Trisomy 18: Changes in sex ratio during intrauterine life

To the Editor: Among live-births generally more males are born than females. In Switzerland sex ratio of healthy newborns was 1.04 (male to female: 37,340 to 35,742) in 2004 [Bundesamt für Statistik, 2005]. Considering the higher prenatal loss rate of males, Huether [1990] extrapolated a sex ratio at conception for whites of approximately 1.15. Trisomy 18 is a multiple malformation syndrome; many of the live-born children die during the first days of life. For fetuses with trisomy 18 diagnosed at amniocentesis (usually at about 16–20 gestational weeks), Hook et al. [1983] calculated a fetal survival coefficient of 0.32. Hook et al. [1989] further observed spontaneous fetal death in 81.8% (9 of 11) of males and 58.3% (7 of 12) of females. Distinctly more females than males are born alive [Ferguson-Smith, 1962; Weber, 1967; Taylor, 1968; Goldstein and Nielson, 1988; Root and Carey, 1994]. Huether et al. [1996] reported a substantial loss of males with trisomy 18 during the second half of fetal development. They reported a prenatal sex ratio of 0.9 for trisomy 18 before and after 16 weeks of gestation and of 0.69 at birth. Not only are more males lost during pregnancy, but also the survival after birth for males seems to be significantly shorter than that of females [Weber, 1967; Van Dyke and Allen, 1990; Root and Carey, 1994; Rasmussen et al., 2003; Niedrist et al., 2006]. Since most of the first trimester pregnancies with trisomy 18 are terminated after prenatal diagnosis, we know little about what would have been the spontaneous outcome for the different sexes. In order to get all available information about the change of the sex ratio in pregnancies with trisomy 18, we collected data about the sex distribution at the time of cytogenetic diagnosis. In this study, we included not only the prenatally examined cases but also the postnatally detected cases (spontaneous abortions, stillborns, and live-borns) in whom the gestational age at birth was known in order not to miss sex differences in the third trimester of pregnancy. We collected data on 352 cases with trisomy 18 from two cytogenetic laboratories in Zurich between January 1964 and May 2003. Part of this study has already been published elsewhere [Niedrist et al., 2006]. For the present study the information about the time of gestation is based on the data given by the referring doctors on the application form and the information about the karyotype on the cytogenetic laboratory results. The time of the chorionic villus sampling or amniocentesis in our study was decided by the referring doctors and we collected the data retrospectively. Cases of mosaic trisomy 18 were excluded. In prenatal cases we registered the gestational age at the time of the first cytogenetic examination. If there has been no prenatal intervention, we used, in live- or stillborn children, the gestational age at delivery. For statistical analysis we used SPSS version 13 [2000–2006]. Approval by an ethics committee or Institutional Review Board was not required by the authors' institution for the present study. In 320 cases of trisomy 18 the week of gestation at the time of cytogenetic intervention and the sex were known (Fig. 1). In 194 cases the cytogenetic diagnosis was obtained prenatally (94 males, 100 females) and in 126 cases (41 males, 85 females) the diagnosis was performed postnatally. In 32 cases either the week of gestation at diagnosis/birth or the sex was not known: 14 males, 15 females, 3 cases of unknown sex, 6 prenatal diagnoses, 26 postnatal diagnoses. Diagrammatic representation of males versus females at different stages of pregnancy. Data from 320 cases with trisomy 18 from Niedrist et al. [2006]. In the present study population 103 females and 58 males with trisomy 18 were born alive (sex ratio 0.56) [published data in Niedrist et al., 2006]. This sex ratio of just 36% males (95% CI: 28.6–43.4% males) is significantly different from the sex ratio of 1.04 of healthy newborns in Switzerland 2004 (binomial test, P < 0.05). For cases with trisomy 18 diagnosed between 10 and 160/7 weeks of gestation, the sex ratio was 1.02 (60 males/59 females), between 161/7 and 280/7 weeks it was 0.82 (27/33), and between 281/7 and 43 weeks it was 0.52 (48/93). The sex ratios in the different time periods are significantly different (Pearson chi-square P < 0.05). The sex ratios before (66 males, 68 females) and after (69 males, 117 females) 18 weeks of gestation are significantly different as well (Pearson chi-square P < 0.05). The sex ratio in our study before 18 weeks of gestation is not significantly different from the sex ratio at conception for whites of approximately 1.15 or the sex ratio for healthy newborns in Switzerland 2004 of 1.04 (binomial test, P > 0.05). On the basis of the fetal death rate for trisomy 18 after amniocentesis of about 68% [Hook et al., 1983] and the live-born sex ratio in our study group of 0.56, we calculated a risk of about 60% of fetal demise for female fetuses after 18 weeks of gestation till live-birth, while for male fetuses, this rate would be about 77% (Table I). The odds ratio is 2.2. When calculating the risk for fetal demise between 18 weeks of gestation and live-birth, we used the referral data of Hook et al. [1983]. They collected data from cytogenetically confirmed trisomy 18 pregnancies, as we did in our study. They based their figures on pregnancies where onset of the observation was defined as usually 16–20 gestational weeks. These results were similar to Hook [1978]; Hook et al. [1989]; and Embleton et al. [1996]. In the literature, there are studies with a lower risk for fetal demise in trisomy 18 [Won et al., 2005], but these studies were either not based on cytogenetic laboratory reporting, or the median onset of observation is later in pregnancy. The present study confirms the observation by Hook et al. [1989] and the results from Huether et al. [1996] and presents more details in risk calculation for the different sexes in prenatal diagnosis of trisomy 18. We conclude on the basis of this study that in trisomy 18 after 18 weeks of gestation, the risk for intrauterine demise in males is twice as high as the risk for intrauterine demise in females. In postnatal survival studies there has already been shown a twofold risk for death in males compared with females [Niedrist et al., 2006]. Our results now show that this risk already exists prenatally after 18 weeks of gestation and that there is a continuous twofold higher prenatal and postnatal loss of male versus female probands with trisomy 18. No specific pattern for timing of loss in trisomy 18 fetuses after diagnostic amniocentesis was shown by Won et al. [2005]. We further conclude that the probability of intrauterine death for fetuses diagnosed with trisomy 18 during the first and early second trimester of pregnancy must be provided differently for female versus male fetuses. This statistic must be considered at genetic counseling when prenatal diagnosis discloses fetal trisomy 18. There are no reports in the literature about differences in the number of male and female cases with trisomy 18 at conception. We assume that the time of cytogenetic examination in a pregnancy with fetal trisomy 18 is independent from the sex of the fetus. To the best of our knowledge, there is no hint in the literature for differences in the incidence and severity of malformations between males and females with trisomy 18, which secondary to fetal ultrasonographic anomalies, would cause earlier or more frequent diagnosis in male as compared to female fetuses. The time of gestation in our study is based on the data obtained from the referring gynecologists and the determination of the sex on the karyotype; for both, we assume a high accuracy. In just 32 cases either the week of gestation at diagnosis/birth or the sex was not known. We assume these cases would have had no or only small influence on the results of the present study. Up to now the reasons for the excessive loss of males before and after birth are not known. Further studies on this issue are necessary.