ADULT UROLOGY

CME ARTICLE

DO PATIENTS WITH A HIGHER BODY MASS INDEX HAVE A

GREATER RISK OF ADVANCED-STAGE RENAL CELL
CARCINOMA?
LUIGI SCHIPS, RICHARD ZIGEUNER, KATJA LIPSKY, FRANZ QUEHENBERGER,
MICHAEL SALFELLNER, SUSANNE WINKLER, KARL PUMMER, AND GERHART HUBMER

ABSTRACT
Objectives. To evaluate whether patients with a higher body mass index (BMI) are at elevated risk of an
advanced tumor stage for renal cell carcinoma at diagnosis. A high BMI has recently been proved to be
associated with advanced tumor stages for some malignant diseases.
Methods. From January 1994 to December 2000, 693 operations for renal cell carcinoma were performed
in 683 patients at our institution. Ten patients underwent surgery twice for bilateral tumors. Of the 683
patients, 417 were men and 266 women. The mean age at surgery was 62.2 years, and the mean tumor
diameter was 5.2 cm. Seventy-eight percent of the patients were asymptomatic at tumor diagnosis. The
following parameters were evaluated with regard to a possible correlation to tumor stage and/or tumor
diameter: BMI, presence of symptoms, age, sex, hemoglobin, lactate dehydrogenase, erythrocyte sedimen-
tation rate, serum cholesterol, and triglycerides. For statistical analysis, the Spearman rank correlation test
was used.
Results. The mean BMI was 26.8 4.4 (range 16.9 to 44.3). Statistical analysis showed a significant
positive correlation between advanced T stage and the presence of symptoms (P 0.0001), erythrocyte
sedimentation rate (P 0.0001), lactate dehydrogenase (P 0.0015), and age (P 0.046), and an inverse
correlation with hemoglobin (P 0.0001) and serum cholesterol (P 0.0001). For all other investigated
parameters, including BMI, no significant correlation could be proved.
Conclusions. Our data indicate that obese patients are not at greater risk of advanced tumor stages of renal
cell carcinoma at the time of diagnosis compared with a population of normal weight. UROLOGY 62:
437441, 2003. 2003 Elsevier Inc.

T umors of the kidney represent about 2% to 3%

of new cases of cancer every year. About 80%
of tumors originate from the renal parenchyma.
The incidence of this tumor has recently in-
creased.1 To explain the rise in incidence of renal
cell carcinoma (RCC), many risk factors have been
considered, including nutritional2 4 and hormon-
al5 parameters, hypertension,6 and a family history
of RCC not related to von-Hippel-Lindau disease,7
as well as smoking or coffee drinking.6,8 10
From the Department of Urology, University Hospital, Karl
Franzens University Graz; and Department of Urology and Insti-
tute of Medical Statistics, University Hospital of Graz, Graz,
Austria
Reprint requests: Luigi Schips, M.D., Department of Urology,
University Hospital, Karl Franzens University Graz, Auenbrug-
gerplatz 7, Graz A-8036, Austria
Submitted: December 13, 2002, accepted (with revisions):
March 27, 2003
An association of a high body mass index (BMI)
with an increased incidence for RCC has been de-
scribed by several investigators.5,6,8,9,11 The hypo-
thetical reasons for this elevated risk include in-
creased levels of estrogens and insulin in the case
of obesity, a higher concentration of growth factors
in the adipose tissue, an abnormality in the metab-
olism of cholesterol, and also alterations in the im-
mune system.5 The impact of BMI on the incidence
of RCC has been reported to be stronger in women
than in men by some investigators,5,1113 and an
association independent of sex has been described
by others.8,14,15
Additionally, BMI has recently been proved to
correlate significantly with advanced tumor stages
in patients with breast cancer16 and also prostate
cancer at our department.17 Furthermore, the de-
tection of early-stage RCC by routine ultrasonog-
raphy in asymptomatic patients may be more diffi-

2003 ELSEVIER INC. 0090-4295/03/$30.00

ALL RIGHTS RESERVED doi:10.1016/S0090-4295(03)00380-7 437
cult in the case of obesity, with a possible
subsequent delay in diagnosis and treatment. On
the other hand, a better prognosis in obese patients
with RCC compared with a population of normal
weight has been reported,18 despite the increased
incidence. Between these contradicting observa-
tions, data are lacking regarding the tumor stage
and grade at diagnosis in overweight versus normal
weight patients with RCC that to our knowledge
has not yet been investigated. The aim of this study
was to investigate a possible association between a
high BMI and an advanced tumor stage in patients
with RCC. The rationale to investigate this ques-
tion was based on both the strong evidence of in-
creased RCC incidence in obese patients and a pos-
itive correlation with BMI and tumor stage in other
cancers, such as breast and prostate cancer.
MATERIAL AND METHODS
From January 1994 to December 2000, 693 operations for
RCC were performed in 683 patients at our department; 10
patients underwent surgery twice because of bilateral RCC.
The preoperative evaluation included medical history, physi-
cal investigation, renal ultrasonography, computed tomogra-
phy of the kidneys, as well as routine laboratory investigation
and chest x-ray. Magnetic resonance imaging was used as an
alternative imaging method in the case of renal failure or con-
traindications for the use of iodine-containing contrast agents.
Renal angiography, excretory urography, or bone scans were
not routinely performed. The operations included radical ne-
phrectomy in 633 (91.3%) and nephron-sparing surgery in 60
(8.7%) cases. Radical nephrectomy was performed outside
Gerotas fascia as described by Robson. Extended lymph node
dissection was not performed; however, pathologically en-
larged lymph nodes were resected when present.
The pathologic records of all patients were re-evaluated and
corrected in accordance with the 1997 edition of the TNM
system for patients treated before 1997. Tumor grade was as-
sessed according to the World Health Organization grading
system. The BMI, as described by Quetelet, which is calculated
by dividing the patients weight (in kilograms) by the square of
the patients height (in meters), was determined using the
weight and height of the patients at admission to the hospital.
Patients with a BMI between 25 and 30 were considered over-
weight and those with a BMI of 30 or more were considered
obese.
The following parameters were evaluated for a possible cor-
relation with stage and/or tumor diameter: BMI, presence of
symptoms, age, sex, hemoglobin, lactate dehydrogenase
(LDH), erythrocyte sedimentation rate (ESR), serum choles-
terol, and triglycerides. In our routine laboratory analyses,
only total serum cholesterol is investigated, low-density-li-
poprotein and high-density-lipoprotein cholesterol levels
were not routinely analyzed. For statistical analysis, the Spear-
man rank correlation test was used with regard to both patho-
logic T stage and tumor grade. Institutional review board ap-
proval and informed consent are not required at our
institution for retrospective studies dealing only with routine
clinical parameters obtained at patient admission or during
treatment.
RESULTS
Of 683 patients, 417 (61%) were men and 266
(39%) were women, and the mean age was 62.2
438
years (range 16 to 88). At diagnosis, 533 patients
(78%) were asymptomatic, and 150 patients (22%)
presented with symptoms. RCC was diagnosed by
routine ultrasonography in the asymptomatic pa-
tients. The most frequent symptoms leading to
RCC diagnosis were gross hematuria in 65, flank
pain in 29, and weight loss in 18 patients.
In 693 operations for RCC, the histologic evalu-
ation (TNM-1997) revealed pathologic Stage T1 in
396 cases (57.1%), pT2 in 42 (6%), pT3a in 125
(18%), pT3b in 110 (16%), pT3c in 10 (1.4%), and
pT4 in 7 (1%). Stage was not available in 3 cases
(0.5%). Of the 693 operations, 107 specimens
(15.4%) were grade 1, 483 (69.7%) were grade 2,
99 (14.3%) were grade 3, and 3 (0.5%) were grade
4; in 1 patient (0.1%), the grade was not reported.
The mean SD and median tumor diameter was
5.2 2.9 cm and 4.5 cm, respectively (range 0.8 to
22). Lymph node metastases were found in 24
(3.5%) and distant metastases in 16 (2.3%) pa-
tients. The mean patient height and weight was
1.69 m and 77 kg, respectively. The mean BMI
(normal range 18.5 to 25) was 26.8 4.4 (range
16.9 to 44.3). The BMI was not available because of
lacking data in 74 (10.8%) of 683 patients. Of 609
assessable patients, a normal BMI was found in 236
(38.7%), an overweight BMI (25 to 30) in 248
(40.7%), and an obese BMI (greater than 30) in 123
(20.2%). Only 18 (2.6%) of 683 patients reported
weight loss as a symptom, and a BMI of less than
18.5 (by definition indicates underweight) was
present only in 3 patients (0.5%).
The statistical analysis results in Table I show
that weight (P 0.61), height (P 0.829), and,
consequently, BMI (P 0.82) did not correlate
with an advanced T stage or tumor diameter in
patients with RCC. Also, no statistically significant
correlation with BMI was noted for grade (r
0.063, P 0.122). Sex also showed no correla-
tion with T stage (P 0.57). On the contrary, a
statistically significant positive correlation was
found between pathologic T stage and the presence
of symptoms, ESR, LDH, and age, and a statistically
significant inverse correlation was noted with he-
moglobin and total serum cholesterol. A nonstatis-
tically significant tendency was noted toward an
inverse correlation with serum triglycerides (P
0.062).
The mean and median follow-up was 44 and 46
months, respectively. Fourteen patients were lost
to follow-up. Overall, 141 patients (20.6%) died,
including 86 (12.6%) of metastatic RCC; 29 (4.3%)
were alive with metastatic disease.
COMMENT
The incidence of RCC has shown a continuous
increase during the past 15 to 20 years in Europe
UROLOGY 62 (3), 2003
 TABLE I. Statistical results of correlation with T stage
Spearman Rank Correlation Test
Patients
Median SD Range r P Value (n)
Height (m) 170 9 145198 0.01 0.82 612
Weight (kg) 77 14.8 39170 0.02 0.61 640
BMI 26.2 4.4 16.844.2 0.01 0.82 609
Hemoglobin (g/100 mL) 14.4 5.9 7.315.8 0.16 0.0001 647
LDH (mg/100 mL) 160 57.5 74675 0.13 0.0015 643
ESR 1st hr (mm/hr) 14 27.7 0130 0.25 0.0001 627
ESR 2nd hr (mm/hr) 32 32.5 0150 0.26 0.0001 626
Cholesterol (mg/100 mL) 215 45.7 88374 0.2 0.0001 645
Triglycerides (mg/100 mL) 117 109.5 251850 0.07 0.062 646
Age (yr) 63.1 11.4 16.187.6 0.08 0.046 683
Sex 0.02 0.58 683
Symptoms 0.25 0.0001 683
KEY: BMI body mass index; LDH lactate dehydrogenase; ESR erythrocyte sedimentation rate.

and North America.1 This observation may be at
least in part a result of the widespread use of rou-
tine ultrasonography, which is responsible for the
detection of asymptomatic and frequently early-
stage tumors.19,20 In our series of almost 700 pa-
tients, 80% were detected without symptoms. In
contrast to the United States, routine abdominal
ultrasonography is very frequently performed in
Europe, even in patients without symptoms, not
only by radiologists, but also by urologists, inter-
nists, and general surgeons, leading to a high rate
of incidentally detected kidney tumors. Usually,
incidentally detected renal masses undergo surgi-
cal exploration unless the patient is not suitable for
general anesthesia because of severe comorbidity.
We routinely resect kidney tumors even in ad-
vanced stages. Therefore, the patients admitted to
our department were not selected but represented
the normal epidemiology. An increased risk of the
development of RCC in the obese has been re-
ported by several investigators.5,6,8,9 This observa-
tion is underlined by a mean BMI of 26.8 (normal
less than 25) in our series, which is higher than in
the general population. On the basis of this in-
creased incidence, it seems reasonable to assume a
correlation also exists between BMI and tumor
stage. A high BMI can be caused by either obesity
or muscle mass. However, because the mean age of
our patients was older than 60 years, the likelihood
of muscle mass being responsible for a high BMI
can be estimated as low and without statistical im-
pact. We are well aware that the BMI is only a
rough measure to define overweight and obesity
and that more exact parameters are available such
as the waist/hip ratio. However, because this was a
retrospective study and only patients weight and
height were routinely recorded at patient admis-
sion, the BMI was the only possible parameter to
determine those who were overweight and obese.
For the same reason, data regarding the patients
diet, smoking, and family history were lacking for
most patients and therefore were not included in
our study, a limitation of our results. On the other
hand, the assessment of the kidneys by ultrasonog-
raphy in obese patients might be more difficult,
with the possibility of missing small tumors and a
consequent delay in diagnosis and treatment to a
more advanced tumor stage. The findings in our
own series disprove these concerns, because no
correlation between pT stage and BMI was found.
As almost two thirds of our patients presented with
early-stage RCC, one might argue that the lack of a
correlation with BMI could have been due to low
statistical power. However, more than one third of
patients presented with advanced stage, and a
number greater than 250 can be assumed to be
sufficient to prove an association if present. The
power of the study can be assessed as follows: the
BMI values were approximately normally distrib-
uted, with mean SD of 26.8 4.4. Assuming a t
test with 200 patients with Stage 1 and 200 patients
with Stage 3a-b would give 80% power for a 1.2-
point difference in BMI. The Spearman correlation
coincided with the results of the Wilcoxon test
when the two groups were compared.
The Wilcoxon two-sample test has 95% effi-
ciency compared with the t test. Adding more pa-
tients in more pT stages would increase the power
if the relation between pT and BMI is monotone.
We had 396 patients with pT1 and 235 patients
with pT 3a-b in this study. Therefore, the study had
at least 80% power to detect a BMI difference of 1.2
points between these pT stages. Because we in-
tended to investigate the correlation of BMI with
tumor size at diagnosis, and tumor stage and grade
are available only after treatment, it was not possi-

UROLOGY 62 (3), 2003 439

ble to correct for stage and grade to see whether
obesity was an independent variable.
Because the TNM system for RCC does not pro-
vide a reliable association between tumor stage and
tumor volume (Stage pT1 tumor sizes range up to a
diameter of 7 cm and Stage pT3a tumor sizes can be
much smaller), we also investigated the association
between BMI and tumor diameter and failed to de-
tect a correlation as well. One might argue that the
results were flawed by the patients with advanced
tumor stage already presenting with tumor ca-
chexia. However, only 18 (2.6%) of 683 patients
reported weight loss as a symptom, and a BMI of
less than 18.5, which by definition means under-
weight, was present in only 0.5% of patients; a BMI
of more than 25 (overweight) was found in more
than 60%. A clear statistically significant correla-
tion could be proved between pT stage and symp-
toms, ESR, hemoglobin, cholesterol, and LDH.
However, this finding was not very surprising, be-
cause these associations are well established.21 Be-
cause we re-evaluated all the patients charts, we
decided to include these routine laboratory param-
eters, although they only confirmed older data of
patients presenting with advanced stage and do not
add much information, except for the validity of
these findings for asymptomatic patients as well.
Because the data regarding low-density-lipopro-
tein and high-density-lipoprotein cholesterol were
not available, we cannot provide information
about which lipoprotein is responsible for the in-
verse correlation between total serum cholesterol
and T stage. However, comparable data have been
reported for other cancer types. Knekt et al.,22 as
well as Eichholzer et al.,23 reported associations
between decreased serum cholesterol levels and
cancer progression and mortality. Whether this in-
verse correlation results from decreased synthesis,
increased uptake by the tumor (as reported by
Rudling and Collins24), or increased catabolism of
low-density-lipoprotein, as described by Henriks-
son et al.25 and Wannamethee et al.26 cannot cur-
rently be sufficiently answered. A large variety of
parameters, including some of those included in
our study, have been investigated with respect to
the prognosis of patients with RCC2730; however,
a correlation with tumor stage to our knowledge
has not yet been published. Although there is evi-
dence that obesity represents a risk factor for the
development of RCC,5,6,8,9,12,14,15 we were unable
to prove an association with advanced tumor stage
despite a sample size of more than 600 patients
assessable with respect to BMI, which can be re-
garded as an adequate number. Therefore, obesity,
which is associated with a large number of health
problems, does not inhibit or even compromise the
detection of asymptomatic early-stage RCC.
440
CONCLUSIONS
The pathologic T stage of RCC showed a signifi-
cant positive correlation with the presence of
symptoms, levels of LDH and ESR, and age, as well
as a negative correlation with hemoglobin and se-
rum cholesterol. A correlation between BMI and
tumor stage or tumor diameter could not be proved
in our series. Overweight patients developing RCC
are not at a greater risk of presenting with an ad-
vanced stage at diagnosis compared with patients
of normal weight.
REFERENCES
1. Chow WH, Devesa SS, Warren JL, et al: Rising inci-
dence of renal cell carcinoma in the United States. JAMA 281:
1628 1631, 1999.
2. Wolk A, Gridley G, Niwa S, et al: International renal
cell cancer study. VII. Role of diet. Int J Cancer 65: 6773,
1996.
3. Wolk A, Lindblad P, and Adami HO: Nutrition and
renal cell cancer. Cancer Causes Control 7: 518, 1996.
4. Chow WH, Gridley G, McLaughlin JK, et al: Protein
intake and risk of renal cell cancer. J Natl Cancer Inst 86:
11311139, 1994.
5. Moyad MA: Obesity, interrelated mechanisms, and ex-
posures and kidney cancer. Semin Urol Oncol 19: 270 279,
2001.
6. Chow WH, Gridley G, Fraumeni JF, et al: Obesity, hy-
pertension and the risk of kidney cancer in men. N Engl J Med
343: 13051311, 2000.
7. Schlehofer B, Pommer W, Mellengaard A, et al: Inter-
national renal-cell-cancer study. VI. The role of medical family
history. Int J Cancer 66: 723726, 1996.
8. Asal NR, Risser DR, Kadamani S, et al: Risk factors in
renal cell carcinoma: I. Methodology, demographics, tobacco,
beverage use, and obesity. Cancer Detect Prev 11: 359 377,
1988.
9. Yu MC, Mack TM, Hanisch R, et al: Cigarette smoking,
obesity diuretic use, and coffee consumption as risk factors for
renal cell cancer. J Natl Cancer Inst 77: 351356, 1986.
10. Oh WK, Manola J, Renshaw AA, et al: Smoking and
alcohol use may be risk factors for poorer outcome in patients
with clear cell renal cell carcinoma. Urology 55: 3135, 2000.
11. Tavani A, and La Vecchia C: Epidemiology of renal cell
carcinoma. J Nephrol 10: 93106, 1997.
12. Chow WH, McLaughlin JK, Mandel JS, et al: Obesity
and risk of renal cell cancer. Cancer Epidemiol Biomarkers
Prev 5: 1721, 1996.
13. Mellemgaard A, Moller H, Olsen JH, et al: Increased risk
of renal cell carcinoma among obese women. J Natl Cancer
Inst 83: 15811582, 1991.
14. Bergstrom A, Hsieh CC, Lindblad P, et al: Obesity and
renal cell cancera quantitative review. Br J Cancer 85: 984
990, 2001.
15. Shapiro JA, Williams MA, and Weiss NS: Body mass
index and risk of renal cell carcinoma. Epidemiology 10: 188
191, 1999.
16. Cui Y, Whiteman MK, Flaws JA, et al: Body mass and
stage of breast cancer at diagnosis. Int J Cancer 98: 279 283,
2002.
17. Pummer K, Stettner H, Trummer H, et al: Body mass
index is an independent predictor for organ confined prostate
cancer (abstract). J Urol 163(suppl): 183, 2000.
18. Yu ML, Asal NR, and Geyer JR: Later recurrence and
longer survival among obese patients with renal cell carci-
noma. Cancer 68: 1648 1655, 1991.
19. Tsui KH, Shvarts O, Smith RB, et al: Renal cell carci-
UROLOGY 62 (3), 2003
noma: prognostic significance of incidentally detected tumors.
J Urol 163: 426 430, 2000.
20. Luciani LG, Cestari R, and Tallarigo C: Incidental renal
cell carcinomaage and stage characterisation and clinical
implications: study of 1092 patients (19821997). Urology
56: 58 62, 2000.
21. Chisholm GD: Nephrogenic ridge tumors and their
syndromes. Ann NY Acad Sci 230: 403423, 1974.
22. Knekt P, Reunanen A, Aromaa A, et al: Serum choles-
terol and risk of cancer in a cohort of 39,000 men and women.
J Clin Epidemiol 41: 519 530, 1988.
23. Eichholzer M, Stahelin HB, Gutzwiller F, et al: Associ-
ation of low plasma cholesterol with mortality for cancer at
various sites in men: 17-y follow-up of the prospective Basel
study. Am J Clin Nutr 71: 569 574, 2000.
24. Rudling M, and Collins VP: Low density lipoprotein
receptor and 3-hydroxy-3-methylglutaryl coenzyme A reduc-
tase mRNA levels are coordinately reduced in human renal cell
carcinoma. Biochem Biophys Acta 1299: 7579, 1996.
UROLOGY 62 (3), 2003
25. Henriksson P, Eriksson M, Ericsson S, et al: Hypocho-
lesterolaemia and increased elimination of low-density li-
poproteins in metastatic cancer of the prostate. Lancet 2:
1178 1180, 1989.
26. Wannamethee G, Shaper AG, Whincup PH, et al: Low
serum total cholesterol concentrations and mortality in mid-
dle aged British men. BMJ 311: 409 413, 1995.
27. Lau WK, Cheville JC, Blute ML, et al: Prognostic fea-
tures of pathologic stage T1 renal cell carcinoma after radical
nephrectomy. Urology 59: 532537, 2002.
28. Yayciogly O, Roberts WW, Chan T, et al: Prognostic
assessment of nonmetastatic renal cell carcinoma: a clinically
based model. Urology 58: 141145, 2001.
29. Roosen JU, Engel U, Jensen RH, et al: Renal cell carci-
noma: prognostic factors. Br J Urol 74: 160 164, 1994.
30. Citterio G, Bertuzzi A, Tresoldi M, et al: Prognostic
factors for survival in metastatic renal cell carcinoma: retro-
spective analysis from 109 consecutive patients. Eur Urol 31:
286 291, 1997.
441