Placental thickness at mid-pregnancy as a predictor of Hb Bart's disease.

The measurement of placental thickness can effectively differentiate normal pregnancies from affected pregnancies requiring invasive work-up. The objective was to evaluate the efficacy of placental thickness at mid-pregnancy in predicting fetal Hb Bart's disease in pregnancies at risk. Among 17 254 pregnant women screened for severe thalassaemia between June 1994 and December 1998, 345 pregnancies at risk for having a fetus with Hb Bart's disease underwent ultrasound examinations and cordocentesis at 18-21 gestational weeks. Before cordocentesis, the placental thickness was measured and recorded. The definite fetal diagnosis was performed with high performance liquid chromatography. The efficacy of placental thickness in predicting Hb Bart's disease was evaluated by sensitivity and specificity. Various cut-off values of the placental thickness were used for calculation and the best cut-off value was determined by a receiver-operating characteristic (ROC) curve. Of 345 pregnancies at risk, 70 fetuses with Hb Bart's disease were finally diagnosed. The mean placental thickness (+/-SD) of the normal pregnancies and pregnancies with Hb Bart's fetuses were significantly different, 24.6+/-5.2 mm and 34. 5+/-6.7 mm, respectively (Student's t-test, p<0.001). The sensitivity and specificity of placental thickness in prediction were calculated for various cut-off values. Based on the ROC curve, the best cut-off value was 30 mm (>30 mm considered abnormal), giving a sensitivity of 88.57 per cent, specificity of 90.18 per cent, positive-predictive value of 78.48 per cent and negative-predictive value of 96.87 per cent. For couples at risk, when sonographic placental thickness is normal, the risk of having an Hb Bart's fetus is markedly decreased. The measurement of placental thickness can effectively, though not absolutely, differentiate the normal pregnancies from affected ones requiring invasive work-up.