作者: S.H. Hoeboer , P.J. van der Geest , D. Nieboer , A.B.J. Groeneveld
DOI: 10.1016/J.CMI.2014.12.026
关键词:
摘要: The diagnostic use of procalcitonin for bacterial infections remains a matter debate. Most studies have used ambiguous outcome measures such as sepsis instead infection. We performed systematic review and meta-analysis to investigate the accuracy bacteraemia, proven bloodstream searched all major databases from inception June 2014 original, English language, research articles that studied between positive blood cultures in adult patients. calculated area under summary receiver-operating characteristic (SROC) curves pooled sensitivities specificities. To minimize potential heterogeneity we subgroup analyses. In total, 58 1567 eligible were included provided total 16,514 patients, whom 3420 suffered bacteraemia. overall analysis SROC curve was 0.79. optimal most widely cut-off value 0.5 ng/mL with corresponding sensitivity 76% specificity 69%. analyses lowest found immunocompromised/neutropenic patients (0.71), highest intensive-care (0.88), ranging 66 89% specificities 55 78%. spite study heterogeneity, had fair bacteraemia suspected infection or sepsis. particular low levels can be rule out presence Further is needed on safety efficacy single tool avoid taking cultures.