Participant- and Disease-Related Factors as Independent Predictors of Treatment Outcomes in the RESTORE-IMI 2 Clinical Trial: A Multivariable Regression Analysis.

Abstract

In the RESTORE-IMI 2 trial, imipenem/cilastatin/relebactam (IMI/REL) was noninferior to piperacillin/tazobactam in treating hospital-acquired bacterial pneumonia/ventilator-associated bacterial pneumonia. This post hoc analysis was conducted to determine independent predictors of efficacy outcomes in the RESTORE-IMI 2 trial, to assist in treatment decision making. A stepwise multivariable regression analysis was conducted to identify variables that were independently associated with day 28 all-cause mortality (ACM), favorable clinical response at early follow-up (EFU), and favorable microbiologic response at end of treatment (EOT). The analysis accounted for the number of baseline infecting pathogens and in vitro susceptibility to randomized treatment. Vasopressor use, renal impairment, bacteremia at baseline, and Acute Physiologic Assessment and Chronic Health Evaluation (APACHE) II scores ≥15 were associated with a greater risk of day 28 ACM. A favorable clinical response at EFU was associated with normal renal function, an APACHE II score <15, no vasopressor use, and no bacteremia at baseline. At EOT, a favorable microbiologic response was associated with IMI/REL treatment, normal renal function, no vasopressor use, nonventilated pneumonia at baseline, intensive care unit admission at randomization, monomicrobial infections at baseline, and absence of Acinetobacter calcoaceticus-baumannii complex at baseline. These factors remained significant after accounting for polymicrobial infection and in vitro susceptibility to assigned treatment. This analysis, which accounted for baseline pathogen susceptibility, validated well-recognized patient- and disease-related factors as independent predictors of clinical outcomes. These results lend further support to the noninferiority of IMI/REL to piperacillin/tazobactam and suggests that pathogen eradication may be more likely with IMI/REL. NCT02493764.