Sensitivity and specificity of the diagnostic criteria for intrauterine inflammation or infection proposed by the 2015 National Institute of Child Health and Human Development workshop

To investigate the test characteristics of the diagnostic criteria for intrauterine inflammation or infection or both (triple I) in a cohort of febrile intrapartum women. This was a retrospective cohort study of women ≥ 24 weeks gestation with a temperature ≥ 100.4°F during labor or within one hour postpartum, all of whom had blood cultures sent at a single tertiary hospital in 2016. Women with fetal demise, expectantly managed preterm premature rupture of membranes, or non-obstetrical infectious diagnosis were excluded. We classified women as either having an “isolated fever” or “suspected triple I” as outlined by the proposed criteria. Suspected triple I was defined as the presence of fever ≥ 102.2°F or sustained fever ≥ 100.4°F with either leukocytosis > 15,000 per mm3, fetal tachycardia, or purulent amniotic fluid. Isolated fever included women with single fever ≥ 100.4°F but <102.2°F, and women with sustained fever ≥ 100.4°F but no other maternal or fetal signs of infection. We assessed the sensitivity and specificity of clinical criteria for triple I using either a pathological diagnosis of chorioamnionitis or adverse infectious outcome with or without pathological diagnosis as the gold standard diagnosis. 297 women met inclusion criteria, 201 in the isolated fever group and 96 in the suspected triple I group. Baseline demographic and obstetric characteristics were similar between groups. Rates of pathologic diagnosis of chorioamnionitis (47.1% v. 55.6%; p=0.21) and adverse infectious outcome (14.9% v. 15.6%; p=0.88) were similar between the isolated fever and suspected triple I groups, respectively. The sensitivity and specificity of the suspected triple I diagnostic criteria to predict pathologic chorioamnionitis were 36.0% (95% CI 27.6%-45.1%) and 71.4% (95% CI 62.7%-79.1%), respectively, and to predict adverse infectious outcome were 33.3% (95% CI 20.0%-49.0%) and 67.9% (95% CI 61.7%-73.6%), respectively. The proposed criteria to confirm intraamniotic infection or predict adverse infectious outcomes are neither sensitive nor specific in this cohort. Improving the diagnosis of intraamniotic infection could optimize intrapartum management of fever and minimize unnecessary use of antibiotics in mothers and their neonates.