A retrospective observational cohort study of the clinical epidemiology of bloodstream infections due to carbapenem-resistant Klebsiella pneumoniae in an OXA-48 endemic setting.

This study aimed to characterize the epidemiology and clinical outcomes of patients with bloodstream infections (BSIs) due to carbapenem-resistant Klebsiella pneumoniae (CRKP) in an OXA-48-predominant environment. This was a retrospective single-centre cohort study including all consecutive patients with CRKP BSIs treated between 1 January 2014 and 31 December 2018. Multivariate analysis, subgroup analysis and propensity-score-matched analysis were employed to analyse 30-day mortality as the primary outcome. Clinical cure at day 14 was also analysed for the whole cohort. In total, 124 patients with unique isolates met all the inclusion criteria. OXA-48 was the most common type of carbapenemase (85.5%). Inappropriate therapy was significantly associated with 30-day mortality [70.6% vs 39.7%, adjusted odds ratio (aOR) 4.65, 95% confidence interval (CI) 1.50-14.40, P=0.008] and 14-day clinical failure (78.5% vs 56.2%, aOR 3.14, 95% CI 1.09-9.02, P=0.033) in multivariate analyses. Among those treated appropriately, the 30-day mortality rates were similar in monotherapy and combination therapy arms (OR 2.85, 95% CI 0.68-11.95, P=0.15). INCREMENT CPE mortality score (aOR 1.16, 95% CI 1.01-1.33, P=0.029), sepsis at BSI onset (aOR 2.90, 95% CI 1.02-8.27, P=0.046), and inappropriate therapy (aOR 4.65, 95% CI 1.50-14.40, P=0.008) were identified as independent risk factors for 30-day mortality. Colistin resistance in CRKP had no significant impact on 30-day mortality. These results were also confirmed in all propensity-score-matched analyses and sensitivity analyses. Appropriate regimens were associated with better clinical outcomes than inappropriate therapies for BSIs with CRKP predominantly possessing OXA-48.