The chance of two chromosome abnormalities occurring in one conceptus is very small. However, some authors have suggested that double aneuplodies (DAs) might be more common than the product of their individual frequencies. The nonrandomness of such DA events was considered to be evidence that nondisjunction (NDJ) may be genetically determined. Data collected from the National Down syndrome Cytogenetic Register (NDSCR) in England and Wales and from the literature indicate that the frequencies of all nonmosaic DAs, except for 48,XXY,+21, are lower than expected, probably because of strong intrauterine selection against such pregnancies. Collectively, we identified 52 cases of nonmosaic 48,XXY,+21; 28 cases of 48,XYY,+21; and 14 cases of 48,XXX,+21 in liveborns and 13 cases of 48,XXY,+21; four cases of 48,XYY,+21; and two cases of 48,XXX,+21 after prenatal diagnoses. Among these cases, analysis of the published unbiased cytogenetic surveys of liveborn DS revealed 24 cases of 48,XXY,+21; nine cases of 48,XYY,+21; and seven cases of 48,XXX,+21. These figures are different from the expected proportion of 1:1:1 (P < 0.001), with carriers of XXY overrepresented in the group of carriers of DA. Mechanisms put forth to account for the higher occurrence of 48,XXY,+21 may include greater accessibility of disomic ovum to Y-carrying sperm, and promotion of NDJ in ovum by Y-bearing sperm. 48,XXY,+21 DA was found to be age-dependent, as the proportion of mothers over age 35 (x = 33.0) was increased over the general population. This is in contrast to the apparently age-independent 48,XYY,+21 DA, with a mean maternal age of 24.7 (P < 0.001). Paternal ages were also remarkably different between the groups, with a mean age of 37.9 in 48,XXY,+21 cases and a mean age of 27.9 in 48,XYY,+21 cases (P < 0.01). Maternal age-related factors, rather than genetic predisposition, may play a more important role in the etiology of the most common DA, 48,XXY,+21.

The predominance of females segregating chromosome aberrations to their offspring has been explained mostly by selection disadvantage of unbalanced products of spermatogenesis. However, analysis of data from the literature supports the idea that somatic cells of early female embryos are similar to female germ cells in that they are prone to malsegregation. The goal of this study was to compare the sex ratio (male to female ratio) of carriers of presumably mitotic-occurring chromosome abnormalities to identify any sex biases. In examining the literature, we found a female prevalence in cases of mosaicism associated with uniparental disomy (UPD) (26 male individuals/conceptions and 45 female individuals/conceptions, sex ratio is 0.58, significantly different from 1.06 in newborn population, P = 0.0292). This predominance was highest at gestational age <16 week (8 male and 22 female conceptuses, sex ratio is 0.36, significantly different from expected figure of 1.28, P = 0.0025), which diminished at later stages of fetal development indicating potential correction of trisomies predominantly in females. There is a threefold prevalence of 46,XX/45,X mosaics over 46,XY/45,X mosaics in prenatally diagnosed cases, which also suggests a gender-specific postzygotic chromosome loss. The male prevalence in Prader-Willi syndrome with maternal UPD of chromosome 15 also can be explained by sex-specific trisomy correction, with predominant loss of a maternal chromosome causing biparental inheritance and therefore, complete correction of trisomy in females (without UPD). Finally, there is a female predominance in carriers of chromosome rearrangement with pericentromere break (mosaicism for Robertsonian translocation/isochromosome, centric fission, nonacrocentric isochromosome, and whole arm rearrangement), in both prenatal (21 males and 36 females, sex ratio is 0.58, P < 0.0184) and postnatal ill-defined cases (14 males and 35 females, sex ratio is 0.40, P = 0.001). Thus, the findings presented in this paper suggest that, in addition to reduction in male fertility, and to probable selection against abnormal cell line(s), there are two mechanisms that contribute to female preponderance among carriers of mosaicism: sex-specific chromosome loss and sex-specific centromere instability. The data obtained suggest that females may have gonadal mosaicism for aneuploidies and structural rearrangements more often than males. This may lead to the maternal origin bias in offspring with trisomies or structural rearrangements.

Acrocentric rearrangements are the most common chromosome abnormalities in humans. Carriers of homologous acrocentric rearrangements (Robertsonian translocations (ROBs) between homologous chromosomes and isochromosomes) are at very high risk of having multiple spontaneous abortions and chromosomally abnormal offspring. Parents of fetuses and children with unbalanced homologous acrocentric rearrangements are rarely found to be carriers or mosaic for the same rearrangement. Even though recurrent miscarriages may indicate a carrier parent, carriers are rarely identified. Comparison of non-chromosome 21 homologous rearrangements to rea(21q21q) culled from the literature revealed a 7-fold decrease in the number of mosaic cases among the parents of non-rea(21q21q) offspring. This under-ascertainment in parents may be due to low level mosaicism confined to the gonads, a true biological difference between chromosome 21 rearrangements and other homologous acrocentric rearrangements, or simply to the lack of rigorous clinical investigation of the parental karyotypes to uncover mosaicism. We recommend that polymorphic marker analysis be applied to apparently de novo acrocentric rearrangements to distinguish those resulting from biparental postzygotic formation from those resulting from meiotic formation; the latter of which may indicate a potential carrier parent. Parental chromosomal constitutions could then be screened in a large number of cells and in more than one tissue type to identify mosaicism. Identification of mosaicism allows for accurate genetic counseling and discussion of reproductive options. However, given that mosaicism may be restricted to the gonads, prenatal testing is likely to be desired by the family whether or not mosaicism is found.