A meta-analysis 1 included 12 studies with a total of 1 386 subjects. The accuracy of determination of PCT and CRP levels for the diagnosis of bacterial infection was evaluated. The 12 enrolled studies included a total of 46 neonates, 638 children, and 702 adults in different areas of the hospital. Approximately one-half of the subjects were in intensive care units. The diagnostic criteria and the type of infections varied markedly among the studies.
In the statistical analysis, PCT level was found to be more sensitive 88% (95% CI 80% to 93%) vs. 75% (95% CI 62% to 84%) and more specific 81% (95% CI to 67% to 90%) vs. 67% (95% CI 56% to 77%) than CRP level for differentiating bacterial from non-infective causes of inflammation. The Q value (obtained from SROC curves) for PCT markers was higher (0.82 vs. 0.73). The sensitivity for differentiating bacterial from viral infections was also higher for PCT markers 92% (95% CI 86% to 95%) vs. 86% (95% CI 65% to 95%); the specificities were comparable 73% (95% CI 42% to 91%) vs. 70% (95% CI 19% to 96%). The Q value was higher for PCT markers (0.89 vs. 0.83). PCT markers also had a higher positive likelihood ratio and lower negative likelihood ratio than did CRP markers in both groups.
Comment: The quality of evidence is downgraded by review quality, poor reporting, inconsistency and indirectness (differences in studied patients and diagnostic criteria).
Primary/Secondary Keywords