Trisomy 15 mosaicism and uniparental disomy (UPD) in a liveborn infant

We describe a liveborn infant with uniparental disomy (UPD) with trisomy 15 mosaicism. Third trimester amniocentesis yielded a 46,XX/47,XX,+15 karyotype. Symmetrical growth retardation, distinct craniofacies, congenital heart disease, severe hypotonia and minor skeletal anomalies were noted. The infant died at 6 weeks of life. Peripheral lymphocyte chromosomes were “normal” 46,XX in 100 cells. Parental lymphocyte chromosomes were normal. Skin biopsy showed 47,XX,+15 in 80% of fibroblasts and results were equivalent in fibroblasts from autopsy lung tissue. Molecular analysis revealed maternal uniparental heterodisomy for chromosome 15 in the 46,XX cell line. We describe an emerging phenotype of trisomy 15 mosaicism, confirm that more than one tissue should be studied in all cases of suspected mosaicism, and suggest that UPD be considered in all such cases. © 1996 Wiley-Liss, Inc.     

True trisomy 2 mosaicism in amniocytes and newborn liver associated with multiple system abnormalities

Among 58,000 amniocenteses completed, our laboratories found one case of true cytogenetic trisomy 2 mosaicism in a fetus with multiple abnormalities. In contrast, 11 fetuses phenotypically normal at birth were found to have true trisomy 2 mosaicism in their chorionic villus cells among the 10,500 fetuses tested by chorionic villus sampling (CVS). In our single abnormal case, amniocentesis performed at 19 weeks after finding an elevated maternal serum AFP found two independent cultures with trisomy 2 karyotypes in 8 of 25 and 7 of 31 amniocytes, respectively. Although oligohydramnios was noted by ultrasound, the mother elected to continue the pregnancy. At 26 weeks the fetus had intrauterine growth retardation (IUGR), hydronephrosis, and cardiac abnormalities. When delivered by Cesarean section at 30 weeks, the infant had multiple anomalies and developed necrotizing enterocolitis and severe cholestasis. At 5 months coronal magnetic resonance imaging (MRI) displayed delayed myelination and abnormal brain morphology. The patient also exhibited significant growth failure and developmental delay. Although chromosomes were normal in blood, skin fibroblasts, and ascites fluid cells, 4 of 100 hepatic biopsy fibroblasts were 47,XY,+2. Molecular analysis excluded uniparental disomy (UPD) of chromosome 2 in the 46,XY cell line. This and other reports of rare phenotypically abnormal trisomy 2 mosaic fetuses identified by karyotyping amniocytes emphasizes the substantially higher fetal risk of abnormal development than when trisomy 2 is found only in chorionic villus cells. Am. J. Med. Genet. 72:343–346, 1997. © 1997 Wiley-Liss, Inc.     

Trisomy 7 mosaicism prenatally misdiagnosed and maternal uniparental disomy in a child with pigmentary mosaicism and Russell- Silver syndrome

Prenatal diagnosis of true mosaic trisomy 7 is rare in amniotic fluid and can be misinterpreted as pseudomosaic. The phenotype is highly variable and may be modified by a maternal uniparental disomy of chromosome 7 leading to mild Russell-Silver syndrome (RSS). We report here the third postnatal case of mosaic trisomy 7 with maternal uniparental disomy of chromosome 7 in a boy presenting a mild RSS. Fetal karyotype performed in amniocentesis for intrauterine growth retardation was considered normal. Mosaic trisomy 7 was diagnosed after birth, on fibroblasts karyotype performed for blaschkolinear pigmentary skin anomalies and failure to thrive. Maternal uniparental disomy of chromosome 7 was observed in blood sample. Retrospectively, trisomic 7 cells were identified in one prenatal long-term flask culture revealing a prenatal diagnosis failure. This report emphasizes the difficulty of assessing fetal mosaicism and distinguishing it from pseudomosaicism in cultured amniocytes. It is important to search for uniparental disomy as an indirect clue of trisomy 7 mosaicism and a major prognosis element. Although there are only few prenatal informative cases, detection of trisomy 7 in amniocentesis appears to be associated with a relatively good outcome when maternal uniparental disomy has been ruled out.     

Maternal serum α-fetoprotein screening and fetal chromosome anomalies: Is lowering maternal age for amniocentesis preferable?

We have compared the cytogenetic abnormalities diagnosed prenatally in 1,098 patients referred for amniocentesis because of low maternal serum α-fetoprotein (MSAFP) to those of 445 patients whose indication was elevated MSAFP and those of 361 patients who had amniocentesis for “maternal anxiety.” Autosomal trisomies, sex chromosome aberrations, and various structural rearrangements were detected in all 3 groups and actually exceeded the age-related incidence estimates. The frequency of chromosome anomalies in cases studied because of “maternal anxiety” with no prior screening was similar to that in the group referred for low MSAFP (1.38 and 1.27%, respectively). A relatively higher frequency (2.02%) was detected in the group whose indication was elevated MSAFP. Maternal serum screening is designed primarily to recalculate risk figures for Down syndrome, but not for other major chromosome abnormalities. The concept of prenatal screening for chromosome aberrations must therefore be reevaluated. We think that efforts should be directed at making amniocenteses more accessible to patients who request it. “Lowering” maternal age limits to 30 would encompass a greater proportion of pregnancies at risk and would be a step toward more effective prenatal diagnosis for chromosome abnormalities.     

2052例羊水细胞产前诊断的染色体结果分析

目的探讨孕中期羊膜腔穿刺产前诊断胎儿染色体异常的经验。方法收集、分析孕中期羊水培养核型分析结果资料。主要的产前诊断指征包括高龄、产前血清学筛查高风险、胎儿超声检查异常等。染色体异常包括常染色体非整倍体,性染色体非整倍体及染色体结构异常。结果成功培养羊水细胞2052例,其中指征为高龄孕妇的占16．0％．血清学筛查阳性占75．0％,胎儿超声检查异常占4．5％,其它原因占4．5％。检出率最高的指征为胎儿超声检查异常。共检出75例染色体异常胎儿,其中非整倍体为51例（68．0％）,染色体结构异常24例（32．0％）。结论本研究证明了羊水细胞用于诊断染色体异常胎儿的作用,其结果可用于临床遗传咨询。     

Maternal serum alpha-fetoprotein screening and fetal chromosome anomalies: is lowering maternal age for amniocentesis preferable?

We have compared the cytogenetic abnormalities diagnosed prenatally in 1,098 patients referred for amniocentesis because of low maternal serum alpha-fetoprotein (MSAFP) to those of 445 patients whose indication was elevated MSAFP and those of 361 patients who had amniocentesis for "maternal anxiety." Autosomal trisomies, sex chromosome aberrations, and various structural rearrangements were detected in all 3 groups and actually exceeded the age-related incidence estimates. The frequency of chromosome anomalies in cases studied because of "maternal anxiety" with no prior screening was similar to that in the group referred for low MSAFP (1.38 and 1.27%, respectively). A relatively higher frequency (2.02%) was detected in the group whose indication was elevated MSAFP. Maternal serum screening is designed primarily to recalculate risk figures for Down syndrome, but not for other major chromosome abnormalities. The concept of prenatal screening for chromosome aberrations must therefore be reevaluated. We think that efforts should be directed at making amniocenteses more accessible to patients who request it. "Lowering" maternal age limits to 30 would encompass a greater proportion of pregnancies at risk and would be a step toward more effective prenatal diagnosis for chromosome abnormalities.     

Trisomy 16 and trisomy 16 mosaicism: A review

A review of all prenatal and postnatal diagnoses of trisomy 16 and trisomy 16 mosaicism was carried out in the context of the current understanding of confined placental mosaicism and uniparental disomy (UPD). The prenatal detection of trisomy 16 cells is associated with a high probability of fetal death, preterm delivery, intrauterine growth retardation, and fetal anomalies. Birth defects were typical of those seen in nonmosaic partial duplications of chromosome 16. Surprisingly, anomalies were sometimes limited to a single organ and included some relatively common isolated defects such as a ventricular septal defect, hypospadias, imperforate anus, inguinal hernia, and clubfoot. The risk for abnormality appeared to be higher in those pregnancies in which trisomy 16 cells were identified in amniotic fluid compared to the detection in chorionic villi samples. Contrary to nonmosaic trisomy 16 with an excess of males, mosaic trisomy 16 shows an excess of female karyotypes. Following the prenatal detection of trisomy 16 cells, aneuploid cells are almost never found in fetal or neonatal lymphocytes. Studies on fibroblasts also often fail to confirm the presence of the abnormal cell line even in cases in which multiple anomalies are present. It is likely that trisomy 16 cells are sometimes present in the early developing embryo even though subsequent cytogenetic studies on fetal or neonatal tissues may not detect any aneuploid cells. UPD can be excluded as a mechanism for those anomalies that are common to mosaic trisomy 16 and nonmosaic partial duplications. The term “occult mosaicism” is suggested to describe the situation in which the presence of an abnormal cell line is suspected on the basis of clinical data but unproven by laboratory analysis. Am. J. Med. Genet. 79:121–133, 1998. © 1998 Wiley-Liss, Inc.     

Prenatal diagnosis policy without routine amniocentesis in pregnancies with a positive family history for neural tube defects

The recurrence risk for neural tube defects in pregnancies of women with a family history of neural tube defects greatly exceeds the general population risk. In these high risk pregnancies, we used a prenatal diagnostic method differing from that usually employed, relying mainly on the results of maternal serum alpha-fetoprotein (MSAFP) and ultrasound examination, without routine amniocentesis. During the 6 years reviewed in this study, this method was applied in 539 pregnancies. Of a total of 20 neural tube defects, 19 were detected using this combination of MSAFP, ultrasound, and selective amniocentesis, and the authors estimate that about 8-10 spontaneous abortions were avoided because only 28 amniocenteses were carried out instead of 539. The risk of recurrence was found to be lower than that experienced earlier.     

Interpretation of chromosome mosaicism and discrepancies in chorionic villi studies

In 3,000 chorionic villi studies (CVS) 33 cases of mosaicism and 7 false-positive cell lines in all cells were seen. The mosaic cell lines were caused by aneuploidy of autosomes (13x), sex chromosomes (9x), and structural anomalies (11x). Mosaics of fetal origin were only 4 cases of trisomy 21 and one 47,XXY mosaic. In 7 cases abnormal karyotype of non-fetal origin was seen in all cells in direct studies, including trisomy 16 (3x) and trisomy 18 (2x). The combined use of direct CVS and cell cultures always uncovered the non-fetal origin of chromosome abnormalities and the study of cultured cells in all cases could have prevented 5 terminations. Complete follow-up studies demonstrated no false-negative results. Therefore, CVS can be nearly 100% accurate when both direct studies and cultures are examined in cases of mosaicism and other cell lines of possible non-fetal origin, such as trisomy 16, trisomy 18, translocation (21;21), and 45,X cells.     

Confirmation of mosaicism and uniparental disomy in amniocytes, after detection of mosaic chromosome abnormalities in chorionic villi

Chromosome mosaicism is detected in about 1-2% of chorionic villi samples (CVS), and may be due to a postzygotic nondisjunction event generating a trisomic cell line in an initially normal conceptus (mitotic origin) or the postzygotic loss of one chromosome in an initially trisomic conceptus (meiotic origin and trisomy rescue). Depending on the distribution of the abnormal cell line, the mosaic can be confined to the placenta (CPM) or generalised to the fetus (TFM, true fetal mosaicism). Trisomy rescue could theoretically be associated with a 33.3% probability of uniparental disomy (UPD) in the fetus. The aim of this study was to determine the risk of fetal involvement in a cohort of numerical and structural chromosome mosaics revealed in chorionic villi by means of combined direct and long-term culture analyses; we also determined the incidence of UPD associated with mosaic aneuploidies and supernumerary markers involving imprinted chromosomes. A total of 273 of a consecutive series of 15,109 CVS evaluated during a period of 5 years showed a mosaic condition in direct preparations and/or long-term cultures; confirmatory amniocentesis was performed in 203 cases. The abnormal cell line was extended to the fetus in 12.8% cases in terms of structural and numerical abnormalities involving autosomes and sex chromosomes; the risk of TFM varied and depended on the placental tissue distribution of the abnormal cell line. One of the 51 cases in which the mosaic involved an imprinted chromosome showed UPD, thus indicating a risk of 1.96%.     

Mosaic trisomy 16 in a thriving infant; maternal heterodisomy for chromosome 16

Lindor NM, Jalal SM, Thibodeau SN, Bonde D, Sauser KL, Karnes PS. Mosaic trisomy 16 in a thriving infant; maternal heterodisomy for chromosome 16
Clin Genet 1993: 44: 185–189. © Munksgaard, 1993
Trisomy 16 is the most common trisomy in spontaneous abortions and is usually, if not always, lethal in the nonmosaic state. We report a liveborn infant with trisomy 16 mosaicism first diagnosed by amniocentesis at 20 weeks gestation. At birth, the infant was growth retarded and mildly dysmorphic. At age 14 months she was developmentally normal and had facial asymmetry. Her length, weight and head circumference were normal. Pure trisomy 16 was found in cells from the placenta. A normal female karyotype was found in lymphocytes from the infant. Skin fibroblasts revealed a trisomy 16 karyotype in 6 of 30 cells. Molecular analysis showed maternal uniparental heterodisomy, indicating that the trisomic conceptus arose from a nondisjunction of maternal meiosis. Fibroblasts may be the tissue of choice for detection of low-level trisomy 16 mosaicism.     

Cerebellar hypoplasia, zonular cataract, and peripheral neuropathy in trisomy 17 mosaicism

Trisomy 17 mosaicism in liveborns is an extremely rare chromosomal abnormality, with only three cases reported in the literature. Here we describe a 7-year-old boy with trisomy 17 mosaicism. The chromosome abnormality was detected by amniocentesis and was confirmed postnatally in cultured skin fibroblasts. The main clinical features were mental retardation and growth reduction, peripheral motor and sensory neuropathy, hypoplastic cerebellar vermis, zonular cataract, and body asymmetry. In our patient, and in the three earlier described cases, the additional chromosome 17 was detected in skin fibroblasts, not in peripheral lymphocytes. Molecular investigations excluded uniparental disomy of chromosome 17 in our patient. The extra chromosome 17 probably originated from a postzygotic mitotic nondisjunction of the maternal chromosome 17. In most cases of trisomy 17 mosaicism detected in amniocytes the chromosome abnormality seems to be confined to extra-embryonic tissues and clinically normal children are born. If, however, there are also ultrasound abnormalities, the possibility of fetal trisomy 17 mosaicism should certainly be considered. If postnatal karyotyping is limited to blood the diagnosis of trisomy 17 mosaicism could easily be missed. Therefore, we recommend chromosome analysis to be based on cultured skin fibroblasts in all cases where mental retardation is accompanied by postnatal growth retardation, body asymmetry, peripheral neuropathy, and cerebellar hypoplasia or zonular cataract.     

Maternal uniparental isodisomy 20 in a foetus with trisomy 20 mosaicism: clinical,cytogenetic and molecular analysis

The clinical significance of trisomy 20 mosaicism detected prenatally remains uncertain due to the rarity of liveborn cases with inconsistent clinical findings, and lack of long-term follow-up and outcome. We describe a case of true trisomy 20 mosaicism in a liveborn girl with maternal uniparental isodisomy of chromosome 20 in the diploid blood cells. Trisomy 20 mosaicism was originally detected in amniotic fluid (98%) and was confirmed in the term placenta (100%), as well as in the blood (10%) and urine sediment (100%) of the neonate. There was intrauterine and postnatal growth retardation, but otherwise the newborn manifested no gross abnormalities. At 9 months of age moderate psychomotor retardation, central hypotonia with peripheral hypertonia, numerous minor morphogenetic variants, marked kyphosis, and extensive Mongolian spot were observed. To our knowledge this represents the first case of trisomy 20 mosaicism detected prenatally and confirmed in different tissues of the newborn, where uniparental disomy was demonstrated in the diploid cell line. The clinical and laboratory findings in our patient are compared with those of five previously reported cases of UPD20, suggesting that maternal UPD20 might be associated with a characteristic phenotype.     

Prenatal detection of double aneuploidy trisomy 10/monosomy X in a liveborn twin with exclusively monosomy X in blood

Both double aneuploidy and trisomy 10 are rare chromosome findings. All five published cases of trisomy 10 in liveborns were found to be mosaic with an euploid cell line. In a liveborn female twin, double aneuploidy mosaicism 47,XX, + 10/45,X was detected prenatally by amniocentesis performed because of severe intrauterine growth retardation and malformations. Chromosome analysis from neonatal lymphocyte cultures revealed exclusively the 45,X cell line. Double aneuploidy mosaicism trisomy 10/monosomy X was confirmed from skin fibroblasts. The child died at the age of 7 weeks. This is the first reported case of double aneuploidy involving trisomy 10, and the first case of trisomy 10 without a normal cell line in a liveborn. Prenatal diagnosis of trisomy 10 in a liveborn has not been published so far. The case illustrates that in specific cases amniotic fluid cells may reflect the karyotype of the fetus better than blood.     

Dilemma of trisomy 20 mosaicism detected prenatally: Is it an innocent finding?

The clinical significance of mosaicism trisomy 20 detected prenatally following amniocentesis remains uncertain, due to the rarity of liveborn cases with inconsistent clinical findings, the short postnatal follow-up, and failure in evaluating other fetal tissues for the presence of the trisomy. We report on a 15 month-old 46,XX chromosome constitution in white blood cells, while skin fibroblasts demonstrated trisomy 20 mosaicism (54%) by fluorescence in situ hybridization (FISH) analysis. Clinical examination of the baby showed only minor phenotypic signs (bilateral epicanthal folds, delayed closure of fontanel with no other gross anomalies), but demonstrated a considerable developmental delay in gross and fine motor skills along with hypotonicity. This is the second oldest described liveborn with trisomy 20 mosaicism confirmed in skin fibroblasts. This cytogenetic aberration along with her developmental delay suggests that the two findings are related and that aberration affects various fetal tissues and is not confined to extra-embryonic tissue as suggested previously. Yet, an undiagnosed condition may be the cause of the child's developmental delay. Based on this case and following a review of the literature we suggest that when mosaic trisomy 20 is identified in amniocytes, further evaluation is required. Cord blood should be analyzed preferably by FISH. During counseling the parents should be advised of an additional risk, such as developmental delay, even when fetal cord karyotype and detailed ultrasonic scan are normal. Am. J. Med. Genet. 77:72–75, 1998. © 1998 Wiley-Liss, Inc.     

Confined placental mosaicism.

In most pregnancies the chromosomal complement detected in the fetus is also present in the placenta. The detection of an identical chromosomal complement in both the fetus and its placenta has always been expected as both develop from the same zygote. However, in approximately 2% of viable pregnancies studied by chorionic villus sampling (CVS) at 9 to 11 weeks of gestation, the cytogenetic abnormality, most often trisomy, is confined to the placenta. This phenomenon is known as confined placental mosaicism (CPM). It was first described by Kalousek and Dill in term placentas of infants born with unexplained intrauterine growth restriction (IUGR). Contrary to generalised mosaicism, which is characterised by the presence of two or more karyotypically different cell lines within both the fetus and its placenta, CPM represents tissue specific chromosomal mosaicism affecting the placenta only. The diagnosis of CPM is most commonly made when, after the diagnosis of chromosomal mosaicism in a CVS sample, the second prenatal testing (amniotic fluid culture or fetal blood culture analysis) shows a normal diploid karyotype.     

Pre- and postnatal findings in trisomy 17 mosaicism

Trisomy 17 mosaicism is one of the rarest autosomal trisomies in humans. Thus far, only 23 cases have been described, most of them detected prenatally. In only five instances has mosaicism been demonstrated in lymphocytes and/or fibroblasts postnatally, and only in these have multiple congenital anomalies (MCA), facial dysmorphisms, and mental retardation been reported. Patients with trisomy 17 mosaicism at amniocentesis and a normal karyotype in blood and fibroblasts (n = 17) were always healthy. Here, we report on pre- and postnatal clinical, cytogenetic, molecular-cytogenetic, and molecular findings in four patients with trisomy 17 mosaicism. The first case was detected in cultured but not in short-term chorionic villi and amniocytes. Due to MCA on prenatal ultrasound examination the pregnancy was terminated. The second patient is a 13-month-old healthy boy, in whom low level trisomy 17 mosaicism was detected in cultured chorionic villi only. The third patient is a 2-year-old girl with growth retardation, developmental delay, MCA, and trisomy 17 mosaicism in amniocytes, fibroblasts, and placenta, but not in blood and buccal smear. The fourth patient is a 9-year-old boy with growth and mental retardation, sensoneurinal hearing loss, and MCA. Cytogenetic analyses showed trisomy 17 mosaicism in amniocytes, skin fibroblasts, and urinary sediment cells, whereas in blood and buccal smear a 46,XY karyotype was found. Molecular investigations in all four cases indicated biparental inheritance of chromosome 17. Formation of trisomy was most likely due to a maternal meiosis I error in Patient 1 and a postzygotic non-disjunction of the paternal chromosome 17 in Patient 4. Cerebellar malformations, reported in two cases from the literature and in two reported here may be a specific feature of trisomy 17 mosaicism. Since the aberration has rarely been reported in lymphocytes, chordocentesis is not indicated in prenatal diagnosis. Prenatal genetic counseling for trisomy 17 mosaicism in chorionic villi or amniocytes should consider that the clinical significance remains uncertain.     

Cytogenetic evidence for enhanced selective miscarriage of trisomy 21 pregnancies with advancing maternal age.

The effect of advancing maternal age on the risk of death of fetuses with certain chromosome abnormalities has been tested by comparing their frequency at the time of chorionic villus sampling (CVS) with that at amniocentesis. The frequency of chromosome abnormalities among women whose sole risk factor for a chromosome abnormality was advanced maternal age (> or = 35 years old) was determined in a pooled group of 15,147 CVS cases, of whom > 1/3 were from the initial 7,500 CVS cases at the University of California, San Francisco, and compared with a pooled group of 74,851 amniocentesis cases collected from the literature. The frequency of trisomy 21 not only increased with advancing maternal age as expected, but the slope of the increase was about 25% greater in the CVS group than in the amniocentesis group (P = 0.08 for the difference in slopes by a logistic statistical model and P = 0.04 by a normit model). Similar patterns were seen for trisomies 18 and 13, but the P values for the differences in slopes were much higher. These results suggest that the miscarriage rate of trisomy 21 during the gestational interval studied is selectively greater with advancing maternal age. The basis for the enhanced selective loss of trisomy 21 with maternal age may be a reduced ability of the ageing "maternal compartment" to compensate for abnormal conceptuses.     

Variable clinical expression of mosaic trisomy 16 in the newborn infant

Trisomy 16 is common in embryos and fetuses aborted early during development. Mosaicism for trisomy 16 is sometimes encountered during prenatal diagnosis, particularly with chorionic villi biopsy specimens, and, until recently, was thought to be confined to the placenta. However, recently, several liveborn infants with trisomy 16 mosaicism have been described. We report on an additional liveborn infant with trisomy 16 mosaicism and compare the clinical findings with those of the previously reported cases in an attempt to delineate a mosaic trisomy 16 syndrome. Cytogenetic analysis from our patient showed that there was a different proportion of abnormal cells in different tissues and that the anomaly was undetectable in blood lymphocyte cultures. This observation was consistent with some of the previous reports. DNA analysis of parents and child was carried out using a polymorphic dinucleotide marker that maps to the long arm of chromosome 16. This analysis showed that the extra chromosome 16 in the infant was maternal in origin and suggested that the nondisjunction was probably a first meiotic division error. Our results suggest that an investigation of multiple tissues is required before concluding that mosaicism is confined to the placenta. We conclude that a finding of trisomy 16 mosaicism at prenatal diagnosis should be regarded with extreme caution. This diagnosis may be associated with a highly variable phenotype that may occasionally be compatible with extrauterine life. © 1993 Wiley-Liss, Inc.     

European collaborative research on mosaicism in CVS (EUCROMIC)—fetal and extrafetal cell lineages in 192 gestations with CVS mosaicism involving single autosomal trisomy

Cytogenetic information on cells from cytotrophoblast, villus mesenchyme, and one or more fetal tissues was available for 192 gestations with mosaicism or non-mosaic fetoplacental discrepancy involving a single autosomal trisomy in the chorionic villus sample (CVS), registered in a collaborative study (EUCROMIC) during the period 1986–1994. In order to identify predictors of confined placental mosaicism (CPM), generalized mosaicism and/or uniparental disomy (UPD), distribution of the mosaic and non-mosaic aneuploid cell lines in the different fetal and extrafetal cell lineages were analyzed. Data were related to existing hypotheses on mechanisms leading to fetoplacental discrepancies and early extraembryonic cell differentiation. Trisomy 21 mosaicism was the one most frequently confirmed in the fetus. Non-mosaic trisomy 13, 18, and 21 in the villus mesenchyme indicated the presence of a trisomic cell line in the fetus proper. Non-mosaic trisomy 2, 7, and 16 in villus mesenchyme was always found with concomitant mosaic or non-mosaic trisomy in the cytotrophoblast, but was never recovered in the fetus. Mosaic trisomy 3, 7, and 20 was predominantly restricted to the cytotrophoblast, mosaic trisomy 2 to the villus mesenchyme. Trisomies 15 and 16 were most often found in both cytotrophoblast and villus mesenchyme and not in fetal cells. This supports the hypothesis that mosaicism/discrepancy for trisomies 15 and 16 results more often than for the other trisomies from trisomic zygote rescue, enhancing their risk for UPD. We recommend, due to the risk of fetal trisomy, amniocentesis in all gestations involving mosaic autosomal trisomy in villus mesenchyme. In gestations with mosaic or non-mosaic autosomal trisomy in both cytotrophoblast and villus mesenchyme we recommend, in order to exclude fetal trisomy and/or UPD, depending on the chromosome involved, further examination by amniocentesis, ultrasound and/or test for UPD. We also recommend, due to a small but not negligible risk of false negative and false positive diagnoses, not to solely use direct preparation. Am. J. Med. Genet. 70:179–187, 1997. © 1997 Wiley-Liss, Inc.