Elevated maternal serum alpha-fetoprotein levels: what is the risk of fetal aneuploidy?

OBJECTIVE: Our purpose was to present the findings of a project to determine the efficacy of including routine fetal karyotyping in the investigation of an elevated maternal serum alpha-fetoprotein concentration. STUDY DESIGN: Targeted ultrasonographic examinations were performed in 658 patients with elevated maternal serum alpha-fetoprotein levels. The scans were normal in 557 women, of whom 427 consented to amniocentesis; 435 fetuses were karyotyped. In the 101 patients with abnormal ultrasonographic examinations 75 had fetal karyotyping. RESULTS: In the 435 fetuses with normal scans, two had karyotypic anomalies, a 47,XYY and an inherited balanced translocation. Three fetuses with normal karyotypes and high amniotic fluid alpha-fetoprotein levels had congenital nephrosis. In the 101 patients with abnormal scans, 75 fetuses were karyotyped. There were four aneuploidies. Among the 26 patients with abnormal scans who declined amniocentesis one fetus with multiple anomalies was karyotyped after delivery and triploidy was discovered. CONCLUSIONS: These results provide little justification for including fetal karyotyping in the investigation of elevated maternal serum alpha-fetoprotein when the targeted ultrasonographic examination is normal. When it is abnormal, selective rather than routine karyotyping is more appropriate.

Congenital nephrosis: detection of index cases through maternal serum alpha-fetoprotein screening.

Congenital nephrosis is an autosomal recessive disorder with an incidence of 1 in 8000 in Finland, but it is quite rare in non-Finnish populations. In families known to be at risk, prenatal detection is possible by means of maternal serum and/or amniotic fluid alpha-fetoprotein levels. We report the antenatal diagnosis of four cases of congenital nephrosis, three of which were index cases, through maternal serum alpha-fetoprotein screening. The diagnosis was confirmed at birth in two infants. Two patients elected to terminate their pregnancies, and the diagnoses were confirmed pathologically (obliteration of foot processes on electron microscopy of fetal glomeruli) in both. In cases of elevated maternal serum alpha-fetoprotein, with unexplained and marked elevations of amniotic fluid alpha-fetoprotein and normal acetylcholinesterase levels, the diagnosis of congenital nephrosis must be considered regardless of ethnic origin.

The role of ultrasonography and amniocentesis in the evaluation of pregnancies at risk for neural tube defects.

To assess the relative efficacy of amniocentesis versus targeted (detailed) ultrasonography, 225 patients referred because of an elevated maternal serum alpha-fetoprotein level (79.6%) or a family history of neural tube defect (20.4%) were evaluated. Ultrasonographic examination alone detected all 26 fetal abnormalities (11 cases of anencephaly, 10 cases of open spina bifida, and five other anomalies). Twenty-eight patients declined amniocentesis; all had normal pregnancy outcomes. Of the 167 patients with apparently normal fetal anatomy by sonography, seven had elevated alpha-fetoprotein levels but no acetylcholinesterase in the amniotic fluid. Six of these pregnancies resulted in normal infants; one infant had congenital nephrosis. The remaining 160 patients had normal sonograms with normal amniotic fluid alpha-fetoprotein levels and no fetal malformations at delivery. Although these results suggest that targeted ultrasonography by experienced personnel is a reasonable alternative to amniocentesis in evaluations for neural tube defects, the availability, cost-effectiveness, and diagnostic accuracy of this approach must be well documented in large prospective studies.