Non-invasive prenatal testing for aneuploidy: current status and future prospects.

Non-invasive prenatal testing (NIPT) for aneuploidy using cell-free DNA in maternal plasma is revolutionizing prenatal screening and diagnosis. We review NIPT in the context of established screening and invasive technologies, the range of cytogenetic abnormalities detectable, cost, counseling and ethical issues. Current NIPT approaches involve whole-genome sequencing, targeted sequencing and assessment of single nucleotide polymorphism (SNP) differences between mother and fetus. Clinical trials have demonstrated the efficacy of NIPT for Down and Edwards syndromes, and possibly Patau syndrome, in high-risk women. Universal NIPT is not cost-effective, but using NIPT contingently in women found at moderate or high risk by conventional screening is cost-effective. Positive NIPT results must be confirmed using invasive techniques. Established screening, fetal ultrasound and invasive procedures with microarray testing allow the detection of a broad range of additional abnormalities not yet detectable by NIPT. NIPT approaches that take advantage of SNP information potentially allow the identification of parent of origin for imbalances, triploidy, uniparental disomy and consanguinity, and separate evaluation of dizygotic twins. Fetal fraction enrichment, improved sequencing and selected analysis of the most informative sequences should result in tests for additional chromosomal abnormalities. Providing adequate prenatal counseling poses a substantial challenge given the broad range of prenatal testing options now available.

Late first-trimester invasive prenatal diagnosis: results of an international randomized trial.

OBJECTIVE: To assess, in a randomized trial, the safety and accuracy of amniocentesis and transabdominal chorionic villus sampling (CVS) performed at 11-14 weeks of gestation, given that this time frame is increasingly relevant to early trisomy screening. METHODS: We compared amniocentesis with CVS from 77 to 104 days of gestation in a randomized trial in a predominantly advanced maternal age population. Before randomization, the feasibility of both procedures was confirmed by ultrasonography, and experienced operators performed sampling under ultrasound guidance; conventional cytogenetic analysis was employed. The primary outcome measure was a composite of fetal loss plus preterm delivery before 28 weeks of gestation in cytogenetically normal pregnancies. RESULTS: We randomized 3,775 women into 2 groups (1,914 to CVS; 1,861 to amniocentesis), which were comparable at baseline. More than 99.6% had the assigned procedure, and 99.9% were followed through delivery. In contrast to previous thinking, in the cytogenetically normal cohort (n = 3,698), no difference in primary study outcome was observed: 2.1% (95% confidence interval 1.5, 2.8) for CVS and 2.3% (95% confidence interval, 1.7, 3.1) for amniocentesis. However, spontaneous losses before 20 weeks and procedure-related, indicated terminations combined were increased in the amniocentesis group (P =.07, relative risk 1.74). We found a 4-fold increase in the rate of talipes equinovarus after amniocentesis (P =.02) overall and in week 13 (P =.03, relative risk = 4.65), but data were insufficient to determine this risk in week 14. CONCLUSION: Amniocentesis at 13 weeks carries a significantly increased risk of talipes equinovarus compared with CVS and also suggests an increase in early, unintended pregnancy loss. LEVEL OF EVIDENCE: I

New molecular techniques for chromosome analysis.

The advent of molecular genetic technology has significantly advanced knowledge about the structure of chromosomes and their behaviour during meiosis and mitosis, as well as delineating cytogenetic aberrations that cannot be identified by conventional chromosome analysis. Molecular cytogenetics, the visualization of genetic loci using the dynamic recombinant technology of fluorescence in situ hybridization (FISH), now provides the obstetrician and gynaecologist with increasingly important diagnostic and prognostic information heretofore unavailable. The technical principles underlying FISH are briefly discussed. Emphasis is placed on the clinical applications of FISH and technologies derived from FISH, in particular comparative genome hybridization, microdissection FISH and multiplex FISH. These technologies play increasingly significant roles in preimplantation and prenatal genetic diagnosis, in the identification of microdeletion syndromes, cryptic translocations and marker chromosomes, and in defining chromosome mosaicism. FISH and related technologies also constitute essential diagnostic modalities in follow-up of organ transplantation, in a variety of haematological disorders and in determining the amplification of oncogenes associated with specific forms of cancer and neoplasia.

Recurrent triploidy of maternal origin.

We report the occurrence of triploid preimplantation embryos following in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) in a woman with two previously-identified triploid conceptuses which spontaneously underwent fetal demise at 10 and 23 weeks' gestation. An error in maternal meiosis II is proposed as the most likely cause.

The clinical application of interphase FISH in prenatal diagnosis.

Fluorescence in situ hybridization (FISH) for five chromosomes (13, 18, 21, X and Y) detected 87 of 107 (81%) of the chromosome aberrations identified by conventional chromosome analysis applied to fetal interphase cells obtained by chorionic villus sampling or amniocentesis. The choice of FISH was solely determined by prospective parents after formal genetic counselling concerning the advantages and disadvantages of FISH analysis. Excluding known familial chromosome aberrations, if FISH analysis revealed normal signals, there was an overall residual risk of 1 in 149 for an undetectable chromosome aberration. This risk varied according to the indication for prenatal diagnosis: 1 in 177 for women of advanced maternal age; 1 in 60 for women at increased risk for Down syndrome based on maternal serum screening; and, 1 in 43 for women whose ultrasound examination revealed fetal anomalies. There were 20 cases of discordance between the FISH results and standard karyotype analysis: four were the outcome of a failure to apply the appropriate FISH probe; 16 were not detectable by the available FISH probes. Of these 16, nine were chromosome abnormalities with clinical significance and seven were familial. If FISH is to become a standard part of prenatal genetic diagnosis, genetic counselling that is sensitive to patient health needs must be based on accurate information about the biological and obstetrical implications of the results of FISH analysis.

At the crossroads: the intersection of the Internet and clinical oncology.

The Internet is rapidly becoming a third party in the doctor-patient relationship. The World Wide Web, electronic mail (e-mail), and discussion groups have dramatically increased the quantity of medical and health information available to patients, who, in turn, vary greatly in their understanding of that newly discovered information. This article reviews the advantages and disadvantages of the Internet for both oncology patients and physicians. This forms the background for a discussion of three steps that clinical oncologists and other health care professionals can take to direct and control the potential of the Internet so as to optimize patient care. These steps include: (1) finding out what type of cancer information is being disseminated on the Web; (2) using Internet-derived material that patients bring to the clinic as a stepping-stone for patient education; and (3) becoming an active participant on the Web. Each of these strategies requires health professionals to be proactive. The appropriate and effective use of the Internet, as well as its boundaries, are rapidly expanding in medicine and are likely to co-evolve with changes in patient-health care provider relationships.

Hybridization of chromosome 18 alpha-satellite DNA probe to chromosome 22.

Fluorescence in situ hybridization (FISH) of uncultured chorionic villus diploid cells with a chromosome 18 alpha-satellite DNA probe (D18Z1) revealed a third small signal in addition to two large signals. FISH analysis of diploid metaphase cells from cultured chorionic villus cells and from maternal lymphocytes revealed that the third signal resulted from hybridization to the centromere of chromosome 22. This is the first report of a variant involving D18Z1 detected by FISH and of hybridization of alpha-satellite from a sub-metacentric chromosome to the centromere of an acrocentric chromosome. We propose that this inherited variant resulted from insertion of chromosome 18 specific alpha-satellite DNA sequences into the centromeric region of chromosome 22.

Insulin receptor substrate-1 is expressed at high levels in all cells of the peri-implantation mouse embryo.

Insulin and insulinlike growth factors are important for embryonic growth and metabolism. Intracellular transduction for these factors has not been studied in the preimplantation mouse embryo. Peri-implantation mouse embryos synthesize insulinlike growth factor (IGF)-II ligand, insulin receptor, IGF-I receptor, and IGF-II receptor and respond to IGF-II, IGF-I, and insulin metabolically and mitogenically. Maternal tissues in the oviduct and uterus are also sources of IGF-I and insulin. Signaling of IGFs occurs through insulin receptor substrate (IRS)-1 and IRS-2. This paper shows that IRS-1 mRNA and protein are highly expressed in preimplantation mouse embryos, in embryonic cell lines, and in cultured blastocyst outgrowths. IRS-1 mRNA and protein are detected in embryo-derived cell lines cultured to produce the three cell lineages (stem cells, endoderm, and trophoblast cells). IRS-1 mRNA is detected by reverse transcription-polymerase chain reaction (RT-PCR) in the E3.5 blastocyst before implantation and in F9 teratocarcinoma stem cells and parietal endoderm cells. IRS-1 mRNA is detected by Northern blot hybridization at high levels in stem cells and in differentiated progeny of F9 cells and C3H/NE trophectoderm cells. IRS-1 protein was detected in these cell lines and in an overexpressing CHO-IRS-1 fibroblast cell line by immunocytochemistry. Cultured blastocyst outgrowths are a model for implantation events of the trophoblast/placenta lineage and endoderm/yolk sac lineage. In the blastocyst outgrowth, IRS-1 protein is detected in inner cell mass cells (ICM cells), primitive endoderm, parietal endoderm, and trophectoderm cells. These data suggest that IRS-1 is expressed in all cell lineages of the peri-implantation mouse embryo and mediates some effects of insulin and IGFs at this stage.

Isolated fetal pyelectasis: assessment of risk for postnatal uropathy and Down syndrome.

Eight-two consecutive fetuses with ultrasound evidence of isolated pyelectasis (defined as dilation in the antero-posterior renal pelvic dimension of > or = 4 mm) were prospectively followed to determine the risk of postnatal uropathy and Down syndrome. In 98 (60%) kidneys, isolated pyelectasis was shown to be the first manifestation of a pathophysiological process that evolved into a gamut of postnatal uropathies (defined as urological conditions requiring remedial surgery or extended medical surveillance). Data quantifying the risk for postnatal uropathy in fetuses with varying degrees of isolated pyelectasis, at different gestational ages, are presented in figure format to facilitate prenatal counselling. Bivariate analysis showed that the evolution of isolated pyelectasis to uropathy was statistically significant when in utero progression was noted or in conjunction with other findings including contralateral pyelectasis (p < 0.01), male gender (p < 0.01) and increased kidney length (p < 0.001). Importantly, 55% of the infants requiring corrective surgery demonstrated in utero progression of pyelectasis (p < 0.002). Serial ultrasound examinations were necessary to evaluate progression or regression in the extent of pyelectasis. Finally, isolated pyelectasis was associated with an increase risk for Down syndrome, beginning at maternal age of 31 years, in the interval of 16-20 week's gestation.

Current practices of commercial cryobanks in screening prospective donors for genetic disease and reproductive risk.

OBJECTIVE: To determine how the screening practices of commercial semen banks vary from published guidelines, which factors influence cryobanks to exclude prospective semen donors for genetic reasons, and the current role of clinical geneticists/genetic counselors in evaluating prospective semen donors. DESIGN: The genetic screening of prospective donors by commercial semen banks was evaluated using written questionnaires completed by bank directors. Responses were analyzed to determine exclusion criteria, adherence to published guidelines, and contribution of genetic professionals. SETTING AND PARTICIPANTS: Semen banks were selected on the basis of membership in the American Association of Tissue Banks and commercial use of semen for artificial insemination by donor. MAIN OUTCOME MEASURE: Semen bank practices as reported by commercial semen bank directors. RESULTS: Of 37 eligible banks, 16 responded. All screen prospective donors by medical/family history and physical examination, 94% have upper age limits; 63% examine for minor physical defects; 56% routinely karyotype; 81% screen men of ethnic groups at risk for Tay Sachs disease, sickle cell disease and thalassemia; 19% screen all donors; 25% screen all donors for cystic fibrosis and 50% only screen if family history positive. Donor rejection was based on three criteria: mode of inheritance of familial disorder, severity of disease, and availability of carrier/confirmatory testing of donor genotype. Ten of 16 banks have no genetic professional on staff. CONCLUSION: Commercial semen banks primarily rely on family history as the major exclusion criterion in genetic screening of donors. Considerable differences exist among semen bank practices in accordance with guidelines published by national agencies. Genetic professionals have a minimal effect overall on evaluation of semen donors.

Isolated hyperechoic fetal bowel: significance and implications for management.

OBJECTIVE: The objective of this study was to determine the significance of isolated hyperechoic fetal bowel. STUDY DESIGN: Forty-five cases with prospective, ultrasonographic diagnosis of isolated hyperechoic fetal bowel were reviewed. Fetal variables, including aneuploidy, deoxyribonucleic acid studies for cystic fibrosis, congenital infection, growth retardation, and intrauterine death were reported. RESULTS: Thirty-four of the 45 cases (76%) resulted in live-born infants without detected abnormalities. However, hyperechoic bowel was associated with cystic fibrosis in two cases (4%), congenital infection in two cases (4%), and fetal alcohol syndrome in one case. Termination of pregnancy was elected in three cases and intrauterine fetal death occurred in three cases (7%). Growth retardation was observed in five of 39 (13%) live-born infants. CONCLUSION: Isolated hyperechoic fetal bowel is associated with significant pathologic disorders. Women whose fetuses are diagnosed as having isolated hyperechoic bowel should be offered additional prenatal diagnostic options, including maternal serologic studies for congenital infection, fetal karyotype, and deoxyribonucleic acid testing for cystic fibrosis. In addition, continuing ultrasonographic evaluation of fetal growth and antenatal biophysical assessment should be considered.

Isolated choroid plexus cyst(s): an indication for amniocentesis.

OBJECTIVE: Our purpose was to prospectively evaluate the risk of chromosomal abnormalities associated with isolated choroid plexus cyst(s) in gravid women undergoing second-trimester ultrasonographic examination. STUDY DESIGN: During a 24-month period 9100 pregnant women underwent midtrimester ultrasonographic evaluation. Women with a fetal diagnosis of choroid plexus cyst(s) were offered amniocentesis and a repeat examination in 4 to 6 weeks. RESULTS: A diagnosis of choroid plexus cyst(s) was made in 102 fetuses (1.1%). In four of these fetuses multiple congenital anomalies were noted. Three of the four fetuses had a chromosomal abnormality, two trisomy 18 and one unbalanced translocation, t(3;13). In the remaining 98 fetuses the choroid plexus cysts were isolated findings, that is, there were no other ultrasonographically detected anomalies. Seventy-five of these 98 fetuses underwent amniocentesis. An abnormal karyotype was identified in four fetuses: three had Down syndrome (two trisomy 21 and one unbalanced translocation, t[14;21]), and one trisomy 18. The offspring of the 23 patients in which amniocentesis was declined were phenotypically normal. CONCLUSIONS: In our prospective study the risk of chromosomal abnormality with isolated choroid plexus cyst(s) was 1:25, a risk that exceeds the 1:200 risk of pregnancy loss after amniocentesis and the 1:126 and 1:260 risk for aneuploidy and Down syndrome, respectively, in a 35-year-old pregnant women during the midtrimester. These findings indicate that amniocentesis should be offered to pregnant women in the presence of isolated fetal choroid plexus cyst(s).

Cell-type-specific adhesion mechanisms mediated by fibronectin adsorbed to chemically derivatized substrata.

Plasma fibronectin (pFN) adhesion mechanisms on inert substrata were evaluated for murine fibroblasts (3T3) and human neuroblastoma (Platt) cells using glass coverslips chemically derivatized with a self-assembled monolayer of aliphatic chains terminated with a specific endgroup to interact with adsorbed pFN: [CH3], [SH], [SCOCH3], [NH2], [SO3H], or underivatized glass [SiOH]. All surfaces bound similar amounts of pFN and facilitated attachment of both cell types within narrow ranges. However, spreading/differentiation responses of cells differed considerably among the surfaces. While 3T3 cells spread and developed microfilament stress fibers comparably on all surfaces, inclusion of an RGDS-containing synthetic peptide in the medium revealed variation in resistance to stress fiber formation mediated by an RGDS-recognizing integrin: [NH2] greater than [CH3] much greater than [SiOH],[SH],[SCOCH3]. Different patterns of neurite formation were observed for neuroblastoma cells: [SiOH], [SO3H] greater than [SCOCH3],[SH] much greater than [CH3] greater than [NH2]. Similarity in cell responses to both [CH3] and [NH2] surfaces argues against a pattern dependent upon the hydrophobicity of substrata. When pFN was diluted to a subsaturable concentration with albumin for adsorption, neuroblastoma responses changed significantly from those observed on pFN-saturated surfaces, for both spreading and neurite generation: [NH2],[SO3H] much greater than [SH], [SCOCH3] greater than [SiOH],[CH3]. Responses to the pFN: albumin mixture were markedly improved from responses after sequential adsorptions, demonstrating "optimization" of pFN conformation (not merely binding) by coadsorption of albumin molecules. In most cases, the [NH2] surface yielded responses distinctively different from the other surfaces. Overall, these data suggest many variations in the conformation of pFN molecules adsorbed to specific inert surfaces, as well as variations in the responses of cell surface receptors to conformationally specific pFNs. They also reveal cell-type-specific changes in differentiated cell responses on derivatized substrata, mediated by different classes of cell surface receptors for the two cell types, and provide optimism for regulating FN-dependent adhesion mechanisms in either positive or negative contexts on biomaterial surfaces derivatized with one or more of these chemical end-groups.

A randomized comparison of transcervical and transabdominal chorionic-villus sampling. The U.S. National Institute of Child Health and Human Development Chorionic-Villus Sampling and Amniocentesis Study Group.

BACKGROUND: Chorionic-villus sampling is done in early pregnancy to obtain fetal cells for the prenatal diagnosis of genetic and chromosomal defects. Transcervical chorionic-villus sampling has been shown to be safe and effective in national trials. Recently, an alternative transabdominal technique has been suggested as potentially easier and safer. METHODS: From April 1987 through September 1989, we prospectively compared transcervical and transabdominal chorionic-villus sampling in 3999 women with singleton pregnancies in whom the risk of a genetically abnormal fetus was increased. Women between 7 and 12 weeks of gestation underwent ultrasonographic evaluation of placental and uterine position. Those with active vaginal infections, active bleeding, or cervical polyps were excluded. If the obstetrician thought either sampling procedure was acceptable, the woman was asked to consent to random assignment to one of the two procedures. Both groups were followed to determine the outcome of pregnancy and the rate of spontaneous fetal loss after chorionic-villus sampling. RESULTS: Among the 3999 women who entered the study, sampling was attempted in 3873 (97 percent), 1944 of whom had been assigned to undergo transcervical sampling and 1929 to undergo transabdominal sampling. Of these 3873 women, sampling was eventually successful in 3863. Sampling was successful after a single insertion of the sampling instrument in 94 percent of the transabdominal procedures and 90 percent of the transcervical procedures. Among the women with cytogenetically normal pregnancies who had sampling because of maternal age, the rate of spontaneous fetal loss through 28 weeks of pregnancy was 2.5 percent in the transcervical-sampling group and 2.3 percent in the transabdominal-sampling group (difference, 0.26 percent; 95 percent confidence interval, -0.5 to 1.0 percent). CONCLUSIONS: Transabdominal and transcervical chorionic-villus sampling appear to be equally safe procedures for first-trimester diagnosis of fetal abnormalities.

Cytogenetic results from the U.S. Collaborative Study on CVS.

Cytogenetic data are presented for 11,473 chorionic villus sampling (CVS) procedures from nine centres in the U.S. NICHD collaborative study. A successful cytogenetic diagnosis was obtained in 99.7 per cent of cases, with data obtained from the direct method only (26 per cent), culture method only (42 per cent), or a combination of both (32 per cent). A total of 1.1 per cent of patients had a second CVS or amniocentesis procedure for reasons related to the cytogenetic diagnostic procedure, including laboratory failures (27 cases), maternal cell contamination (4 cases), or mosaic or ambiguous cytogenetic results (98 cases). There were no diagnostic errors involving trisomies for chromosomes 21, 18, and 13. For sex chromosome aneuploidies, one patient terminated her pregnancy on the basis of non-mosaic 47,XXX in the direct method prior to the availability of results from cultured cells. Subsequent analysis of the CVS cultures and fetal tissues showed only normal female cells. Other false-positive predictions involving non-mosaic aneuploidies (n = 13) were observed in the direct or culture method, but these cases involved rare aneuploidies: four cases of tetraploidy, two cases of trisomy 7, and one case each of trisomies 3, 8, 11, 15, 16, 20, and 22. This indicates that rare aneuploidies observed in the direct or culture method should be subjected to follow-up by amniocentesis. Two cases of unbalanced structural abnormalities detected in the direct method were not confirmed in cultured CVS or amniotic fluid. In addition, one structural rearrangement was misinterpreted as unbalanced from the direct method, leading to pregnancy termination prior to results from cultured cells showing a balanced, inherited translocation. False-negative results (n = 8) were observed only in the direct method, including one non-mosaic fetal abnormality (trisomy 18) detected by the culture method and seven cases of fetal mosaicism (all detected by the culture method). Mosaicism was observed in 0.8 per cent of all cases, while pseudomosaicism (including single trisomic cells) was observed in 1.6 per cent of cases. Mosaicism was observed with equal frequency in the direct and culture methods, but was confirmed as fetal mosaicism more often in cases from the culture method (24 per cent) than in cases from the direct method (10 per cent). The overall rate of maternal cell contamination was 1.8 per cent for the culture method, but there was only one case of incorrect sex prediction due to complete maternal cell contamination which resulted in the birth of a normal male.(ABSTRACT TRUNCATED AT 400 WORDS)

First-trimester biochemical and molecular diagnoses using chorionic villi: high accuracy in the U.S. collaborative study.

The accuracy of biochemical and molecular prenatal diagnoses using chorionic villi as the fetal source was assessed by seven centres participating in the NICHD collaborative study on the safety and accuracy of chorionic villus sampling (CVS) and amniocentesis. Of 601 pregnancies studied, biochemical methods were used to determine the diagnosis in 283 fetuses at risk for 35 different metabolic disorders. Fifteen different lysosomal storage diseases accounted for 81 per cent of the biochemical prenatal diagnoses performed, with 57 per cent of these pregnancies at risk for Tay-Sachs disease. No errors were made in the biochemical diagnoses that predicted affected or unaffected fetuses. However, the diagnoses of certain disorders (e.g., mucopolysacchariodosis type IH, metachromatic leukodystrophy, and Krabbe disease) occasionally required confirmatory studies in cultured amniocytes because the enzyme results were inconclusive in direct and/or cultured villi or due to the presence of a pseudodeficiency allele. Of these, only the diagnosis of a fetus at risk for Krabbe disease remained inconclusive after special studies to discriminate between mutant and pseudo-deficiency alleles. Recombinant DNA techniques were used to predict the diagnosis of 318 fetuses at risk for 16 different disorders in which the defective disease gene could be detected either directly or by linkage analysis to a nearby polymorphic marker. Of these, 32 per cent were for haemoglobinopathies, 25 per cent for cystic fibrosis, 24 per cent for Duchenne or Becker muscular dystrophy, and 7 per cent for haemophilias. Pregnancies at risk for known disorders with specific molecular lesions (e.g., sickle cell disease) were accurately diagnosed in direct and/or cultured villi. Diagnoses requiring analyses with closely linked polymorphic markers were occasionally uniformative or inconclusive. Maternal contamination was not reported in any biochemical or molecular-based diagnosis. These studies document the high accuracy and rapidity of both biochemical and mutation-specific prenatal diagnoses with direct and cultured chorionic villi.

Risk factors associated with transcervical CVS losses.

Factors found to be associated with pregnancy loss after transcervical CVS were race (higher for non-white), history of spontaneous abortion, unplanned pregnancy, history of spotting or bleeding during the pregnancy prior to CVS, and placental position (higher for fundal or lateral locations). Whether the increase in loss risk is due to the factor, per se, or the factor plus the CVS cannot be determined due to the lack of appropriate control data.

The risk and efficacy of chorionic villus sampling in multiple gestations.

Chorionic villus sampling (CVS) in the first trimester of pregnancy provides a safe and effective method for the early prenatal diagnosis of cytogenetic abnormalities in multiple gestations. In this multicentre study involving 126 twin and 2 triplet gestations primarily at risk because of advanced maternal age, the overall success rate of obtaining an adequate villus sample from each fetus was 99.2 per cent. For women of advanced maternal age, the rate of combined losses of chromosomally normal fetuses due to spontaneous abortion, stillbirths, and neonatal deaths was 5.0 per cent, compared with a 4.0 per cent total loss rate following CVS in singleton pregnancies derived from the same population (Rhoads et al., 1989). There was a 100 per cent success rate in obtaining a cytogenetic analysis; a cytogenetic abnormality was present in five of the multiple gestations (3.9 per cent) and involved seven fetuses (2.7 per cent). There were no diagnostic errors and no cases of normal cytogenetic diagnosis followed by the birth of a cytogenetically abnormal newborn. Based on cases of XX/XY admixture, cell contamination derived either from maternal decidua or the other twin occurred in 6 of 256 samples (2.3 per cent), giving an overall estimate of the frequency of cell contamination of 4.6 per cent; these cases did not present a diagnostic problem. However, there were two cases (0.8 per cent) in which the fetal sex was incorrect, due either to complete maternal cell contamination or to the possibility that in error one twin was sampled twice.