The association between alpha-fetoprotein and βhcg levels prior to and following chorionic villus sampling in cases that spontaneously miscarried

Maternal serum alpha-fetoprotein (AFP) and beta-human chorionic gonadotropin (βhCG) measurements taken prior to chorionic villus sampling (CVS) in 21 patients who subsequently miscarried were compared with measurements in a control group of 113 patients with uneventful pregnancies. Patients with AFP levels of 10 iu/ml or more prior to the CVS had a 4·3 times greater risk of miscarriage (95 per cent confidence interval 1·3–13·6). AFP levels obtained 1 week after the CVS in the 13 patients with late miscarriages were higher than in the control group (P = 0·06). Patients miscarrying had a greater rise in AFP (P = 0·06) and a greater fall in βhCG levels (P = 0·04) following the CVS procedure, compared with the control subjects. Each 10-unit change in the difference between AFP or βhCG levels prior to and 1 week following the CVS was associated with a significantly increased risk for late miscarriage. Elevated maternal serum AFP levels early in pregnancy and changes in AFP and βhCG levels following CVS may predict an increased risk for subsequent miscarriage.     

Rapid detection of numerical aberrations of chromosomes 13, 18 and 21 in chorionic mesenchymal cells

We have devised and evaluated a rapid screening method for the detection of numerical aberrations of chromosomes13, 18 and 21 in chorionic villus cells. We used non-radioactive in situ hybridization (ISH) with three chromosome-specific probes on overnight-attached mesenchymal cells from chorionic villi. A blind study was performed of 47 karyotypically normal samples, one triploid sample, two samples trisomic for chromosome 21, and two samples from a fetus with putative mosaicism (46/47, +21). All samples were hybridized with the chromosome 18- and 21-specific probes. Thirty samples were additionally hybridized with the chromosome 13-specific probe. The test could be completed within 3-4 days of sampling. In samples disomic with respect to the probed chromosomes, an average of 2 per cent (range 0-9 per cent) had three hybridization signals. By contrast, in the samples trisomic for the probed chromosome(s), 57 per cent (chromosome 13), 51 per cent (chromosome 18), and an average of 74 per cent (55-86 per cent) (chromosome 21) of the nuclei exhibited three signals. In the putative mosaic samples, the number of nuclei with three chromosome 21-specific signals ranged from 41 to 69 per cent. We conclude that this technique rapidly and clearly distinguishes between normal and trisomic/triploid samples, and consequently may be of use in future prenatal diagnosis.     

Superovulation fails to increase human blastocyst yield after uterine lavage

Uterine lavage affords the potential for non-invasive human blastocyst recovery, with obvious potential for preimplantation genetic diagnosis. In an effort to duplicate in women the multiple blastocyst recovery per cycle that can be achieved in several other species, we initiated a programme in which fertile women underwent superovulation, followed by lavage and embryo collection. We superovulated 15 fertile women, aged 21–40, in 29 cycles using one of four regimens. Insemination was by either intercourse or artificial intracervical donor insemination with cryopreserved sperm from men of proven fertility. In 28 of 29 cycles, the uterus was lavaged daily for 1, 2, or 3 days between 5 and 10 days after human chorionic gonadotropin (hCG) administration or luteinizing hormone (LH) surge. Almost total fluid volume was recovered in every lavage. There were no retained pregnancies and no complications. Surprisingly, only two morulae, one blastocyst, and four unfertilized ova were recovered. Thus, alterations in ovulation induction, insemination timing, or lavage techniques must be contemplated in order to increase the blastocyst yield and thus fulfil the potential of uterine lavage for preimplantation diagnosis.