Attributable mortality of carbapenem-resistant Klebsiella pneumoniae infections in a prospective matched cohort study in Italy, 2012-2013

A. Hoxha, T. Kärki, C. Giambi, C. Montano, A. Sisto, A. Bella, F. D'Ancona, G. A. Tura, A. Rossi, M. F. Pedna, M. Minghetti, M. Zoli, M. Chiesa, A. Carminati, P. Ceccarelli, E. Costa, M. Tejada, F. Castaldo, L. Carpinelli, M. NelliP. Sansone, T. Chiossone, N. Cenderello, P. fabbri, M. P. Crisalli, A. Poli, E. Fornai, F. Niccolini, P. Pecile, L. Magistri, E. Mantengoli, I. Bianco, A. Gambi, M. A. Pompeo, F. Silverj, A. di Girolamo, F. de Vita, A. Argentieri, S. Gatti, F. Milano, M. Franchino, C. Siciliano, P. A. Dusi, M. Dotta, M. Assensi, S. Montoro, C. Farina, M. Cosentino, F. Vailati

Research output: Contribution to journalArticlepeer-review


Background: In Italy, infections with carbapenem-resistant Klebsiella pneumoniae (CRKP) have increased markedly since 2009, creating unprecedented problems in healthcare settings and limiting treatment options for infected patients. Aim: To assess the attributable mortality due to CRKP in ten Italian hospitals and to describe the clinical characteristics of patients with an invasive CRKP and carbapenem-susceptible K. pneumoniae (CSKP) infection. Methods: We conducted a matched cohort study, and calculated crude and attributable mortality for CRKP. The attributable mortality was calculated by subtracting the crude mortality rate of the patients with CSKP from the crude mortality rate of the patients with CRKP. We also described the clinical characteristics of CRKP and CSKP patients and analysed the determinants of mortality by using conditional Poisson regression. Findings: The study included 98 patients, 49 with CRKP and 49 with CSKP. CRKP patients had undergone more invasive procedures and also tended to have more serious conditions, measured by higher Simplified Acute Physiology Score II. The attributable mortality of CRKP at 30 days was 41%. CRKP patients were three times more likely to die within 30 days [matched incidence rate ratio (mIRR): 3.0; 95% confidence interval (CI): 1.5-6.1]. Adjusting for potential confounders, the risk remained the same (adjusted mIRR: 3.0; 95% CI: 1.3-7.1). Conclusion: CRKP infection had a marked effect on patient mortality, even after adjusting for other patient characteristics. To control the spread of CRKP we recommend prioritization of control measures in hospitals where CRKP is found.

Original languageEnglish
Pages (from-to)61-66
Number of pages6
JournalJournal of Hospital Infection
Issue number1
Publication statusPublished - Jan 1 2016


  • Antibiotic resistance
  • Carbapenems
  • Klebsiella pneumoniae
  • Mortality

ASJC Scopus subject areas

  • Microbiology (medical)
  • Infectious Diseases

Fingerprint Dive into the research topics of 'Attributable mortality of carbapenem-resistant Klebsiella pneumoniae infections in a prospective matched cohort study in Italy, 2012-2013'. Together they form a unique fingerprint.

Cite this