The accuracy of the summated amniotic fluid index in evaluating amniotic fluid volume in twin pregnancies

OBJECTIVE: Our purpose was to determine the accuracy of the summated amniotic fluid index designed to estimate the total amniotic fluid volume in twin pregnancies. STUDY DESIGN: The summated amniotic fluid index was measured in 62 normal diamniotic twin pregnancies by adding the deepest vertical pockets in the four quadrants. Actual amniotic fluid volume was then determined in all 124 amniotic sacs by amniocentesis and a dye-dilution technique. For data analysis, amniotic fluid volumes were classified by percentile with use of previously reported norms. RESULTS: There were significant differences in the percentile distribution of amniotic fluid volume as estimated by the summated amniotic fluid index and the actual volume as determined by dye dilution (p < 0.001). The summated amniotic fluid index has a sensitivity of only 13% in predicting amniotic sac volume. CONCLUSION: The summated amniotic fluid index is a poor predictor of intertwin differences in amniotic fluid volume and cannot identify twin pairs al risk for oligohydramnios and hycramnios.