Association of Follow-up Blood Cultures With Mortality in Patients With Gram-Negative Bloodstream Infections: A Systematic Review and Meta-analysis

IMPORTANCE: Obtaining follow-up blood cultures (FUBCs) in patients with Staphylococcus aureus bloodstream infection (BSI) is standard practice, although its utility in patients with gram-negative bacterial BSI (GN-BSI) is unclear. OBJECTIVE: To examine whether obtaining FUBCs is associated with decreased mortality (key question [KQ] 1) and whether positive vs negative FUBCs are associated with increased mortality (KQ2). DATA SOURCES: MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, Web of Science, and gray literature were searched from inception to March 11, 2022. STUDY SELECTION: Two investigators used predefined eligibility criteria to independently screen titles, abstracts, and relevant full texts. Randomized clinical trials or observational studies that matched or statistically adjusted for differences in, at minimum, level of acute illness between patients in the intervention (eg, FUBCs obtained) and control (eg, FUBCs not obtained) groups were included in primary analyses. Articles published in languages other than English were excluded. DATA EXTRACTION AND SYNTHESIS: Data abstraction and quality assessments were performed by one investigator and verified by a second investigator. Risk of bias was assessed with the Newcastle-Ottawa Scale. Effect sizes were pooled using random-effects models. The study followed the Meta-analysis of Observational Studies in Epidemiology (MOOSE) reporting guideline. MAIN OUTCOMES AND MEASURES: Mortality before hospital discharge or up to 30 days from the index blood culture. RESULTS: From 3495 studies, 15 were included (all nonrandomized). In the 5 studies (n = 4378 patients) that met criteria for the KQ1 primary analysis, obtaining FUBCs was associated with decreased mortality (hazard ratio, 0.56; 95% CI, 0.45-0.71). For KQ2, 2 studies met criteria for the primary analysis (ie, matched or statistically adjusted for differences in patients with positive vs negative FUBCs), so an exploratory meta-analysis of all 9 studies that investigated KQ2 (n = 3243 patients) was performed. Positive FUBCs were associated with increased mortality relative to negative blood cultures (odds ratio, 2.27; 95% CI, 1.54-3.34). Limitations of the literature included a lack of randomized studies and few patient subgroup analyses. CONCLUSIONS AND RELEVANCE: In this systematic review and meta-analysis, obtaining FUBCs in patients with GN-BSI was associated with decreased mortality. The benefit of FUBCs may stem from identification of patients with positive FUBCs, which was a poor prognostic marker.