[Saliva, serum estriol, randomized urinary estriol/creatine in prenatal monitoring]

Saliva, randomized urine and blood samples were collected simultaneously from 461 cases of late pregnancies, and saliva, serum estriol(SaE3, SeE3) and randomized urinary estriol/creatinine (RUE3/C) were measured by radioimmunoassay and biochemical method. The coefficient of correlation among SaE3,SeE3 and RUE3/C level was close in either normal or high-risk pregnancies (r = 0.4990-0.7266, P < 0.01). The coefficient of correlation between SaE3 and SeE3 was closest in normal and high-risk pregnancies (r = 0.7266 and r = 0.6901, respectively). The deviation of false negative rate(FN) among these three indexes was not significant. However, the false positive rate (FP) of SaE3 and SeE3 in antenatal monitoring was lower significantly than that of RUE3/C, 91.1% FP RUE3/C cases were corrected by SaE3 or SeE3 test. The correction rate of RUE3/C test (84.6%) in prenatal monitoring was notably lower than that of SaE3 and SeE3 tests (95.7% and 95.4%, respectively, p < 0.01). The results suggest that SaE3 is more valuable than SeE3 fetal monitoring. Combined RUE3/C and SaE3 should replace combined RUE3/C and SeE3 in prenatal monitoring.