To examine whether surfactant protein A (SP-A) in postnatal serum can predict the development of respiratory distress syndrome (RDS), we analyzed the relationship between serum concentrations of SP-A and the risk of RDS using sera from neonates within 24 h after birth. A total of 104 blood samples including 23 samples from newborn infants with RDS were obtained. SP-A content in sera was measured with an enzyme-linked immunosorbent assay system consisting of a standard of native SP-A and two monoclonal antibodies against human SP-A. The level of serum SP-A increased with advancing gestation. Since the mean level of serum SP-A in patients with RDS (3.8 ng/ml) was significantly lower than those without RDS (12.0 ng/ml) (P<0.001), we calculated the diagnostic index values at various cutoff points and chose cutoff values to predict the risk of RDS. Maximum diagnostic value of 85% was obtained at a cutoff point of 3.8 ng/ml (sensitivity 57% and specificity 93%). We also chose a cutoff value of 2.1 ng/ml for definitive diagnosis of RDS, and 8.3 ng/ml for exclusive diagnosis of RDS. The adjusted odds ratios of RDS was significantly elevated when SP-A concentration in serum was under the cutoff values. The presence of SP-A in cord blood serum was also confirmed by immunoblotting analysis. We emphasize the value of SP-A examination in cord blood or postnatal serum from infants who exhibited respiratory difficulties at birth. We believe that our results are consistent with the hypothesis that SP-A is a useful serum marker in predicting the development of RDS.