Surgical infections with enterococcus: outcome in patients treated with ertapenem versus piperacillin-tazobactam

The pathogenicity of Enterococcus in polymicrobial surgical infections is controversial. The objective of this analysis was two-fold. The impact of Enterococcus on clinical outcome was assessed in adults with complicated intra-abdominal (IAI), complicated skin and skin structure (CSSSI), or acute pelvic (PI) infection treated with ertapenem or piperacillin-tazobactam, which is more active in vitro against enterococci than ertapenem. Baseline characteristics were identified that were associated with Enterococcus infection and with treatment failure.

This analysis included 1,558 patients treated in three randomized, triple-blind studies. Of these patients, 223 had Enterococcus in initial cultures: 125 of 623 (20%) with IAI, 28 of 529 (5%) with CSSSI, and 70 of 406 (17%) with PI. Logistic regression models were fit to assess each objective.

The cure rates for the two treatment groups were similar in each of the three studies, regardless of the presence or absence of Enterococcus. Cure rates for both treatment groups combined were significantly lower in patients with Enterococcus than without Enterococcus for IAI (76% [69/91] versus 87% [264/305], OR 2.3 [95% CI, 1.2-4.1], P = 0.009) and CSSSI (58% [11/19] versus 84% [241/287], OR 3.8 [95% CI, 1.5-10.0], P = 0.010); but for PI, rates were similar (96% [50/52] versus 92% [188/205], OR 0.4 [95% CI, 0.1-1.9], P = 0.220). Characteristics predictive of the presence of Enterococcus were Pseudomonas aeruginosa as a baseline pathogen for IAI, older age, and the presence of a complicating underlying disease for CSSSI, and infection severity rated moderate rather than severe for PI. The strongest predictors of treatment failure were >2 days postoperative infection at study entry for patients with IAI and older age for patients with CSSSI.

Choice of antimicrobial therapy did not affect cure rates in patients with or without Enterococcus. The strongest predictors of failure were postoperative infection at study entry in patients with IAI and older age in patients with CSSSI.