Main content area

High delayed mortality after the first episode of Clostridium difficile infection

Cózar, Alberto, Ramos-Martínez, Antonio, Merino, Esperanza, Martínez -García, Cristina, Shaw, Evelyn, Marrodán, Teresa, Calbo, Esther, Bereciartúa, Elena, Sánchez- Muñoz, Luis A., Salavert, Miguel, Pérez-Rodríguez, M. Teresa, García, Dácil, Bravo-Ferrer, J. María, Gálvez-Acebal, Juan, Henríquez, César, Cuquet, Jordi, Gil-Campesino, Helena, Torres, Luis, Sánchez-Porto, Antonio, Royuela, Ana, Cobo, Javier, Romero, José, Muriel, Alfonso, Giner, Livia, Federico, Ferrere, Rocío, Martínez-Ruiz S., Martos, Purificación, Sardiña, Cristina, Elena, Aguirre, Badía, Cristina, Perales, Alfonso, De Santos-Castro, Pedro A., Miguel, AngelBratos-Pérez, Cuellar, Sandra, Eva, Gonzalez, llinares, Pedro, Laura, Castelo, Morales, Isabel, Delgado-Iribarren, Alberto, Mairal, Pilar
Anaerobe 2019 v.57 pp. 93-98
Clostridium difficile, comorbidity, diabetes mellitus, heart failure, immunosuppression, models, mortality, neoplasms, patients, prediction, regression analysis, renal failure
Clostridium difficile infection (CDI) is characterized by a high delayed and unrelated mortality. Predicting delayed mortality in CDI patients could allow the implementation of interventions that could reduce these events. A prospective multicentric study was carried out to investigate prognostic factors associated with mortality. It was based on a cohort (July 2015 to February 2016) of 295 patients presenting with CDI. Logistic regression was used and the model was calibrated using the Hosmer-Lemeshow test.The mortality rate at 75 days in our series was 18%. Age (>65 years), comorbidity (defined by heart failure, diabetes mellitus with any organ lesion, renal failure, active neoplasia or immunosuppression) and fecal incontinence at clinical presentation were associated with delayed (75-day) mortality. When present, each of the aforementioned variables added one point to the score. Mortalities with 0, 1, 2 and 3 points were 0%, 9.4%, 18.5% and 38.2%, respectively. The area under the ROC curve was 0.743, and the Hosmer-Lemeshow goodness-of-fit test p value was 0.875. Therefore, the prediction of high delayed mortality in CDI patients by our scoring system could promote measures for increasing survival in suitable cases.