Predictors of outcome in ICU patients with septic shock caused by Klebsiella pneumoniae carbapenemase–producing K. pneumoniae

Abstract

The aim of this study was to identify factors associated with mortality in intensive care unit patients with Klebsiella pneumoniae carbapenemase–producing K. pneumoniae (KPC-Kp) septic shock. A retrospective analysis of intensive care unit patients with KPC-Kp infection and septic shock observed in a large teaching hospital from November 2010 to December 2014 was performed. A total of 111 patients were included in the study. The most frequent source of infection was unknown-focus bacteraemia in 53 patients (47.7%). The rate of resistance to colistin was 51.3%; 30-day mortality was reported for 44 patients (39.6%). Surviving patients were more frequently treated with an initial therapy (within 24 hours) including two or more antibiotics displaying in vitro activity against the isolated KPC-Kp strain (41.8 vs. 18.1%, p 0.01) and were also more likely to receive a definitive therapy including two or more in vitro active antibiotics (85.1 vs. 15.9%, p <0.001). Cox regression analysis revealed that a colistin-containing antibiotic regimen (hazard ratio (HR) 0.21, confidence interval (CI) 95% 0.05–0.72, p <0.001), use of two or more in vitro active antibiotics as definite therapy (HR 0.08, CI 95% 0.02–0.21, p <0.001) and control of removable source of infection (HR 0.14, CI 95% 0.04–0.25, p <0.001) were associated with favourable outcome; colistin resistance (HR 8.09, CI 95% 3.14–11.23, p 0.001) and intra-abdominal source of infection (HR 2.92, CI 95% 2.11–4.12, p 0.002) were associated with death. In conclusion, use of a definitive therapy with at least two antibiotics displaying in vitro activity against the KPC-Kp isolates was the most important determinant of favourable outcome, whilst isolation of colistin-resistant strains was associated with death in septic patients with KPC-Kp infection.