Low amniotic fluid glucose levels are a specific but not a sensitive marker for subclinical intrauterine infections in patients in preterm labor with intact membranes

Abstract

Objective: The purpose of this study was to evaluate the ability of a low amniotic fluid glucose level, as defined by previously published thresholds, to identify a subclinical intrauterine infection in women in preterm labor with intact membranes.

Study design: Amniotic fluid glucose levels, amniotic fluid cultures, and placental histologic characteristics were determined in women with preterm labor (n = 55) and in preterm controls (n = 58). These amniotic fluid glucose levels were compared between women with and without intrauterine infection using the Mann-Whitney U test. The sensitivity, specificity, and positive and negative predictive values for the identification of an intrauterine infection were determined with four different previously published amniotic fluid glucose threshold levels and the threshold from receiver-operator characteristic curve analysis of the study data.

Results: The median amniotic fluid glucose level was significantly lower in patients with positive amniotic fluid cultures (intraamniotic infection group) but was not lower in the group of patients with only histologic chorioamnionitis (extra-amniotic infection group), compared with noninfected controls. The sensitivity of a low amniotic fluid glucose level to detect both types of intrauterine infection ranged from 41% to 55%, depending on the threshold used. The specificity of a low amniotic fluid glucose level to detect an intrauterine infection ranged from 94% to 100%, the positive predictive value ranged from 86% to 100%, and the negative predictive value ranged from 70% to 80% with the same thresholds.

Conclusion: Low amniotic fluid glucose levels are a specific but not a sensitive marker for either intraamniotic or extra-amniotic intrauterine infections in patients with preterm labor.

PIP: Between August, 1991, and November, 1992, in North Carolina, obstetricians performed amniocentesis on 55 women in preterm labor with intact membranes at 24-34 weeks gestation to test for a subclinical intrauterine infection. They compared the amniocentesis results with those of 58 controls at the same gestational age to determine whether low amniotic fluid glucose levels would identify intraamniotic infection or extraamniotic infection. Women with intraamniotic infection had a considerably lower median amniotic fluid glucose level than did those with extraamniotic infection (chorioamnionitis), those with preterm labor but no infection, and the controls (2 vs. 25.5, 33, and 34 mg/dl, respectively; p = .017 for chorioamnionitis group and p .001 for both groups with no infection). The researchers used 4 different threshold values to determine sensitivity, specificity, positive predictive value, and negative predictive value (= or 5 mg/dl, = or 10 mg/dl, = or 14 mg/dl, and = or 16 mg/dl). Depending on the threshold level, the sensitivity of a low amniotic fluid glucose level to detect intraamniotic infection and chorioamnionitis varied from 41 to 55%. The specificity of a low amniotic fluid glucose level to detect these infections was 94-100%. The positive and negative predictive values were 86-100% and 70-80%, respectively. These findings show that a low amniotic fluid glucose level can correctly identify patients in preterm labor who do not have intraamniotic infection or chorioamnionitis, but it does not correctly identify preterm labor patients with intrauterine or extrauterine infection.