Editor—The finding of a mixture of both 46,XX and 46,XY cells in amniotic fluid culture has been frequently described. In the great majority of cases, the finding is followed by the birth of a normal male infant, leading to a consensus that the finding is the result of contamination with maternal cells in a normal male fetus. There are, however, several other possible explanations for a 46,XX/46,XY karyotype at amniocentesis. These include the presence of cells from an undiagnosed twin pregnancy, cross contamination in the laboratory, the presence of cells from a “vanished” male twin, and true fetal chimerism.1 Reassuringly, all previously reported cases where 46,XX/46,XY has been found at amniocentesis have resulted in the birth of a child with a normal genital phenotype.

We present a case where a 46,XX/46,XY karyotype was found on amniocentesis. This resulted in the birth of an infant with true chimerism and an abnormal genital phenotype. We describe the possible outcomes when a 46,XX/46,XY karyotype is found at amniocentesis and discuss the implications for counselling.

A couple were seen for genetic counselling after an amniocentesis, performed because of advanced maternal age, showed a karyotype 46,XX[30]/46,XY[4]. Thirty clones from four independent cultures showed 46,XX cells, and four clones from two of these cultures showed 46,XY cells. In a fifth dispersed culture, of 60 cells examined, 58 were 46,XX and two were 46,XY. Laboratory error was very unlikely because the karyotype was found in two independently handled samples. An ultrasound examination showed male genitalia with normal appearance.

The couple were counselled regarding the possible clinical outcomes. It was explained that the 46,XX cells might be of maternal origin, in which case the child would be a normal male. The possibility of true chimerism was also raised. Severe genital ambiguity was considered unlikely because of the ultrasound findings, but the possibility of either normal or abnormal genitalia was discussed. It was explained that genital surgery may be necessary, and that there were risks of infertility and gonadoblastoma associated with true chimerism. It was also emphasised that because of the normal autosome and sex chromosome content in both cell lines, there was no reason to expect intellectual compromise. The couple decided against further investigation.

The infant was examined after delivery and had minor genital abnormalities. There was a well developed penis with a short ventral length and a penoscrotal web. The left gonad was present in the scrotum, measuring 2-3 ml, and the right gonad was not palpable. Ultrasound examination failed to locate the right gonad. A micturating cystourethrogram showed grade II vesicoureteric reflux on the right without obstruction. An HCG stimulation test showed normal testosterone production. A peripheral blood karyotype showed 46,XX[25]/46,XY[5].

At the age of 6 months the child underwent exploratory laparoscopy. A normal testis and vas was present on the left, with an ovary and uterus present on the right. The ovary was prophylactically removed.

Cases of true 46,XX/46,XY chimerism are usually ascertained in early childhood during investigation of the abnormal appearance of external genitalia. 46,XX/46,XY chimerism is known to have a variety of presentations, including ambiguous genitalia, hypospadias, gynaecomastia, and inguinal herniae,2 and is responsible for about 13% of cases of true hermaphroditism.3 The true frequency of the genotype, however, is unknown because of the bias in ascertainment. Rare cases with normal male and normal female phenotype have also been ascertained as an incidental finding.2 4 5

The mechanism for chimerism is uncertain, but possibilities include double fertilisation of an ovum and its first or second polar body, or the secondary fusion of two independently fertilised ova. Two recent studies6 7 using DNA fingerprinting techniques have shown the tetragametic origin of cases of XX/XY hermaphroditism, proving that these cases of chimerism originated by the fusion of two fertilised gametes.

The finding of a mixture of 46,XX and 46,XY cells at amniocentesis has been reported to occur at a frequency of about 1.5 per 1000.8-10 The vast majority of these cases are the result of contamination by maternal cells in an otherwise normal male fetus.1 10 Given that maternal contamination would only be detected if the fetus were male, maternal cell contamination is expected to occur at double this rate, that is, at about 3 cases per 1000. Maternal cell contamination is usually detected by the finding of a 46,XX/46,XY karyotype, although cases where 100% of cells were 46,XX have been discovered only after the unexpected delivery of a normal male.8 The maternal cells are thought to arise from the outgrowth of cells from small fragments of maternal tissue removed by the amniocentesis needle. Maternal cell contamination is more likely to occur where the first few millilitres of fluid are not discarded, when a large needle is used for the procedure, and when the fluid is blood stained.8

This case serves as a reminder that not all cases of 46,XX/46,XY karyotype at amniocentesis can be explained by maternal cell contamination where the fetus is a normal male. We have identified six other published cases where there has been an alternative explanation for the presence of the dual cell line at prenatal diagnosis.

The first two cases involved twin pregnancies.10 11 In both cases a 46,XX/46,XY karyotype was found at amniocentesis of the female twin. The XY cells were considered to have arisen from cross contamination from the other (male) twin. This situation would not present a diagnostic dilemma at prenatal diagnosis unless the twin pregnancy was previously undiagnosed.

Three cases have been reported where a 46,XX/46,XY result on amniocentesis has been followed by the delivery of a normal XX female1 10 The source of the XY cell line in these cases has not been explained. It has been hypothesised that the XY cells may have arisen from a male twin which had since died and been resorbed. The so called “vanishing twin” has been proposed in other situations as an explanation for discordance between prenatal and postnatal cytogenetic analysis.12-14 Another possible explanation is that the 46,XY cell line was actually derived from the fetus, but could not be found in the tissues that were sampled after birth.

In the sixth case, an amniocentesis performed at 16 weeks’ gestation showed 46,XX and 46,XY cells in a 4:1 ratio.2 All samples from all flasks showed a similar ratio of 46,XX and 46,XY cells. Maternal contamination was considered unlikely and the couple was counselled that there was a risk of abnormal sexual development. The infant was born with male genitalia that were normal in external appearance. A testosterone level and pelvic ultrasound were also normal. Postnatal karyotype on blood confirmed the 46,XX/46,XY chimerism, with 46,XY cells predominating. The authors believed this to be the first documented prospectively ascertained human chimera.

Our case represents the second report of the prenatal diagnosis of a true XX/XY chimera and the first case with true hermaphroditism.

When a 46,XX/46,XY karyotype is found at amniocentesis, the result can be confusion and anxiety for both parents and physicians. The amniocentesis will usually have been performed for other reasons, such as advanced maternal age, and the result is unexpected. Fortunately, the vast majority of cases have been shown to result from the contamination of the specimen by maternal cells. These pregnancies would be expected to result in the birth of a genetically and phenotypically normal male baby. Including our case, there are only seven reported cases where a 46,XX/46,XY karyotype at amniocentesis has resulted from a cause other than maternal contamination. It is possible that some other cases of true fetal chimerism have been wrongly attributed to maternal contamination. Nonetheless, the fact that no phenotypically abnormal infants have previously been reported suggests that true fetal chimerism at prenatal diagnosis is a rare event.

In assessing the potential outcomes of a pregnancy where 46,XX/46,XY has been discovered, six possible explanations for the occurrence should all be considered (table 1). The sixth possibility, that of a human parthogenetic chimera (composed of a biparental XY cell line and an XX cell line containing only maternally derived chromosomes) has never been described at amniocentesis, but has been reported in a child15 and potentially could be encountered prenatally. All possibilities other than maternal cell contamination are rare.

We suggest the following approach to the further assessment of a 46,XX/46,XY karyotype at amniocentesis. Firstly, the possibility of a twin pregnancy should have already been excluded by ultrasound at the time of the amniocentesis. It may also be possible to exclude laboratory cross contamination as a cause, depending on the method used by the laboratory in processing the specimen. The level of mosaicism present should also be assessed; the presence of 46,XX cells in single culture or clone (level I or II “mosaicism”) is almost certain to represent maternal cell contamination. An ultrasound examination showing male genitalia or a history of a traumatic procedure would add weight to this conclusion.

In cases where there are two or more 46,XX cells present, distributed over two or more flasks (level III “mosaicism”), the initial investigation should be a detailed ultrasound with particular attention to the genitalia. The appearance of male genitalia is consistent with maternal cell contamination as the cause of the 46,XX cells, although a true chimera could also have this appearance. If the genitalia appear female or ambiguous, careful consideration should be given to alternative explanations.

The choice of subsequent investigation will depend on the facilities available. The situation might be clarified by repeating the amniocentesis, which would be expected to show 46,XY unless true chimerism was present. The disadvantages of this approach are, firstly, the small risk involved in repeating the procedure and, secondly, the likelihood that the risk of maternal contamination complicating the second procedure is greater than would be expected by chance alone.16 An alternative approach is to use polymorphic microsatellite markers to determine the parental origin of the X chromosomes.17 Comparison of chromosome polymorphisms as visualised on banded karyotypes of maternal blood and amniotic fluid may also yield distinct patterns.18 If all three X chromosomes are shown to be maternal in origin, then maternal cell contamination is almost certainly present, notwithstanding the remote possibility that the fetus is a parthogenetic chimera. The presence of a non-parental X chromosome strongly suggests that laboratory cross contamination has occurred. If both maternal and paternal X chromosomes are found, the possibilities are, firstly, a female fetus with the presence of male cells of unknown origin (the so called “vanishing twin”) and, secondly, true fetal chimerism. Fetal blood sampling may be of benefit in confirming true fetal chimerism, although a finding of 46,XX in blood will still not exclude chimerism in other tissues.

The difficulty in differentiating true chimerism from the situation of a female with male cells of unknown origin may relate to the hypothesis that these two phenomena have identical genetic origins. It is possible that both conditions originate from dizygotic male and female twins. In the case of true fetal chimerism, the two embryos fuse to become one embryo, whereas in the second case, the male twin dies leaving residual cells in the amnion. There would therefore be no genetic way of differentiating the two situations.

If true chimerism cannot easily be excluded, the possibility of true hermaphroditism should be discussed with the couple. Relevant details include the issues of sexual ambiguity, genital surgery, possible infertility and gonadal tumours. The expectation of normal intelligence is an important component of counselling. It must be emphasised, however, that hermaphroditism is an unlikely outcome. Further invasive prenatal testing should be used judiciously according to the ultrasound findings, the availability of DNA and chromosome polymorphism studies, and the wishes of the couple. Where the possibility of true hermaphroditism cannot be excluded, consultation with an endocrinologist may be of benefit.