Abstract

Objective: We aimed to find a single numeric cut-off point in order to predict respiratory distress syndrome (RDS) and transient tachypnea of the newborn (TTN), through quantification of the lamellar bodies (LBs) count.

Materials and methods: Two hundred and twenty-seven single pregnancies delivered within 48 h from amniocentesis in a III level university hospital in a period between 1996 and 2007. LBs counts were determined by amniotic fluid (AF) sampling and after centrifugation in a commercially available Coulter counter. LBs counts were performed in order to predict the risk of neonatal RDS and also evaluated against the risk of TTN by means of the most suitable receiver operator characteristic (ROC) curve in order to predict RDS and TTN.

Results: The best LBs cut-off to significantly predict TTN was ≤35,000/μL [sensitivity 86%, specificity 75%, positive predictive value (PPV) 22% and negative predictive value (NPV) at 96%]. An LBs count ≤32,000/μL was the best compromise between sensitivity and specificity in predicting RDS or TTN [sensitivity 86%, specificity 83%, PPV 37% and NPV at 97%].

Conclusions: The quantification of LBs is a reliable method, simple to execute, useful in predicting RDS, and may be useful in evaluating suspicion of TTN.