Early discontinuation vs. continuation of empirical antibiotics in patients with cancer and febrile neutropenia: a systematic review and meta-analysis of randomized controlled trials.

BACKGROUND

Empirical broad-spectrum antibiotics are standard care for febrile neutropenia, but prolonged treatment until neutrophil recovery may increase antimicrobial resistance, toxicity, Clostridioides difficile infection, fungal superinfection, and healthcare burden. Whether antibiotics can be safely discontinued before neutrophil recovery remains uncertain.

METHODS

We systematically reviewed randomized controlled trials comparing early or short-course discontinuation of empirical antibacterial therapy with continuation until neutrophil recovery or an extended fixed course in patients with cancer and febrile neutropenia. MEDLINE, Embase, Cochrane CENTRAL, and ClinicalTrials.gov were searched from inception. Data were pooled using random-effects models. Risk of bias was assessed using the Cochrane RoB 2 tool. Co-primary outcomes were days of antibiotic therapy and all-cause mortality. The review was registered in PROSPERO (CRD420261434782).

RESULTS

Fifteen trials involving 1,692 participants (1,679 contributing to the mortality analysis) were included. Early discontinuation reduced antibiotic exposure by a mean of 3.43 days (95% CI - 5.59 to -1.27; I2 = 87.1%), with a larger reduction in children (-5.04 days, 95% CI - 8.55 to -1.54) than adults (-1.52 days, 95% CI - 3.49 to 0.44; subgroup p = 0.0002). All-cause mortality did not differ between groups (Peto odds ratio 1.23, 95% CI 0.64 to 2.36), and the absolute risk difference across all trials was null (0.00, 95% CI - 0.01 to 0.01; I2 = 0%). No significant differences were observed in new bloodstream infections, treatment failures, or fever recurrences.

CONCLUSION

In clinically stable, afebrile patients with cancer-related febrile neutropenia, early antibiotic discontinuation shortened antibiotic exposure, particularly in children, without a detectable increase in mortality or major infectious complications. Larger adult trials are needed to confirm safety.

SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/view/CRD420261434782, CRD420261434782.