Objective: Our purpose was to determine whether detection of insulin-like growth factor-binding protein-1 in vaginal secretions could be used in the diagnosis of fetal membrane rupture.
Study design: Consenting patients (n = 105) with complaints suspicious of membrane rupture between 24 and 42 weeks of gestation who had no evidence of placenta previa were enrolled in the study. The diagnosis of membrane rupture required at least two of the following findings on vaginal examination: pooling of fluid, positive Nitrazine paper (Bristol-Myers Squibb, Cherry Hill, N.J.) test, or microscopic evidence of ferning. A swab of the posterior vaginal fornix was obtained, placed in sample buffer, and analyzed for insulin-like growth factor-binding protein-1 by immunoassay. Data analysis included chi 2 analysis, Student t test, or Mann-Whitney U test and linear regression and receiver operating characteristic curve analysis.
Results: A total of 78 (74.3%) patients met the criteria for membrane rupture. There was a highly significant difference in mean vaginal insulin-like growth factor-binding protein-1 concentrations between patients with and without clinical evidence of membrane rupture (553.6 +/- 731.4 micrograms/L vs 3.0 +/- 7.3 micrograms/L, p = 0.0002). Receiver operating characteristic curve analysis demonstrated that the optimal identification of patients with membrane rupture was achieved with an insulin-like growth factor-binding protein-1 value > 3 micrograms/L (sensitivity 74.4%, 95% confidence interval 64.7% to 84.0%; specificity 92.6%, 95% confidence interval 82.7% to 102.5%; positive predictive value 96.7%, 95% confidence interval 92.1% to 101.2%; negative predictive value 55.6%, 95% confidence interval 41.0% to 70.1%).
Conclusions: The presence of vaginal insulin-like growth factor-binding protein-1 is highly predictive of membrane rupture, identifying 74.4% of affected patients with a very low false-positive rate.