterally articulated laparoscopic stapling devices (ETS FLEX 35, Ethicon Endosurgery, Cincinnati, OH) (Reflex AEC 35, Richard-Allan, Richland, MI). However, the achieval of an adequate distal margin (even in tumors of the proximal third of the rectum) still depends on the index-thumb palpation of the distal edge of the lesion. Furthermore, colorectal anastomosis after laparoscopic high anterior resection of the rectum should not be carried out completely laparoscopically whenever operating on cancer since the totally intracorporeal approach [3] may not permit washout of the proximal colon end [21]. Long-term cure rates True recurrence, survival, and death rates after laparoscopic surgery for colorectal cancer remain unknown, although some retrospective papers [5, 12, 18] reported early outcomes comparable with conventional surgery. This view is supported by the short-term outcome analysis of a small randomized trial [16]. Five-year results of a nonrandomized prospective study showed no significant differences in 191 patients undergoing a totally intracorporeal technique [9]. Conclusion Although laparoscopic resection of colorectal cancer may be technically possible, it must be kept in mind that adenocarcinoma of the large bowel is curable by open surgery in about 50% of the cases. This includes the considerable achievements in terms of long-term survival rates reported by the Registry of Repeat Resection of Colorectal Hepatic Metastases [7]. Moreover, since the diagnosis of synchronous colorectal neoplasms is established intraoperatively in

76% of the patients [1], the loss of manual palpation is not reassuring. The future of laparoscopic surgery for cure of colorectal cancer will not become more certain as a result of recommendations to excise trocar sites [13], use the cytotoxic agent povidone-iodine [10], salvage intraperitoneal chemotherapy [16], or perform with a gasless technique. Until we know more about the long-term results, laparoscopic radical surgery for colon cancer should not be performed based on surgeon or patient preference, but, preferable, inside randomized studies, or possibly within prospective series designed for strict follow-up and audit. References
1. Adloff M, Arnaud JP, Bergamaschi R, Schloegel M (1989) Synchronous carcinoma of the colon and rectum: prognostic and therapeutic implications. Am J Surg 157: 299302 2. Allardyce R, Morreau P, Bagshaw P (1996) Tumor cell distribution following laparoscopic colectomy in a porcine model. Dis Colon Rectum 39: S47S52 3. Bergamaschi R, Arnaud JP (1997) Intracorporeal colorectal anastomosis following laparoscopic left colon resection. Surg Endosc (in press) 4. Bouvy ND, Marquet RL, Hamming JF, Jeekel J, Bonjer HJ (1996) Laparoscopic surgery in the rat. Surg Endosc 10: 490494 5. Clinical outcomes of Surgical Therapy (COST) Study Group: Fleshman JW, Nelson H, Peters WR, Kim HC, Larach S, Boorse RR, Ambroze W, Legget P, Bleday R, Stryker S, Christenson B, Wexner S, Senagore A, Rattner D, Sutton J, Fine AP (1996) Early results of laparoscopic surgery for colorectal cancer: retrospective analysis of 372 patients treated by Clinical outcomes of Surgical Therapy (COST) Study Group. Dis Colon Rectum 39: S53S58 6. Darzi A, Lewis C, Menzies-Gow N, Guillou PJ, Monson JRT (1995) Laparoscopic abdominoperineal excision of the rectum. Surg Endosc 9: 414417 7. Fernandez-Trigo V, Shamsa F, and other members of the Repeat Hepatic Resection Registry: Aldrete J, Andersson R, Arnaud JP, Bergamaschi R, Ciferri E, Connolly D, Coppa G, Crucitti F, Dagradi A, Elias D, Gazzaniga M, Herrera L, Hodgson WJB, Hughes KS, Hohenberger P, Ilstrup D, Kemeny M, Margarit C, Meyers W, Murio JE, Nagorney DM, Nicoli N, Nims T, Que F, Scheele J, Schlag P, Stangl R, Steves M, Sugarbaker PH, van de Velde CJH, Vidal-Jove J (1994) Repeat liver resections from colorectal metastasis. In: Sugarbaker PH (ed) Hepatobiliary cancer. Kluwer Academic, Norwell, MA, pp 185196 8. Forde KA, Hulten L (1996) Laparoscopy in colorectal surgery (editorial). Surg Endosc 10: 10391040 9. Franklin ME, Rosenthal D, Abrego-Medina D, Dorman JP, Glass JL, Norem R, Diaz A (1996) Prospective comparison of open vs. laparoscopic colon surgery for carcinoma. Dis Colon Rectum 39: S35S46 10. Hewett PJ, Thomas WM, King G, Eaton M (1996) Intraperitoneal cell movement during abdominal carbon dioxide insufflation and laparoscopy. Dis Colon Rectum 39: S62S66 11. Hirata N, Kawamoto K, Ueyama T, Iwashita I, Masuda K (1994) Endoscopic ultrasonography in the assessment of colonic wall invasion by adjacent diseases. Abdom Imaging 19: 2126 12. Hoffman GC, Baker JW, Doxey JB, Hubbard GW, Ruffin WK, Wishner JA (1996) Minimal invasive surgery for colorectal cancer. Ann Surg 223: 790798 13. Hubens G, Pauwels M, Hubens A, Vermeulen P, Van Marck E, Eyskens E (1996) The influence of a pneumoperitoneum on the peritoneal implantation of free intraperitoneal colon cancer cells. Surg Endosc 10: 809812 14. Jacobi AC, Sabat R, Bo hm B, Zieren HU, Volk HD, Mu ller JM (1997) Pneumoperitoneum with carbon dioxide stimulates growth of malignant colonic cells. Surgery 121: 7278 15. Jacquet P, Averbach AM, Stephens AD, Sugarbaker PH (1995) Cancer recurrence following laparoscopic colectomy. Dis Colon Rectum 38: 11101114 16. Jacquet P, Sugarbaker PH (1996) Wound recurrence after laparoscopic colectomy for cancer. New rationale for intraoperative chemotherapy. Surg Endosc 10: 295296

799 17. Lacy AM, Garc a-Valdecasas JC, Pique JM, Delgado S, Campo E, Bordas JM, Taura P, Grande L, Fuster J, Pacheco JL, Visa J (1995) Short-term outcome analysis of a randomized study comparing laparoscopic vs open colectomy for colon cancer. Surg Endosc 9: 1101 1105 18. Lord SA, Larach SW, Ferrara A, Williamson PR, Lago CP, Lube MW (1996) Laparoscopic resections for colorectal carcinoma. Dis Colon Rectum 39: 148154 19. Mathew G, Watson DI, Rofe AM, Baigrie CF, Ellis T, Jamieson GG (1996) Wound metastases following laparoscopic and open surgery for abdominal cancer in a rat model Br J Surg 83: 10871090 20. Monson JRT, Darzi A, Carey PD, Guillou PJ (1992) Prospective evaluation of laparoscopic assisted colectomy in an unselected group of patients. Lancet 340: 831833 21. Monson JRT, Hill ADK, Darzi A (1995) Laparoscopic colonic surgery. Br J Surg 82: 150157 22. Pfeifer J, Wexner SD, Reissman P, Bernstein M, Nogueras JJ, Singh S, Weiss E (1995) Laparoscopic vs open colon surgery. Costs and outcome. Surg Endosc 9: 13221326 23. Phillips EH, Franklin M, Carroll BJ, Fallas MJ, Ramos R, Rosenthal D (1992) Laparoscopic colectomy. Ann Surg 216: 703707 24. Reissman P, Cohen S, Weiss E, Wexner SD (1996) Laparoscopic colorectal surgery: ascending the learning curve. World J Surg 20: 277282 25. Slim K, Pezet D, Riff Y, Clark E, Chipponi J (1995) High morbidity rate after converted laparoscopic colorectal surgery. Br J Surg 82: 14061408 26. Vukasin P, Ortega AE, Greene FL, Steele GD, Simons AJ, Anthone GJ, Weston LA, Beart RW (1996) Wound recurrence following laparoscopic colon cancer resection. Dis Colon Rectum 39: S20S23 27. Wexner SD, Cohen SM (1995) Port site metastases after laparoscopic colorectal surgery for cure of malignancy. Br J Surg 82: 295298 28. Whelan RL, Sellers GJ, Allendorf JD, Laird D, Bessler MD, Nowygrod R, Treat MR (1996) Trocar site recurrence is unlikely to result from aerosolization of tumor cells. Dis Colon Rectum 39: S7S13

R. Bergamaschi H. E. Myrvold
Section of Gastroenterology Department of Surgery University Hospital of Trondheim Olav Kyrres gt 17 N-7006 Trondheim Norway

Surg Endosc (1997) 11: 805808

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Development of a fogless scope and its analysis using infrared radiation pyrometer
D. Hashimoto,1 M. Shouji2
1 2

Department of Surgery, Tokyo Metropolitan Police Hospital, 2-10-41 Fujimi, Chiyoda-ku, Tokyo 102, Japan Department of Mechanical Engineering, Tokyo University, 7-3-1 Hongou, Bunkyo-ku, Tokyo 113, Japan

Received: 11 September 1996/Accepted: 14 December 1996

Abstract Background: In laparoscopic surgery, fogging of the lens tip instantly degrades the quality of image and interrupts the flow of surgical maneuvers. External warming devices prevent fogging for a certain length of time, but use of them is troublesome at best. These problems prompted us to develop a fogless mechanism built into the scope so that the fogless image would be continuous and trouble-free. We then conducted several experiments that demonstrate the precise effect it has on the lens tip. Methods: The infrared radiation pyrometer was used to compare the temperature distribution as well as the temperature change against time of each lens tip of the fogless and ordinary laparoscopes. Results: The temperature of the lens tip of the ordinary scope did not change, but that of the fogless scope rose to 41C in 184208 s. Conclusions: The fogless mechanism effectively warms the lens tip to a safe temperature that prevents fogging in the intraabdominal environment. Key words: Fogless scope Anti-fogging Rigid scope Laparoscopic surgery

Sometimes during the operation, the scope must be retracted from the abdominal cavity to clean off any bodily fluids that were accidentally smeared on the lens. The temperature drop caused by this action is substantial enough that another warming process becomes necessary, thus causing an interruption to the flow of surgery. During an emergency situation in a laparoscopic procedure such as arterial bleeding, fogging can potentially be devastating since proper warming of the scope is time consuming. The handling of such an emergency becomes unnecessarily complicated. It is, therefore, evident that a mechanism which provides a clear, fogless view quickly and continuously would contribute greatly to ameliorating the safety of laparoscopic surgery. The key concepts and development of the fogless scope When considering the phenomenon of fogging, we noticed that it occurs only at the lens tip (not the lenses inside the scope) and that it does not occur when the lens is warmer than the body temperature. With this concept in mind, we developed a fogless laparoscope that has a small metallic lens holder at its tip (Fig. 1). The holder warms the lens tip within a few minutes (approximately 3 min) after the light source is turned on by converting the light energy of the optic fibers into heat. As long as the light source is set at maximum output, the temperature of the metallic holder and the adjacent lens reaches approximately 41C, so fog formation can be avoided. Once the set temperature is achieved, a clear view of the intracorporeal space is provided throughout the operation even when the scope is temporarily retracted from the body. Since there is no need to rewarm the scope via hot water or any other device, the operation can proceed smoothly. The Fogless Laparoscope (1) is an original rigid laparoscope designed and manufactured by Shinko Optical, Inc. (Japan patent 5-90607). The scope consists of an optical fiber bundle surrounding the central hollow tube with relay lenses inside. The glass fiber bundle (light guide) provides

Fogging at the lens tip of the rigid scope jeopardizes the safety of the laparoscopic procedure as the view of the operative field becomes unclear. Fogging most often occurs when the lens tip is introduced into the abdominal cavity where the temperature and humidity are higher than the extracorporeal environment. This is the reason why the scope is commonly warmed before inserting it into the abdominal cavity. Heating is usually done by soaking the tip of the scope in hot water or by using special warming devices. Still, the essential problem of fogging remains unsolved even with the use of rather costly antifogging machines.
Correspondence to: D. Hashimoto

806

Fig. 1. Illustration of the fogless scope tip. The illuminating optical fibers running on the outer part of the scope warm the lens tip by bridging the metal cap which surrounds the lens. Once the temperature exceeds body temperature, the fogging of the lens tip is avoided completely. Fig. 2. The different rates of temperature changes corresponding to the different levels of power source. When the power of the light source is too low, the antifog mechanism does not work.

the illumination of the abdominal cavity, and the central lens system (image guide) transmits the image to the charge-coupled device (CCD) camera. The unique feature of this scope is the metallic lens holder, which has two notches that bridge over the illuminating optical fiber. The light source is metallic halide, 160 W. This is not a cold light, so enough heat energy can be tapped to heat the lens to 41C. Since the light source should be kept at or near 100% power to maintain the fogless effect, some may worry about white out of the video image. This phenomenon occurs when the CCD receives too much light from an object, and the image of that particular object becomes unrecognizable. Most of the recent laparoscopic systems include an image processor which automatically adjusts the light level of the image. The only financial investment required to achieve foglessness is the price of replacing the scope itself. Since the fogless mechanism is very simple, its cost compared to the ordinary laparoscope is negligible.

Study design First, varying temperaturetime curves from different light source power levels were compared to each other in terms of time required to reach body temperature. The fogless laparoscope was placed before the infrared radiation pyrometer, and the central point of the lens tip was chosen for digital display and graphic plotting. The connected light source was turned on at power levels of 100%, 50%, and 25%. The metal-halide lamp used in this experiment was 160 W. In the second part of the experiment, an ordinary nonfogless laparoscope was placed before the infrared radiation pyrometer to obtain control data. The tip of the laparoscope was visualized through the thermotracer, and the appropriate thermocolor settings were made. The temperature change and distribution were assessed by the real-time computer graphic display of the color changes. The central point of the lens was chosen for the digital display and graphic plotting of the precise temperatures. The connected light source was turned on to 100% power of 160 W, and the data was retrieved from the computer in real time. The fogless laparoscope was then placed before the infrared radiation pyrometer, using similar conditions to those for the nonfogless laparoscope. The central point of the lens was chosen for graphic analysis. To demonstrate that inserting the laparoscope into an open syringe filled with gauze speeds up the warming process, a third experiment was performed. Once the fogless laparoscope was placed before the thermotracer with similar settings to those of previous experiments, the gauze-filled syringe was placed over the tip of the scope. The light source was then turned on, and the temperature of the lens was measured by quickly removing the syringe at the desired time. This was repeated at different intervals so that an accurate temperaturetime graph could be drawn. Results For the first part of the experiment, the graph shows that for all power levels, the temperature of the lens increases logarithmically with time, and once the power is off, the temperature decreases exponentially. At t 360 s, the light source power of 100% warmed the lens tip to 44C, but the

Analysis using infrared radiation pyrometer The infrared radiation pyrometer (the NEC Thermotracer TH3100 series) is a noncontact high-sensitivity infrared radiation thermometer which analyzes data using a Thermotracer Processing Program which runs on a personal computer. This device senses infrared energy radiated from surrounding objects and converts it into an optical scan image that represents surface temperature distribution. The color scale (bluered) is assigned to a specified temperature range, and the image with appropriate color distribution appears on the computer screen at real time. For our study, blue was assigned to the temperature of 20C and red was assigned to 50C. The other temperatures in between were assigned to appropriate shades of visible light spectrum. The precise temperature for any specified point on the image can be recorded to construct the temperature-time curve. The infrared radiation pyrometer was used to determine the temperature change of the lens tip of the fogless laparoscope, and the results were compared to those of an ordinary laparoscope. We had found that inserting the tip into an open syringe stuffed with gauze significantly speeds up the warming process of the fogless laparoscopic lens. This anecdotal fact, which we noticed every time the fogless laparoscope was used, was verified using this sophisticated thermometer.

807

Fig. 3. Plot a represents the temperaturetime curve of ordinary laparoscope and plot b represents that of a fogless laparoscope. The accompanying infrared radiation pyrometer picture of the two scope tips demonstrates the respective temperature distributions at t 144 s.

power of 50% barely reached body temperature, and that of 25% never reached it even after 360 s (Fig. 2). The power of 50% warmed the lens to 37C, and the power of 25% warmed the lens to 32C. Secondary, the lens of the fogless laparoscope achieved the optimal temperature of 41C at 210 s after full power of the light source began, while the temperature of the ordinary laparoscope increased about 3C after 360 s (Fig. 3). The maximal temperature (equilibrium) of the fogless laparoscope was 43.6C, and the ordinary laparoscope reached maximal temperature of 30C at equilibrium. The third experiment was to determine whether or not placing a syringe filled with gauze over the laparoscope tip speeds up the warming process. The time required to achieve a lens temperature of 41C was slightly less than 40 s. The rate of lens warming increased by more than fivefold when compared to normal warming with the fogless mechanism (Fig. 4). All of the experiments were conducted in a laboratory where the temperature was 2224C.

Fig. 4. The temperaturetime curve of the lens tip when the fogless scope is covered by a syringe filled with ordinary gauze. The temperature of the lens reaches body temperature in less than 30 s.

Discussion In laparoscopic surgery, keeping clear intracorporeal vision has always been a problem. This is an important issue because clear vision is an absolute necessity for success of the surgery. One of the major culprits causing deterioration of clear vision is fogging. In the operating room, the temperature is usually around 2026C, and the humidity is about 4060%. The intraabdominal environment, however, is about 38C, with humidity of 90100%. When the laparoscope, which has been in a cool environment, is suddenly inserted into the warm and humid abdominal cavity, the lens tip immediately fogs up, and the monitor screen whites out. Once fogging occurs, the lens must be either wiped, or soaked in water, or blow dried before a clear view can be

restored. The fogging does not go away spontaneously even if one waits for a long time. Previously, several attempts have been made to solve this fogging problem, but most of the solutions do not solve the root of the problem. Some examples include inserting the lens tip into hot water or a warming device each time the scope is retracted from the abdomen. This is troublesome because the required additional steps may interrupt the flow of the surgical procedure. Even if the additional steps may not be bothersome in a controlled setting, these steps may potentially become a lethal obstacle when an emergency situation arises. Another example involves smearing antifogging fluid onto the lens before the procedure. This may be worrisome because a foreign substance (in this case, fluid) comes in contact with the operative field. A scope with a built-in irrigation system, which can deal with fogging as well as soiling of the lens, has also been created. The disadvantage of lens irrigation is that the water droplets at the lens tip can easily interfere with the clarity of the vision.

808

The fogless laparoscope we developed takes a completely simple and novel approach to the problem of fogging. The fogless mechanism built into the tip of the scope does not require the surgeon or any other member of the surgical team to take additional steps. Once the appropriate temperature is reached, worry-free fogless vision is achieved even if the scope is temporarily retracted from the intraabdominal space. In addition because this optimal temperature is only slightly higher than the normal body temperature, there is no worry of burn injury from accidentally touching the tissues. The additional warming inside a syringe filled with gauze significantly speeds up the prewarming step. This is most likely due to the reflection of light source from the

white gauze directly hitting the lens tip. We find that the extra step compensates very well, since the time saving is fivefold. The experiments using the infrared radiation pyrometer proved that the fogless mechanism effectively warms the tip of the scope to a safe temperature that prevents fogging in the intraabdominal environment.

Reference
1. Hashimoto D (1995) Advanced techniques in gasless laparoscopic surgery (abdominal wall lifting with subcutaneous wiring). World Scientific, Singapore

Surg Endosc (1997) 11: 834837

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic gastrojejunostomy and endoscopic biliary stent placement for palliation of incurable gastric outlet obstruction with cholestasis
I. B. Brune,1 H. Feussner,1 H. Neuhaus,2 M. Classen,2 J.-R. Siewert1
1 2

Chirurgische Klinik und Poliklinik der Technischen Universita t Mu nchen, Ismaninger Strae 22, D-81675 Munich, Germany II. Medizinische Klinik und Poliklinik der Technischen Universita t Mu nchen, Ismaninger Strae 22, D-81675 Munich, Germany

Received: 7 May 1996/Accepted: 12 December 1996

Abstract Background: For patients with incurable malignant gastric outlet obstruction and cholestasis, laparoscopic gastrojejunostomy combined with endoscopic biliary stent placement seems to offer a minimally invasive palliation. Methods: We retrospectively analyzed the data of 16 patients submitted to laparoscopic gastrojejunostomy. Laparoscopic gastroenterostomy was performed as an antecolic, side-to-side gastrojejunostomy with enteroenterostomy. In 12 patients cholestasis was relieved preoperatively by stent placement via endoscopy (n 6, 37.5%), percutaneous access (n 5, 31%) or bilioenteric anastomosis (n 1, 6.25%). One patient needed a percutaneous Yamakawa prosthesis postoperatively. Results: Mean operative time was 126 min. There were no intraoperative complications. In one patient conversion to open surgery became necessary because of extensive adhesions. The only postoperative complication was bleeding from a trocar site requiring reintervention; there was no mortality. Median postoperative hospital stay was 7 days. Delayed gastric emptying was observed in 3 (18.7%) patients. Median survival was 87 days after the operation. All patients died from their primary disease but could maintain oral intake during the remaining survival time. Conclusions: We conclude that laparoscopic gastrojejunostomy and endoscopic or percutaneous biliary stenting provide a good functional result while impairing the quality of life only to a minimal extent. Key words: Gastric outlet obstruction Laparoscopic gastrojejunostomy Endoscopic biliary stent replacement

Most patients with incurable, malignant gastric outlet obCorrespondence to: I. B. Brune

struction also present with cholestasis due to compression or infiltration of the distal common bile duct. A few decades ago, the only therapeutic possibility for palliation and relief from symptoms was gastroenterostomy and bilioenteric anatomosis via laparotomy [28]. This operation mostly provided satisfying functional results but is quite invasive. It is related to a considerable period of hospitalization and to pain, discomfort, and possible complications that can accompany laparotomy. In the last few years, endoscopic and percutaneous techniques [1, 9] for introduction of biliary wallstents have been developed by gastroenterologists [10] in search of less-invasive palliation modalities of cholestasis. In the surgical field, the effort to reduce operative trauma resulted in development of minor access or laparoscopic operations. While laparoscopic surgery for malignant disease with curative intention is still discussed controversially [7, 8] palliative laparoscopic procedures are of special benefit for the patient who in general has a very short life expectancy. The advantages of laparoscopic surgery, i.e., reduced pain, a fast recovery, and short hospital stay, allow the surgical procedure to interfere as little as possible with quality of life. If an obstruction is found to be located in the esophagus or cephalad part of the stomach, laparoscopic placement of a feeding catheter into the stomach or jejunum is a minimally invasive alternative for those patients in which endoscopy is not possible because the stenosis cannot be passed with the endoscope [11, 15, 22]. If obstruction occurs in the distal third of the stomach or the duodenum, laparoscopic gastroenterostomy can assure rapid relief of symptoms and allow the patient to be discharged within a few days under oral nutrition [2, 12, 13, 18, 23, 29]. Our current policy is to submit patients with malignant, incurable cholestasis and gastric outlet obstruction to endoscopic placement of a biliary stent for relief from jaundice, followed by laparoscopic gastroenterostomy to allow further oral nutrition.

835

Patients and methods

The data on 16 patients who were submitted to laparoscopic gastroenterostomy at the Surgical Department of the Technical University, Munich, between 1993 and 1995 were retrospectively analyzed. The outcome variables assessed included operative time and intraoperative complications, procedure-related morbidity and mortality, postoperative hospital stay, overall survival, and the ability to sustain oral nutrition during the survival period. A follow-up to determine survival after the operative procedure could be obtained in all patients. There were nine men and seven women with a mean age of 66.5 years (range 3787). In all patients indication for the gastrointestinal bypass was given by gastric outlet obstruction due to incurable malignant disease. The primary disease was pancreatic carcinoma (n 7), gallbladder carcinoma (n 4), distal gastric carcinoma (n 2), Klatskin tumor (n 1), carcinoma of the papilla (n 1), and intraabdominal lymph node metastasis from a bronchial carcinoma (n 1). Inoperability of the pancreatic tumors was assessed by endoscopy with endoluminal ultrasound, ERCP, CT scan, angiography, and MRI. Workup of the other patients did not include angiography and MRI. Biopsies were obtained whenever possible. Impaired gastric emptying was ascertained clinically as well as radiologically and/or by endoscopy. Patients with extensive previous abdominal surgery or uncorrectable coagulopathy were not considered suitable for the laparoscopic procedure. Because of the short life expectancy, mainly plastic stents were used for biliary palliation. When endoscopic biliary stent implantation was not possible in the presence of biliary and duodenal obstruction, a percutaneous transhepatic tube with a diameter of 1418 Fr was inserted through the common bile duct and the duodenal stenosis with the tip reaching the duodenojejunal junction. This tube (a Yamakawa prosthesis) was blocked at the skin level, providing internal biliary drainage [16]. Whenever this was not possible, the patient was referred to open gastroenterostomy with bilioenteric bypass. Only four patients did not present with initial cholestasis (two gastric carcinoma, one distal pancreatic carcinoma, and one gallbladder carcinoma). The other 12 patients (80%) showed biliary obstruction that was treated preoperatively by biliary stent placement via endoscopy (n 6) or a percutaneous access (n 5). In one patient open bilioenteric bypass had been perviously performed at another institution 4 months before admission for gastroenterostomy. One patient with gallbladder carcinoma developed cholestasis on the 6th postoperative day and was treated by percutaneous placement of a Yamakawa prosthesis. During the study period, three patients had to be referred to open gastroenterostomy with bilioenteric bypass: two because of extensive previous surgery and one because of the inability to place a biliary stent endoscopically or percutaneously. Cholestasis parameters (bilirubin, -GT, and alkaline phosphatase [AP]) were compared the day before endoscopic stent placement and the day before the operation to evaluate relief from biliary obstruction. Laparoscopic gastrojejunostomy was performed as an antecolic, isoperistaltic side-to-side anastomosis with an enteroenterostomy 25 cm below the gastrojejunostomy to assure sufficient biliary drainage from the afferent loop and avoid exposure of the gastric anastomosis to bile acids. This is the procedure we are using in open surgery. The operation was performed under general anesthesia with the patient in supine position. The stomach and bladder were catheterized and antibiotic single-shot prophylaxis with 1 g ceftriaxone was given. Four trocars (three 10 mm and one 12 mm) were placed as described in Fig. 1. As the laparoscope must sometimes be moved to the upper ports for creation of the enteroenterostomy, we always use 10-mm trocars. The table was tilted in a 30 Trendelenburg position and the transverse colon and omentum were swept cephalad to identify the ligament of Treitz. From there, the first mobile jejunal loop was brought up in an antecolic way to the stomach, where it was fixed to the antrum at the anterior greater curvature with two stay sutures. The openings for insertion of the stapler jaws were made in the stomach and small bowel using electrocoagulation (Fig. 2). A 30-mm Endo-GIA stapler (USSC, Norwalk, CT) was introduced through the 12-mm trocar in the right middle abdomen. The jaws of the instrument were inserted in the stomach and small bowel (Fig. 3). Care had be taken to place the stapler on the antimesenteric side in the small bowel to avoid bleeding into the anastomosis. Two subsequent applications of the 30-mm stapler were necessary to ensure a sufficiently wide diameter of the gastroenterostomy. Alternatively, a 60-mm stapling device may be used, the handling of which may be cumbersome due to instrument length. The nasogastric tube was placed in the afferent loop below the anastomosis under laparoscopic control. The remaining opening was closed by manual

Fig. 1. Trocar placement for laparoscopic gastroenterostomy.

suture (Fig. 4). The same technique was used for the enteroenterostomy, where one application of the 30-mm Endo-Gia usually provided a sufficiently wide anastomosis. Leakage of both anastomoses was excluded at the end of the procedure by instillation of methylene blue via the nasogastric tube under laparoscopic control. A Robinson drain may be left at the anatomosis. A liquid diet was started on the 3rd postoperative day and progressively increased to normal oral nutrition. Radiographic control of the anastomosis with water-soluble contrast medium was performed on the 4th postoperative day to assess patency and exclude leakage.

Results Median interval between biliary stent placement and gastroenterostomy was 70 days (3630). In eight patients biliary stenting was performed during the same hospital stay as gastroenterostomy. Preoperative relief from cholestasis is reflected in a decrease of bilirubin, -GT, and AP as shown in Table 1. Mean operative time was 126 min (70210). No major intraoperative complications were noted. In one patient, conversion to laparotomy became necessary due to technical difficulties caused by severe adhesions; the further postoperative course of this patient was uneventful. The Robinson drain was left in place for a mean of 4.7 days (28). There was no postoperative mortality. One patient (6.25%) required operative reintervention for postoperative hemorrhage from a trocar incision. Postoperative radiographic control of the anastomosis showed delayed gastric emptying (DLE) in three patients (18.75%) due to swelling and edema of the anastomotic region. Clinically, the signs of DLE were only mild in these patients who all could be discharged under oral nutrition. There was no anastomotic leakage or intraabdominal abscess. Median postoperative hospital stay was 7 days (332). At follow-up, median survival was revealed to be 87 days (31259). There was no readmission for complications related to stent placement or the operative procedure. All

836

Fig. 2. The jejunum is fixed to the greater curvature with stay sutures and openings for insertion of stapler jaws are created using electrocoagulation. Fig. 3. Stapler jaws are placed in stomach and small bowel. Fig. 4. Closure of opening left after stapler removal of manual suture.

patients had died from their underlying disease and had been able to sustain oral nutrition for the whole surviving period.

Table 1. Decrease of cholestasis parameter after placement of a biliary stent (mean values) Cholestasis parameter Total bilirubin (mg/dl) -GT (U/l) AP (U/l) Before stent 15.1 (2.935.6) 574 (3111,670) 1,093 (4302,250) After stent 5.6 (1.07.5) 250 (31588) 401 (154658)

Discussion Patients with incurable malignant disease presenting with biliary obstruction and/or impaired gastric emptying have a very poor prognosis [3, 5, 19]. The only therapeutic objective is rapid relief from symptoms with as little interference with the quality of life as possible. In the last decades, surgical palliation with bilioenteric bypass and gastroenterostomy has been the method of choice for patients presenting with both symptoms [3, 14, 27]. Although providing quite satisfactory results, open gastroenterostomy has been associated with delayed gastric emptying in up to 16% [3]. It also carries the risk and discomfort of a laparotomy. The mortality of this operation ranges between 8% and 17%; the morbidity is around 25%, with a rate of reintervention of 8% [3, 14]. Survival is between 130 and 200 days [3, 14]. Whether to perform a gastric bypass in principle, i.e., in all patients undergoing bilioenteric anastomosis regardless of whether they present the clinical symptoms of gastric outlet obstruction or not, or to add gastroenterostomy of necessity only in patients with impaired gastric emptying has been discussed by many authors [3, 14, 20, 27]. Neuberger et al. state that after biliary bypass only in 474 patients, secondary reoperation for gastric obstruction was necessary in 12% and carried a 25% mortality while complication rates after combined biliary and gastric bypass were not significantly higher than after biliary bypass alone; they therefore advocate adding gastric bypass in principle. Sarr et al., find similar results with 13% reoperation. Other authors [3, 26] found morbidity after prophylactic bypass to be increased and therefore do not recommend it. Since reduction of the operative trauma has gained considerable importance with the introduction of laparoscopic surgery, indications for this technique have been expanded. As gastroenterostomy only requires reconstruction, the problem of organ retrieval from the abdominal cavity did

not arise. While laparoscopic surgery for curative purposes in abdominal malignancies is still in discussion, the value of laparoscopy as a palliative procedure is generally recognized. Therefore, several authors have started to perform gastroenterostomy routinely via laparoscopic access [2, 12, 13, 18, 23, 29]. Surgical treatment of biliary obstruction with bilioenteric anastomosis was the only alternative until the development of endoscopic stent techniques. For open biliary bypass, Rooij et al. have published a postoperative mortality of 5.3%, a complication rate of 18.4%, reinterventions in 5.2%, and a mean survival of 244 days. Redmond et al. noted a mortality of 2%, morbidity of 35%, and reinterventions in 2% of patients. Neuberger et al. have observed a mortality of 27%, a procedure-related morbidity of 13%, a reintervention rate of 12%, and a mean survival of 279 in a study including 474 patients with bilioenteric bypass. Endoscopic or percutaneous biliary stenting offers several advantages over surgical biliary bypass. Usually it will be performed via endoscopic retrograde cholangiography (ERC). Whenever passage of the endoscope through the pylorus and proximal duodenum is not possible because of tumor compression and infiltration, a percutaneous access under radiological control is attempted. It does not require general anesthesia and is not as invasive as a laparotomy, but the method is also associated with certain risks. Even though the self-expandable metal wallstents seem to obstruct less than previously described polyethylene stents [9], the rate of recurrent jaundice is ranging from 5% [9, 24] to 24% [6]. This complication can mostly be treated by reendoscopy. When the endoscopic approach fails due to duodenal obstruction, insertion of a percutaneous transhepatic tube (Yamakawa prosthesis) bridging the biliary and duo-

837

denal obstruction provides an effective internal biliary drainage [17]. Complication rates of biliary stenting are between 15% [21] and 31% [6]. The procedure-related mortality rate is 14% [10, 24]. This compares favorably with open surgical biliary bypass, especially when survival is expected to be less than 6 months [25]. In an attempt to achieve optimal palliation with minimal discomfort and risk for the patient, we have adopted the strategy of submitting patients with malignant gastric outlet obstruction and cholestasis to biliary stenting followed by laparoscopic gastroenterostomy. Our experience showed that preoperative biliary stenting provided fast relief from jaundice. The operative time was somewhat longer than it would have been in open surgery but, at slightly more than 2 h it is still acceptable. There was no intraoperative complication: Conversion to laparotomy was not to be considered a complication but rather a decision assuring safety for the patient because of the technical difficulty. The procedure-related morbidity requiring reintervention was 6.6% being a postoperative hemorrhage from a trocar incision. The mild delayed gastric emptying observed in three patients (18.75%) occurred with a similar frequency to that described for open surgery. The follow-up showed that gastrointestinal passage remained unhampered for the rest of the survival period in all patients. As laparoscopic biliodigestive anastomosis is feasible but technically extremely demanding [12], the patients presenting with cholestasis due to obstruction of the common bile duct in whom preoperative endoscopic placement of a stent is not possible should be referred to open surgery. Also, patients with portal hypertension and coagulation disorders are not very suitable candidates for the laparoscopic procedure. Previous abdominal surgery is a relative contraindication. In any case, the surgeon must have sufficient experience in laparoscopy, especially with laparoscopic suturing techniques. Even though our actual experience is still limited, our preliminary results seem to demonstrate that our therapeutic policy, which consists in preoperative biliary stenting and consecutive laparoscopic gastroenterostomy, achieves results that are at least equal to those of conventional surgery. At the same time, the patient has the benefit of the laparoscopic procedure, which, as has been proven for other operations [4], is associated with less pain and a shorter recovery time. This is of major importance since life expectancy is very short for these patients. It will hopefully be confirmed by further studies including a larger number of patients. Within these limits, laparoscopic gastrojejunostomy for palliation of impaired gastric emptying seems to be the method of choice in the hands of an experienced laparoscopic surgeon.

References
1. Bethge N, Wagner HJ, Knyrim K, Zimmermann HB, Starck E, Pausch J, Vakil N (1992) Technical failure of biliary metal stent deployment in a series of 116 applications. Endoscopy 24: 395400 2. Brune IB, Scho nleben K (1992) Laparoskopische Seit-zu-Seit GastroJejunostomie. Chirurg 63: 577580 3. de Rooij PD, Rogatko A, Brennan MF (1991) Evaluation of palliative surgical procedures in unresectable pancreatic cancer. Br J Surg 78: 10531058

4. Eypasch E, Ure B, Spangenberger W, Troidl H (1993) BefindlichkeitLebensqualita t: Schlu sselbegriff zur Bewertung der endoskopischen Chirurgie. In: Brune IB, Scho nleben K (eds) Laparo-Endoskopische Chirurgie. Marseille, pp 4354 5. Finch MD, Butler JA (1990) Palliation for nonpancreatic malignant obstruction of the biliary tract. Surg Gynecol Obstet 170: 437440 6. Glattli A, Stain SC, Baer HU, Schweizer W, Triller J, Blumgart LH (1993) Unresectable malignant biliary obstruction: treatment by selfexpandable biliary endoprosthesis. HPB Surg 6: 175184 7. Goh PMY, Kum CK (1993) Laparoscopic Billroth II Gastrectomy: a review. Surg Oncol 2 Suppl 1: 1318 8. Goh PMY, Tekant Y, Kum CK, Isaac J, Ngoi SS (1992) Totally intraabdominal laparoscopic Billroth II gastrectomy. Surg Endosc 6: 160 9. Huibregtse K, Carr Locke DL, Cremer M, Domschke W, Fockens P, Foerster E, Hagenmu ller F, Hatfield AR, Lefevre JF, Liguory CL et al. (1992) Biliary stent occlusiona problem solved with self-expanding metal stents? Endoscopy 24: 391394 10. Hyoty MK, Nordback IH (1990) Biliary stent or surgical bypass in unresectable pancreatic cancer with obstructive jaundice. Acta Chir Scand 156: 391396 11. Lathrop JC, Felix EJ, Lauber D (1991) Laparoscopic Janeway gastrostomy utilizing an endoscopic stapling device. J Laparoendosc Surg 1: 335339 12. Mouiel J, Katkhouda N, White S, Dumas R (1992) Endolaparoscopic palliation of pancreatic cancer. Surg Laparosc Endosc 2: 241243 13. Nathanson LK (1993) Laparoscopic cholecysto-jejunostomy and gastroenterostomy for malignant disease. Surg Oncol 2 Suppl 1: 1924 14. Neuberger TJ, Wade TP, Swope TJ, Virgo KS, Johnson FE (1993) Palliative operations for pancreatic cancer in the hospitals of the US Department of Veterans Affairs from 1987 to 1991. Am J Surg 166: 632637 15. Neufang T et al. (1992) Die perkutane laparoskopische Gastrostomie (PLG). Minim Invasive Chirurg 1: 3134 16. Neuhaus H, Hoffmann W, Classen M (1993) Nutzen und Risiken der perkutanen transhepatischen Cholangioskopie. Dtsch Med Wochenschr 118: 574581 17. Neuhaus H, Hoffmann W, Classen M (1993) Percutaneous management of bile duct carcinoma. Gastrointest Endosc 39: 326 18. Rangraj MS, Mehta M, Zale G, Maffucci L, Herz B (1994) Laparoscopic gastro-jejunostomy: a case presentation. J Laparoendosc Surg 4: 8187 19. Redmond HP, Stuart R, Hofmann KP, Collins PG, Gorey TF (1991) Carcinoma of the head of the pancreas. Surg Gynecol Obstet 172: 186190 20. Rothlin M, Schob O, Weder W, Buchmann P, Largiader F (1993) Gastroenterostomy par principle or de ne cessite in palliative therapy of pancreatic cancer. Helv Chir Acta 60: 111115 21. Salomonowitz EK, Adam A, Antonucci F, Stuckmann G, Zollikofer CL (1992) Malignant biliary obstruction: treatment with selfexpandable stainless steel endoprosthesis. Cardiovasc Intervent Radiol 15: 351355 22. Shallmann R (1991) Laparoscopic percutaneous gastrostomy. Gastrointest Endosc 37: 493494 23. Sosa JL, Zalewski M, Puente I (1994) Laparoscopic gastrojejunostomy technique: case report. Laparoendosc Surg 4: 215220 24. Stoker J, Lameris JS (1993) Complications of percutaneously inserted biliary wallstents. J Vasc Interv Radiol 4: 767772 25. Van den Bosch RP, Van der Schelling GP, Klinkenbijl JH, Mulder PG, Van Blankenstein M, Jeekel J (1994) Guidelines for the application of surgery and endoprosthesis in the palliation of obstructive jaundice in advanced cancer of the pancreas. Ann Surg 219: 1824 26. Van der Schelling GP, Van den Bosch RP, Klinkenbijl JH, Mulder PG, Jeekel J (1993) Is there a place for gastroenterostomy in patients with advanced cancer of the head of the pancreas? World J Surg 17: 128 132 27. Wade TP, Neuberger TJ, Swope TJ, Virgo KS, Johnson FE (1994) Pancreatic cancer palliation: using tumor stage to select appropriate operation. Am J Surg 167: 208213 28. Watanapa P, Williamson RCN (1992) Surgical palliation for pancreatic cancer: developments during the past two decades. Br J Surg 79: 820 29. Wilson RG, Varma JS (1992) Laparoscopic gastro-enterostomy for malignant duodenal obstruction. Br J Surg 79: 1348

Surg Endosc (1997) 11: 868874

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Systems and instruments for laparoscopic surgery without pneumoperitoneum

C. N. Gutt, J. Daume, B. Schaeff, V. Paolucci
Department of General Surgery, Johann-Wolfgang-Goethe-University, Theodor-Stern-Kai 7, D-60590 Frankfurt/Main, Germany Received: 19 November 1996/Accepted: 30 December 1997

Abstract. The insufflation of carbon dioxide into the peritoneal cavity is a routine technique of abdominal exposure in laparoscopic surgery. Because of adverse physiological effects and technical disadvantages of the pneumoperitoneum, alternative methods of abdominal wall lifting have been explored recently. Two groups of retraction systems exist: intraabdominal lifting and subcutaneous lifting of the abdominal wall. Some systems require additional pneumoperitoneum, because the extent of intraabdominal exposure is not sufficient. Other systems are working completely without gas insufflation. Two systems combine abdominal wall lifting with pressure on the internal organs. Every method allows the use of standard laparoscopic instruments, which originally were designed for a regular pneumoperitoneum. The use of a low-pressure pneumoperitoneum in combination with mechanical augmentation of the peritoneal cavity reduces physiological disadvantages of laparoscopy. But technical advantages, such as combination of laparoscopic and open techniques, can be realized only without gas insufflation. Conventional instruments have been designed to fit the ergonomical needs of isopneumic laparoscopy and to be employed with simple valveless cannulae. Key words: Laparoscopy without pneumoperitoneum Abdominal wall lifting Conventional instruments Isopneumic laparoscopy

In addition to physiological consequences of the pneumoperitoneum, there are technical disadvantages. Instruments have to be introduced into the abdominal cavity through sealed trocars in order to maintain the applied intraabdominal pressure throughout the procedure. This adds another difficulty to the construction of these instruments. The construction of these instruments is hampered by their small diameter and length, because their shaft has to be circular and fit into the trocar. In laparoscopy without pneumoperitoneum conventional instruments for open surgery and simple rubber trocars can be used because sealing is not necessary in absence of intraabdominal pressure. In addition, there is the possibility to use instruments without laparoscopic equivalents, which could not be designed to fit through sealed trocars [8]. Thus, alternative methods for mechanical distension of the abdominal wall have been explored. The following retraction systems were developed to provide exposure for adequate visualization of the viscera in combination with a low-pressure pneumoperitoneum or even without pneumoperitoneum. Mechanical distension of the abdominal cavity is achieved with intraabdominal and subcutaneous lifting. Two intraabdominal systems combine traction of the abdominal wall with pressure on the internal organs. Every system allows the use of all laparoscopic instruments originally designed for pneumoperitoneum. Only in the absence of pneumoperitoneum can conventional instruments be used.

The use of laparoscopic methods in surgery depends essentially on the establishment of an intraperitoneal space which allows good exposure of the region of interest. For this purpose insufflation of CO2 gas is routinely used for extension of the abdominal wall. Laparoscopy with a pneumoperitoneum causes a restriction of the instruments freedom of movement and can be the reason for typical complications.
Correspondence to: C. N. Gutt

Intraabdominal retraction systems

Abdominal Cavity Expander-System ACE-WISAP (Semm), T-shaped endoscopic retractor (Gazayerli) and Sling (Banting and Cusheri) Method: Point lifting Requirements: Permanent low-pressure pneumoperitoneum Materials: Reusable stainless steel

869

The Abdominal Cavity Expander [14] is a T-shaped retractor which helps to maintain the volume inside the abdominal cavityfor example, in the case of loss of gas during laparoscopy with pneumoperitoneum. It can be introduced through a 5-mm as well as 10-mm port and will avoid the need to convert to laparotomy. A similar T-shaped fan is described by Gazayerli [6] and used to provide an isolated area of lift for additional exposure during laparoscopy with pneumoperitoneum. The Sling [2] retracts the upper abdominal wall and the falciforme ligament together and consists of a flexible plastic tube. The two ends of the tube are tied together and attached to a framework above the operating table. Tube placement is achieved with a metal introducer. Winch retractor (Araki), U-shaped retractor (Kitano), coathanger (Maher) Method: Linear lifting Requirements: Initial pneumoperitoneum Materials: Reusable stainless steel The Winch retractor [1] consists of a curved Kirschner wire attached to the adjustable chain of two table-based winches. The wire is inserted percutaneously through all layers of the abdominal wall into the peritoneal cavity. During laparoscopic cholecystectomy the pneumoperitoneum is decompressed after the gallbladder is retracted by grasping the fundus. The U-shaped retractor [11] is introduced into the abdominal cavity with a guide tube under endoscopic control after initial installation of a pneumoperitoneum. A framework and winches provide abdominal wall retraction. After pneumoperitoneum the Coathanger-shaped retractor is inserted into the abdominal cavity through an incision lateral to the epigastric vessels and exits lateral to the epigastric vessels at the other side of the abdomen. The retractor is fixed to a chain by which it is elevated. Pelvi-Snake (Volz) and Suspendor 3-X (Mouret) Method: Planar lifting Requirements: Initial pneumoperitoneum Materials: Reusable stainless steel The Pelvi-Snake [16] (Fig. 1) consists of a steel spring which is screwed into the abdomen 5 cm below the umbilicus through a 3-mm incision. With a diameter of 15 cm it covers an area of approximately 150 cm2 inside the abdomen. A simple metal rod which easily rests loose down on a frame fixed to the is responsible for clamping the Pelvi Snake. Via minilaparotomy the Suspendor 3-X [5], a gallow-shaped metal bar, is inserted for additional security after a pneumoperitoneum is built up. The base is fastened laterally to the table. A screw system allows regulation of the tent-shaped abdominal wall elevation. Peritoneal cavity augmentation PCA (Schaller) Method: Linear lifting Requirements: Initial pneumoperitoneum, permanent lowpressure pneumoperitoneum

Materials: Reusable steel, reusable pressure body The Peritoneal Cavity Augmentation [15] (Fig. 2) applies traction to the abdominal wall and at the same time mechanical pressure on internal organs. Two sleeves are introduced percutaneously near the operative field under endoscopic control. An internal connector is attached under light compression of the abdominal wall. At the tips of two metal rods, which are conducted in the sleeves, a flat translucent pressure body is coupled and placed on the internal organs under view. A suspension at the operating table allows lifting the abdominal wall. Spreading trocars (Dragojevic) Method: Point lifting Requirements: Initial air CO2 insufflation Materials: Reusable steel After initial air CO2 gas insufflation a safety trocar is placed. Following abdominal inspection the safety trocar is substituted by a trocar with a spreading device [4]. When introducing a reduction socket into the trocar, the splitted tip is advanced, deploys, and a stable fixture in the abdominal wall emerges. After having attached the first trocar to a gallow, additional spreading trocars are placed and suspended. The desired pulling force can be applied separately. Laparolift (Chin) Method: Planar lifting Requirements: Isopneumic conditions Materials: Disposable plastic/metal fan The Laparolift [3] contains a special retraction device made of high-strength plastic with two blades which is introduced over a 2-cm incision into the abdominal cavity and fanned to conform to the abdominal wall. This retraction device (Laparofan) is attached to the tip of an electricpowered liftarm, which raises and lowers with the push of a button. Besides the Laparofan the laparoscope is introduced through the same incision. Modular retraction system MORES (Gutt) Method: Planar lifting Requirements: Isopneumic conditions Materials: Reusable steel The Modular Retraction System [9] (Fig. 3) is introduced via a 2-cm minilaparotomy. For abdominal wall retraction two lifting parts of different size and shape are assembled to create a intraperitoneal frame. Using combinations of these parts the retractor can be adjusted to different abdominal quadrants and the patients individual anatomy. After correct positioning under endoscopic control it is attached to an outside mechanical lifting arm which is mounted on the operating table. According to the needs of exposure a translucent plastic membrane can be unfolded and placed for posterior organ retraction using the same access.

870

Fig. 1. Pelvi-Snake (Volz). A steel spring which is screwed into the abdomen. Fig. 2. Peritoneal Cavity Augmentation PCA (Schaller). At the two metal rods an internal connector and a pressure body are connected. Fig. 3. Modular Retraction System (Gutt). A Abdominal wall lifting and posterior organ retraction using the same access. B Two lifting elements of different size and shape are assembled. The translucent plastic membrane is unfolded and attached for posterior organ retraction. Fig. 4. Laparo Tenser System with subcutaneous needles (Lucini). Two convex-shaped subcutaneous needles attached to a mechanical lifting arm.

871

Subcutaneous retraction systems Tent-shaped wiring (Nagai) Method: Linear lifting Requirements: Isopneumic conditions Materials: Reusable stainless steel The system consists of two wire loops which anchor subcutaneously in the skin. The wires are attached to lifting handles, and a winching device is tightened and raised for upward traction. The raising of the wire loops produces single-point, tent-shaped lifting of the skin in two places [13]. Subcutaneous wiring (Hashimoto) Method: Linear/planar lifting Requirements: Isopneumic conditions Materials: Reusable stainless steel Subcutaneous wires with a defined shape are tunnelled and suspended with winches to a Kent retractor to achieve abdominal wall lifting. Sutures are used on both sides as hangers to lift up the subcutaneous wires. If the lifting effect is not satisfying a thin metal plate with multiple holes is placed between the subcutaneous wires for better balanced lifting [10]. Laparo tenser (Lucini) Method: Planar lifting Requirements: Isopneumic conditions Materials: Reusable stainless steel The system is consisting of an abdominal wall retractor with two convex-shaped subcutaneous needles which can be attached to a mechanical lifting arm. To improve exposure low-pressure CO2 insufflation can be associated (Fig. 4). Instruments Laparoscopic Laparoscopic surgery with mechanical distension of the abdominal wall allows the use of every laparoscopic instrument, without exception, which originally was designed for a pneumoperitoneum. The construction of these instruments is hampered by their small diameter and length, because in the case of a pneumoperitoneum the shaft must fit into sealed trocars. Usually these instruments comprise distally located mechanisms for single- or twin-action jaws. Every action of the jaws must be transmitted over a central rod in the shaft from a hinge system at the instruments handle. Sturdier, hingeless instruments open their outside-bent jaws by an in-and-out movement of the central rod of an additional pushing tube. In contrast to conventional instruments, where hand power is deflected over only one joint in laparoscopic instruments, power is deflected over two hinge

Table 1. Instruments for gasless laparoscopy Ring forceps Grasper Right angle Scissors Swab on a stick Needle holder Metzenbaum scissors Intestinal clamp Suction-irrigation Pool suction

systems. This leads to reduced tactile feedback and restriction of movement. The instrument use can be awkward, because during development of the handles, ergonomic criteria often were not taken into account. High mechanical strain demands a high standard in material quality to avoid instrument breakdown. For reasons of sterilization these instruments should be dismantlable. These highly technical conditions are accompanied by great expense [12]. Nevertheless, a standard laparoscopic instrument set should always be available. Laparoscopic microscissors and graspers are very useful in certain situations during laparoscopic procedures without pneumoperitoneum.

Conventional In absence of a pneumoperitoneum, simple rubber cannulae can be used. Theres no difficulty in introducing instruments with a curved distal end and larger diameter, e.g., the gallbladder grasper or the right angle, due to the flexibility of the trocar. It is advantageous that instruments without any comparable laparoscopic equivalent can be used. Table 1 shows instruments which have been used in gasless laparoscopic operations. Sufficient length of these instruments is an important factor for their use in gasless laparoscopy. Conventional instruments consist of only two parts, which are combined with a simple hinge. Their easy but robust construction has been established in surgery for many decades. In contrast to laparoscopic instruments, there is only one deflection of power. Due to this fact the surgeon has a better tactile sense, because the instrument is merely a prolongation of his finger. Using conventional instruments in gasless laparoscopy can mean that the opening of the jaw is restricted because the joint of the instrument concipated for open surgery is too far distal and interferes with the abdominal wall. This leads to a limited range of instrument movement. Specially designed conventional instruments for gasless laparoscopy have prolonged branches and a more proximally located joint. During the operation the surgeon will achieve the highest possible freedom of movement if the joint is positioned in the middle of the rubber trocar. The instrument set can be adjusted to the individual anatomy by using varied branch length and angle. Nevertheless, the ideal position of the rubber trocars relative to the operating field is a substantial prerequisite for efficient instrument handling. Furthermore, in selecting length, form, and joint position of an instrument, ergonomic criteria have to be considered, because they have great influence on the surgeons freedom of movement and fatigue. For instance, a

872

Fig. 5. Conventional forceps and ring forceps (Aesculap, Tuttlingen, Germany) allow effective clamping and retraction of the gallbladder. Fig. 6. Conventional right angle forceps (Aesculap). Grasping, dissecting, and effective hemostasis with electric cautery. Fig. 7. Different instrument tips of insulated conventional right angle forceps (Aesculap). Fig. 8. Conventional needle holder (Aesculap). Hinges allow various operation activity depths without hindrance while opening the branches. Fig. 9. Suction (Aesculap). The curved shape improves the intraabdominal range of motion.

permanent flexion of the surgeons wrist restricts instrument range and leads to early fatigue [7, 8]. Forceps and ring forceps with varied branch length and angle for clamping and retraction of the gallbladder are as effective as in open surgery (Fig. 5). Dissecting and grasping of fine structures can be achieved with the right angle (Fig. 6). Isolation of the instrument allows hemostasis with electric cautery. Instruments with tips in different sizes are available (Fig. 7). Branches of the needle holder are tight fitting and bent (Fig. 8). The placement of the hinge allows various operation activity depths by using rubber trocars without any hindrance while opening the branches. The

Suction has a curved shape to improve the intraabdominal range of motion. For effective hemostasis using electric cautery, the instrument is isolated (Fig. 9). The absence of a pneumoperitoneum allows an unlimited use of suction and irrigation without change of the intraoperative exposure.

Conclusion Laparoscopy with pneumoperitoneum facilitates a good view and a sufficient range of motion in the abdominal cavity through the extension of the abdominal wall. Com-

873

plications, which are related to gas insufflation, are rare but predictable. Physiological consequences and the extent of possible oncological disadvantages are still unclear. Given these considerations, alternative methods for mechanical distension of the abdominal wall have been explored. Several systems require an additional permanent low-pressure pneumoperitoneum, because the extent of intraabdominal exposure is not sufficient [2, 6, 14, 15]. The concept of laparoscopy with mechanical wall lifting in combination with a low-pressure pneumoperitoneum differs from conventional laparoscopy with a pneumoperitoneum only in how one achieves an intraabdominal space for adequate vision of the viscera. The expense of the installation of a pneumoperitoneum is expanded by additional system setup for mechanical retraction. It may help to prevent typical complications and provide additional exposure in isolated areas during laparoscopy with reduced gas insufflation. But there are no technical advantages with regard to instrumentation. The instruments still need to be introduced through sealed trocars to maintain the intraabdominal pressure. The construction of these instruments is hampered by their small diameter and length, because their shaft has to fit into the trocars. The tactile sensitivity of the surgeon using laparoscopic instruments with a hinge system or hingeless instruments is reduced to a great extent because of their axial-located mechanism, which is governed by an in-andout movement of a central rod. The action of the jaw needs to be transmitted either by two hinge systems, or by a hingeless system with elastic jaws. Action with both instrument designs has high friction and reduced force feedback. In combination with a low-pressure pneumoperitoneum, only laparoscopic instruments are useful. Several intraabdominal systems require an initial pneumoperitoneum for a safe introduction of intraabdominal retraction devices under endoscopic guidance [1, 5, 11]. These techniques do not minimize the risk of needle injuries. After system setup the pneumoperitoneum is decompressed. Abdominal wall lifting mostly creates in these systems a more tent-shaped suspension with limited intraabdominal exposure. The PelviSnake [16] (Fig. 1) creates a more dome-shaped suspension, but its application is limited to the pelvic region. Beside optic and instrument ports these systems require additional access for mechanical retraction. Conventional surgical techniques can be applied after system setup. Spreading trocars [4] combine intraperitoneal access with lifting the abdominal wall. But there is no technical advantage with regard to instrumentation. The Peritoneal Cavity Augmentation System [15] (Fig. 2) combines a low-pressure pneumoperitoneum with abdominal wall traction and mechanical pressure to internal organs and provides good exposure. Procedures without specimen removal can be performed with small laparoscopic instruments under local anesthesia. The concept of this retraction system focuses on a further reduction of invasiveness in combination with gas-insufflation. The concept of laparoscopy without pneumoperitoneum eliminates the need for intraabdominal pressure and gas tightness. Intraabdominal systems, the disposable fanshaped retractor [11] and the Modular Retraction System [9] (Fig. 3), are inserted under direct digital guidance together with the optic trocar over a small incision. Additional access for system setup is not required. Both systems provide a

Fig. 10. Methods to provide intraabdominal exposure.

planar suspension of the abdominal wall. The disposable fan retractor [3] is available in different designs for each abdominal quadrant. The intraperitoneal frame of the MORES retractor improves lateral exposure and a more domeshaped suspension with reduced tenting effects. The combination of its lifting parts provides individual adjustment to the patients anatomy and creates a retractor shape for every abdominal region. Compared to a pneumoperitoneum the Modular Retraction System (Gutt) allows application of mechanical pressure on internal organs. According to the needs of vision a translucent plastic membrane can be unfolded and placed for posterior organ retraction. Subcutaneous retraction systems do not require initial pneumoperitoneum; however, to improve exposure, CO2 insufflation can be added. In the absence of pneumoperitoneum, conventional instruments can be used. Setting up the system of subcutaneous wiring [14] is, in comparison to the Laparo Tenser (Fig. 4), more complicated. Looking only at the point of intraabdominal exposure, no mechanical retraction system can compete with a CO2 pneumoperitoneum. The advantage is in combination with microsurgical laparoscopic instruments performing limited procedures without specimen removal under local anesthesia. On the other hand, if a slightly limited exposure to the region of interest is accepted in the absence of intraperitoneal insufflation, surgical qualities which have been lost with the establishment of laparoscopic techniques are regained. From the practical standpoint the surgeon is allowed to combine the advantages of endoscopic surgery, such as magnified video-endoscopic visualization, with well-known open surgical techniques to create a symbiosis of proved methods (Fig. 10). For example, the use of conventional ring forceps, lap sponges, and insulated, bent right angles increases the radius of manipulation and improves the exposurefor example, of the gallbladder and the triangle of Callot. Preparation of fine structures and tissue with the surgeons habitual sense of touch is more efficient [7]. Instead of time consuming knot-tying techniques and expensive staplers, extracorporal knot-tying has proven to be a

874

fast and easy procedure. The surgeons finger palpating tissue and organ surface is the most sensitive instrument. In laparoscopy under isopneumic conditions there is no technical hindrance for the direct touch of the surgeons hand. Careful specimen removal is facilitated by using conventional forceps under constant laparoscopic control. Borders between laparoscopy and open surgery can be overcome. Shorter operating time and a limited technical expenditure is cost efficient. This may offer more patients the possible advantages of a less-invasive treatment. References
1. Araki K, Namikawa K, Yamamoto H, Mizutani J, Doiguchi M, Arai M, Yamaguchi T, Uno K, Ido Y, Hayashi N, Ogawa M (1993) Abdominal wall retraction during laparoscopic cholecystectomy. World J Surg 17: 105108 2. Banting S, Shimi G, Vander Velpen G, Cushieri A (1993) Abdominal wall lift: low-pressure pneumoperitoneum laparoscopic surgery. Surg Endosc 7: 5759 3. Chin AK, Moll FH, McColl MB (1993) Mechanical peritoneal retraction as a replacement for carbon dioxide pneumoperitoneum. J Am Assoc Gynecol Laparosc 1: 6266 4. Dragojevic B, Tomic D (1996) Multifunctional trocars as suspension devices for gas-free laparoscopic operations. Min Invas Ther Allied Technol 5: 9598 5. Francois Y, Mouret P (1992) Suspenseur de paroi et coelio-chirurgie. J Chir 129: 492493 6. Gazayerli MM (1991) The Gazayerli endoscopic retractor model 1. Surg Laparosc Endosc 1: 98100

7. Gutt CN, Held S, Voepel H, Paolucci V (1996) Instruments for gasless laparoscopic surgery. Min Invas Ther Allied Technol 5: 307312 8. Gutt CN, Voepel H, Linker R, Bamberg W (1996) Conventional surgical instruments for gasless laparoscopy. In: Paolucci V, Schaeff B (eds) Gasless laparoscopy in general surgery and gynecology. Thieme, Stuttgart, pp 138144 9. Gutt CN, Heinz P, Held S, Paolucci V, Encke A (1996) Modular retraction system (MORES) for gasless laparoscopy. Surg Endosc 10 (5): 584 10. Hashimoto D, Nayeem SA, Kajiwara S, Hoshino T (1993) Laparoscopic cholecystectomy: an approach without pneumoperitoneum. Surg Endosc 7: 5456 11. Kitano S, Iso Y, Tomikawa M, Moriyama M, Sugimachi K (1993) A prospective randomized trial comparing pneumoperitoneum and Ushaped retractor elevation for laparoscopic cholecystectomy. Surg Endosc 7: 311314 12. Melzer A, Buess G, Cushieri A (1992) Instruments for endoscopic surgery. In: Cushieri A, Buess G, Perissat J (eds) Operative manual of endoscopic surgery. Springer, Berlin, pp 1436 13. Nagai H, Konodo Y, Yasuda T, Kasahara K, Kanazawa K (1993) An abdominal wall-lift method of laparoscopic cholecystectomy without pneumoperitoneal insufflation. Surg Laparosc Endosc 3: 175179 14. Semm K, Lehmann-Willenbrock (1996) Pelvioscopy and laparoscopy without overpressurethe aspiration pneumoperitoneum. In: Paolucci V, Schaeff B (eds) Gasless laparoscopy in general surgery and gynecology. Thieme, Stuttgart, pp 2933 15. Schaller G, Engelke V, Manegold BC (1996) Mechanical augmentation of the peritoneal cavity in laparoscopic surgerya new instrument set. Min Invas Ther Allied Technol 5: 2124 16. Volz J, Ko ster S, Wei M (1996) Developments in gasless gynecologic pelioscopy. In: Paolucci V, Schaeff B (eds) Gasless laparoscopy in general surgery and gynecology. Thieme, Stuttgart, pp 108114

Surg Endosc (1997) 11: 877

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Endoscopic right thyroid lobectomy

Endocrine surgery is well defined by C. Proye: Attention a ` ce que vous enlevez, mais attention plus encore a ` ce que vous laissez [3]! Thus, in order to maximize the benefit from image magnification in the identification of the inferior laryngeal and parathyroid glands, we have successfully attempted a new endoscopic approach to thyroid surgery. A 30-year-old woman, operated 3 months previously for a malignant endocrine neoplasm of the body of the pancreas, was admitted with a 4-mm hypoechogenic nodule of the right thyroid lobe. Thyroid hormone values were normal; fine-needle aspiration biopsy was compatible with adenoma. Other investigations for multiple endocrine neoplasm I were negative. The patient underwent endoscopic right thyroid lobectomy on 8 July 1996. The lateral approach according to J. Marescauxs open technique was employed [2]. Three laparoscopic trocars were inserted under the platysma muscle at the anterior margin of the sternomastoid muscle: at the jugular notch (5 mm), at the angle of the mandible (5 mm), and midway between the other two, approximately 4 cm above the clavicle. A 30 endoscope was used. Carbon dioxide was insufflated at low pressure to create a space for the dissection. The medial margin of the sternocleidomastoid was dissected and the muscle was retracted laterally with external stitches. The inferior thyroid artery and inferior laryngeal nerve were easily visualized and dissected. The arm of a wall-lifter (Laparo Tenser, CHIO-MED s.r.l., Treviglio, Bergamo, Italy) was inserted through the jugular trocar to enhance the exposure without excessive CO2 insufflation. The strap muscles were divided with bipolar coagulation; the upper pole of the gland was freed and the external branch of the superior laryngeal nerve was identified and preserved. The superior and inferior thyroid vessels were clipped and divided. The two parathyroid glands were visualized and preserved. The inferior laryngeal nerve was dissected with an ultrasound dissector (Harmonic Scalpel, UltraCision, Smithfield, USA). The nerve was followed for all its cervical course up to the inferior constrictor of the pharynx. With the same instrument, small perineural vessels were divided and the gland was freed from its posterior attachments. Finally, section of the isthmus was performed with the bipolar coagulator. The specimen, measuring 3 cm by 2 cm, was inserted in a plastic bag (Endopouch, Ethicon Endo Surgery, USA) and extracted through the 10mm access. A suction drain was left in place and removed after 12 h. The procedure took 4 h and 45 min; blood loss was virtually nil. There were no intraoperative complications, and specifically no hypercarbia. The patient was ambulant on the same night of the operation. She had a normal voice and could drink sips of water. She didnt need analgesic medication. Surgical enfisema could be felt in the subcutaneous tissue limited to the neck, causing little or no discomfort. In this respect our experience differs from that described by Gagner following endoscopic parathyroidectomy [1]. We believe this is explained by the combined use of a wall lifter with low-pressure CO2 insufflation. The absence of hypercarbia can also be explained in this way. The patient was discharged home on the 2nd postoperative day. The cosmetic result was very satisfactory. Pathological examination showed the presence of a 3-mm papillary microcarcinoma, moderately differentiated with focal invasion of the capsule. Thyroid endoscopic approach is feasible, with adherence to the rules of endocrine surgery and with very good cosmetic result. Image magnification permits an excellent view of nervous and vascular structures and parathyroid glands. The difficulties encountered were due to the excessive length of instruments designed for laparoscopic surgery: the availability of shorter and more ergonomic instruments should facilitate the procedure. References
1. Gagner M (1996) Endoscopic subtotal parathyroidectomy in patients with primary hyperparathyroidism. Br J Surg 83: 875 2. Marescaux J (1991) Thyro de. In: Proye C, Dubost C, et al. (eds) Endocrinologie chirurgicale. Medsi/McGraw Hill, Paris, pp 1353 3. Proye C (1991) Principles ge ne raux de traitement en chirurgie endocrinienne. In: Proye C, Dubost C, et al. (eds) Endocrinologie chirurgicale. Medsi/McGraw Hill, Paris, pp 111

C. S. G. Hu scher S. Chiodini C. Napolitano A. Recher

Department of General Surgery Vallecamonica Hospital 25040 Esine (BS) Italy

History
Surg Endosc (1997) 11: 856863

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Exposure of the operative field in laparoscopic surgery

V. Paolucci, B. Schaeff, C. N. Gutt, G. S. Litynski
Department of General Surgery, Johann Wolfgang Goethe University, Theodor Stern-Kai, D-60590 Frankfurt am Main, Germany Received: 12 June 1996/Accepted: 8 January 1997

Abstract. Endoscopic surgery, as a result of over 90 years of investigation, has now become the most innovative part of general surgery; every procedure in the thoracic and abdominopelvic cavity, intraperitoneal or extraperitoneal, has been reviewed for feasibility. The basic principles in the management of surgical patients, however, have not changed: adequate exposure and good lighting remain important and may become more important with endoscopic techniques. Historical review shows the dependence of advances in laparoscopy upon technical development in the field of intraabdominal exposure as the result of two objectives: namely abdominal wall displacement and bowel retraction. Key words: Pneumoperitoneum Mechanical abdominal wall lifting Balloon dissection Isopneumic laparoscopy The main prerequisite for exposure in the abdominal cavity and at the same time one of the limiting factors in the development of laparoscopy was, throughout history, the pneumoperitoneum. The therapeutic intraperitoneal injection of air, oxygen, or nitrous oxide in the treatment of tuberculosis of the peritoneum began to be quite popular during the last century [15]. At the beginning of the 20th century, some principles of this technique and instruments were standardized. Based on these prior experiences, the German surgeon Georg Kelling (18661945, Dresden), in his experiments on living dogs, insufflated the abdominal cavity with air to a pressure of up to 100 mmHg and hoped in this way to develop a treatment of intraabdominal and intestinal bleeding. He called this Lufttamponade (air-tamponade) (Fig. 1) [16]. Two out of 20 dogs died during the experiment and the reason for their sudden death was not quite clear to Kelling. He couldnt know that high pressure in the abdomen kills by reducing the respiratory function and stopping the venous return to the heart. He tried twice to practice his

Correspondence to: V. Paolucci

new method on humans but, fortunately, they refused to be treated by air-tamponade. In 1901 Kelling published the results of his experiment in a German medical journal, where he also mentioned Koelioskopie for the first time as a closed-cavity endoscopy on a living dog by means of a 6-mm Nitze cystoscope [12]. He probably never performed a Koelioskopie on humans; nevertheless, laparoscopy was born, as a side effect of an unsuccessful experiment. Kelling also observed that during Koelioscopie the abdominal organs were much smaller than normally. This is not surprising if we know that this surgeon created a pneumoperitoneum of 5060 mmHg and once even 100 mmHg! This was the first description of laparoscopic retraction of the abdominal organs. Concerned about the problem of exposure and its consequences, the Swedish internist Hans Christian Jacobaeus (18791937) restricted laparoscopy at the beginning of the century to patients with ascites, but in the following years he expanded the intraabdominal space through the insufflation of room air without any control upon pressure [10]. Roger Korbsch of Oberhausen, Germany [13], after 8 years of experience and 300 laparoscopies, suggested in 1927 the use of CO2 instead of air for the inflation of the abdomen because it is easily and very quickly absorbed. He also recommended that the intraabdominal pressure should not exceed 15 cm H2O. For a long time, however, air was generally used in preference to oxygen, carbon dioxide, nitrogen, or other gases, as a medium of inflation, because it is always available and free. Anderson represented the typical point of view, writing in 1937 [2]: At first various gases were employed but later, filtered air; now we are using unfiltered air. The papers of Ruddock and Kalk played an important role in the establishment of air pneumoperitoneum. In 1957 Ruddock still recommended unfiltered air by means of a baumanometer bulb. Air makes a perfect visualization medium, and objects appear in their natural proportion and color without distortion [25]. Five years later Heinz Kalk (18951973), in his second book about laparoscopy, still preferred the injection of room air by using of a 500-ml syringe, still without pressure control [11] (Fig. 2).

857

Fig. 2. Syringe for creating pneumoperitoneum [11].

Various gases have been used or can be theoretically used:

Room air
Fig. 1. Kellings apparatus for creating air-tamponade (1901) [3].

The pathophysiology of pneumoperitoneum remained unclear for a long time. Respiratory distress and possible cardiac embarrassment were dependent upon whether the patient had a relaxed abdomen (had had children), was obese, etc. The injection of room air by syringe without intraabdominal pressure control was hazardous, and this was, perhaps, one reason for the slow development of laparoscopy. The rediscovery of laparoscopy in the United States in the late 1960s led to many studies on blood gas change during pneumoperitoneum. In 1969 Alexander and Brown found average changes in arterial pH and carbon dioxide. They recommended using nitrous oxide for creating pneumoperitoneum [1]. However, in 1974 Phillips found out that CO2 was the gas of choice for 91% and N2O for only 6% of the members of the AAGL [20]. The best exposure of the operative field in laparoscopic surgery can be obtained by means of insufflation of the abdomen with gas. Safe insertion of trocars and freedom of manipulation of the instruments are also important goals of exposure through gas insufflation. The ideal gas Currently, laparoscopy implies the use of CO2 pneumoperitoneum. However, the ideal gas for exposure in the abdominal cavity and retroperitoneum during endoscopic procedures has not been found. Solubility, diffusibility, combustibility, and possible pharmacologic side effects (for example, on lung and heart function, infections, as well as tumor growth) are the parameters of choice for use in humans.

Room air consists of up to 79% nitrogen. In various countries [29] it is still used for pneumoperitoneum, but is generally avoided because of its extremely low solubility in blood. Indeed, nitrogen is insoluble in water as well as chemically, and there is some danger of air embolism. Gas remaining in the abdominal cavity will not be absorbed and it is radiologically visible, as a subphrenic sickle, with the patients having typical complaints [22].

Oxygen (O2) Plain oxygen is a physiological gas, quite tissue compatible, 10 times as difficult to absorb as CO2 gas, and very explosive. Especially when used in cauterization, the formation of hydrogen and explosive gas, through electrolytic splitting, cost some patients and physicians their lives. It has been abandoned as an insufflant for laparoscopy [6].

Nitrous oxide (N2O) Nitrous oxide (laughing gas) has an anesthetic effect. For this reason, it was mostly used in diagnostic upper abdominal laparoscopy under local anesthesia (N2O-Pneu by Siede, System Semm 1967). It is less suitable for operative laparoscopy. On one hand, it dissociates with increasing temperature in N and O2 (laughing gas danger); on the other hand, it can trigger a large number of undesirable systemic effects. Nitrous oxide, usually considered safe for diagnostic laparoscopy, has been associated with fatal explosions during the use of electrocautery [5, 23]. A recent experimental work, however [9], did not confirm the risk of combustion in the presence of N2O under normal conditions.

858

Fig. 3. Physical effects of an intraabdominal pressure of 12 mmHg/cm2.

Fig. 4. Palmers peritoneostat (1947) [17].

Inert gases Helium, xenon, and argon are inert, water-insoluble monoatomic gases. Under pneumoperitoneum conditions using these gases, the troublesome fogging up of the scope is not possiblea very small advantage because, on the other hand, the danger of gas embolism is high. Helium has been suggested as an alternative for pneumoperitoneum because it is not flammable, is nontoxic, and is physiologically inert. It has been used clinically for insufflation [3]. Experimental studies (lit. in [29]) proved helium to be dangerous as the agent for gas embolism in the dog. Furthermore, intravascular helium in humans (secondary to rupture of a heliumfilled intraaortic balloon) has been reported to produce neurologic deficits, probably because of vascular occlusion by gas bubbles (lit. in [29]). In addition, the gas is expensive and noneconomical, meaning commercially not available. Any authors pointed out advantages of pneumoperitoneum with helium relative to CO2 [21]. Further investigations must be done before we abandon this interesting medium. Carbon dioxide (CO2) Carbon dioxide is a common product of metabolism. It dissolves very well in the blood fluid physically and chemically. Therefore, the danger of gas embolism is very low. CO2 doesnt support combustion. So this gas is commonly used for pneumoperitoneum. Adverse pharmacologic effects are of concern. Besides hypercarbia, which is well tolerated in most patients, a positive effect on tumor cell growth and on bacterial infections seems to have been experimentally proven [28]. Recently the use of 32C heated CO2 for insufflation became possible. The use of warm CO2 produced a significant reduction in postoperative shoulder and diaphragm pain [14]. Whether heating could also modify other undesired pharmacologic effects of carbon dioxide is now under investigation. The ideal gas pressure The higher the pressure, the better the view. This longheld axiom doesnt point out that, under pneumoperitoneum

conditions, potential danger to the patients safety and optimal operative exposure essentially have an antinomic relationship. Comparing the abdominal cavity with a hemisphere of 30 cm in diameter, at a pressure of 12 mmHg/cm2 the force applied onto the total surface of the parietal peritoneum will reach approximately 33 kps. The abdominal wall will be lifted with a force of approximately 11 kp. The remaining forces will work in equal parts laterally onto the diaphragm, resulting in a further enlargement of the intraabdominal space and downward onto the viscera resulting in a retraction of bowel loops and stomach (see Kellings observation) (Fig. 3). Undesired but unavoidable side effects are reduction of the respiratory minute volume and compression of the vena cava with obstruction of venous blood return to the heart. These side effects limit the overpressure in the abdominal cavity to approximately 12 mmHg. This results in a good expansion of the gas bubble. Furthermore, higher pressures hardly lead to an effective enlargement of the gasfilled abdominal cavity, even in obese patients. Whenever the extension reserve of the abdominal muscles is reached, the abdominal wall resistance to additional expansion increases overproportionally. In this context, it must be remembered that intraabdominal pressure is a function of abdominal wall muscle tone. A complete and constant muscle relaxation is not only the presupposition for optimal intraoperative view; prevention of gas-pressure-related complications is also necessary. In order to ensure a constant pressure of 12 mmHg in the abdominal cavity, an automatic gas insufflation device became necessary.

The CO2-insufflators The first laparoscopist to make pneumoperitoneum the subject of scientific study was Roul Palmer (19041985) in Paris. In 1943 Palmer built his first insufflator [17] (Fig. 4). It produced pneumoperitoneum at the rate of 300 to 500 ml of carbon dioxide per minute. Palmer emphasized that the intraabdominal pressures should not exceed 25 mmHg. In the late 1950s, Frangenheim designed his first insuf-

859

Fig. 5. Monofilar-Bivalent-System acc. to Semm (WISAP, Sauerlach b. Mnchen, Germany). This system must deliver 7 l of carbon dioxide gas per minute. Due to the Monofilar-Bivalent System (MBS) gas insufflation and static pressure estimations may be measured through the same opening in the Veress needle, or later, in the trocar sheath.

Fig. 6. Abdominal Cavity Expander (ACE-WISAP) after Semm. This Tshaped retractor takes over the one-point lifting of the abdominal wall in addition to the pneumoperitoneum.

flator prototype; he modified for that purpose the CO2 part of an anesthetic machine [16]. In 1963 Kurt Semm (b. 1927) finally developed the automatic insufflator CO2-Pneu that monitored abdominal pressure and gas flow and allowed more complex laparoscopic procedures to be performed. Since exchange of multiple instruments and specimen extraction were associated with loss of gas, Semm in 1979 perfected his apparatus, creating an electronic insufflator; fully monitored pneumoperitoneum became possible [27]. The foundament for the automatic regulation is the Monofilar-Bivalent-System (MBS). The gas refill stream regulates itself by measuring the actual intraabdominal pressure through the same opening. The intermittent electronical measurement of the static pressure, meaning the actual intraabdominal gas pressure (CO2-PNEU-Electronic), provides the consistancy of the gas bubble size through regulation of the gas refill stream, in case of gas loss (Fig. 5). Occasionally it will meet certain limits though. In many cases, CO2 overpressure laparoscopy is still the adequate procedure. For patients who are cardially impaired or have hiatal hernias, for example, an intraabdominal pressure of 12 mmHg may already be too high. In addition, extensive opening of the retroperitoneum, e.g., in parailiac lymph-node dissection, could cause cardiopulmonary problems. Injured small veins, not bleeding, due to the overpressure, could absorb large amounts of gas. In other cases a pressure of 12 mmHg is not sufficient to create adequate sight in the abdomen, e.g., in very obese patients, in patients with intestinal distension, in the case of extensive abdominal adhesions, or when for anesthesiological reasons sufficient muscle relaxation is not possible. Especially at the lateral borderlines of the gas dome, handling ability is reduced, so efficient operating is not possible. There could be problems in maintaining the pressure in the abdominal cavity as well. The reason for this could be

insufficient sealing of the trocars or the necessity for suction of large amounts of fluid (blood, irrigation liquid) within a short period of time. Especially during minimally invasive surgery of the upper abdomen, the abdominal cavity collapses very quickly after a loss of pressure.

Mechanical abdominal-wall lifting systems To compensate for this loss of pressure as well as to optimize the intraoperative view in a particular region of the peritoneum, the concept of additional point lifting of the abdominal wall was developed (Semm, Gazayerli) (Fig. 6). Point lifting involves the application of a retractor with a hinged distal blade, 7.5 cm in length, to pull up the abdominal wall at a single point and augment the lift of gas insufflation whenever necessary. The point lifting system is not a replacement for pneumoperitoneum; rather, it is used as an adjunct to CO2 insufflation. Isopneumic laparoscopy The increasing number of laparoscopies between the 1960s and the 1990s led to numerous reports about the complications of laparoscopy. Much data has been accumulated; the most frequent complications were specifically associated with creation and maintenance of the pneumoperitoneum. Pneumoperitoneum can be replaced by alternative methods of abdominal wall lifting for the purpose of creating intraabdominal space to work under isopneumic conditions. The first one who had the idea of viewing the abdomen without pneumoperitoneum was Bernheim in 1911 [15]. In the modern era it was Mu he, who performed laparoscopic cholecystectomy without pneumoperitoneum as early as 1985: after the first six operations we changed the method

860

Fig. 7. A The single-point as well as the linear lifting of the abdominal wall results in a narrow pyramid or tent which provides insufficient space inside. B A better work space is achieved following the planar concept of abdominal wall retraction. Fig. 8. A The planar lifting system results in a trapezoidal cavity, with the lateral wall dropping off the distal end of the fan blades. B The electrically powered arm (Laparolift Origin, Menlo Park, CA, USA) is secured on the side of the table at shoulder level and, connected with the intraabdominal fan retractor (Laparofan), lifts the upper abdomen in a shape similar to a truncated pyramid. The abdominal parts toward the ribs bow are not retracted at all. Fig. 9. Final setup of abdominal wall lifting with subcutaneous wiring and adjustable plate lifting according to Hashimoto. The trapezoidal shape of the retracted abdominal wall is evident. Fig. 10. Under particular circumstances, the lowered ceiling and angled sides of every kind of mechanical abdominal lifting, together with bowel distension, may really disturb the operative view and the freedom of manipulation. Fig. 11. Four types of curved retractor blades for each abdominal quadrant. They fit to the Laparolift arm origin, and are inserted into the abdominal cavity.

861

and the remaining 88 patients were operated on using a simplified approach, namely laparoscopic cholecystectomy without pneumoperitoneum [15]. Finally, in the early 1990s, numerous physicians working independently published their experiences about the so-called gasless technique. [8, 19] Mechanical abdominal wall lifting has evolved from early concepts of additional point lifting (Semm, Gazayerli) through the realization of linear lifting (Bainting and Cuschieri, Araki, Maher, Nagai, Kitano) to the final solution of planar lifting (Mouret, Hashimoto, Chin and Moll, Gutt and Paolucci). Particularly in Japan, investigators developed retractors that provide a line of lift, along cables or wires inserted through the abdominal wall in two locations and suspended from rigid frames or winches above the patient. In some cases, cables are inserted at multiple points to provide sufficient lift for endoscopic exposure. The planar concept of abdominal wall retraction combines multiple point or linear lifting schemes into a single retractor that elevates the abdomen in a planar fashion (Fig. 7A,B). The so-called Laparolift (Origin Medsystems, Inc., Menlo Park, CA, USA) is used most often. In the original version it employs a two-legged retractor which is inserted via a 15-mm incision and fans out to form a triangular plane within the abdomen (Fig. 8A). The fan retractor is elevated by a motorized mechanical arm, which controls the amount of abdominal wall displacement achieved during the procedure. Advantages of mechanical lifting include conventional surgical instrument use, a lack of complications due to pneumoperitoneum, the potential for laparoscopy under regional or local anesthesia, the ability to apply constant suction aspiration, finger access for transabdominal palpation, and procedural cost savings associated with the use of simple valveless access ports [18]. Limitations noted during laparoscopic abdominal lift procedures generally involve insufficient visualization and working space. All of the current isopneumic systems produce pyramidal or trapezoidalshaped distension of the abdominal wall (Figs. 8B and 9). In this way exposure is limited to the abdominal quadrant of interest. Therefore inspection of the entire abdomen requires multiple repositioning of the retractors during operative procedures. Furthermore, lateral exposure may be difficult to obtain, particularly in obese patients. While no pressure is exerted on the diaphragm, the cranial exposure (esophageal hiatus, left liver lobe, splenic region) is insufficient [19]. Pneumoperitoneum provides dome-shaped exposure which is given in all four abdominal quadrants, allowing inspection of the entire abdomen at once without time-consuming repositioning of the retraction device; the lowered ceiling and angled sides of mechanical abdominal lifting may be claustrophobic for the surgeon (Fig. 10). Also, the introduction of retractors with curved hooks, especially designed for each abdominal quadrant (Fig. 11), didnt solve the problem of exposition in obese patients or those with strong abdominal muscles. Balloon technology Application of balloon technology to endoscopic surgery began with retroperitoneal dissection for urologic indica-

tions, using a surgical glove tied to a rubber catheter and inflated with a pneumatic squeeze bulb [7]. This concept has been successfully applied to the technique of preperitoneal balloon dissection for all existing extraperitoneal endoscopic procedures. That was the background for the extended balloon applications described below. Inflatable balloon lift Recently, the development of an inflatable balloon lift (Fig. 12A,B) addressed some of the limitations identified in the available abdominal wall retraction systems [4]. The balloon lift provides a large surface area of contact with the abdominal wall, decreasing the pressure exerted on the tissue. In addition, compressibility of the inflated structure imparts softness to the lifting mechanism, resulting in atraumatic abdominal wall displacement. The plane of lift includes the abdominal wall superior and inferior to the entry point. The balloon lift provides a laparoscopic cavity with better lateral exposure. The expanded intraabdominal space formed by the inflatable lift permits bowel to migrate from the operative field, resulting in improved, but not yet optimal, surgical access. Bowel loop retraction therefore remains an unsolved problem in isopneumic laparoscopy. While under isopneumic conditions the compression of the intestinal lumen is missing, bowel distension becomes disturbing. The bulk of the intestinal mass has to be displaced from the visual field, requiring ancillary bowel retractors. Only two of 18 known mechanical lift systems try to address this problem, combining abdominal wall traction with simultaneous retraction of the viscera [19] (see article in this issue); in this way, partially obviating limitations of work space through blowing intestinal loops and stomach. Inflatable retractor Another way to obtain a better exposition of the operative field under isopneumic conditions involves use of inflatable retractors [4]. An inflatable retractor with a broad, flat surface is useful for control of distended bowel which has migrated to the operative site. A balloon retractor is particularly applicable to gasless laparoscopy, as it allows a large retraction surface to be introduced through a small incision. Unlike currently available bowel retractors of very old conception, which contain multiple blades (Fig. 13), the inflatable retractor consists of a single continuous surface. Bowel may creep through the spaces between separate retractor blades; this does not occur with the balloon retractor, which has the shape of an elliptical disc. The balloon surface is planar with the axis of the handle. This instrument is very useful in controlling bowel distention by applying compression to the surface of the bowel mass (Fig. 14). A second version employs a balloon surface perpendicular to the axis of the handle. This retractor can be used to rake in loops of bowel and displace it from the surgical site. Structural balloon A large, space-occupying balloon may be used to provide simultaneous abdominal wall displacement and bowel re

862

Fig. 12. A The inflatable balloon lift consists of an inflatable, flat, circular disc with a central hole to insert the laparoscope. The deflated balloon device is encased in a perforated sleeve to present a smooth profile at insertion. B The balloon lift is suspended from the mechanical arm by a cable system and extends radially from the point of insertion. It provides a better lateral exposure. Fig. 13. Robinson and Fiske (Santa Barbara, CA) 1941 presented this retractor for displacing viscera during peritoneoscopic examination [24]. The close resemblance to the modern retractors is very impressive. Top to bottom: blades of retractor in collapsed position; blades of retractor spread; collapsed blades covered with ligated latex rubber bag; retractor blades spread within elastic latex bag.

Fig. 14. Inflatable retractor. Multiple baffles are built into the balloon, to permit a higher inflation pressure and increased balloon rigidity. The high-strength nonelastomeric material used in balloon fabrication allows it to assume a thin profile for functional organ displacement without excessive retractor bulk. Air inflation of the balloon retractor provides a lightweight, cushioned contact surface for atraumatic contact with bowel and intraabdominal organs (ORIGIN Medsystems, Inc., Menlo Park, CA, USA). Fig. 15. A An example of a structural balloon (Origin Medsystems Inc.). B Schematic representation of the intraabdominal situation. The internal cavity of the balloon is inflated to form a laparoscopic working space. The outside struts are then inflated to maintain the space as instruments are passed through the window areas on the top and the front side.

traction in laparoscopic surgery (Fig. 15A,B). The structural balloon incorporates two components: one component is a large, single-cavity balloon which provides initial displacement of the abdominal wall, and the second component is an

inflatable strut system which maintains the peritoneal working cavity. Together, these systems should replace gas insufflation and permit laparoscopic surgery to occur without the use of valved trocar ports. The single-cavity balloon is

863

inflated first, to separate the abdominal wall from the underlying organs. The strut system is a separately inflatable framework bonded to the outside of the single-cavity balloon. It is inflated after the large inner balloon has formed the working cavity. The function of the strut system is to maintain the working cavity while allowing surgical instruments to pass through the membrane of the deflated inner balloon. Access incisions made between the inflatable struts allow instrument passage to intraperitoneal organs. The structural balloon is an intraabdominal analog of the mechanical lifting system. The supporting structure is contained within the belly, as opposed to outside lifting devices. Its probable advantages compared with present devices include the absence of outside structures which may impede instrument movement, the increased intraperitoneal contact area of lift, the softness of the strut system, and the ability of the structure to compress bowel and form a larger working space. Theoretical disadvantages of the structural balloon include the potential for balloon puncture during instrument use with subsequent loss of exposure, limitations on locations of instrument access sites, and the extra time required to accomplish instrument placement. Until now the structural balloon has been used only for extraperitoneal procedures. Conclusion The over-90-year-long developments outlined above address limitations in presently available techniques of exposure of the operative field in laparoscopic surgery. In general, the goal of ongoing research efforts is to provide improved peritoneal exposure for visualization and surgical manipulation, with a high degree of gentleness in tissue retraction and without local or general side effects. The search for the ideal gas for pneumoperitoneum is not yet completed. A revisitation of nitrous oxide and helium from the point of view of the modern minimally invasive surgical procedure seems to be justified. Abdominal-wall lifting techniques, without compromising desirable features incorporated in contemporary devices, have to solve the central problem of insufficient exposure in extreme situations. Inflatable structures seem to offer considerable potential. Advancements in material properties and processing have allowed balloon structures to support high loads while retaining desirable atraumatic surface characteristics. Laparoscopic surgery has reached a plateau. Continued research will allow novel clinical applications to be created, and new technical challenges to be met. This will undoubtedly extend the effectiveness of therapy in the field of minimal access surgery. References
1. Alexander GD, Brown EM (1969) Physiologic alterations during pelvic laparoscopy. Am J Obstet Gynecol 105: 10781081 2. Anderson ET (1937) Peritoneoscopy. Am J Surg 35: 136139 3. Bongard FS, Pianim N, Liu SY (1991) Using helium for insufflation during laparoscopy. JAMA 266: 3131

4. Chin AK, Moll FH (1996) Future technical developments in abdominal wall lifting. In: Paolucci V, Schaeff B (eds) Gasless laparoscopy in general surgery and gynecology. Thieme Medical, New York, pp 150 155 5. El-Kady AA, Abd-el-Razek M (1976) Intraperitoneal explosion during female sterilisation by laparoscopic electrocoagulation. A case report. Int J Gynaecol Obstet 14: 487488 6. Fevers C (1933) Die Laparoskopie mit dem Cystoskop. Med Klinik 31: 10421045 7. Gaur DD (1992) Laparoscopic operative retroperitoneoscopy: use of a new device. J Urol 148: 11371138 8. Gazayerli MM (1991) The Gazayerli endoscopic retractor model 1: brief clinical report. Surg Laparosc Endosc 1: 98100 9. Hunter JG, Staheli J, Oddsdottir M, Trus T (1995) Nitrous oxide pneumoperitoneum revisited. Is there a risk of combustion? Surg Endosc 9: 501504 10. Jacobaeus HC (1910) Ueber die Mo glichkeit die Zystoskopie bei Untersuchung sero ser Ho hlungen anzuwenden. Mu nch Med Wochenschr 57: 20902092 11. Kalk H, Wildhirt E (1962) Lehrbuch und Atlas der Laparoskopie und Leberpunktion. 2 Auflage. Georg Thieme Verlag, Stuttgart 12. Kelling G (1901) Die Tamponade der Bauchho hle mit Luft zur Stillung lebensgefa hrlicher Intestinalblutungen. Mu nch Med Wochenschr 48: 14801483, 15351538 13. Korbsch R (1927) Lehrbuchund Atlas der Laparo- und Thorakoskopie. JF Lehmanns Verlag, Mu nchen 14. Korell M (1996) Postoperative pain intensity after laparoscopy. In: Paolucci V, Schaeff B (eds) Gasless laparoscopy in general surgery and gynecology. Thieme Medical, New York, pp 3438 15. Litynski GS (1996) Highlights in the history of Laparoscopy. Barbara Bernert Verlag, Frankfurt am Main 16. Litynski GS, Schaeff B, Paolucci V (1996) Vom Pneumoperitoneum bis zur Koelioskopie. Georg Kelling (18661945)der Pionier der Laparoskopie. Chirurg 67: 283287 17. Palmer R (1947) Instrumentation et technique de la coelioscopie gyne cologique. Gyne col Obste t (Paris) 46: 420431 18. Paolucci V, Gutt CN, Schaeff B, Encke A (1995) Gasless laparoscopy in abdominal surgery. Surg Endosc 9: 497500 19. Paolucci V, Schaeff B, Gutt CN (1995) Gasless laparoscopywhy and how? Min Invas Ther 4: 165172 20. Phillips JM, Keith D, Keith L (1974) Gynecological laparoscopy 1973: the state of the art. In: Phillips JM (ed) Gynecological laparoscopy: principles and techniques. Stratton Intercontinental Medical Book, Baltimore, pp 113 21. Rademaker BMP, Bannenberg JJG, Kalkman CJ, Meyer DW (1995) Effects of pneumoperitoneum with helium on hemodynamics and oxygen transport: a comparison with carbon dioxide. J Laparoendosc Surg 5: 1520 22. Riedel HH, Semm K (1979) Das postpelviskopische (laparoskopische) Schmetzsyndrom. Arch Gyna k 228: 283284 23. Robinson JS, Thompson JM, Wood AW (1979) Fire and explosion hazards in operating theatres: a reply and new evidence. Br J Anaesth 51: 90 24. Robinson S, Fiske LG (1941) An instrument for retraction of viscera during peritoneoscopy. West J Surg Obstet Gynecol 71: 284288 25. Ruddock JC (1957) Peritoneoscopy: a critical clinical review. Surg Clin North Am 37: 12491260 26. Semm K (1979) New methods of pelviscopy (gynecologic laparoscopy) for myomectomy, ovariectomy, tubectomy and adnectomy. Endoscopy 2: 8593 27. Semm K, Lehmann-Willenbrock E (1996) Pelviscopy and laparoscopy without overpressurethe aspiration pneumoperitoneum. In: Paolucci V, Schaeff B (eds) Gasless laparoscopy in general surgery and gynecology. Thieme Medical, New York, pp 2933 28. Volz J, Ko ster S (1996) The effects of pneumoperitoneum on intraperitoneal tumor implantation in nude mice. Gynaecol Endosc 5: 193 196 29. Wolf JS, Carrier S, Stoller ML (1994) Gas embolism: helium is more lethal than carbon dioxide. J Laparoendosc Surg 4: 173177

Surg Endosc (1997) 11: 848849

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic splenectomy for splenic hamartoma

T. Yoshizumi,1 Y. Iso,1 C. Yasunaga,1 S. Kitano,2 K. Sugimachi3
1 2

Department of Surgery, Saiseikai Yahata General Hospital, 5-9-27 Harunomachi, Yahata Higashi-ku, Kitakyushu 805, Japan Department of Surgery I, Oita Medical University, 1-1 Hazama-machi, Oita 879-55, Japan 3 Department of Surgery II, Faculty of Medicine, Kyushu University, Fukuoka 812-82, Japan Received: 12 January 1996/Accepted: 22 March 1996

Abstract. Advances in imaging techniques have made preoperative diagnosis of splenic tumors possible. A case of successful laparoscopic splenectomy for splenic hamartoma is described here and the indications of this technique are discussed. Key words: Laparoscopic splenectomy Splenic hamartoma Indication

Laparoscopic splenectomy has been performed safely for the treatment of immune thrombocytopenia purpura (ITP) [2, 3, 5, 7] and hereditary spherocytosis [5]. This procedure, however, has not been described for the treatment of splenic tumors. Herein, we report a case of successful laparoscopic splenectomy for splenic hamartoma.

Case report
A 45-year-old man with a 6-month history of left upper quadrant abdominal and back pain was admitted to our hospital for the evaluation of a splenic mass detected by ultrasonography. Physical examination did not reveal an abdominal mass, hepatomegaly, or lymphadenopathy. Laboratory data were as follows: WBC 4,200/mm3 with no atypical cells, hematocrit 43.3%, platelets 186 103/mm3, serum carcinoembryonic antigen 1.7 (normal <5) ng/ml, and serum carbohydrate antigen 19-9 10 (normal <60) U/ml. Blood chemistry and other laboratory data were within normal limits. Ultrasonography demonstrated an 8 5 cm isoechoic mass in the midportion of the spleen. Computed tomography (CT) demonstrated an indiscrete mass without calcification (Fig. 1). Gallium scintigraphy failed to exhibit any abnormal uptake in the spleen. Selective splenic angiography revealed a well-defined tumor stain with neovascularity which occupied most of the spleen (Fig. 2). The splenic artery and vein were dilated. Barium enema and upper gastrointestinal series were negative.

Based on the preoperative workup, the splenic tumor was thought to be benign. The most common benign tumor of the spleen is the hemangioma [1, 6]. Despite its benign nature, these tumors do have the potential to rupture or undergo sarcomatous degeneration [6]. We therefore judged a splenectomy to be indicated. After obtaining informed consent from both the patient and his family, we prepared the patient for laparoscopic splenectomy. The operative techniques paralleled those described by Hashizume et al. [5]. Briefly, the patient was placed in the right lateral semidecubitus and Fowlers position under general anesthesia. A 10-mm trocar was inserted through a supraumbilical skin incision, and carbon dioxide was insufflated to create and maintain pneumoperiotoneum of 1012 mmHg. Three other trocars (one 12-mm and two 10-mm) were then placed along the left costal margin under direct visualization. Ligamentous attachments around the spleen were either dissected by an electrocautery or divided between metal clips. The splenic artery and vein were exposed near the hilus and individually doubly ligated with 2-0 silk. The splenic hilus was then divided using an Endo GIA 30 (US Surgical Co., Ltd., Norwalk, CT). The resected spleen was placed in a nylon surgical bag and delivered out of the peritoneal cavity through a left subcostal incision which was extended laterally for 3 cm. The operation time was 305 min and the estimated blood loss was 450 ml. The spleen weighed 192 g and measured 8.0 5.0 5.5 cm. The cut section exhibited a 6.0 3.8 cm well-circumscribed mass, which was diagnosed histologically as a hamartoma (Fig. 3). His postoperative recovery was uneventful. The patient returned to work on postoperative day 38 and has remained well.

Discussion Laparoscopic general surgical procedures have been performed increasingly after the report of laparoscopic cholecystectomy [4, 8, 9]. Splenectomy using laparoscopic devices was first reported in 1992 [2, 3], and the indications for this technique remain controversial [2, 3, 5, 7]. Carroll et al. proposed that this procedure should not be performed for patients with splenomegaly, uncontrolled bleeding diathesis, or overt malignant disease [2]. Delaitre and Maignien reported that laparoscopic splenectomy should be restricted to patients with hematologic splenic diseases without splenomegaly [3]. This is based on their method of fracturing the spleen within a bag to avoid an extension of the skin

Correspondence to: T. Yoshizumi, Department of Surgery II, Faculty of Medicine, Kyushu University, Fukuoka 812-82, Japan

849

Fig. 3. Cut surface of the resected spleen. The tumor is well demarcated and exhibits hemorrhage and marked vascularity.

lymphoma, leukemia, angiosarcoma, and metastatic tumors are extremely rare. Recent advances in imaging techniques have allowed preoperative detection of splenic tumors. Such tumors are diagnosed during a workup for splenomegaly or for the evaluation of other organs. As to laparoscopic splenectomy for tumors, the spleen needs to be removed intact to avoid peritoneal dissemination of tumor cells and to allow adequate histological evaluation, especially when malignancy cannot be ruled out. We believe that laparoscopic techniques can be performed using this principle for most splenic tumors. References
1. Bostick W (1945) Primary splenic neoplasm. Am J Pathol 21: 1143 1165 2. Carroll BJ, Philips EH, Semel CJ, Fallas M, Morgenstern L (1992) Laparoscopic splenectomy. Surg Endosc 6: 183185 3. Delaitre B, Maignien B (1992) Laparoscopic splenectomytechnical aspects. Surg Endosc 6: 305308 4. Dubois F, Berthelot G, Levard H (1989) Cholecystectomie par coelicoscopie. Presse Med 18: 980982 5. Hashizume M, Sugimachi K, Kitano S, Shimada M, Baba H, Ueno K, Ohta M, Tomikawa M (1994) Laparoscopic splenectomy. Am J Surg 167: 611614 6. Husni E (1961) The clinical course of splenic hemangioma. Arch Surg 83: 681688 7. Lefor A, Melvin W, Bailey R, Flowers J (1993) Laparoscopic splenectomy in the management of immune thrombocytopenia purpura. Surgery 114: 613618 8. Reddick EJ, Olsen DO (1989) Laparoscopic laser cholecystectomy. Surg Endosc 3: 131133 9. Roesel R (1989) Experience with mini-invasive endoscopic removal of gallstones. Surg Endosc 3: 124125

Fig. 1. Preoperative computed tomogram demonstrates an indiscrete low density mass (arrows) in the slightly enlarged spleen. Fig. 2. Selective splenic digital subtraction angiography demonstrates a large hyperevascular mass (arrows) which occupies most of the spleen.

incision. Lefor et al., on the contrary, support the extension of the incision to allow complete pathologic evaluation of the specimen and to avoid inadvertent splenosis by spillage of splenic tissue into the peritoneal cavity [7]. Splenic tumors are rare in which cysts and benign tumors including hemangioma, lymphangioma, and hamartoma are the main pathology, while malignant tumors such as

Surg Endosc (1997) 11: 850851

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic splenectomy in a Jehovahs Witness with profound anemia

G. S. Ferzli, J. B. Hurwitz, M. A. Fiorillo, N. E. Hayek, F. A. Dysarz, T. Kiel
Department of Laparoendoscopic Surgery, Staten Island University Hospital, 78 Cromwell Avenue, Staten Island, NY 10304, USA Received: 29 March 1996/Accepted: 4 June 1996

Abstract. Open surgery in a severely anemic patient may be complicated by a substantial blood loss from a large incision and subsequent poor wound healing secondary to the anemia. We report our success in performing a splenectomy laparoscopically in a profoundly anemic patient. A 50-year-old white male Jehovahs Witness who was HIV positive was referred for splenectomy after he developed profound, worsening anemia secondary to hypersplenism that was refractory to medical management. His preoperative hemoglobin and hematocrit levels were 2.7 g/dl and 8.8%, respectively, but his religious beliefs precluded transfusion. A laparoscopic splenectomy by the posterior gastric approach was performed. The patient tolerated the surgery well and experienced no additional morbidity. On postoperative day 7, his hemoglobin and hematocrit were 6.8 g/dl and 22%, respectively. We conclude that laparoscopic splenectomy is an attractive procedure in a severely anemic patient who requires splenectomy and refuses blood transfusion. Key words: Laparoscopic splenectomy Anemia Jehovahs Witness

and hematocrit levels were 2.7 g/dl and 8.8%, respectively, and had been decreasing. His platelet count was 22,000/ml. Preoperative correction of his anemia was not possible because the patient refused blood transfusions or splenic irradiation owing to his religious convictions. Splenectomy was performed laparoscopically, by a posterior gastric approach [4]. The entire operation lasted 96 min. Estimated blood loss was 20 cm3. A cell saver was available but not needed. Splenic volume was 3,252 cm3. The patient was discharged on postoperative day 4. On postoperative day 7, his hemoglobin and hematocrit were 6.8 g/dl and 22%, respectively. He has continued to do well, with no additional morbidity.

Discussion Operations on severely anemic patients should only be considered in life-threatening situations. Jehovahs Witnesses generally refuse all allogenic blood products and any autologous blood that has been separated from the body. These prohibitions still allow the majority to accept the use of cardiopulmonary bypass, dialysis, and intraoperative blood salvage and reinfusion. In the last case a closed circuit must be maintained [3]. As a matter of conscience some may accept products that contain blood products such as immune globulin. Epoetin alfa, which contains a small amount of human albumin, is acceptable to most. Every effort should be made to augment red cell volume before an operation is undertaken. Two widely used protocols involve iron dextran and epoetin (EpogenAmgen, Thousand Oaks, CA) [6]. The patient presented in this report had been treated with these protocols both at our institution and at the referring one, but paradoxically his red cell volume had decreased. Due to the absence of further options acceptable to him, splenectomy was considered to be lifesaving. Though not performed with ideal preoperative preparation, the case demonstrated two important points. The first was the ability of a severely anemic patient to tolerate prolonged CO2 insufflation at a pressure of 15 mmHg. We are not aware of any literature that addresses the effect of pneumoperitoneum in profound anemia; however, the general response to pneumoperitoneum has been widely discussed. No significant alteration in any hemodynamic parameter was noted in this patient either intra- and postoperatively.

Stimulated by the worldwide success of laparoscopic cholecystectomy, laparoscopic splenectomy (LS) was introduced to the general surgical community in 1992 [1]. LS is an attractive alternative to the standard open approach, and many variations on the original method have been described [2, 5]. This report describes a new application for LS. Case report
A 50-year-old white male Jehovahs Witness was referred to our institution for splenectomy; he was HIV positive and had profound anemia secondary to hypersplenism caused by B-cell hyperplasia. His admission hemoglobin

Correspondence to: G. S. Ferzli

851

The second point of note in this case relates to the appearance of bleeding, as viewed laparoscopically, in a severely anemic patient. Fresh bleeding appears as pink irrigation fluid. Significant bleeding can thus occur without the operator visualizing blood or dark red fluid. If this blood loss goes unrecognized, the patient may lose significant intravascular volume, and because of the preexisting anemia, significant oxygen-carrying capacity. The ultimate role of laparoscopic splenectomy has yet to be defined. This report demonstrates that the procedure can be safely performed in a severely anemic patient. It may be the preferred method when additional blood loss from a large incision must be avoided and when postoperative wound healing may be compromised by the anemia.

References
1. Carroll BJ, Phillips EH, Semel CJ, Fallas M, Morgenstern L (1992) Laparoscopic splenectomy. Surg Endosc 6: 183185 2. Delaitre B, Maignien B, Icard P (1992) Laparoscopic splenectomy. Br J Surg 79: 1334 3. Dixon JL, Smalley MG (1981) Jehovahs Witnesses. The surgical/ ethical challenge. JAMA 246: 24712472 4. Ferzli GS, Fiorillo MA (1995) A posterior gastric approach to laparoscopic splenectomy. Surg Endosc 9: 10171019 5. Hashizume M, Sugimachi K, Kitano S (1994) Laparoscopic splenectomy. Am J Surg 167: 611614 6. Spence RK (1995) Surgical red blood cell transfusion practice policies. Am J Surg 170: 312

Surg Endosc (1997) 11: 819824

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Percutaneous fetal access and uterine closure for fetoscopic surgery

Lessons learned from 16 consecutive procedures in pregnant sheep
T. Kohl,1 Z. Szabo,2 K. Suda,3 T. M. Quinn,2 E. Petrossian,1 M. R. Harrison,2 F. L. Hanley1
1 2

Division of Pediatric Cardiothoracic Surgery, University of California, 505 Parnassus Avenue, San Francisco, CA 94143, USA Division of Pediatric Surgery and Fetal Treatment Center, University of California, 505 Parnassus Avenue, San Francisco, CA 94143, USA 3 Division of Pediatric Cardiology, University of California, 505 Parnassus Avenue, San Francisco, CA 94143, USA Received: 18 September 1996/Accepted: 12 December 1996

Abstract Background: Maternal morbidity and preterm labor from fetal surgery might be minimized by a percutaneous technique for fetal access and uterine closure. Methods: In each of 16 ewes, we inserted three trocars percutaneously into the amniotic cavity using ultrasound and fetoscopic guidance. In six ewes, percutaneous uterine closure after the procedure was attempted. We assessed feasibility and acute complications of our technique during surgery and at autopsy. Results: We achieved percutaneous fetal access in 14 ewes and closed the uterus percutaneously in all six ewes attempted. Fetal injury was related to amnioinfusion or fixation of chorioamniotic membranes. Other complications were trocar dislodgment and damage to uterine wall and chorioamniotic membranes. The latter complication was prevented using balloon-tipped trocars. Conclusions: Percutaneous intraamniotic access and uterine closure for fetoscopic surgery can be achieved reliably with little maternal and fetal morbidity in sheep. Minor modifications are desired to apply this approach in humans. Key words: Fetoscopy Fetal surgery Percutaneous fetal access Uterine closure Sheep

transuterine trocar placement via small uterine incisions [24]. The development of a solely percutaneous technique not requiring maternal laparotomy should even further decrease maternal morbidity from fetoscopic surgery. Therefore, the purpose of this experimental study was to assess the feasibility and acute complications of percutaneous fetal access and uterine closure for fetoscopic surgery in pregnant sheep. Methods Surgical preparation
We studied the feasibility and acute complications of percutaneous fetal access and uterine closure in a total of 16 ewes between 86 and 121 days of gestation (term 145 days). The study protocol had been approved by the local Committee on Animal Research and was performed according to institutional guidelines. We sedated each ewe with an intramuscular injection of ketamine hydrochloride (1020 mg/kg) and intubated and administered general anesthesia with 0.52.0% halothane and 100% oxygen. We positioned the ewe supine and performed a detailed transabdominal ultrasound study (Sonos 1500, Hewlett-Packard, Palo Alto, CA) to determine fetal number and position, amniotic fluid volume, as well as the fetal abdominal insertion site and placental origin of the umbilical cord.

Amnioinfusion
If the amniotic fluid volume was judged too small to permit safe placement of the initial trocar, we performed amnioinfusion with saline. Under ultrasound guidance, we inserted an 18-gauge needle into the amniotic cavity and confirmed free intraamniotic needle position by unhindered amniotic fluid withdrawal and injection of a small amount of air. We then infused 500800 ml warm normal saline at rates of approximately 150 ml/min into the amniotic cavity until a satisfactory amniotic fluid pocket was obtained.

Despite remarkable advances in the development of open fetal surgery, the suppression of preterm labor after hysterotomy remains the main obstacle for this approach in human fetuses [1]. Conversely, fetoscopic operative approaches have the potential to reduce the incidence of preterm labor because fetal access is achieved by maternal laparotomy and
Correspondence to: Thomas Kohl, Department of Pediatric Cardiology, University of Mu nster Medical School, Albert Schweitzer Str. 33, 48129 Mu nster-Germany

Trocar insertion
In each animal, we inserted three 5-mm trocars into the amniotic cavity. In order to enter the amniotic sac during trocar insertion, the chorioamniotic membranes were fixed to the uterine wall with T-fasteners (Ross Product Division-Abbot Laboratories, Columbus, OH). We then incised the maternal skin and fascia 35 mm in the center of the exit sites of the T-fastener

820 sutures. While pulling up on the sutures, we inserted the trocar through the small incision into the amniotic cavity. The trocars were secured inside the amniotic cavity by either a radially dilating balloon (Entec Corporation, Madison, CT) or flanges (Dexide Inc., Forth Worth, TX) attached to the trocar tip. We monitored the insertion of the initial T-fasteners and trocar with transabdominal ultrasound (Fig. 1) and placement of subsequent devices under direct observation with a 5-mm, 0 fetoscope connected to a xenon light source (Karl Storz, Culver City, CA) (Fig. 2). In order to prevent injury to maternal intraabdominal contents and uterine vessels, we improved the visibility of trocar insertion by low pressure (48 cm H2O) gaseous insufflation of the amniotic cavity and transillumination of the uterine and abdominal walls.

was no injury to fetal abdominal contents. Once percutaneous access was established, the malpositioned T-fasteners were recovered without difficulty. Trocar insertion During the 16 procedures, we inserted a total of 27 balloontipped trocars in nine ewes and a total of 15 flanged trocars in five ewes (three trocars per ewe). During three procedures, intrauterine manipulation with laparoscopic instruments resulted in balloon rupture of one of the balloontipped trocars (3/27 balloon-tipped trocars 11.1%). In one of these procedures, the trocar dislodged from the amniotic cavity into the maternal abdomen, thus deflating the amniotic cavity. As a result, we had to abandon the fetoscopic approach. In the other two cases of balloon rupture, we exchanged the trocars without jeopardizing the fetoscopic approach. In five ewes, manipulation of the flanged trocars resulted in notable damage to the uterine wall and tearing of the chorioamniotic membranes. In three of these ewes, the chorioamniotic membranes separated from the uterine wall and slipped off the trocar tip, interfering with intraamniotic instrument insertion and insufflation. Conversely, we did not observe damage to the uterine wall or chorioamniotic membranes from manipulation of balloon-tipped trocars. In two ewes, insertion of one trocar (2/42 trocars 4.8%) resulted in bleeding of small cotyledonal vessels into the chorioamniotic space and hematoma formation. At autopsy about 20 cc of blood was recovered from the chorioamniotic space in each case. We failed to achieve satisfactory percutaneous fetal access and had to convert to an open operative approach in two ewes. In the first ewe, we placed the initial trocar into the chorioamniotic space and were not able to advance the device farther into the amniotic cavity. To avoid further chorioamniotic membrane separation, we performed maternal laparotomy, and placed all trocars into the exteriorized uterus. In the second ewe, although we placed the initial trocar into the amniotic cavity, fetal visualization and surgical manipulation were prevented by a large intraamniotic membrane. This membrane had not been appreciated at the initial ultrasound study and also necessitated conversion to the open operative approach. Uterine closure We closed the uterine trocar insertion sites percutaneously by tying the T-fastener sutures across the trocar insertion sites in all of the six ewes in which this approach was tested. In two of these ewes, the T-fasteners were inserted through the major omentum which we subsequently tied onto the external uterine wall. The six ewes were allowed to continue gestation for 1 day to 2 weeks before elective termination. During this period we did not observe intraabdominal or transvaginal amniotic fluid loss as assessed by transabdominal ultrasound and daily observation. Discussion This study in pregnant sheep demonstrates that after an initial learning phase percutaneous fetal access and uterine

Uterine closure
In six ewes, we assessed a method of percutaneous uterine closure of the trocar insertion sites after completion of our operations (Fig. 3). We disengaged the trocars, withdrew them from the amniotic cavity into the maternal abdomen, and engaged them again. We then insufflated the maternal abdomen and released the T-fastener sutures into the maternal abdomen. We closed the uterine trocar insertion sites by tying the T-fastener sutures across the wall defect with laparoscopic instruments. After uterine closure, we removed the trocars from the abdominal cavity and closed the abdominal trocar insertion sites with simple interrupted sutures.

Study variables
We recorded the number of animals in which percutaneous fetal access was successful and inspected the maternal abdomen for bleeding or injury to other organs. We assessed adverse fetal effects from amnioinfusion or T-fastener insertion and determined the incidence and etiology of trocar dislodgment during a fetoscopic procedure. After completion of our study, we sacrificed the ewes with pentobarbital overdose. At autopsy, we examined the uterine trocar insertion sites for injury to the uterine wall and integrity of the chorioamniotic membranes. In the six ewes in which uterine closure after fetoscopic surgery was tested, we assessed the trocar insertion sites for intraabdominal or transvaginal amniotic fluid leakage by direct observation and transabdominal ultrasound.

Results We performed successful percutaneous fetal access in 14 of the 16 ewes (87.5%) and inserted a total of 42 trocars (three trocars per ewe) into the amniotic cavity without chorioamniotic membrane separation. We did not observe injuries to maternal viscera or significant bleeding complications in mother or fetus from percutaneous fetal access or uterine closure. Amnioinfusion and T-fastener insertion To facilitate trocar insertion we performed amnioinfusion in 11 of the 15 ewes (73.3%) and observed complications from this procedure in one fetus. Despite having confirmed a free intraamniotic needle position by ultrasound, saline was injected into a hindleg of this fetus. In three studies, we punctured fetal parts during ultrasound-guided T-fastener insertion (3/42 T-fasteners 7.1%). In two studies, we inserted a T-fastener into a fetal hindleg (Fig. 4). In one of our early studies, one T-fastener was inserted into the fetal abdomen resulting in perforation of the abdominal wall during ultrasound-guided placement of the first trocar (1/42 trocars 2.4%). However, there

821

Fig. 1. Ultrasound-guided insertion of initial T-fasteners and trocar (A). To prevent chorioamniotic membrane separation during trocar insertion, the membranes were fixed to the uterine wall using T-fasteners released from an 18-gauge needle. The needle (N) can be recognized inside the amniotic cavity (B). Appropriate intraamniotic needle position was confirmed with amniotic fluid withdrawal and air injection (C) before

the T-fastener (TF) is released (D). Then the maternal skin and fascia were incised 35 mm in the center of the exit sites of the T-fastener sutures (E). While pulling up on the sutures, the trocar was inserted into the small incision (F). The trocars were secured in the amniotic cavity by either a radially dilating balloon or flanges attached to the trocar tip.

822

Fig. 2. Fetoscopy-guided insertion of T-fasteners and trocar. Whereas the insertion of the initial T-fasteners and trocar were monitored with transabdominal ultrasound, insertion of subsequent devices was accomplished under direct observation with a 5-mm, 0 fetoscope connected to a xenon light source. Two T-fasteners, the 18-gauge needle (A) and an expanded 5-mm flanged trocar (B), can be recognized inside the amniotic cavity. In order to prevent injury to maternal intraabdominal contents and uterine vessels, low-pressure gaseous insufflation of the amniotic cavity and transillumination of the uterine and abdominal walls were performed. Fig. 3. Percutaneous uterine closure of the trocar insertion sites after completion of the operations. The trocars are disengaged, withdrawn from the amniotic cavity into the maternal abdomen, and engaged again (A). The maternal abdomen is insufflated and the T-fastener sutures are released into the maternal abdomen. Uterine closure is achieved by tying the T-fastener sutures across the uterine trocar insertion sites with laparoscopic instruments (B). After closure of all trocar insertion sites (C), the trocars are removed from the abdominal cavity and the abdominal trocar insertion sites are closed with a single stitch.

closure for fetoscopic surgery can be achieved reliably with little maternal and fetal morbidity. The percutaneous approach can be readily applied to various animal models to advance fetal surgery from open to percutaneous fetoscopic procedures; only a minor technical modification (i.e., absorbable T-fasteners) is desired to utilize this approach in humans. In those, the percutaneous approach should substantially decrease maternal morbidity from fetal surgical procedures since laparotomy and hysterotomy are not required. As a consequence, percutaneous fetal access may

also result in less preterm labor and premature delivery than the open operative approach. Technique of percutaneous fetal access Intraoperative maternal transabdominal ultrasound monitoring, T-fastener fixation of the chorioamniotic membranes, and trocars that can be engaged in the amniotic cavity were

823

To avoid larger maternal or fetal vessels underneath the insertion sites, Doppler color flow mapping with low Nyquist limits during ultrasound-guided T-fastener and trocar insertion is useful. Once the initial trocar is placed into the amniotic cavity, direct fetoscopic observation and transillumination of the uterine and maternal abdominal walls with the fetoscopic light source serve the same purpose. Nevertheless, small uterine and placental vessels may not be avoided. Despite Doppler color flow mapping and direct fetoscopic observation, trocar insertion resulted in bleeding of small cotyledonal vessels into the chorioamniotic space and hematoma formation during two of our procedures. This minor complication is invited by the architecture of the ovine placenta with cotyledons scattered over the interior uterine wall and may not be encountered in humans in whom the placenta is localized.

Feasibility and complications of percutaneous fetal access during a surgical procedure During our operations, the percutaneously placed trocars did not need to be supported by an assistant because they were stabilized by the abdominal wall. This advantage is lost if the percutaneous setup cannot be achieved and the trocars need to be placed into the exteriorized uterus. Whereas avoidance of chorioamniotic membrane separation is the key to successful percutaneous fetal access, this complication can result from trocar manipulation during the actual operation. In three of five studies, intraoperative manipulation of flanged trocars resulted in notable damage to the uterine wall and chorioamniotic membranes. The membranes separated from the uterine wall, slipped off the trocar tip, and interfered with instrument insertion or insufflation. This complication was prevented by balloon-tipped trocars in subsequent procedures. The broad, soft base of the balloon protects the uterine wall and chorioamniotic membranes by distributing shear stress and traction forces from instrument manipulation over a larger uterine surface area. Yet the balloon-tipped trocars are prone to balloon rupture, which was observed during three of nine procedures in which these trocars were used. This complication resulted in loss of the percutaneous fetoscopic setup in one case, necessitating conversion to an open operative approach. Despite the risk of balloon rupture, we currently favor balloontipped trocars because they were less traumatic in our study, and we recommend frequent balloon assessment during the procedure.

Fig. 4. Complications of percutaneous fetal access. In two studies, fetal limbs were punctured during ultrasound-guided T-fastener insertion. (A) shows a T-fastener inserted into a fetal hindleg. Once the percutaneous fetal access was established, the malpositioned T-fasteners were easily recovered. In five ewes, manipulation of a trocar with radially expanding flanges resulted in notable damage to the uterine wall and tearing of the chorioamniotic membranes (B). In three of these ewes, the chorioamniotic membranes slipped off the trocar tip and subsequently interfered with intraamniotic instrument insertion and insufflation. We did not observe this complication with the use of balloon-tipped trocars in subsequent procedures.

the prerequisites for our technique. Successful insertion of the initial T-fasteners and trocar depends on the identification of a sufficiently large pocket of amniotic fluid; the accurate definition of fetal lie by maternal transabdominal ultrasound permits selection of the most strategic insertion site if several adequate pockets are available. Amnioinfusion is commonly required to increase the size of amniotic fluid pockets. Careful ultrasound examination prior to insertion of the initial T-fasteners and trocar is important to recognize any membranes dividing the amniotic cavity. Failure to appreciate amniotic cavity septation may result in unfavorable intraamniotic or extraamniotic position of the initial trocar and can necessitate conversion to an open operative approach. Augmenting intramniotic fluid volume and shifting of fetal position, however, may provide alternative entry sites for the initial trocar in the presence of amniotic cavity septation.

Feasibility and complications of percutaneous uterine closure after fetoscopic surgery In tying the T-fastener sutures across the uterine trocar insertion sites utilizing laparoscopy, we found a novel and effective method for percutaneous closure of the uterus. We did not observe postoperative intraabdominal or transvaginal amniotic fluid leakage using this technique. Leakage, however, was commonly observed in earlier experiments of our group when 510-mm trocars were removed without closing their insertion sites.

824

Limitations of percutaneous fetal access and uterine closure Our study is limited to the assessment of acute complications from percutaneous fetal access and uterine closure in sheep. Because the ovine uterus is remarkably resistant to preterm labor, we cannot extrapolate the incidence of postoperative preterm labor and premature delivery that may arise from our approach in humans. However, a recent study in rhesus monkeys has shown that preterm labor is remarkably reduced after maternal laparotomy and endoscopic access for fetal surgery even in the absence of tokolysis [3, 5]. A similarly favorable result might be expected for the even less invasive percutaneous approach. To perform percutaneous fetal access and uterine closure in humans, the placenta ought to be located posteriorly. In case of an anterior placenta, laparotomy and transuterine placement of the trocars may be required for safe trocar insertion. In addition, absorbable T-fasteners are desired to avoid foreign body retention.
Acknowledgment. We gratefully acknowledge the expert technical assistance of Mr. Vincente G. Lapiz as well as the support of Boston Scientific Corporation, Sunnyvale, CA; Karl Storz America, Culver City, CA; Entec

Corporation, Madison, CT; Ross Product Division, Abbot Laboratories, Columbus, OH; Wilfried Vogeler, MD, Essen, Germany; Deniz Kececioglu, MD, Mu nster, Germany; Hilfe fu r das Herzkranke Kind, e.V., Mu nster, Germany. Dr. Kohl is a research fellow supported by a research grant (Ko 1484/2-1) of the Deutsche Forschungsgemeinschaft (DFG)Bonn/Germany.

References
1. Estes JM, Szabo Z, Harrison MR (1992) Techniques for in utero endoscopic surgery. A new approach for fetal intervention. Surg Endosc 6(5): 215218 2. Feitz WF, Steegers EA, Aarnink RG, Arts T, De Vries JD, Van der Wildt B (1996) Endoscopic intrauterine fetal therapy: a monkey model. Urology 47(1): 118119 3. Harrison MR, Adzick NS, Flake AW, VanderWall KJ, Bealer JF, Howell LJ, Farrel JA, Filly RA, Rosen MA (1996) Fetendo clipfetal endoscopic tracheal clip procedure in a human fetus. Annual Meeting of the International Fetal Medicine and Surgery Society, Capri, Italy 4. Longaker MT, Golbus MS, Filly RA, Rosen MA, Chang SW, Harrison MR (1991) Maternal outcome after open fetal surgery. A review of the first 17 human cases. JAMA 265(6): 737741 5. van der Wildt B, Luks FI, Steegers EA, Deprest JA, Peers KH (1995) Absence of electrical uterine activity after endoscopic access for fetal surgery in the rhesus monkey [letter]. Eur J Obstet Gynecol Reprod Biol 58(2): 213214

Surg Endosc (1997) 11: 825829

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Adhesion formation in laparoscopic inguinal hernia repair

V. L. Vader,1 D. M. Vogt,1 K. A. Zucker,1 J. P. Thilstead,2 M. J. Curet1
1 2

Department of Surgery, University of New Mexico, School of Medicine, 2211 Lomas Boulevard, NE, Albuquerque, NM 87131-5341, USA New Mexico Department of Agriculture, 700 Camino de Salud, NE, Albuquerque, NM 87106, USA

Received: 22 August 1996/Accepted: 13 November 1996

Abstract Background: In laparoscopic inguinal hernia repair controversy exists concerning the most appropriate repair method and implant material to use if intraabdominal adhesions are to be minimized. Methods. In 108 pigs, we implanted three different types of mesh by both the TAPP (transabdominal preperitoneal) and Onlay (prosthesis placed directly upon the peritoneum) methods. Specimens were harvested in three time periods and adhesion formation was compared. Results. Average adhesions at 3 days were TAPP 18% and Onlay 49% (p < 0.001). At 3 weeks average adhesions were TAPP 8% and Onlay 23% (p < 0.04). Three-month figures were TAPP 1% and Onlay 13% (p < 0.001). In contrast, there were no differences in adhesion formation due to material type in any of the three time periods (all p > 0.17). Conclusions: A peritoneal covering over a laparoscopic inguinal implant significantly reduced adhesions. Prosthetic material type did not affect adhesion formation in this study. Key words: Adhesion Laparoscopic Inguinal hernia Peritoneum

The era of contemporary surgical repair of inguinal hernias was ushered in by Eduardo Bassini over a century ago and today approximately 500,000 inguinal hernia operations are performed in the United States each year [13]. Repair methods continue to be developed and modified to address such issues as high recurrence rates, prolonged delay in return to function, as well as post-operative pain and discomfort. To this end, laparoscopic hernia repair was introduced in 1990 by Schultz [16]. Advantages of the laparoscopic techniques over conventional methods are less postoperative pain, decreased need for narcotic analgesics, improved cosmesis, and no tension on the surgery site, therefore allowing earlier return to function without jeopardizing the repair [1, 3, 4,
Correspondence to: M. J. Curet

810, 17, 20]. Disadvantages of a laparoscopic approach include the need for general anesthesia, lack of long-term follow-up data, and more importantly, the risk of adhesions to the intraabdominal prosthetic material [14] with subsequent erosion or fistulization into adjacent visceral structures. Currently nearly every laparoscopic hernia repair requires insertion of a prosthetic material which is secured intraabdominally to the inguinal floor with staples or sutures. This material may be simply secured over the peritoneum (Onlay technique) or placed beneath the peritoneum of the inguinal floor (the TAPP or transabdominal preperitoneal approach). A third laparoscopic method, the extraperitoneal approach, also results in mesh placed beneath the peritoneum of the inguinal floor, but without entering the peritoneal cavity. While the Onlay technique is faster and easier to perform, the TAPP and the extraperitoneal repairs have been proposed to minimize contact between the implant material and the viscera, hopefully reducing adhesions. Although the TAPP and the extraperitoneal repairs are executed differently, the end result is similar in that the peritoneum is elevated from its underlying support structures and interposed between the prosthesis and the visceral structures. One of the concerns with these latter two approaches is that the peritoneum over the mesh may not actually remain functional in preventing adhesions due to the disruption in its blood supply and placement on top of a reactive foreign body. Previous small-scale animal studies have not adequately addressed this issue [2, 15]. Although they have suggested that the peritoneum does continue to provide a protective effect in spite of the extensive surgical dissection, statistical significance is lacking. We proposed a large-scale animal study to address this issue. We hypothesized that the peritoneum does remain functional in preventing adhesions in spite of the trauma of surgical dissection. This should be evidenced by short- and long-term differences in adhesion formation between the TAPP and Onlay methods of laparoscopic inguinal hernia repair. We elected not to include examination of the extraperitoneal approach in our study due to sample size limits, the fact that the extraperitoneal method is difficult to perform in the

826 Table 1. Subject groupings by time and implantation method108 total pigs TAPP method 3 days 3 weeks 3 months Polypropylene ePTFE mesh ePTFE solid Polypropylene ePTFE mesh ePTFE solid Polypropylene ePTFE mesh ePTFE solid 6 6 6 6 6 6 6 6 6 pigs pigs pigs pigs pigs pigs pigs pigs pigs Onlay method Polypropylene ePTFE mesh ePTFE solid Polypropylene ePTFE mesh ePTFE solid Polypropylene ePTFE mesh ePTFE solid 6 6 6 6 6 6 6 6 6 pigs pigs pigs pigs pigs pigs pigs pigs pigs prosthetic patch of either polypropylene, or ePTFE solid, or ePTFE mesh was then secured to the underlying tissues using surgical (titanium) fascial staples (Endopath Multifeed Stapler, Ethicon Endosurgery, Cincinnati, OH). Sites of attachment included the inferior aspect of the rectus sheath, the iliopubic tract, and Coopers ligament. The prosthetic material was then completely covered with peritoneum reapproximated using fascial staples. In the 54 Onlay subjects a 1-cm area of peritoneum was dissected away from Coopers ligament to facilitate anchoring to this critical structure. No further peritoneal dissection was done. A 3 5 cm prosthetic patch of either polypropylene, or ePTFE solid, or ePTFE mesh was placed over the direct and indirect hernia spaces and then secured to Coopers ligament and the peritoneum over the inferior rectus sheath, iliopubic tract, and transversalis fascia with endoscopic staples. Following placement of the prosthetic material, the laparoscopic cannulas were removed and the carbon dioxide was expelled from the peritoneal cavity in all animals. All three fascial defects were closed. The pigs were awakened and extubated. They were cared for in either a standard laboratory setting (3-day and 3-week subjects) or returned to a local farm (3-month subjects) until they were euthanized with pentobarbital 390 mg/ml (Euthasol, Delmerva Laboratories, Inc., Midlothian, VA, 23113) at 80 mg/kg in their respective time periods. Inspection of the implant site at necropsy was accomplished via wide exposure of the lower abdomen and pelvis. A full-thickness incision in the abdominal wall began at the left caudal lateral aspect of the abdomen, curved anteriorly to the xiphoid, and then continued in a curve to the right caudal lateral aspect of the abdomen. The large abdominal flap thus created was reflected caudally to provide excellent exposure of the previous surgical area without manipulating the surgical site proper. The implanted area was then visually inspected and adhesions were rated by two observers. Numerical scores were given to correlate with the percent surface area of the graft or surgical site that was covered by adhesions. In the Onlay procedures adhesions were to the graft itself. In TAPP animals, adhesions mostly involved the overlying peritoneum, but occasionally, in 3-day specimens, a small amount of mesh was not fully covered by peritoneum and adhesions could involve mesh itself in addition to the overlying peritoneum. Scores by the two observers tended to agree within five percent points. Where scores differed, the average was used. After gross inspection, rating of adhesions, and photographing the exposed implanted area, the surgical site was excised in toto using a full-thickness technique and submitted for histological evaluation of adhesion formation. In the first part of the analysis, all TAPP animals, regardless of mesh type, were evaluated together and groups were separated only by the three time periods. The Onlay animals were evaluated similarly. Average percent of prostheses covered by adhesions was compared between the TAPP and Onlay specimens in all three time periods separately. A Wilcoxon test was used due to the shape of our data (non-normal distribution due to multiple occurrences of zero percent adhesions), which calls for the use of a nonparametric test. In the second part of the analysis, to determine if implant type had a bearing on adhesions irrespective of method of implantation, a Fischers exact test was employed. For this a 2 3 table was constructed comparing the three mesh types vs adhesions or no adhesions. TAPP and Onlay specimens were pooled into the three mesh types. In this manner, each time period was examined separately as well as in a combined time grouping. A binary outcome for this Fischers exact test was defined here: 05% coverage of patch by adhesions was considered no adhesions, and >5% coverage of patch was considered positive for adhesions. A logistic regression analysis was also performed to confirm results of the Wilcoxon test and the Fischers exact test.

porcine model, and because we felt that results would likely be comparable to those of the TAPP procedure for reasons mentioned above. Additionally, since prosthesis choice may also have an impact on adhesions, we used three different implant materials in our study. Our second hypothesis is that mesh type will affect adhesion formation.

Methods and materials

An adult porcine model was utilized because the inguinal architecture of the pig is the closest nonprimate animal model to the human. In addition this species is readily available. Several different implant materials with different properties and reactivities are currently in use. We chose for implantation the two most common of these, as well as a prototype for a new material. The three different implant materials used were: 1. Polypropylene mesh (Marlex) first utilized for hernia repair in 1963 by Usher [19]. 2. Expanded polytetrafluroethylene solid tissue patch (ePTFE or Gore-Tex solid tissue patch) introduced by W. L. Gore and Associates, Flagstaff, Arizona, in 1974. 3. A customized Gore-Tex mesh developed by the Surgical Endoscopy Department at the University of New Mexico [20]. The customized mesh is created by passing the Gore-Tex solid tissue patch through a standard skin graft meshing device (Zimmer Inc.) using a 1:1.5 template to generate a fenestrated material. As shown in Table 1, 108 pigs were randomly assigned into 18 groups. One-half of these 108 animals underwent a TAPP placement of mesh and one half underwent an Onlay procedure. These two groups were further divided into the different mesh materials and different time points. Designated animals were sacrificed at 3 days, 3 weeks, and 3 months to monitor development of adhesions and tissue reaction. The choice of these time periods was made on the basis of Milligans classic study of peritoneal adhesion formation [12]. Milligan found that the initial fibrinous reaction in a peritoneal insult begins to be replaced by an invasion of fibroblasts and collagen at 34 days and increased collagen is laid down in the 2-week to 2-month period. Noncastrated pigs of average age 5 months and average size approximately 20 kg were anesthetized using 4 mg/kg intramuscularly of Telazol (Fort Dodge Laboratories, Fort Dodge, IA) and 2 mg/kg intramuscularly of xylazine (Sedazine, Fort Dodge Laboratories, Fort Dodge, IA) followed by intubation and maintenance on halothane and oxygen administered by ventilator (Ohio, Madison, WI). The subjects were prepped and draped in a sterile fashion for survival surgery. A pneumoperitoneum to 15 mgHg was established using a standard small Veress needle (Endopath Ultra Veress Needle, Ethicon Endosurgery, Cincinnati, OH) inserted just above the umbilicus. The Veress needle was then removed and replaced with a 10/11mm laparoscopic cannula (Tristar Trocar, Ethicon Endosurgery, Cincinnati, OH). A video laparoscope (Olympus America Inc., Melville, NY) was introduced into the peritoneal cavity, two accessory cannulas were inserted into the right and left midabdomen under direct visualization, and the pelvic anatomy was identified. In the 54 animals in the TAPP arm of the study, a horizontal incision was made 1.5 cm cephalad to the internal ring and a preperitoneal pocket was created. This area was large enough to encompass both the potential indirect and direct hernia spaces. A 3 5 cm

Results All animals survived surgery and were free of complications such as bowel obstruction, abscesses, or fistulas. The difference in adhesions between the TAPP and Onlay animals was statistically significant at 3 days (Wilcoxon test, p < 0.001), 3 weeks (Wilcoxon test, p < 0.04), and 3 months (Wilcoxon test, p < 0.001) (Table 2). The Fischers exact test indicated no difference in adhesion formation due to material type in any of the three time periods individually or in the pooled time group (Fishers exact test, all p > 0.17). A logistic regression of the

827

Fig. 1. Difference in percent adhesions over time, p < 0.001.

Table 2. Percent of prosthesis covered by adhesions in three time periods Combined mesh types 3-day adhesion avg. 3-week adhesion avg. 3-month adhesion avg. * p < 0.001 ** p < 0.04 *** p < 0.001 TAPP method 17.5% 7.52 8.3% 4.12 1.4% 1.39 Onlay method 48.6% 7.39* 22.7% 7.35** 13.1% 3.87***

binary adhesion outcome vs mesh types, three time periods, and two implantation methods bears out the conclusions of the Wilcoxon test and the Fischers exact test. An odds ratio of 8.5 (95% CI 3.123) was shown for adhesions if the Onlay method was used vs the TAPP, controlling for time and patch type. In addition, we found a significant trend for reduced adhesions, as defined by percent coverage of mesh with adhesions, over time (logistic regression, p < 0.001). This is demonstrated graphically in Fig. 1. When we analyzed for frequency of adhesions, defined by the binary outcome of adhesion vs no adhesion with >5% adhesions being the inclusion criteria, we found the same trend. Discussion The advantages and disadvantages of laparoscopic vs conventional herniorrhaphy, including postoperative pain, recurrence rates, return to function, complication rates, and cost, have been argued elsewhere. This study did not address these issues. Rather, in the event that laparoscopic repair is elected over conventional repair, this study helps

clarify the controversy concerning which method of laparoscopic inguinal mesh implantation and which prosthetic material is more likely to result in the least intraabdominal adhesion formation. Adhesions are considered by many clinicians to be one of the more serious complications potentially encountered in laparoscopic repairs. By using the animal model, a controlled, detailed assessment of adhesion formation is possible. The choice of time periods for evaluation of implants, 3 days, 3 weeks, and 3 months, was based on Milligans findings [12] that the initial fibrinous reaction in a peritoneal insult begins to be replaced by an invasion of fibroblasts and collagen at 34 days and increased collagen is laid down in the 2-week to 2-month period. Three main types of laparoscopic inguinal hernia repair are currently in use: the Onlay, the TAPP, and the extraperitoneal repair [5, 6, 8, 11, 20]. In the Onlay method, the implant is placed on top of the peritoneum intraabdominally and comes in direct contact with the visceral structures. In a TAPP procedure a large area of peritoneum is dissected free of its blood supply to make a space for the implant. The peritoneum is then reapposed over the graft. The extraperitoneal approach involves an incision through the skin, subcutaneous tissues, muscle, and fascia of the anterior abdomen followed by dissection between the fascia and the peritoneum over the hernial defect. In the pocket provided by this dissection, a prosthetic mesh is secured over the direct and indirect hernial space. The Onlay method does not use the peritoneum as a barrier between the implanted prosthesis and the viscera, while the TAPP and the extraperitoneal repairs do. The extraperitoneal approach, where the abdomen is never entered, preserves the peritoneum and interposes it between the mesh and the abdomen contents much like the TAPP procedure. For this reason we felt that, in respect to adhesion formation, results would be similar between the TAPP and the extraperitoneal technique. In addition, the extraperitoneal repair is difficult to perform in the pig. For these reasons the extraperitoneal method was not modeled in this study. Thus we compared only differences in adhesion formation between the Onlay method and the TAPP method. Two previous small-scale, short-term animal studies have been reported addressing the controversy of optimum prosthesis placement in laparoscopic inguinal hernia repair. One study involved six dogs evaluated at 6 weeks follow-up [15]. In this study each of six dogs had polypropylene mesh implanted within the abdominal cavity by the TAPP, the Onlay, and the extraperitoneal method, and adhesion formation was compared. A second study consisted of 15 pigs [2], each of which had polypropylene implants placed intraabdominally by TAPP on the right side and Onlay on the left side. Subjects were evaluated for adhesions at 2 weeks postoperatively. Both of these studies hinted at the possibility of reduced adhesions when a layer of peritoneum was interposed between the abdominal contents and the prosthesis but both of these studies lacked statistical importance and the authors suggested further animal studies. In contrast to these studies, our study showed a clear, statistically significant advantage to the TAPP procedure over the Onlay approach. Although several articles referring to clinical aspects of laparoscopic repair of inguinal hernias have suggested using the Onlay technique since it is technically

828

easier and faster to perform [7, 20], the statistically significant results of our animal study in respect to adhesion formation indicate that the TAPP is preferable to the Onlay technique when laparoscopic repair is elected. As the preperitoneal approach was not modeled in this study for reasons already alluded to, we cannot reach a conclusion on the likelihood of avoiding adhesions with this method. We postulated that results of the TAPP and extraperitoneal methods of mesh placement would be similar in regard to adhesion formation since both interpose a layer of peritoneum between mesh and abdominal viscera. However, because the adhesions in our TAPP animals were primarily to the abdominal surface of the peritoneum, and in the preperitoneal approach the abdominal side of the peritoneum is not manipulated, one could speculate that the preperitoneal approach may afford an additional advantage in decreasing adhesions. This may warrant future study. A second area of interest in this study is the decrease in adhesions over time in both the TAPP and Onlay groups. This was true if analyzing for percent coverage of mesh by adhesions or for frequency of adhesions. The significance of this in laparoscopic surgery, regarding the timing of complications due to adhesion formation in relation to this observation, is unknown. Our study followed animals to 3 months postoperatively, which, at the onset of the study, we felt could be considered long term. One could speculate that the trend toward decreased adhesions with time would continue, but it is unknown how long. Perhaps over time as both groups approached zero adhesions, the statistical difference between TAPP and Onlay methods of implantation would disappear. As Milligan mentions in his study on the pathogenesis of peritoneal adhesion formation [12], the difficulty in assessing the progress of adhesions lies in attempting to interpret a dynamic process from a series of static appearances. The correlation between our selected time periods and the decline in adhesion formation may prove useful to others when future studies must choose a single time period to investigate. Another controversy in laparoscopic inguinal hernia repair involves the selection of the mesh type to be implanted. Some authors advocate polypropylene prosthesis due to its uniformity and reliability in stimulating scar formation, as well as due to its openings, which allow the surgeon to see through it and hopefully avoid vital structures while stapling [7]. Others feel strongly that the ePTFE prosthesis is superior as it elicits a negligible tissue reaction and has a low predisposition to infection [8]. Controlling for implantation method, in this study of 108 pigs we found no statistical difference in adhesion formation with any of the mesh types. This was true at 3 days, 3 weeks, and 3 months. This indicates that, at least in relation to adhesion formation, choice of mesh may be only a matter of surgeon preference. Several limitations of this study warrant discussion. The first and possibly most important is the power analysis of the negative results of the Fischers exact test to determine if prosthesis type had a bearing on adhesion formation. The statement that there is no difference in adhesions due to mesh type is limited by a lack of statistical power. One would need a total of 171 pigs to have 80% power. Our sample size of 108 pigs has only 60% power to detect a twofold increase or decrease in the percentage of adhesions covering 5% or more of the area of the patch. On the other

hand, in the logistic regression, time and method were significant while type of prosthesis was not. A second limitation of this study is the lack of interpretable histological data. We designed this study to approximate the procedures currently in use on human patients. This involved stapling the mesh in place, and, in the TAPP technique, also stapling the peritoneum closed over the prosthesis. For obvious reasons the histological blocks could not be processed without staple removal. Unfortunately, the removal of the staples prior to histological processing proved to be significantly detrimental to the architecture of the specimens. All specimens were evaluated histologically but results were inconsistent and inconclusive. The 3-day samples, in particular, where tissue associations were still very tenuous, were quite disrupted. Future studies might consider the use of punch biopsy specimens taken of representative areas in the removed tissues in question. A third limitation is that which has been previously noted by Fitzgibbons et al. [7] in their study on Onlay procedures performed on porcine subjects concerning the quadruped posture of the experimental animal vs the upright biped posture of man. This difference from the human posture could potentially be given skewed results. As Fitzgibbons noted in his study, and in our study as well, most adhesions were between the implanted prosthesis and the bladder. Fitzgibbons thought to control for this by also placing mesh on the anterior abdominal wall. We elected not to do this as we were primarily interested in the reactions in the inguinal region. Additionally, we must consider that in interpreting these results, there will be species variability in tissue reactions. Conclusions The peritoneum does perform a significant protective function when used to cover prosthetic material in laparoscopic inguinal hernia repair. This was evidenced by a statistically significant lower adhesion rate in the TAPP animals vs Onlay. Adhesions were most prevalent at 3 days and decreased with time, although differences between Onlay and TAPP were still significant at 3 months. Further, we found that the type of prosthetic material used did not significantly affect adhesion formation. This finding, however, is tempered by a power of only 60%. The results of this study suggest that laparoscopic inguinal hernia repair methods which maintain the peritoneum interposed between the prosthesis and the abdominal contents may be the preferred clinical techniques because of fewer short- and long-term intraabdominal adhesions. Further studies comparing the adhesion formation between the various prosthetic implants used in laparoscopic inguinal hernia repair are needed to truly assess potential differences in adhesion formation due to mesh type.
Acknowledgment. The authors wish to thank Dr. Clifford Qualls, professor of mathematics and statistics and research biostatistician at the University of New Mexico, for his help with the statistical analyses. The authors also express appreciation to W. L. Gore and Associates and to the Ethicon Endosurgery University-based Education and Research Center.

References
1. Arregui ME, Davis CJ, Yucel O, Nagan RF (1992) Laparoscopic mesh repair of inguinal hernia using a preperitoneal approach: a preliminary report. Surg Laparosc Endosc 2(1): 5358 2. Attwood SEA, Caldwell MTP, McDermott M, Stephens RB (1994)

829 Adhesions after laparoscopic inguinal hernia repaira comparison of extra versus intra peritoneal placement of a polypropylene mesh in an animal model. Surg Endosc 8: 777780 Campos L, Sipes E (1993) Laparoscopic hernia repair: use of a fenestrated PTFE graft with endo-clips. Surg Laparosc Endosc 3(1): 3538 Corbitt JD Jr (1991) Laparoscopic herniorrhaphy. Surg Laparosc Endosc 1(1): 2325 Dion Y-M, Morin J (1992) Laparoscopic inguinal herniorrhaphy. Can J Surg 35(2): 209212 Ferzli GS, Massad A, Albert P (1992) Extraperitoneal endoscopic inguinal hernia repair. J Laparoendosc Surg 2(6): 281286 Fitzgibbons RJ Jr, Salerno GM, Filipi CJ, Hunter WJ, Watson P (1994) A laparoscopic intraperitoneal onlay mesh technique for the repair of an indirect inguinal hernia. Ann Surg 219(2): 144156 Hanafy M (1993) Laparoscopic hernia repair: a review. Min Invas Ther 2: 229236 Kavic MS (1993) Laparoscopic hernia repair. Surg Endosc 7: 163167 MacFadyen BV, Arregui ME, Corbitt JD, Filipi CJ, Fitzgibbons RJ Jr, Franklin ME, McKernan JB, Olsen DO, Phillips EH, Rosenthal D, Schultz LS, Sewell RW, Smoot RT, Spaw AT, Toy FK, Waddell RL, Zucker KA (1993) Complications of laparoscopic herniorrhaphy. Surg Endosc 7: 155158 McKernan JB, Laws HL (1993) Laparoscopic repair of inguinal hernias using a totally extraperitoneal prosthetic approach. Surg Endosc 7: 2628 12. Milligan DW, Raftery AT (1974) Observations on the pathogenesis of peritoneal adhesions: a light and electron microscopical study. Br J Surg 61: 274280 13. Polister P, Cunico E (eds) (1989) Socio-economic factbook for surgery. American College of Surgeons, Chicago, IL, pp 2542 14. Salerno GM, Fitzgibbons RJ Jr, Corbitt JD Jr, Hart RO, Filipi CJ (1993) Laparoscopic inguinal hernia repair. In: Zucker KA (ed) Surgical laparoscopy update. Quality Medical, St Louis, MO, pp 373395 15. Schlechter B, Marks J, Shillingstad RB, Ponsky JL (1994) Intraabdominal mesh prosthesis in a canine model. Surg Endosc 8: 127129 16. Schultz L, Graber J, Pietrafitta J, Hickok D (1990) Laser laparoscopic herniorrhaphy: a clinical trial, preliminary results. J Laparoendosc Surg 1(1): 4145 17. Seid AS, Deutsch H, Jacobson A (1992) Laparoscopic herniorrhaphy. Surg Laparosc Endosc 2(1): 5960 18. Toy FK, Smoot RT Jr (1991) Toy-Smoot laparoscopic hernioplasty. Surg Laparosc Endosc 1(3): 151155 19. Usher FC (1963) Hernia repair with knitted polypropylene mesh. Surg Gynecol Obstet 117: 239240 20. Vogt DM, Curet MJ, Pitcher DE, Martin DT, Zucker KA (1995) Preliminary results of a prospective randomized trial of laparoscopic onlay versus conventional inguinal herniorrhaphy. Am J Surg 169: 8490

3. 4. 5. 6. 7. 8. 9. 10.

11.