Comparison of the Accuracy of Fetal Pulmonary Maturity Testing Using Transvaginally and Transabdominally Collected Amniotic Fluid
1993; Informa; Volume: 2; Issue: 6 Linguagem: Inglês
10.3109/14767059309017252
ISSN1476-7058
AutoresHelen Y. How, Carolyn C. Marcell, Joseph A. Spinnato, Joseph A. Spinnato,
Tópico(s)Preterm Birth and Chorioamnionitis
ResumoTo evaluate the reliability of transvaginally collected amniotic fluid in the assessment of fetal pulmonary maturity.A retrospective chart review was conducted on mother-infant pairs in which delivery occurred before 3 7 weeks gestation and a lecithin/sphingomyelin ratio and phosphatidylglycerol had been determined within 1 week of the delivery.Of 299 pairs, 154 had amniotic fluid collected transvaginally and 145 transabdominally. A lecithin/ sphingomyelin ratio of ≥2.0 predicted the absence of hyaline membrane disease with 99% accuracy in the transabdominal group compared with 96% in the transvaginal group. The false-negative rate of the lecithin/sphingomyelin ratio was 65% in the transabdominal group and 78% in the transvaginal group. The sensitivity of phosphatidylglycerol to predict the absence of hyaline membrane disease was approximately half that of a mature lecithin/sphingomyelin ratio (≥2), i.e., 48% versus 80% in the transabdominal group, 26% versus 50% in the transvaginal group. Only 2% (3/153) of premature infants ≥33 weeks gestational age developed hyaline membrane disease; none was in the transvaginal group.A lecithin/sphingomyelin ratio of ≥2.0 using a transvaginally collected sample reliably excludes hyaline membrane disease. The higher false-negative rate of phosphatidylglycerol and the high positive predictive value of the lecithin/sphingomyelin ratio in both transabdominally and transvaginally collected specimens make the lecithin/sphingomyelin ratio a more clinically useful test for difficult management decisions involving preterm pregnancy. With ruptured membranes at ≥33 weeks, fetal maturity tests are at best confirmatory and may suggest disease risk where none exists.
Referência(s)