Ultrasonographic assessment of amniotic fluid does not reflect actual amniotic fluid volume, ,

OBJECTIVE: Our objective was to compare the ability of two methods of amniotic fluid assessment (two-diameter amniotic fluid pocket versus the amniotic fluid index) to predict oligohydramnios (actual amniotic fluid volume <500 ml) or polyhydramnios (actual amniotic fluid volume >1500 ml). STUDY DESIGN: The amniotic fluid index and the two-diameter amniotic fluid pocket were assessed before amniocentesis and determination of amniotic fluid volume with the dye (aminohippurate sodium)–dilution technique. To assess the detection of either oligohydramnios or polyhydramnios, the areas under the receiver-operator characteristic curves (±SE) were estimated by the point-to-point trapezoidal method of integration. Prediction limits were calculated by regression analysis of amniotic fluid index or two-diameter amniotic fluid pocket versus actual amniotic fluid volume and determination of 95th percentile ranges for amniotic fluid volume. RESULTS: We studied 144 patients with a mean (±SD) gestational age of 31.7 ± 5.5 weeks; mean (±SD) amniotic flluid index and two-diameter amniotic fluid pocket were 12.6 ± 6.1 cm and 21.2 ± 18.4 cm2, respectively. Mean (±SD) actual amniotic fluid volume was 722 ± 735 ml (range 101 to 4318 ml). The areas under the four receiver-operator characteristic curves were not significantly different from the nondiagnostic line (p < 0.05). Regression slopes (r values) for amniotic fluid index and two-diameter amniotic fluid pocket versus actual amniotic fluid volume were 0.34 and 0.23, respectively. Calculation of the prediction limit for 95% confidence that oligohydramnios is absent requires that the amniotic fluid index be 30 cm and the two-dimension amniotic fluid pocket be 90 cm2, both thresholds of which are currently considered to represent clinical polyhydramnios. CONCLUSIONS: Both amniotic fluid index and two-dimension amniotic fluid pocket appear to be inaccurate predictors of actual oligohydramnios or polyhydramnios when compared with dye-dilution calculations of actual amniotic fluid volume. (Am J Obstet Gynecol 1997;177:291-7.)