Epidemiology and predictive factors for early and late mortality in Candida bloodstream infections: A population-based surveillance in Spain

M. Puig-Asensio, B. Padilla, J. Garnacho-Montero, O. Zaragoza, J. M. Aguado, R. Zaragoza, M. Montejo, P. Muñoz, I. Ruiz-Camps, M. Cuenca-Estrella, B. Almirante, Jesús Guinea, José Ramón Pãno, Julio García, Carlos García, Jesús Fortún, Pilar Martín, Elia Gómez, Pablo Ryan, Carolina CampeloIgnacio de los Santos, Buenaventura Buendía, Beatriz Perez, Mercedes Alonso, Francisca Sanz, Paloma Merino, Fernando González, Miguel Gorgolas, Ignacio Gadea, Juan Emilio Losa, Alberto Delgado-Iribarren, Antonio Ramos, Yolanda Romero, Isabel Sánchez, Jeśus Rodriguez-Bãno, Ana Isabel Suarez, Ana Loza, Estrella Martín-Mazuelos, Ana Isabel Aller, Maite Ruiz, Carlos Ortiz, Mónica Chávez, Fernando L. Maroto, Miguel Salavert, Javier Pemán, José Blanquer, David Navarro, Vicente Abril, Concepción Gimeno, Juan José Camarena, Silvia Hernáez, Guillermo Ezpeleta, Elena Bereciartua, José L. Hernández, Rosa Ana Rivas, Rafael Ayarza, Ana Ma Planes, José Mensa, Manel Almela, Mercé Gurgui, Ferran Sánchez-Reus, Joaquin Martinez-Montauti, Montserrat Sierra, Juan Pablo Horcajada, Luisa Sorli, Juliá Gómez, Amadeu Gené, Mireia Urrea, Maricela Valerio, Mario Fernández-Ruiz, Ana Díaz-Martín, Francesc Puchades, Alessandra Mularoni

Research output: Contribution to journalArticleResearchpeer-review

217 Citations (Scopus)


A prospective, multicentre, population-based surveillance programme for Candida bloodstream infections was implemented in five metropolitan areas of Spain to determine its incidence and the prevalence of antifungal resistance, and to identify predictors of death. Between May 2010 and April 2011, Candida isolates were centralized to a reference laboratory for species identification by DNA sequencing and for susceptibility testing by EUCAST reference procedure. Prognostic factors associated with early (0-7 days) and late (8-30 days) death were analysed using logistic regression modelling. We detected 773 episodes: annual incidence of 8.1 cases/100 000 inhabitants, 0.89/1000 admissions and 1.36/10 000 patient-days. Highest incidence was found in infants younger than 1 year (96.4/100 000 inhabitants). Candida albicans was the predominant species (45.4%), followed by Candida parapsilosis (24.9%), Candida glabrata (13.4%) and Candida tropicalis (7.7%). Overall, 79% of Candida isolates were susceptible to fluconazole. Cumulative mortality at 7 and 30 days after the first episode of candidaemia was 12.8% and 30.6%, respectively. Multivariate analysis showed that therapeutic measures within the first 48 h may improve early mortality: antifungal treatment (OR 0.51, 95% CI 0.27-0.95) and central venous catheter removal (OR 0.43, 95% CI 0.21-0.87). Predictors of late death included host factors (e.g. patients' comorbid status and signs of organ dysfunction), primary source (OR 1.63, 95% CI 1.03-2.61), and severe sepsis or septic shock (OR 1.77, 95% CI 1.05-3.00). In Spain, the proportion of Candida isolates non-susceptible to fluconazole is higher than in previous reports. Early mortality may be improved with strict adherence to guidelines. © 2013 European Society of Clinical Microbiology and Infectious Diseases.
Original languageEnglish
JournalClinical Microbiology and Infection
Issue number4
Publication statusPublished - 1 Jan 2014


  • Antifungal resistance
  • Candida bloodstream infections
  • Early mortality
  • Epidemiology
  • Prognostic factors
  • Surveillance


Dive into the research topics of 'Epidemiology and predictive factors for early and late mortality in Candida bloodstream infections: A population-based surveillance in Spain'. Together they form a unique fingerprint.

Cite this