Midtrimester prelabor rupture of membranes (PPROM) can delay lung development, resulting in pulmonary hypoplasia. Because of the high perinatal mortality and morbidity of pulmonary hypoplasia, accurate prediction of the probability of its occurrence before 24 weeks is important both for optional management and for counseling of parents regarding the option of termination of pregnancy. In a recent meta-analysis, gestational age at PPROM was a significantly better predictor of hypoplasia than latency period until delivery or degree of oligohydramnios. No systematic meta-analysis has assessed the predictive capacity of biometric parameters for the presence of hypoplasia after midtrimester PPROM.
This systematic meta-analysis investigated the capacity of biometric parameters assessed with ultrasound or magnetic resonance imaging (MRI) to predict pulmonary hypoplasia following midtrimester PPROM. Studies included in the analysis reported both outcomes of pregnancies complicated by PPROM between 14 and 27 completed weeks of gestational age and any ultrasound or MRI parameter used to predict pulmonary hypoplasia. Each selected study allowed the construction of a 2 × 2 table comparing at least one of the parameters with the occurrence of pulmonary hypoplasia. The methodological quality of the selected studies and the sensitivity and specificity of the predictive tests for the occurrence of lethal hypoplasia were determined. To estimate the overall performance of the imaging parameters, a bivariate metaregression model was used to calculate summary estimates of sensitivity and specificity for predictive values and to fit a summary receiver operating characteristic curve.
A total of 13 studies were identified that reported on the prediction of lethal pulmonary hypoplasia. Sensitivity and/or specificity were low in all but one study. The chest circumference/abdominal circumference ratio and other parameters had limited accuracy based on their estimated summary receiver operating characteristic curves in the prediction of pulmonary hypoplasia. The studies had several methodological weaknesses. There was verification bias in many of the studies either because the test was being used in patient management or because of a lack of blinding of the observers. Moreover, only a single center was used in each study (which limited the sample size because of the rarity of PPROM).
These findings show that biometric parameters have limited accuracy in the prediction of pulmonary hypoplasia among women with midtrimester PPROM.