Twin pregnancies in the second trimester in women in an alpha-fetoprotein screening program: sonographic evaluation...

We correlated sonographic findings with fetal outcomes in women with unsuspected twin pregnancies who had sonography in the second trimester as part of a screening program for maternal serum alpha-fetoprotein (MSAFP) level and history of neural tube defect.

The study group consisted of 97 women with twin pregnancies who participated in a screening program for MSAFP level and history of neural tube defect. Seventy-three had normal MSAFP levels, 21 had elevated MSAFP levels, and two had low MSAFP levels. One patient had a family history of anencephaly. All 97 patients had sonography during their second trimester of pregnancy. Sonographic findings were reviewed retrospectively for information on gestational age, fetal anomalies, sex of the fetus, location of the placenta, presence and thickness of a dividing membrane, and interpretation of amnionicity and chorionicity. Information on fetal outcome included gestational age at delivery, survival, birth weight, sex, congenital anomalies, obstetric complications, amnionicity, chorionicity, and placental abnormalities.

Amnionicity and chorionicity were correctly detected on sonograms in 44 (90%) of 49 diamniotic-dichorionic gestations, 23 (72%) of 32 diamniotic-monochorionic gestations, and two (50%) of four monoamniotic-monochorionic gestations. Fetal anomalies were present at delivery in five neonates and had been correctly detected at sonography in one (hemivertebra); one fetus with duodenal atresia had abnormal sonographic findings in the third trimester. Missed anomalies included absent forearm, cleft lip and palate, and imperforate anus. Sex of the fetuses was correctly predicted on the basis of sonographic findings in 40 of 43 pairs. Nine twin pairs had possible twin-twin transfusion syndrome suspected sonographically on the basis of abnormal fluid volumes, discrepant growth measurements, and abnormal findings on Doppler studies. Outcomes included two confirmed cases of the syndrome (two survivors, two deaths) and three probable cases (six deaths); four pregnancies resulted in eight survivors who were delivered after 34.4 weeks' gestation and had birth weights in the 25th percentile or higher. Survival rates for diamniotic-dichorionic, diamniotic-monochorionic, and monoamniotic-monochorionic gestations were 90%, 91%, and 50%, respectively. Fetuses in women with MSAFP levels greater than 4.5 multiples of the median and with monochorionic placentation had lower survival rates than fetuses in women with normal MSAFP levels and monochorionic placentation (67% vs 96%). Half the fetuses delivered after 20 weeks' gestation had birth-weight discordance of less than 10%. Premature deliveries occurred in 56% of pregnancies.

The results suggest that (1) sonography is useful in predicting placentation, (2) placentation may be helpful in predicting fetal outcome, (3) increased MSAFP levels correlate with increased perinatal mortality in diamniotic-monochorionic pregnancies, and (4) caution should be taken in diagnosing and determining prognosis for suspected twin-twin transfusion syndrome in the second trimester.