Does moderate renal impairment affect clinical outcomes in complicated intra-abdominal and complicated urinary tract infections? Analysis of two randomized controlled trials with ceftolozane/tazobactam

Abstract

Objectives: For reasons not well understood, antibacterials can yield lower cure rates in renally impaired patients. We explored this subject for the novel antibacterial ceftolozane/tazobactam. Methods: ASPECT-complicated intra-abdominal infections (cIAIs) and ASPECT-complicated urinary tract infections (cUTIs) were randomized, double-blinded clinical trials. Analyses in moderate [creatinine clearance (CLCR) 30-50 mL/min] and mild/no (CLCR > 50 mL/min) renal impairment (RI) patients were pre-specified as exploratory endpoints in the statistical analysis plans. We also explored variables potentially impacting outcomes in these subgroups. Clinicaltrials.gov NCT01445665/NCT01445678 and NCT01345929/NCT01345955. Results: At baseline, 4.5% (36/806) of cIAI patients and 7.3% (58/795) of cUTI patients had moderate RI. Moderate RI patients were older, had more comorbid conditions and had higher APACHE-II scores. In the cIAI microbiological intent-to-treat population, response rates were 48% and 69% in moderate RI patients receiving ceftolozane/tazobactam and meropenem, respectively; among moderate RI cIAI patients considered treatment failures, indeterminate responses were more frequent with ceftolozane/tazobactam (39%; 9/23) than meropenem (8%; 1/13). In the cUTI microbiological modified intent-to-treat population, response rates were 81% and 78% in moderate RI patients receiving ceftolozane/tazobactam and levofloxacin, respectively. In both studies, response rates in moderate RI patients were similar between treatment arms in microbiologically evaluable populations, which excluded indeterminate responses due to missing data/protocol deviations (cIAI: 72.7% ceftolozane/tazobactam versus 71.4% meropenem; cUTI: 87% ceftolozane/tazobactam versus 80% levofloxacin). Conclusions: Regardless of treatment, clinical cure rates in cIAI and cUTI were lower in moderate versus mild/no RI patients. In moderate RI cIAI patients, numerical differences in response rates between treatments were attributable to imbalances in the numerical patients deemed indeterminate.