A Matched Case-Case-Control Study of the Impact of Clinical Outcomes and Risk Factors of Patients with IMP-Type Carbapenemase-Producing Carbapenem-Resistant Enterobacteriaceae in Japan

Abstract

IMP-type carbapenemase, found in various Gram-negative bacteria, has been increasingly detected worldwide. We aimed to study the outcomes and risk factors for acquisition of IMP-type carbapenemase-producing carbapenem-resistant Enterobacteriaceae (IMP-CRE), as this has not been evaluated in detail. We conducted a matched case-case-control study of patients from whom IMP-CRE isolates were obtained. All patients who tested positive for IMP-CRE were included and matched with those having carbapenem-susceptible Enterobacteriaceae (CSE) and with controls at a ratio of 1:1:2. The risk factors for acquisition between CRE and CSE groups, and mortality which were calculated using multivariate logistic regression model with weighting according to the inverse probability of propensity scores were compared. In total, 192 patients (96 CRE and CSE each; 130 Enterobacter cloacae and 62 Klebsiella spp.) were included. IMP-11 type was present in 43 patients, IMP-1 in 33, and IMP-60 and IMP-66 in 1; 31 (32.3%) patients with CRE and 34 (35.4%) with CSE developed infections. Multivariate analysis identified the following independent risk factors: gastrostomy, history of intravenous therapy or hemodialysis, and previous exposure to broad spectrum β-lactam antibiotics, including penicillin with β-lactamase inhibitors, cephalosporins, and carbapenems. On propensity score-adjusted analysis, mortality in the CRE and CSE groups (15.0% versus 19.5%, respectively) were similar. We found that IMP-CRE may not contribute to worsened clinical outcomes compared to CSE, and gastrostomy, previous intravenous therapy, hemodialysis, and broad-spectrum antimicrobial exposure were identified as risk factors for CRE isolation. Fluoroquinolone and aminoglycosides are potentially useful antibiotics for IMP-CRE infections.