Human Reproduction vol.13 no.3 pp.

576582, 1998

Chromosome abnormalities in 447 couples undergoing intracytoplasmic sperm injectionprevalence, types, sex distribution and reproductive relevance

D.Meschede1,4, B.Lemcke1, J.R.Exeler1, Ch.De Geyter2, H.M.Behre2,3, E.Nieschlag3 and J.Horst1

1Institute

of Human Genetics, 2Womens Hospital and 3Institute of Reproductive Medicine of the University, D-48149 Munster, Germany whom correspondence should be addressed at: Institute of Human Genetics, Vesaliusweg 12-14, D-48149 Munster, Germany

4To

Chromosomal abnormalities are thought to be a major contributor to the genetic risks of infertility treatment by intracytoplasmic sperm injection (ICSI). Apart from abnormalities arising de novo, abnormal karyotypes in pregnancies conceived through assisted reproductive technology may be directly derived from predisposing parental aberrations. In a prospective study we have analysed the chromosomes of 868 male and female patients prior to planned ICSI treatment. A total of 33 aberrant karyotypes was diagnosed, corresponding to an abnormality rate of 7.6% per couple or 3.8% per individual studied. Even though male factor infertility was twice as common as female factor infertility in this cohort, 24 of the chromosomal abnormalities were found among the women. Lowlevel mosaicism for numerical sex chromosome anomalies was diagnosed in 20 individuals, and one patient had the triple X karyotype. With respect to structural chromosomal anomalies, we found six reciprocal and three Robertsonian translocations, two paracentric inversions and one marker chromosome. Many of the aberrations that we diagnosed could be classied as carrying only a small to moderate reproductive risk. Given the high rate of abnormal karyotypes among the female subjects, we suggest that not only the males, but both partners should be routinely karyotyped prior to ICSI. Key words: chromosome analysis/genetic counselling/ infertility/intracytoplasmic sperm injection

in children born after ICSI arise de novo without a predisposing parental chromosomal aberration (Van Opstal et al., 1997), others may be derived from a predisposing abnormality present in one of the parents (Kremer et al., 1997). Particular concern has been voiced about the fact that many infertile patients treated with ICSI might be carriers of chromosomal abnormalities that pose a genetic risk to their offspring (Bui and Wramsby, 1996; Rosenbusch et al., 1996). It is well documented that compared to unselected male newborns, infertile men have a higher prevalence of abnormal karyotypes (Chandley, 1984; Retief et al., 1984; Bourrouillou et al., 1985; De Braekeleer and Dao, 1991; Pandiyan and Jequier, 1996). This is particularly true for patients with the most pronounced impairments of fertility, such as azoospermia and severe oligozoospermia. Recent reports demonstrate that among infertile men enrolled in in-vitro fertilization (IVF) or ICSI programmes, the rates of chromosomal aberrations are markedly increased above the population baseline (Hens et al., 1988; Lange et al., 1993; Baschat et al., 1996b; Peschka et al., 1996; Testart et al., 1996; Mau et al., 1997). The present study, conducted in a large and prospectively ascertained cohort karyotyped prior to intracytoplasmic sperm injection, adds further to the database on the cytogenetics of ICSI patients. We report a surprisingly high rate of chromosomal abnormalities in the female partners of these couples, a nding relevant to the design of appropriate diagnostic strategies for the future. We analyse in detail how the observed chromosomal abnormalities are relevant in terms of genetic risk that results for these patients offspring. Materials and methods
Patient recruitment Patients for this study were prospectively selected from August 1994, when we established a genetic counselling programme for infertile couples considering ICSI therapy at Munster University Hospital or outside institutions. All couples were recommended to have a chromosome analysis performed for both partners prior to the planned treatment. Cytogenetic methods Chromosome analysis was carried out on cultured peripheral lymphocytes using standard techniques (Benn and Perle, 1992; Gosden et al., 1992). At least 11 cells were karyotyped, and in cases of suspected mosaicism the number of analysed metaphases was increased up to a total of 150. A resolution of 400 bands per haploid karyotype was achieved as a minimum, and often a more detailed structural analysis at the 550700 band level could be done. The routine analysis was performed on G-banded chromosome preparations. To characterize structural abnormalities or polymorphisms, additional Q-, C- or NORbanded preparations were studied as required (Benn and Perle, 1995).
European Society for Human Reproduction and Embryology

Introduction Intracytoplasmic sperm injection (ICSI) is now widely acknowledged as the most effective therapeutic approach to severe male factor infertility (Palermo et al., 1996a; Van Steirteghem et al., 1996). However, the question of possible adverse effects of this technique still attracts considerable attention, and the genetic safety issue plays a pre-eminent role in that debate (Engel and Schmid, 1995; Meschede et al., 1995; Patrizio, 1995; Baschat et al., 1996a; Chandley and Hargreave, 1996; Persson et al., 1996). Some of the genetic anomalies observed 576

Chromosome abnormalities in ICSI patients

In selected cases, uorescence in-situ hybridization (FISH) with various probes was employed to typify abnormal or variant chromosomes (Buckle and Kearney, 1994; Cremer et al., 1995). The origin of one marker chromosome was determined by chromosomal microdissection with consecutive reverse chromosome painting (Muller-Navia et al., 1995). All chromosomal abnormalities were reported in accordance with the current international standard nomenclature (Mitelman, 1995). Structural rearrangements conned to a single cell were disregarded, as were most single cell numerical abnormalities. This is common practice in routine cytogenetic analysis. We analysed more metaphases (usually 50) when single cells trisomic for chromosomes 8, 9, 13, 18 or 21 were observed, and in all cases with the loss or gain of sex chromosomes in single cells. If the extended analysis showed at least one more cell with the same type of abnormality this was regarded as representing true mosaicism. Since in low-level mosaicism for 45,X there is often a second abnormal cell line, other single cell abnormalities involving the sex chromosomes were also registered in such cases. Risk analysis and genetic counselling For genetic counselling the observed chromosomal abnormalities were grouped into four classes. Category A comprised anomalies that carry a low risk for giving rise to an aneuploid pregnancy; category B covered anomalies that implicate an increased risk for spontaneous pregnancy losses (abortion, stillbirth) secondary to aneuploidy, but not for a liveborn child with an unbalanced karyotype; category C included anomalies with a high risk both for abortion or stillbirth due to chromosomal abnormality and for aneuploidy in viable offspring; and category D included anomalies not unequivocally classiable with categories A through C and posing a potential (albeit probably small) risk for viable offspring with aneuploidy or clinically adverse uniparental disomy (Ledbetter and Engel, 1995). Usually, the risk for spontaneous abortions is also increased in this category. The risk analysis was based on experience with naturally conceived pregnancies, as compiled in the computerized database by Schinzel (1994) and reviewed by Gardner and Sutherland (1996). The factors considered in the risk analysis were: (i) history of aneuploid livebirths in the patients family; (ii) empirical risk gures from the published literature and in particular whether any reports exist about livebirths with the type of aneuploidy that the patients chromosomal abnormality could give rise to; (iii) sex of the carrier of the chromosomal anomaly. This factor strongly inuences the risk of unbalanced transmission of structural chromosome anomalies (Daniel et al., 1989); (iv) the phenotypic expression of the chromosome abnormality in its carrier, which is relevant in patients with mosaicism for a marker chromosome; and (v) the possibility of whether the observed chromosomal abnormality could indicate an increased risk for uniparental disomy (Ledbetter and Engel, 1995) in the patients offspring.

Results A total of 447 couples undergoing counselling before planned intracytoplasmic sperm injection was eligible for inclusion into the study. The indication for treatment was male factor infertility in 53.7% of the couples, combined male and female factor infertility in 44.7%, and female factor infertility in 1.4%. Data were lacking for one couple. Of the 894 patients, 25 (2.8%; 14 men and 11 women) rejected a chromosome analysis for personal reasons. One male patient was on continuous low-dose chemotherapy, and no cytogenetic result could be obtained, as three attempts to cultivate his lymphocytes failed.

Overall, 868 of 894 eligible individuals (97.1%; 432 men and 436 women) were karyotyped. Table I summarizes the 33 cases with abnormal chromosome results upon analysis. Twelve structural and 21 numerical anomalies were found. The structural abnormalities comprised ve reciprocal translocations, three Robertsonian translocations, two paracentric inversions, one marker chromosome in mosaicism, and one reciprocal translocation combined with mosaicism for a marker (derivative) chromosome. With a single exception, all numerical aberrations represented lowlevel mosaicism for a normal and from one to three abnormal cell lines. The two marker chromosomes were classied with the structural abnormalities, and two single cell ndings of structural X chromosome aberrations with the numerical anomalies. Nine abnormal karyotypes were found among the 432 men studied, and 24 abnormalities among the 436 female individuals. This corresponds to chromosomal abnormality rates of 2.1 and 5.5% in the male and female subpopulations respectively. The rate of abnormal karyotypes was 3.8% per individual (irrespective of sex) or 7.6% per couple studied. We did not observe any couples where both partners had an aberrant karyotype. The results of the risk analyses are also summarized in Table I. All 21 numerical chromosome anomalies and one paracentric inversion were classied as posing a low risk for a chromosomally abnormal pregnancy (category A). Two of the reciprocal translocations mainly pose a risk for recurrent pregnancy losses, but viable livebirths with aneuploidy appear improbable (category B). Four reciprocal translocations (one combined with mosaicism for a marker chromosome) carry a risk for producing both viable and nonviable unbalanced pregnancies (category C). One paracentric inversion, one marker chromosome (in mosaicism), and all three Robertsonian translocations were classied as probably posing a low risk for aneuploidy or uniparental disomy in liveborn offspring, but a small to moderate risk for these unfavourable outcomes may exist (category D). We are aware of pregnancies in seven of the 33 couples who had a chromosomal anomaly in one of the partners. The true pregnancy rate may be higher, as patients were referred from outside institutions and lost to follow-up after genetic counselling. The daughter of a male carrier of a Robertsonian translocation (case 3) inherited the anomaly in balanced form and is clinically healthy. Case 11 (reciprocal translocation plus marker chromosome) has an ongoing twin pregnancy and did not wish to undergo prenatal karyotyping. The male child of case 16 (47,XXX/45,X/46,XX mosaicism) was born prematurely in gestational week 28 and has not yet been examined. The pregnancies of one inversion carrier (case 4), one carrier of a reciprocal translocation (case 9) and two patients with low-level sex chromosome mosaicism (cases 19 and 24) were reported to us as uneventful, but no follow-up information or karyotypes of children are available. For all 12 patients diagnosed as carrying a structural chromosome aberration, we recommended a family study to trace the origin of the anomaly and identify other relatives with an abnormal karyotype. The families of six patients did not make use of this offer. Only one parent of cases 2 and 11 577

D.Meschede et al.

Table I. Summary of all cases with structural (cases 112) and numerical (cases 1333) chromosome abnormalities Case no. Sex 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 m m m m m m m f f f f f m m f f f f f f f f f f f f f f f f f f f Age 29 34 31 36 34 32 29 27 30 28 29 33 32 33 37 30 32 35 34 32 26 37 36 31 33 37 29 34 35 39 38 42 32 Karyotype 46,XY,t(2;21)(q33;q22) 46,XY,inv(12)(q15q24.1) 45,XY,der(13;15)(q10;q10) 46,XY,inv(11)(q22q24) 47,XY, mar/48,XY, mar, mar/46,XY 45,XY,der(13;14)(q10;q10)mat 46,XY,t(5;8)(q22;q24.1) 46,XX,t(2;12)(p21;q13) 46,XX,t(10;14)(q11.2;q32.1) 45,XX,der(13;14)(q10;q10) 47,XX,t(3;15)(p21.3;q26), der(2)(:p11.1q11.2:)/ 46,XX,t(3;15)(p21.3;q26) 46,XX,t(10;12)(p11.1;p13.3) 47,XXY/46,XY 45,X/46,XY 45,X/46,XX 47,XXX/45,X/46,XX 45,X/46,XX 45,X/46,XX 45,X/46,XX 45,X/47,XXX/49,XXXXX/46,XX 45,X/48,XXXX/48,XX,i(Xq),i(Xq)/46,XX 47,XXX 45,X/46,XX 45,X/46,XX 45,X/48,XXXX/46,XX 47,XXX/46,XX 45,X/46,XX 45,X/46,XXq-/46,XX 45,X/47,XXX/46,XX 45,X/47,XXX/46,XX 45,X/46,XX 45,X/47,XXX/46,XX 45,X/49,XXXXX/46,XX Size of cell linesa Risk to offspringb C D D A D D B B C D C C A A A A A A A A A A A A A A A A A A A A A (A through D) are

15/1/4

13/12 2/28 3/67 2/28 2/1/27 2/23 3/47 3/28 4/2/1/64 3/1/1/45 5/45 3/47 2/1/27 2/47 2/28 4/1/142 2/2/10 2/2/46 2/48 5/1/46 2/1/45

aIn cases with mosaicism the absolute number of cells from each cell line is specied. bWith respect to reproductive risks posed by the various chromosomal anomalies four classes

distinguished. For denition of these classes see text (Materials and methods section).

was alive, and both had a normal result upon chromosome analysis. One sister of case 9 was found to have a normal karyotype, but her parents and brother did not wish to be tested. The Robertsonian translocation of case 6 was also noted in her mother, her three sisters and a nephew. All the sisters and the mother had a history of multiple spontaneous abortions (from one to 10), and one had lost a malformed infant. The brother of case 7 was identied as a carrier of the reciprocal translocation, and a study of the parents is under way. The parents and three of four siblings of case 8 rejected a chromosome analysis, but one sister and her son underwent karyotyping and were diagnosed as balanced carriers of the reciprocal translocation between chromosomes 2 and 12 found in the index patient. Overall, of 11 rst or second degree relatives that we examined, eight had abnormal karyotypes. In all nine couples where the male partner was found to have an abnormal karyotype, male factor infertility was diagnosed; in three of these cases, female factor infertility was also diagnosed. Of 24 couples with a chromosomal abnormality in the female, there was exclusively male factor infertility in nine, and combined male and female factor infertility in 15. Table II summarizes the clinical diagnoses for all patients with an abnormal karyotype. 578

Discussion The rate of chromosomal anomalies detected in our patient cohort was 7.6% per couple or 3.8% per individual studied. For a clinically useful interpretation, these crude gures need to be broken down by the types of anomalies detected, the sex of the carriers, and the risk that the various abnormal karyotypes pose for the desired offspring. Twenty cases with an abnormal cytogenetic result had lowlevel mosaicism for a numerical sex chromosome anomaly. There was a strong preponderance of females among the subjects with this type of anomaly. None of these women had clinical evidence of Turner syndrome. Except for two cases, the aneuploid cell lines constituted 10% of the total number of metaphases studied. These few abnormal cells might represent culture or preparation artefacts. However, that explanation would be difcult to reconcile with the highly signicant skewing of the sex ratio. It is hard to imagine how laboratory artefacts should almost exclusively affect samples from the female, but not the male subjects. Therefore, the abnormal cell lines may indicate low-level mosaicism truely present in these patients lymphocytes. It is a well-established fact that the rate of metaphases aneuploid for the X chromosome increases with a womans advancing age (Fitzgerald and McEwan, 1977;

Chromosome abnormalities in ICSI patients

Table II. Indications for ICSI and clinical diagnoses in cases with abnormal karyotypes Case no. Class of chromosome Sex of carrierb abnormalitya 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 REC INV ROB INV MAR ROB REC REC REC ROB REC MAR REC SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS SCA-MOS m m m m m m m f f f f f m m f f f f f f f f f f f f f f f f f f f Indication for Clinical diagnosis in carrier of ICSIc chromosomal abnormalityd m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m m f f f OAT asthenozoospermia, unilateral orchidectomy oligoteratozoospermia asthenozoospermia, sperm antibodies oligozoospermia OAT, cryptorchidism OAT (no reproductive system abnormalities) endometriosis tubal obstruction corpus luteum insufciency unilateral tubal obstruction OAT, cryptorchidism OAT, cryptorchidism anovulation PCO syndrome gynaecological disorder (no reproductive system abnormalities) (no reproductive system abnormalities) (no reproductive system abnormalities) endometriosis, tubal obstruction (no reproductive system abnormalities) endocrine problems ovarian failure anovulation hyperprolactinaemia (no reproductive system abnormalities) hyperprolactinaemia (no reproductive system abnormalities) anovulation (no reproductive system abnormalities) anovulation, hyperprolactinaemia (no reproductive system abnormalities)

f f f f f f f

f f f f f f f f

aREC reciprocal translocation; INV inversion, ROB Robertsonian translocation; MAR marker chromosome; SCA sex chromosomal anomaly; SCA-M mosaicism for sex chromosomal anomaly. bm male, f female. cm male factor infertility, f female factor infertility. dOAT oligoasthenoteratozoospermia; PCO polycystic ovary syndrome; in some cases only incomplete data are available.

Guttenbach et al., 1995; Surralles et al., 1996), but the aberrant cell line is usually not found in skin broblasts (Horsman et al., 1987). The mean age of the 18 women with sex chromosome mosaicism in our cohort was 34.0 years and thus above the average of 31.6 years calculated for the rest of our female study population. We favour the hypothesis that at least a substantial fraction of those individuals diagnosed with lowlevel mosaicism actually have chromosomally abnormal cells in their blood. This is substantiated by the fact that in three of four patients who provided a second blood sample, the same type of mosaicism was conrmed upon re-examination, albeit usually with a lower proportion of abnormal cells than in the rst sample. The preceding considerations notwithstanding, it remains likely that some cases with apparent low-level mosaicism represent laboratory artifacts. The biological and prognostic signicance of low-level sex chromosome mosaicism is not clear. To postulate a connection between this type of chromosomal aberration and the fertility problem present in our patients would be highly speculative. It is most likely that the karyotype 45,X/46,XX in concurrence with clinical signs of Turner syndrome reects true constitutional mosaicism, and the risks for an abnormal pregnancy

outcome may be substantial (Kaneko et al., 1990). This is possibly different for women with low-level sex chromosome mosaicism and absent Turner syndrome signs, as was the case in all the subjects that we examined. In the relatively small study by Horsman et al. (1987), there was no indication that low-level mosaicism for the sex chromosomes heralded an increased risk for the birth of an aneuploid child. Low-level sex chromosome mosaicism is more commonly encountered among women with recurrent abortions than in controls with a normal reproductive history (Holzgreve et al., 1984; Ostrowski et al., 1993). Therefore, the risk for spontaneous pregnancy losses is probably somewhat increased in women with this type of chromosomal aberration. Gardner and Sutherland (1996, p. 201) suggest that sex chromosome mosaicism with 10% abnormal cells probably does not indicate specic reproductive risks in women without Turner syndrome stigmata. While that appears to be a useful rule of thumb for genetic counselling, we are not aware of sufciently large empirical studies addressing this issue. It may therefore be prudent to offer all patients with low-level sex chromosome mosaicism, especially if conrmed on a repeat sample, a prenatal diagnosis. It must be made clear, 579

D.Meschede et al.

Table III. Cytogenetic studies of patients enrolled in IVF or ICSI programmes Type of treatment Males planned or karyotyped performed (n) Hens et al. (1988) Lange et al. (1993) Baschat et al. (1996b) Mau et al. (1997) Peschka et al. (1996) Testart et al. (1996) Present series IVF, GIFT, ZIFT IVF ICSI ICSI ICSI ICSI ICSI 500 72 32 150 200 261 432 Females karyotyped (n) 500 72 150 200 261 436 Males: abnormal karyotypes (%) 4 (0.8) 2 (2.8) 2 (6.2) 18 (12.0) 6 (3.0) 11 (4.2) 9 (2.1) Females: abnormal karyotypes (%) 9 (1.8) 11 (15.3) 9 6 3 24 (6.0) (3.0) (1.1) (5.5)

IVF in-vitro fertilization; GIFT gamete intra-Fallopian transfer; ZIFT Fallopian transfer; ICSI intracytoplasmic sperm injection.

zygote intra-

however, that it is not proven whether this type of mosaicism does justify exposing pregnancies to the procedure-related risks of invasive prenatal diagnosis. The same consideration applies to women with a 47,XXX karyotype, as diagnosed in one subject from our series. In fact, it is likely that such women have the same chances of euploid offspring as agematched peers (reviewed in Gardner and Sutherland, 1996). Even if the 20 cases of low-level sex chromosome mosaicism are disregarded in the analysis, an almost equal number of abnormal karyotypes among the men and the women in our study remains [seven of 432 males (1.6%) and six of 436 females (1.4%)]. This is an unexpected nding, given the clear preponderance of male fertility problems in this cohort. Testart and coworkers (1996) karyotyped 261 couples prior to ICSI and reported chromosomal abnormality rates of 4.2 and 1.1% in the male and female subpopulations respectively. Similarly, Bonduelle et al. (1996) detected nine male, but only two female carriers of chromosomal aberrations among the parents of 877 children born after ICSI. From a cohort of 150 couples treated with ICSI, Mau et al. (1997) reported chromosomal abnormality rates of 12 and 6% for the male and female subpopulations respectively. There is one recent publication, however, with ndings similar to ours. Peschka and coworkers (1996) diagnosed a total of 12 abnormal karyotypes among 200 couples studied, and of these, six each were found in males and females respectively. It is difcult to reconcile these contradictory data, and it is possible that the divergent observations may be explained by random variations that could cancel themselves out as more data become available. Table III reviews cytogenetic series that included only patients enrolled in either an IVF or an ICSI programme. A comparison with the ~1% chromosomal abnormality rate in unselected newborns (Nielsen and Wohlert, 1991; Jacobs et al., 1992) shows that these patients are predisposed to have abnormal karyotypes. If only couples treated with ICSI are considered, our series has the lowest prevalence of chromosomal anomalies in the male partners, and the second highest rate among the females. The very high abnormality rates communicated by Mau et al. (1997) are partially due to inclusion of aberrant single cell ndings. If these are subtracted, 3.3% aberrations among their female and 7.3% among their male study subjects remain. Interestingly, working on a group 580

of patients undergoing conventional IVF treatment, Lange et al. (1993) obtained results quite similar to ours. Not only were chromosomal aberrations very common among the women of this cohort and more prevalent than in the males, also the types of anomalies observed in the female subcohort strongly overlap with our ndings. The latter is also true for the study by Mau et al. (1997). Robertsonian translocations, as diagnosed in three of our patients, are thought to implicate a low to moderate risk (1 2% or less) for viable offspring with an unbalanced karyotype (Gardner and Sutherland, 1996). It is important to note that all homologous translocations and the non-homologous translocations involving chromosome 21 in female carriers are exceptions to this rule and carry a much higher risk. Robertsonian translocations may also predispose to uniparental disomy (UPD) in the offspring (Ledbetter and Engel, 1995). There are no empirical data available yet, but the risk for this type of genetic anomaly is probably not very high. Nevertheless, the possibility of UPD needs to be considered in genetic counselling, because it can have serious clinical consequences (Ledbetter and Engel, 1995; Kotzot and Braun-Quentin, 1996). This is particularly true for UPD of the chromosomes 14 and 15 (Temple et al., 1991; Pentao et al., 1992; Nicholls, 1993; Cotter et al., 1997). We advise patients with a 13;14, 13;15 or 14;15 Robertsonian translocation that they have a clearly increased risk for spontaneous pregnancy losses, and a risk for viable offspring with either aneuploidy or uniparental disomy probably in the low to moderate range. Invasive prenatal diagnosis including testing for UPD is generally recommended. In terms of their reproductive relevance, reciprocal translocations, inversions, and marker chromosomes need to be assessed on an individual basis. It is beyond the scope of this paper to discuss how risk gures for these types of structural anomalies are derived. For more specic information the reader is referred to several excellent original articles and reviews (Daniel et al., 1989; Cohen et al., 1994; Madan, 1995; Muller-Navia et al., 1995; Pettenati et al., 1995; Gardner and Sutherland, 1996). For many patients, the most serious adverse consequence of a chromosomal abnormality is not a spontaneous pregnancy loss, but the birth of a handicapped aneuploid child. With regard to this outcome, reciprocal translocations carry the highest risk.

Chromosome abnormalities in ICSI patients

This is also reected in the categorization of the chromosomal anomalies detected in our study. Only four reciprocal translocations were classied as posing a high risk for viable aneuploid offspring, while all the other structural aberrations and the numerical anomalies were considered as less dangerous in this regard. It should not go unmentioned that all risk estimations are currently based on the experience with naturally conceived pregnancies. There is no indication so far that IVF or ICSI have a modifying effect in the sense that a chromosomally unfavourable outcome becomes more or less likely. As the database is still small, however, caution must be exercised and patients should be informed that currently the only reliable safeguard against unexpected chromosomal anomalies in ICSI pregnancies is prenatal karyotyping. Several groups have reported follow-up data on pregnancies conceived through ICSI, and some of them include information about chromosomal abnormalities. Preliminary evidence suggests that sex chromosomal anomalies may be more common than in natural pregnancies (Int Veld et al., 1995; Liebaers et al., 1995; Van Opstal et al., 1997), and recently also a considerable number of trisomy 18 and 21 cases were reported (Palermo et al., 1996b; Van Opstal et al., 1997). All these chromosomal anomalies arose de novo, possibly in connection with advanced reproductive age of the female partners or, more speculatively, an increased rate of sperm chromosome aneuploidy among infertile men with normal lymphocyte karyotypes (Moosani et al., 1995; Martin, 1996; Int Veld et al., 1997; Van Opstal et al., 1997). Interestingly however, these preliminary data also show that the risk for a liveborn child with an unbalanced karyotype that arose from a predisposing parental anomaly may be rather low after ICSI treatment. Govaerts and colleagues (1995) mention that out of 55 fetuses karyotyped, three inherited a translocation from one of the parents. Although not explicitly mentioned, it is obvious that all these chromosomal aberrations were passed on in the balanced state. In their most recent follow-up report, Bonduelle et al. (1996) describe ve ICSI pregnancies where a parental anomaly (two inversions, two Robertsonian translocations, one marker chromosome) was inherited by the fetus in the balanced form. The same sort of outcome is reported by Testart et al. (1996) for ve fetuses whose parents either had a Robertsonian translocation or an inversion. The published literature contains only two instances of malsegregation. Baschat et al. (1996b) observed a twin pregnancy where one of the children displayed an unbalanced reciprocal Y;22 translocation inherited from the father. However, in this case, adverse clinical consequences were not be expected because there is only genetically inactive chromatin distal to the translocation breakpoints. Thus, while technically unbalanced, the abnormal karyotype diagnosed in this fetus is compatible with normal health. Our group recently reported on a chromosomally unbalanced ICSI pregnancy with one twin trisomic for almost the complete short arm of chromosome 9 (Meschede et al., 1997). The father has a corresponding balanced reciprocal translocation between chromosomes 1 and 9. To our knowledge, this is the only published case from ICSI series where a clinically relevant chromosomal abnormality in a fetus or child was derived directly from a predisposing parental anomaly.

In summary, there is no doubt that abnormal karyotypes are more prevalent among infertile men and women enrolled in ICSI programmes than in normal control populations. However, it appears as though the risk for a viable pregnancy with a chromosomal abnormality originating from a parental aberration may be not very high even in this population. Further studies to establish empirical risk gures are urgently needed, but due to the relatively small number of ICSI patients with abnormal karyotypes the accumulation of such a database will require considerable time. At least until then, we suggest that karyotyping both partners should be recommended routinely prior to a planned treatment with intracytoplasmic sperm injection.

Acknowledgement
We are grateful to Dr Jutta Muller-Navia (Mainz) for performing the chromosomal microdissection with reverse chromosome painting in case 11.

References
Baschat, A.A., Schwinger, E. and Diedrich, K. (1996a) Assisted reproductive techniques are we avoiding the genetic issues? Hum. Reprod., 11, 926928. Baschat, A.A., Kupker, W., Al Hasani, S. et al. (1996b) Results of cytogenetic analysis in men with severe subfertility prior to intracytoplasmic sperm injection. Hum. Reprod., 11, 330333. Benn, P.A. and Perle, M.A. (1992) Chromosome staining and banding techniques. In Rooney, D.E. and Czepulkowski, B.H. (eds), Human Cytogenetics. A Practical Approach. Vol. I: Constitutional Analysis. IRL Press, Oxford, pp. 91118. Bonduelle, M., Wilikens, A., Buysse, A. et al. (1996) Prospective follow-up study of 877 children born after intracytoplasmic sperm injection (ICSI), with ejaculated, epididymal and testicular spermatozoa and after replacement of cryopreserved embryos obtained after ICSI. Hum. Reprod., 11 (Suppl. 4), 131155. Bourrouillou, G., Dastugue, N. and Colombies, P. (1985) Chromosome studies in 952 males with a sperm count below 10 million/ml. Hum. Genet., 71, 366367. Buckle, V.J. and Kearney, L. (1994) New methods in cytogenetics. Curr. Opin. Genet. Dev., 4, 374382. Bui, T.-H. and Wramsby, H. (1996) Micromanipulative assisted fertilization still clinical research. Hum. Reprod., 11, 925926. Chandley, A.C. (1984) Infertility and chromosome abnormality. Oxf. Rev. Reprod. Biol., 6, 146. Chandley, A.C. and Hargreave, T.B. (1996) Genetic anomaly and ICSI. Hum. Reprod., 11, 930932. Cohen, O., Cans, C., Mermet, M.-A. et al. (1994) Viability thresholds for partial trisomies and monosomies. A study of 1159 viable unbalanced reciprocal translocations. Hum. Genet., 93, 188194. Cotter, P.D., Kaffe, S., McCurdy, L.D. et al. (1997) Paternal uniparental disomy of chromosome 14: a case report and review. Am. J. Med. Genet., 70, 7479. Cremer, T., Jauch, A., Ried, T. et al. (1995) Fluoreszenz-in-situ-Hybridisierung (FISH). Dt. Arztebl., 92, A-15931601. Daniel, A., Hook, E.B. and Wulf, G. (1989) Risks of unbalanced progeny at amniocentesis to carriers of chromosome rearrangements: data from United States and Canadian laboratories. Am. J. Med. Genet., 31, 1453. De Braekeleer, M. and Dao, T.-N. (1991) Cytogenetic studies in male infertility: a review. Hum. Reprod., 6, 245250. Engel, W. and Schmid, M. (1995) Gibt es genetische Risiken der mikroassistierten Reproduktion? Fertilitat, 11, 214228. Fitzgerald, P.H. and McEwan, C. (1977) Total aneuploidy and age-related sex chromosome aneuploidy in cultured lymphocytes of normal men and women. Hum. Genet., 39, 329337. Gardner, R.J.M. and Sutherland, G.R. (1996) Chromosome Abnormalities and Genetic Counselling. Oxford University Press, New York, 478 pp.

581

D.Meschede et al. Gosden, C.M., Davidson, C. and Robertson, M. (1992) Lymphocyte culture. In Rooney, D.E. and Czepulkowski, B.H. (eds), Human Cytogenetics. A Practical Approach. Vol. I: Constitutional Analysis. IRL Press, Oxford, pp. 3154. Govaerts, I., Englert, Y., Vamos, E. et al. (1995) Sex chromosome abnormalities after intracytoplasmatic sperm injection. Lancet, 346, 10951096. Guttenbach, M., Koschorz, B., Bernthaler, U. et al. (1995) Sex chromosome loss and aging: in situ hybridization studies on human interphase nuclei. Am. J. Hum. Genet., 57, 11431150. Hens, L., Bonduelle, M., Liebaers, I. et al. (1988) Chromosome aberrations in 500 couples referred for in-vitro fertilization or related infertility treatment. Hum. Reprod., 3, 451457. Holzgreve, W., Schonberg, S.A., Douglas, R.G. et al. (1984) X-chromosome hyperploidy in couples with multiple spontaneous abortions. Obstet. Gynecol., 63, 237240. Horsman, D.E., Dill, F.J., McGillivray, B.C. et al. (1987) X chromosome aneuploidy in lymphocyte cultures from women with recurrent spontaneous abortions. Am. J. Med. Genet., 28, 981987. Int Veld, P., Brandenburg, H., Verhoeff, A. et al. (1995) Sex chromosomal anomalies and intracytoplasmic sperm injection. Lancet, 346, 773. Int Veld, P.A., Broekmans, F.J.M., de France, H.F. et al. (1997) Intracytoplasmic sperm injection (ICSI) and chromosomally abnormal spermatozoa. Hum. Reprod., 12, 752754. Jacobs, P.A., Browne, C., Gregson, N. et al. (1992) Estimates of the frequency of chromosome abnormalities in unselected newborns using moderate levels of banding. J. Med. Genet., 29, 103108. Kaneko, N., Kawagoe, S. and Hiroi, M. (1990) Turners syndrome review of the literature with reference to successful pregnancy outcome. Gynecol. Obstet. Invest., 29, 8187. Kotzot, D. and Braun-Quentin, C. (1996) Uniparentale Disomie und genomische Pragung (genomic imprinting). Ursache seltener Dysmorphiesyndrome. Monatssch. Kinderheilkd., 114, 786792. Kremer, J.A.M., Tuerlings, J.H.A.M., Meuleman, E.J.H. et al. Microdeletions of the Y chromosome and intracytoplasmic sperm injection: from gene to clinic. Hum. Reprod., 12, 687691. Lange, R., Johannson, G., Hinney, B. et al. (1993) Zytogenetische Untersuchungen bei IVF-Patienten. Fertilitat, 9, 4144. Ledbetter, D.H. and Engel, E. (1995) Uniparental disomy in humans: development of an imprinting map and its implications for prenatal diagnosis. Hum. Mol. Genet., 4, 17571764. Liebaers, I., Bonduelle, M., Van Assche, E. et al. (1995) Sex chromosome abnormalities after intracytoplasmic sperm injection. Lancet, 346, 1095. Madan, K. (1995) Paracentric inversions: a review. Hum. Genet., 96, 503515. Martin, R.H. (1996) The risk of chromosomal anomalies following ICSI. Hum. Reprod., 11, 924925. Mau, U.A., Backert, I.T., Kaiser, P. et al. (1997) Chromosomal ndings in 150 couples referred for genetic counselling prior to intracytoplasmic sperm injection. Hum. Reprod., 12, 930937. Meschede, D., De Geyter, C., Nieschlag, E. et al. (1995) Genetic risk in micromanipulative assisted reproduction. Hum. Reprod., 10, 28802886. Meschede, D., Louwen, F., Eiben, B. et al. (1997) Intracytoplasmic sperm injection pregnancy with fetal trisomy 9p resulting from a balanced paternal translocation. Hum. Reprod., 12, 19131914. Mitelman, F. (ed.) (1995) ISCN 1995 An International System for Human Cytogenetic Nomenclature (1995). Karger, Basel, 114 pp. Moosani, N., Pattinson, H.A., Carter, M.D. et al. (1995) Chromosomal analysis of sperm from men with idiopathic infertility using sperm karyotyping and uorescence in situ hybridization. Fertil. Steril., 64, 811817 Muller-Navia, J., Nebel, A. and Schleiermacher, E. (1995) Complete and precise characterization of marker chromosomes by application of microdissection in prenatal diagnosis. Hum. Genet., 96, 661667. Nicholls, R.D. (1993) Genomic imprinting and uniparental disomy in Angelman and PraderWilli syndromes: a review. Am. J. Med. Genet., 46, 1625. Nielsen, J. and Wohlert, M. (1991) Chromosome abnormalities found among 34910 newborn children: results from a 13-year incidence study in Arhus, Denmark. Hum. Genet., 87, 8183. Ostrowski, R.S., Grass, F., Norton, J. et al. (1993) Evaluation of X chromosomal mosaicism by uorescent in situ hybridization in women who have recurrent spontaneous pregnancy losses. Am. J. Hum. Genet., 53, Abstract 587. Palermo, G.D., Cohen, J. and Rosenwaks, Z. (1996a) Intracytoplasmic sperm injection: a powerful tool to overcome fertilization failure. Fertil. Steril., 65, 899908. Palermo, G.D., Colombero, L.T., Schattman, G.D. et al. (1996b) Evolution of pregnancies and initial follow-up of newborns delivered after intracytoplasmic sperm injection. JAMA, 276, 18931897. Pandiyan, N. and Jequier, A.M. (1996) Mitotic chromosomal anomalies among 1210 infertile men. Hum. Reprod., 11, 26042608. Patrizio, P. (1995) Intracytoplasmic sperm injection (ICSI): potential genetic concerns. Hum. Reprod., 10, 25202523. Pentao, L., Lewis, R.A., Ledbetter, D.H. et al. (1992) Maternal uniparental isodisomy of chromosome 14: association with autosomal recessive rod monochromacy. Am. J. Hum. Genet., 50, 690699. Persson, J.W., Peters, G.B. and Saunders, D.M. (1996) Is ICSI associated with risks for genetic disease? Implications for counselling, practice and research. Hum. Reprod., 11, 921924. Peschka, B., Schwanitz, G., van der Ven, K. et al. (1996) Type and frequency of constitutional chromosome aberrations in couples undergoing ICSI. Hum. Reprod., 11, 224225. Pettenati, M.J., Rao, P.N., Phelan, M.C. et al. (1995) Paracentric inversions in humans: a review of 446 paracentric inversions with presentation of 120 new cases. Am. J. Med. Genet., 55, 171187. Retief, A.E., Van Zyl, J.A., Menkveld, R. et al. (1984) Chromosome studies in 496 infertile males with a sperm count below 10 million/ml. Hum. Genet., 66, 162164. Rosenbusch, B., Strehler, E. and Sterzik, K. (1996) Micro-assisted fertilization and sperm chromosome abnormalities. Hum. Reprod., 11, 928930. Schinzel, A. (1994) Human Cytogenetics Database (Version 1.0). Oxford University Press, Oxford. Surralles, J., Jeppesen, P., Morrison, H. et al. (1996) Analysis of loss of inactive X chromosomes in interphase cells. Am. J. Hum. Genet., 59, 10911096. Temple, I.K., Cockwell, A., Hassold, T. et al. (1991) Maternal uniparental disomy for chromosome 14. J. Med. Genet., 28, 511514. Testart, J., Gautier, E., Brami, C. et al. (1996) Intracytoplasmic sperm injection in infertile patients with structural chromosome abnormalities. Hum. Reprod., 11, 26092612. Van Opstal, D., Los, F.J., Ramlakhan, S. et al. (1997) Determination of the parent of origin in nine cases of prenatally detected chromosome aberrations found after intracytoplasmic sperm injection. Hum. Reprod., 12, 682686. Van Steirteghem, A., Nagy, P., Joris, H. et al. (1996) The development of intracytoplasmic sperm injection. Hum. Reprod., 11 (Suppl. 1), 5972. Received on July 2, 1997; accepted on November 17, 1997

582