INTERNATIONAL GUIDELINES currently recommend that patients with cirrhosis and upper gastrointestinal bleeding receive five to seven days of antibiotic prophylaxis to reduce infection and mortality risk. However, a new systematic review and meta-analysis challenges the evidence supporting this practice, raising important questions about whether prolonged antibiotic use remains necessary in modern clinical care.
The study analysed 14 randomised clinical trials including 1,322 participants, most of whom had variceal bleeding. Trial quality was assessed as low to moderate, and definitions of bacterial infection varied across studies, contributing to a degree of uncertainty in interpretation. Of the included trials, two directly compared shorter courses of antibiotics with standard longer courses, while 12 compared any prophylaxis with none.
The results demonstrated a high probability, 97.3%, that shorter durations of antibiotic prophylaxis, including no treatment, were noninferior to longer durations for all-cause mortality. Findings for secondary outcomes were more nuanced. Shorter courses had a 73.8% probability of noninferiority for early rebleeding, though the credible intervals suggested residual uncertainty. However, shorter durations were associated with an increased risk of reported bacterial infections, though this outcome was inconsistently defined and thus subject to bias.
Subgroup analysis of studies published after 2004, when significant therapeutic advances in cirrhosis and variceal bleeding management were introduced, showed an even stronger signal that shorter or absent antibiotic courses may be sufficient in preventing mortality and rebleeding. Notably, none of the included studies reported antibiotic-related adverse events, leaving an evidence gap in understanding potential harms of prolonged courses.
The findings suggest that the mortality benefit long assumed to underpin current international guidelines may be overstated. While antibiotics remain valuable in preventing infection, the optimal duration of prophylaxis is far from clear. The authors conclude that higher-quality, contemporary randomised trials are urgently needed to guide practice. In the meantime, clinicians should be aware that existing recommendations are not based on robust, up-to-date evidence.
Reference
Prosty C et al. Prophylactic antibiotics for upper gastrointestinal bleeding in patients with cirrhosis: a systematic review and bayesian meta-analysis. JAMA Intern Med. 2025;DOI:10.1001/jamainternmed.2025.3832.
