This post-hoc Bayesian analysis examined whether 7-day antibiotic treatment is non-inferior to 14-day treatment for mortality specifically in patients with *Pseudomonas aeruginosa* bacteremia, compared to other Gram-negative bacteremias.
The shorter (7-day) antibiotic course may behave differently in *P. aeruginosa* bacteremia compared to other Gram-negative bacteremias, suggesting the non-inferiority of 7-day treatment demonstrated broadly may not apply equally to this pathogen.
Post-hoc analysis limits causal inference; Bayesian re-analysis of existing trial data means the *P. aeruginosa* subgroup may be underpowered; findings are hypothesis-generating and not from a dedicated prospective trial.
Clinicians should be cautious about routinely shortening antibiotic courses to 7 days for uncomplicated *Pseudomonas aeruginosa* bacteremia until dedicated prospective trials confirm safety. The authors call for future trials focused specifically on this pathogen.