Prenatal ultrasonic prediction of autopsy-proven pulmonary hypoplasia

Abstract

The sensitivity, specificity, and positive and negative predictive values for ultrasound estimation of fetal chest circumference in relation to autopsy-proven pulmonary hypoplasia are described in 58 singleton pregnancies at risk (39 with associated preterm premature rupture of the membranes at 30 weeks' gestation or less and of 12 or more days' duration, and 19 with associated anomalies). The ultrasonographic diagnosis of pulmonary hypoplasia was based on values at or below the lower limit of the 95% confidence intervals (for predicting a future observation) for normal pregnancies for each of the following tests; chest circumference in relation to gestational age and to femur length, for the chest circumference to abdominal circumference ratio and for a combination of the three measurements. At necropsy, the diagnosis of pulmonary hypoplasia was based on lung weight to birthweight ratio or radial alveolar counts. Pulmonary hypoplasia was diagnosed in 16 cases at autopsy. The clinicians and the pathologist were blinded to the ultrasonographic measurements. All tests performed well with a sensitivity of 0.55 to 0.80, specificity of 0.90 to 1.00, positive predictive value of 0.80 to 1.00, negative predictive value of 0.87 to 0.91, and an overall accuracy of 0.87 to 0.91. We conclude that ultrasound measurement of fetal chest circumference is of value with regard to the management of pregnancies at risk for lethal pulmonary hypoplasia.