BACKGROUND Respiratory distress syndrome of the newborn caused by the fetal lung immaturity is a very serious clinical problem. Different tests of prenatal analysis of amniotic fluid, such as lamellar body count and Clements' test, are available for predicting the fetal lung maturity. METHODS A prospective clinical study was conducted on amniotic fluid samples from 2005 to 2006. The amniotic fluid samples were obtained at the gestational age of 30 to 42 weeks and collected by vaginal amniotomy, amniotomy during Caesarean section and 72 hours before the delivery by amniocentesis. A haematology analyzer (Nikon - Kohden) was used to determine the lamellar body counts. Clements' test involved adding an equal volume of 96% ethanol to the multiple amniotic fluid volume (1:2, 1:4, 1:16. 1:32), followed by shaking and noting the presence of ring of bubbles. After the delivery, we compared the lamellar body count results and Clements' test and the outcome of pregnancies, primarily the development of respiratory distress syndrome. The most specific lamellar body cutoffs for maturity and immaturity were determined according to receiver operating characteristic curves. CONCLUSIONS Out of 232 amniotic fluid samples which were tested, 112 samples were collected after vaginal amniotomy, 88 during the Caesarean delivery and 32 samples by amniocentesis. The overall incidence of respiratory distress syndrome was 14.6%. Receiver operating characteristic curves were used to identify cutoff points for the test. We found that both tests are good screening tests for predicting the fetal lung maturity with the area under the curve of 0.782 in Clements' test and 0.751 in the lamellar body count. Clements' cutoff 2 with sensitivity of 67.6% and specificity of 72.2%, proved best in the prediction of the fetal lung maturity. The lamellar body count cutoff of 42 x 10(3)/microl had the sensitivity of 82.4% and specificity of 64.6% in predicting the fetal lung maturity. CONCLUSIONS Although both tests are good in predicting the fetal lung maturity, the lamellar body count has more advantages, because it is not only more objective, but also inexpensive, easy and fast to do, requires a small sample volume and is universally available.