Diagnostic fetal blood sampling-technique related losses

BACKGROUND

The applications of cordocentesis are limited by its safety. Cordocentesis is performed either freehand or with the aid of a fixed needle guide. Recent reports suggest a loss rate of 1-7% using the freehand technique. The potential influence of technique on the perinatal loss rate has not been examined.

STUDY DESIGN

The procedure-related loss rates of two fetal diagnosis and treatment units, one in the United States (10 operators) and one in Japan (15 operators), who perform diagnostic cordocentesis with the aid of a fixed needle guide is calculated from a prospectively maintained database and compared to the published experience of large centers who use a freehand technique.

RESULTS

A total of 1,260 diagnostic cordocenteses were performed. The mean gestation at sampling was 29.1 +/- 5 weeks. The umbilical vein was punctured in 90%. There were 12 procedure-related losses yielding an overall perinatal loss rate of 0.9%. There was no relationship between the risk of a loss and the number of prior procedures the operator had performed. Losses were more often associated with puncture of the umbilical artery (41.7 vs. 9.2%, p = 0.002). Eleven of 12 losses were associated with a postprocedure bradycardia. Eight fetuses who died had either a trisomy or triploidy, 1 had renal agenesis, 2 had severe early-onset growth restriction and 1 had rhesus disease. For all diagnoses other than a chromosome abnormality and severe fetal growth restriction, the procedure-related loss rate from diagnostic cordocentesis was 0.2% (2/1,021).

CONCLUSION

This study suggests that technique is a variable in the loss rate for cordocentesis.

Efficacy and safety of cordocentesis for prenatal diagnosis

OBJECTIVE

To assess the efficacy and safety of diagnostic cordocentesis.

METHODS

Between January 1991 and May 2004, 2,010 cordocentesis were performed in the outpatient setting in 2,010 women with singleton pregnancies. A fixed needle guide and a 22-gauge percutaneous needle were used and no more than 2 attempts were allowed at 1 visit. In most cases, the umbilical vein was the target vessel. The results of each procedure and pregnancy outcomes were recorded and analyzed.

RESULTS

The most frequent indication for cordocentesis was risk of severe thalassemia (59.0%), followed by a need for rapid karyotyping (30.0%). Most of the procedures (97%) were performed in the free cord loop and the remaining at the cord insertion. The overall success rate was 98.4%, with 80.0% of the successful procedures performed at the first needle insertion and the remaining 20% at the second insertion. Transient bleeding was observed at the puncture site in 19.8% of cases and transient fetal bradycardia in 4.9% of cases. The total fetal loss and cordocentesis-related loss rates within 2 weeks of cordocentesis were 2.7% and 1.0%, respectively, before 24 weeks of gestation and 1.9% and 0.8% after 24 weeks. The other obstetric complications were unremarkable.

CONCLUSIONS

Cordocentesis is a simple, safe, and reliable procedure for prenatal diagnosis.

Guidelines for blood sampling and measurement of pH and blood gas values in obstetrics. Based upon a workshop held in Zurich, Switzerland, March 19, 1993 by an Ad Hoc Committee

Guidelines for the clinical indications for measuring pH and blood gas values in fetal blood, the procedures of blood sampling and measurement and some reference values for the evaluation of the data are given. They cover: prenatal sampling of blood from the umbilical vessels in conjunction with cordocentesis, intra partum sampling of fetal capillary blood by skin puncture of the presenting part, post partum sampling of blood from a clamped section of the umbilical cord and general analytical techniques.

Pallister-Killian syndrome [i(12p)]: first pre-natal diagnosis using cordocentesis in the second trimester confirmed by in situ hybridization

Pallister Killian syndrome (PKS) is the most frequent form of partial autosomal tetrasomy 12p in humans. Sufferers have a mosaic of isochromosome 12p [i(12p)]. We report the first pre-natal diagnosis on fetal blood cells after cordocentesis during the second trimester. The extra chromosome was first diagnosed by in situ hybridization. Fluorescence in situ hybridization (FISH) was used to count the interphase and/or metaphase cells containing the isochromosome. A review of the literature identified 27 other reports of PKS diagnosed pre-natally. We showed that the most consistent pre-natal ultrasound findings include hypertelorism, broad neck, shorts limbs, abnormal hands or feet, diaphragmatic hernia and hydramnios. Recognition of this congenital malformation pattern pre-natally may allow utilization of FISH.

Gene expression in cord blood links genetic risk for neurodevelopmental disorders with maternal psychological distress and adverse childhood outcomes

Prenatal exposure to maternal stress and depression has been identified as a risk factor for adverse behavioral and neurodevelopmental outcomes in early childhood. However, the molecular mechanisms through which maternal psychopathology shapes offspring development remain poorly understood. We applied transcriptome-wide screens to 149 umbilical cord blood samples from neonates born to mothers with posttraumatic stress disorder (PTSD; n = 20), depression (n = 31) and PTSD with comorbid depression (n = 13), compared to carefully matched trauma exposed controls (n = 23) and healthy mothers (n = 62). Analyses by maternal diagnoses revealed a clear pattern of gene expression signatures distinguishing neonates born to mothers with a history of psychopathology from those without. Co-expression network analysis identified distinct gene expression perturbations across maternal diagnoses, including two depression-related modules implicated in axon-guidance and mRNA stability, as well as two PTSD-related modules implicated in TNF signaling and cellular response to stress. Notably, these disease-related modules were enriched with brain-expressed genes and genetic risk loci for autism spectrum disorder and schizophrenia, which may imply a causal role for impaired developmental outcomes. These molecular alterations preceded changes in clinical measures at twenty-four months, including reductions in cognitive and socio-emotional outcomes in affected infants. Collectively, these findings indicate that prenatal exposure to maternal psychological distress induces neuronal, immunological and behavioral abnormalities in affected offspring and support the search for early biomarkers of exposures to adverse in utero environments and the classification of children at risk for impaired development.

Robustness of factor assays following cordocentesis in the prenatal diagnosis of haemophilia and other bleeding disorders

Prenatal diagnosis is the generally accepted option for genetic disorders including haemophilias and other bleeding disorders. Cord blood analysis between 17.4 and 20.6 weeks of gestation was performed in 172 confirmed carriers belonging to families of haemophilia A, haemophilia B, von Willebrand disease (VWD), factor VII and X deficiency; 133 were carriers for haemophilia A, 30 for haemophilia B, six for type 3 VWD, two for FX deficiency and one for FVII deficiency. The approach to the cord was either transabdominal or transamniotic. The volume of blood collected varied between 1 and 2 mL. In case of haemophilias, the diagnosis was offered by factor VIII/IX:C activity and antigen assays wherever required. In case of VWD, the diagnosis was based on von Willebrand factor antigen assays as detected by ELISA along with FVIII:C assay while in cases of FVII and FX deficiency, the diagnosis was based on FVII:C and FX:C respectively. The factor levels were compared with the normal range established in the laboratory for different coagulation factors between 18 and 21 weeks of gestation in women tested for other haematological disorders. Only in two cases, the procedure had to be repeated for reasons of extensive maternal contamination. All the deliveries have been followed up and the diagnoses reconfirmed by repeat clotting factor assays and DNA analysis whenever informative. Simple precautions like collection of fetal blood samples in smaller volumes in separate tubes, assaying multiple coagulation factors in the fetal blood samples helped us to offer diagnoses in all the women analysed. No fetal death or abortion was reported following the procedure. We suggest that accurate fetal blood sampling is a safe technique for the diagnosis of many of the bleeding disorders in places where genetic diagnostic services are not available.

Catecholamine response to ultrasonographically guided percutaneous blood sampling in fetal sheep

OBJECTIVE

The purpose of this study was to investigate the fetal catecholamine and arterial blood gas responses to ultrasonographically guided percutaneous needle aspiration of blood from a fetal cardiac ventricle.

STUDY DESIGN

A crossover trial design was used. Nine pregnant sheep of 120 to 130 days' gestation were stratified to either percutaneous fetal blood sampling or a sham experiment performed on the first day, and the alternative study on the following day. Arterial blood samples were withdrawn from chronically implanted catheters.

RESULTS

Percutaneous fetal blood sampling caused small but statistically significant increases in mean plasma epinephrine and norepinephrine levels 5 seconds after the procedure. Levels thereafter were similar to baseline values. Arterial pH and PCO2 values were unaltered except in one fetus, where blood sampling was followed by bradycardia with acidosis and elevated catecholamine levels.

CONCLUSIONS

Ultrasonographically guided percutaneous fetal blood sampling from a cardiac ventricle in sheep produces a rise in catecholamine levels that return to baseline values within 5 minutes. This technique provides an alternative to chronic catheterization for some experiments in which blood sampling or drug injection is required. The results also indicate that needle insertion produces only a modest and transient stress response in the fetus.

Teaching ultrasound-guided invasive procedures in fetal medicine: learning curves with and without an electronic guidance system

OBJECTIVE

To compare the learning curves of inexperienced junior obstetrics/gynecology registrars for ultrasound-guided invasive procedures on a training model, with and without an electronic guidance system.

STUDY DESIGN

Four junior registrars performed their first 100 procedures on a training model with a new electronic guidance system, and four other junior registrars performed their first 100 procedures on the same training model without using the guidance system. All procedures were performed using a free-hand technique. We evaluated the quality of the procedure, which we defined as the time spent with the entire needle clearly visualized on the screen over the total duration of the procedure. We constructed learning curves for the eight junior registrars for comparative analysis.

RESULTS

Quality of the procedure increased over time for all trainees. The learning curves were significantly steeper for trainees using the electronic guidance system. Trainees using the electronic guidance system performed better in the middle of their learning curve (procedures 25-75). All trainees reached the same level of quality by the end of their 100 procedures.

CONCLUSIONS

The automated electronic guidance system helps faster learning but, after 100 procedures on a training model, both groups reached the same level of quality.

Lutein levels in arterial cord blood correlate with neuroprotein activin A in healthy preterm and term newborns: A trophic role for lutein?

BACKGROUND

Lutein (LT) is a naturally occurring xanthophyll carotenoid most predominant in the central nervous system (CNS), but its neurotrophic role is still debated. We therefore investigated whether cord blood concentrations correlated with a well-established neurobiomarker, namely activin A.

METHODS

We conducted a prospective study on the distribution of LT and activin A in arterial cord blood of healthy preterm (n=50) and term (n=82) newborns according to weeks of gestational age (wGA) and gender.

RESULTS

LT and activin A showed a pattern of concentration characterized by higher levels (P<0.01, for all) at 33-36 wGA followed by a progressive decrease (P<0.01, for all) from 37 onwards with a dip at term. Both LT and activin A were gender-dependent with significantly (P<0.01, for all) higher levels in all recruited females and after sub-grouping for preterm and term births. LT (R=0.33; P<0.001) correlated with wGA at sampling. There were significant positive correlations between lutein and activin A in male (R=0.93; P<0.001) and female (R=0.89; P<0.001) groups.

CONCLUSIONS

The present data showing a correlation between LT and activin A support the notion of a neurotrophic role gender-dependent for LT and open the way to further investigations correlating LT with well-established biochemical markers of CNS development/damage.

Prenatal diagnosis of thrombocytopenia-absent-radius (TAR) syndrome

The prenatal diagnosis of thrombocytopenia-absent-radius (TAR) syndrome using ultrasound and cordocentesis in the 16th week of gestation is established. The sonographic findings detected in this case included bilateral absence of the radius and club hands with normal thumbs and metacarpals. Because of a high index of suspicion for the syndrome, cordocentesis for fetal blood analysis was performed. Thrombocytopenia, with a platelet count of 14,000/mm3, was identified. The pregnancy was electively terminated and subsequent findings confirmed the sonographic diagnosis. This report, to our knowledge, is one of a very limited number of cases published in the literature, in which the prenatal diagnosis of TAR syndrome was made.

Four-dimensional ultrasound guidance of prenatal invasive procedures

OBJECTIVE

Technological advances in ultrasonography have revolutionized prenatal diagnosis and treatment. Here we evaluate the effectiveness of using four-dimensional (4D) ultrasonography to guide prenatal invasive procedures.

PATIENTS AND METHODS

Prenatal invasive procedures using 4D ultrasound were recorded prospectively in 93 cases: 10 amnioinfusions, 50 amniocenteses, 8 chorionic villus samplings (CVS) and 25 cordocenteses. The needle target site was first identified using the two-dimensional (2D) mode, and was then confirmed using the three-dimensional (3D) mode. The needle was inserted under 4D ultrasound guidance. After selecting the needle target site, the true position of the needle was determined in three planes ('real-time 3D targeting').

RESULTS

Using 4D ultrasound guidance, most procedures were performed within 5 min and with a 100% success rate, even in cases involving severe oligohydramnios (amniocentesis), thin placentas (CVS) or narrow umbilical veins (cordocentesis). Moreover, there were no serious complications during or after any procedure.

CONCLUSIONS

4D ultrasonography can be used to guide various prenatal invasive procedures offering clear information in all three planes. It is likely that such imaging will reduce the time taken to complete the procedures and reduce the risks associated with them. Published by John Wiley & Sons, Ltd.

Development of a training model for ultrasound-guided invasive procedures in fetal medicine

Training in ultrasound-guided procedures in fetal medicine is currently available in a few major fetal medicine units. Unlike elective surgical procedures, cordocentesis is usually performed without sedation as an outpatient procedure and active supervision of the trainee intensifies the anxiety already felt by the patient. In addition, the complication rate is higher in the initial few procedures or when the procedure is not regularly practiced. We have developed a comprehensive method of training for transabdominal invasive procedures using a medical model that simulates the in vivo situation.

Cordocentesis in multifetal pregnancies

OBJECTIVE

To describe the experiences in diagnostic cordocentesis in twin pregnancies at midpregnancy

METHODS

The database and medical records of pregnant women attending Maternal Fetal Medicine Unit of the hospital for diagnostic cordocentesis at midpregnancy between January 1989 and September 2006 were retrospectively reviewed.

RESULTS

During 17 years of experience, 4,241 cordocenteses at midpregnancy were performed for prenatal diagnosis, including 59 procedures in 30 multiple pregnancies (29 twins and 1 triplet). The mean gestational age at the time of cordocentesis was 19.5 +/- 1.6 weeks. Success rate of the samplings was 98.3% with one sample was maternal blood contamination. Averaged-time used of the procedures was 8.2 minutes (range 1-45 minutes). The procedure-related complications included transient bleeding at puncture site (8.5%) and transient fetal bradycardia (22.0%). The total fetal loss rate was 10.5% but there was no cordocentesis-related fetal loss (0.0%), defined as a fetal loss within 2 weeks after the procedure.

CONCLUSION

This study may provide a new insight on the safety of cordocentesis in multifetal pregnancies at midpregnancy. The procedure-related fetal loss is not as high as reported in the past. This study suggests cordocentesis be a relatively safe and highly successful in obtaining fetal blood samples.

Ultrasound-guided procedures for prenatal diagnosis and therapy

Ultrasonography has expanded the capabilities of perinatologists to examine,test, and treat the fetus. Amniocentesis and CVS are safe and widely available procedures, which can be used to diagnose a multitude of abnormalities through karyotype analysis and molecular studies. CVS allows earlier diagnosis, but both procedures can provide highly accurate results in the first half of pregnancy. Cordocentesis has fewer indications, but allows direct laboratory testing of fetal blood. Fetocentesis and fetal biopsy are reserved for limited indications, but can play a crucial role in the diagnosis of some conditions, which cannot be assessed less invasively. Fetal transfusion is an important tool in the treatment of isoimmunization, some other forms of fetal anemia, and alloimmune thrombocytopenia. Amnioreduction is a commonly used procedure for the treatment of polyhydramnios and TTTS. Multifetal reduction and selective termination offer previously unavailable options to patients carrying multiple gestations. Fetal shunts can reduce perinatal morbidity and mortality in cases of bladder outlet obstruction and hydrothorax. The limited experience with cord ligation procedures and balloon valvuloplasty suggests these relatively new procedures may serve a greater role in the future as techniques are improved. By providing guidance for all of these procedures, real-time ultrasonography has revolutionized prenatal diagnosis and therapy; it will continue to be a crucial component in evaluating and treating complicated pregnancies.

Four-dimensional ultrasonographic guidance for invasive obstetric procedures

OBJECTIVE

This pilot study evaluated the use of 4-dimensional (4D) real-time ultrasonographic needle guidance for amniocentesis, chorionic villus sampling (CVS), cordocentesis, and intrauterine transfusions.

METHODS

Ninety-nine consecutive procedures were performed with 4D (real-time) multiplanar ultrasonographic imaging. Amniocentesis was done freehand in 3 orthogonal planes of view. Chorionic villus sampling, cordocentesis, and intrauterine transfusions were accomplished with a needle guide and 2 projected orthogonal planes.

RESULTS

Needle tip visualization in the A, B, and C orthogonal planes during amniocentesis was noted in 93%, 63%, and 69% of cases, respectively. When a needle guide was used during CVS and cordocentesis, the needle tip was always seen in the 2 projected orthogonal planes, and no lateralization occurred. Four intrauterine transfusions were done with the 4D technique. The only procedural complication in any patient was bradycardia from vessel spasm during an intrauterine transfusion, requiring a cesarean delivery. There were no statistical differences (P > .05) between the numbers of needle insertions required in the 4D group compared with a historical control group in which 2-dimensional ultrasonographic needle guidance was used.

CONCLUSIONS

In this feasibility study, a real-time 4D needle guidance technique was successfully used to perform amniocentesis, CVS, cordocentesis, and intrauterine transfusion. This appeared to contribute to the accuracy of needle placement by eliminating the lateralization phenomenon when a fixed needle guide attachment was used (for CVS and cordocentesis). Needle tip visualization was seen in each orthogonal plane in most freehand 4D amniocentesis cases. Future developments in 4D ultrasonographic technology may refine the utility of this technique for invasive obstetric procedures.

Detection and measurement of fetomaternal hemorrhage following diagnostic cordocentesis

OBJECTIVE

The aim of this study was to investigate the effect of cordocentesis on fetomaternal hemorrhage (FMH).

METHODS

One hundred and three diagnostic cordocenteses, without any other associated procedure, were performed at 23-40 weeks' gestation. FMH was detected using the Kleihauer-Betke staining of maternal blood taken immediately before and after cordocentesis.

RESULTS

Significant FMH occurred after 40 (38.8%) of the 103 procedures. An increased risk of fetal bleeding was associated with both an anterior placenta (odds ratio (OR) 5.89; 95% confidence interval (CI) 2.27-15.3; p < 0. 001) and a transplacental cordocentesis (OR 37.0; 95% CI 2.15-636; p < 0.001). The volume of FMH was greater after cordocentesis with an anterior placenta (90th percentile 6.20 ml) than after cordocentesis with a lateral (90th percentile 4.58 ml) or posterior placenta (90th percentile 1.35 ml) (p < 0.001). After fetal blood sampling, significant FMH occurred more frequently with a procedure duration of 3 min or more (OR 4.45; 95% CI 1.70-11.7; p = 0.002) and with two or more needle insertions (OR 4.65; 95% CI 1.80-12.1; p = 0.001).

CONCLUSION

FMH following cordocentesis may be related to placental injuries. This event is influenced by placental location, procedure duration and the number of needle insertions.

Midpregnancy cordocentesis training of maternal-fetal medicine fellows

OBJECTIVE

To describe systematic cordocentesis training among maternal-fetal medicine (MFM) fellows.

METHODS

During their 2-year training period, five MFM fellows, who had completed systematic model training, performed 1116 midpregnancy diagnostic cordocentesis procedures (mean, 223 (range, 185-259) procedures each) under expert supervision. The details of the procedures were recorded prospectively. Multiple pregnancies and those with fetal chromosomal or structural anomalies were excluded from analysis. The outcome measures included success rate, duration of the procedure, fetal loss and complications.

RESULTS

Of the 1116 procedures performed, 184 were excluded because of fetal abnormalities; the remaining 932 were available for analysis. Cordocentesis was performed on free-floating umbilical cord (79.3%) and at the placental insertion site (20.7%). There was an immediate complication of transient fetal bradycardia in 10% of cases. Individual success rates ranged from 98.1 to 100% and the mean cumulative success rate had plateaued by approximately 60 procedures. The overall fetal loss rate was 1.3%. The overall mean +/- SD duration of successful procedures was 4.4 +/- 4.7 min, individual mean durations ranging from 3.7 to 5.9 min.

CONCLUSIONS

MFM fellows with systematic training are able to perform cordocentesis with very high success rates, and with an acceptable procedure-related fetal loss rate. An intensive course of preclinical training with the model, and more than 60 procedures on patients under supervision is recommended.

Cordocentesis using the combined technique; needle guide-assisted and free-hand

Two hundred and sixteen diagnostic cordocenteses were performed using the following technique: A guide was used to deliver the distal end of the needle to the immediate vicinity of the umbilical cord, after which the needle was released from the guide and a free-hand technique was used to enter the umbilical cord. The vessel punctured was identified by its sonographic appearance and flow direction using color Doppler technology. All procedure-related losses which occurred within 2 weeks were analyzed. The gestational age at the time of cordocentesis ranged from 18 to 42 weeks. Most punctures (62%) were performed at the placental insertion of the umbilical cord. In 32% of fetuses the free-floating loop was sampled and in 6% the puncture was performed at the site of cord entry into the fetus. Two fetuses died shortly after cordocentesis. One death occurred at 28 weeks in a fetus with severe cytomegalovirus infection. The other death was due to premature rupture of the membranes after the procedure in a very premature fetus. The overall fetal loss rate was 0.93%. In conclusion, the combination of the two cordocentesis techniques appears safe and highly successful in obtaining fetal blood samples.

Study of the fetal and maternal microbiota in pregnant women with intrauterine growth restriction and its relationship with inflammatory biomarkers: A case-control study protocol (SPIRIT compliant)

In general terms, fetal growth restriction (FGR) is considered the impossibility of achieving the genetically determined potential size. In the vast majority of cases, it is related to uteroplacental insufficiency. Although its origin remains unknown and causes are only known in 30% of cases, it is believed to be related to an interaction of environmental and genetic factors with either a fetal or maternal origin. One hypothesis is that alterations in the gastrointestinal microbiota composition, and thus alteration in the immune response, could play a role in FGR development. We performed an observational, prospective study in a subpopulation affected with FGR to elucidate the implications of this microbiota on the FGR condition.A total of 63 fetuses with FGR diagnosed in the third trimester as defined by the Delphi consensus, and 63 fetuses with fetal growth appropriate for gestational age will be recruited. Obstetric and nutritional information will be registered by means of specific questionnaires. We will collect maternal fecal samples between 30 to 36 weeks, intrapartum samples (maternal feces, maternal and cord blood) and postpartum samples (meconium and new-born feces at 6 weeks of life). Samples will be analyzed in the Department of Biochemistry and Molecular Biology II, Nutrition and Food Technology Institute of the University of Granada (UGR), for the determination of the gastrointestinal microbiota composition and its relationship with inflammatory biomarkers.This study will contribute to a better understanding of the influence of gastrointestinal microbiota and related inflammatory biomarkers in the development of FGR.Trial registration: NCT04047966. Registered August 7, 2019, during the recruitment stage. Retrospectively registered. Ongoing research.

[Kell alloimmunization in pregnancy]

INTRODUCTION

Kell alloimmunization is a rare disease, although its incidence is the highest after after anti-D alloimmunization.

METHODS

We report two recent cases and a review of the literature to describe practical management of Kell alloimmunization in pregnancy.

DISCUSSION

When an immunization against the Kell antigen was diagnosed, amniocentesis was performed at 14 weeks gestation to determine the fetal blood group. If the fetus was Kell positive, a first fetal blood sample was drawn at 17 weeks gestation in case of fetal hydrops, and at 20 weeks without fetal hydrops. The diagnosis of anemia led to in utero transfusion. A second fetal blood sample was taken at 8 to 10 days, every two weeks during the second trimester and every three or four weeks during the third trimester. Fetal well-being was assessed with weekly sonography and rates of hemoglobin decline. These measures enable adapting the frequency of fetal blood sampling.

Application of the 'Apt test' in prenatal diagnosis to evaluate the fetal origin of blood obtained by cordocentesis: results of 30 pregnancies

This study aimed to set up a practical lab-side approach to discriminate fetal from maternal blood in samples obtained by cordocentesis. To determine the fetal origin of the blood, a modified Apt test was applied to 30 cases of prenatal diagnosis. A change of colour of the fetal and adult blood during the procedure was the hallmark to assess fetal origin. At the end of 60 s of the test, fetal blood yielded a pink colour whereas adult blood was dark green-brown. The test was repeated in mixtures of fetal and adult blood. The results suggest that the modified Apt test is a practical, quick, inexpensive, and efficient test to determine the origin of blood samples obtained by cordocentesis. However, it should be kept in mind that samples containing a mixture of both fetal and adult blood could also yield a fetal blood reaction. When maternal contamination is suspected, we propose that at least 30 metaphases from different slides should be counted. This could yield fetal as well as maternal chromosomes.