Rapid Phenotypic Antimicrobial Susceptibility Testing and Rapid Molecular Identification with Resistance-Marker Detection for Bloodstream Infections: A Systematic Review with Separate Narrative Syntheses

Background: Rapid antimicrobial susceptibility testing (AST) has been developed to shorten the interval between blood-culture positivity and the availability of actionable susceptibility results in patients with bloodstream infection (BSI). Although rapid AST reduces laboratory turnaround time, its impact on antimicrobial optimization and patient-centered outcomes remains uncertain. This systematic review updated the evidence comparing rapid versus conventional AST in patients with BSI. Methods: MEDLINE (via PubMed), Europe PMC (including PubMed Central and preprints), and ClinicalTrials.gov were searched for records added between 1 October 2020, the end of the Cochrane search, and the prespecified cutoff of 6 October 2026; the documented searches were executed on 23 October 2026 with these date limits and were supplemented by a targeted PubMed search for randomized trials. Embase and CENTRAL were not searched. Randomized and non-randomized comparative studies involving patients with positive blood cultures were eligible. Primary outcomes included time to effective or appropriate therapy, antimicrobial modification, 30-day mortality, and hospital length of stay. The risk of bias in randomized trials was assessed using RoB 2, and the certainty of evidence for key outcomes was assessed using GRADE. Screening, data extraction, and risk-of-bias assessment were performed independently by two reviewers. No new meta-analysis was performed: because outcome definitions and time origins differed across trials, the evidence was synthesized narratively, and pooled estimates from the 2021 Cochrane review are reported as previously published. Results: Of 895 records screened and 27 full-text reports assessed, five new trials were eligible; in total, eleven randomized trials comprising 3689 analyzed participants were included (six carried forward from the 2021 Cochrane review and five identified by the updated search). The number of participants contributing to each outcome varied by outcome. Most trials were judged at low risk of bias or with some concerns, with one older trial judged at high overall risk because of limitations in reporting and selective outcome reporting. Median susceptibility-reporting time was shorter with rapid testing in every trial that reported it, and the median time to the trial-specific antimicrobial-modification outcome was shorter in all six contemporary trials that reported one; however, only RAPIDS-GN, Christensen 2022, and Kim 2026 clearly analyzed all randomized patients, the handling of patients without treatment modification was unclear in the Halperin and FAST trials, and the time-to-event analysis in Caspar 2024 was restricted to patients not receiving optimized therapy at baseline. These are descriptive comparisons of medians with differing outcome definitions and time origins, not pooled estimates. Contemporary phenotypic trials demonstrated substantial turnaround-time reductions; for example, in the Halperin trial, the dRAST system reduced median time from Gram stain to AST report from 25 h to 9 h. Effects on patient-centered outcomes remain uncertain: the previously published Cochrane pooled mortality estimate (RR 1.10, 95% CI 0.82–1.46) is compatible with both appreciable benefit and appreciable harm, no trial was powered for mortality, no trial showed a statistically significant difference in mortality or the composite clinical outcome, and hospital length of stay was statistically significantly shorter in only one trial (Christensen 2022: median 5 versus 7 days; p = 0.035), an exploratory secondary outcome that was not adjusted for multiplicity and should be regarded as hypothesis-generating. GRADE certainty was moderate for AST turnaround time with phenotypic platforms (four phenotypic trials; no genotypic trial contributed to this outcome) and low for antimicrobial modification, mortality, hospital length of stay, and the global composite clinical outcome. Conclusions: Rapid AST improves diagnostic speed and may facilitate earlier antimicrobial modification, but whether faster susceptibility results improve patient-centered clinical outcomes remains uncertain. The benefit may be greatest when rapid AST is integrated with real-time antimicrobial stewardship and targeted to patients at high risk of ineffective empirical therapy or multidrug-resistant infection. This hypothesis is based on trial-level observations rather than on formal subgroup analysis and requires confirmation.