Skip to main content
. 2020 Jun 12;9:141. doi: 10.1186/s13643-020-01389-4

Table 3.

Summary of findings table

Maternal inflammatory markers for chorioamnionitis in preterm prelabour rupture of membranes(PPROM): a systematic review and meta-analysis of diagnostic test accuracy studies
Question In pregnant women with PPROM, can maternal serum inflammatory markers be used to diagnose chorioamnionitis?
Population Pregnant women with PPROM
Studies Any study design where the index test is compared against the reference standard
Index test C-reactive protein (CRP), procalcitonin (PCT) and interleukin 6 (IL6) assessed in maternal serum before delivery
Reference standard Histologic chorioamnionitis (HCA) and/ or funisitis
Prevalence of disease

Median prevalence 50% (range 21–70%, IQR 38 to 57%)

23 studies with a total of 1717 pregnant women with PPROM, 902 of whom had HCA/funisitis

Quality Included studies were generally of poor quality with all studies at high risk of bias in at least one domain (QUADAS-2). There were few studies with high applicability concerns and only in the patient selection domain.
Index test Studies (participants) Sensitivity (95% CI) Specificity (95% CI) Heterogeneity Sensitivity analysis Interpretation: assuming a patient population of 100 pregnant women with PPROM and prevalence of 50%*
Correctly diagnosed cases (TP) Missed cases (FN) Unnecessary interventions (FP) True reassurance of no disease (TN)
CRP at 20 mg/L 5 (252) 59% (47.7–69.0) 83% (74.0–89.2) High heterogeneity despite common cut-off 30 21 9 42
CRP at all cut-offs 17 (1404) 59% (52.0–67.6) 80% Partially explained by nature of cut-off used, sampling interval, risk of bias in the patient selection domain and type of CRP assay Sensitive to gestational age range for study inclusion and year of publication 30 20 10 40
PCT at all cut-offs 6(231) 56% (49.9–68.9) 80% Partially explained by nature of cut-off used and risk of bias in the patient selection domain of QUADAS-2 Sensitive to applicability concerns score in the patient selection domain of QUADAS-2 28 22 10 40
IL6 at all cut-offs 5 (299) 52% (50.0–85.8) 80% Partially explained by sampling interval Sensitive to gestational age range for study 26 24 10 40

The results on this table should not be interpreted in isolation from the results in the main body of the text of the review

*Median prevalence from included studies

Estimate from the summary point from bivariate analysis

‡Sensitivity derived from HSROC analysis assuming a specificity of 80% (false positive rate of 20%)

CRP C-Reactive Protein, PCT Procalcitonin, IL6 Interleukin 6