Perioperative Blood Transfusion is Associated with an Increased Mortality in Older Surgical Patients.

Abstract

More surgical interventions are being performed on octogenarian patients. The aim of this study was to identify factors associated with operative mortality and to determine if perioperative transfusions could affect mortality outcomes in a nonselected series of octogenarian patients undergoing surgery. A descriptive cross-sectional study was performed on a population of 413 consecutive patients over 80 years old, treated surgically, and divided into two groups: transfused and nontransfused patients. The following variables were recorded: sociodemographic characteristics, main diagnoses, surgical procedure and its characteristics, ASA score, Charlson comorbidity index (CCI), National Nosocomial Infection Surveillance (NNIS) index, transfusion requirements, the Clavien-Dindo classification of surgical complications, and operative mortality. The mean age of the patients was 84.5 years (SD + 3.6). Transfused (25.2 %) and nontransfused patients had similar characteristics; except for neoplasia (P <0.001), NNIS (P = 0.008), operative mortality (P = 0.004), and complications according to Clavien-Dindo score (P <0.001). Operative mortality was 20.1 % (83 patients). The predictive variables associated with operative mortality were ASA score (P <0.001), emergency surgery (P <0.001), and blood transfusion (P = 0.004). After adjusting for the variables age, ASA class, NNIS, emergency surgery, and neoplasia, the multivariate analysis showed that the perioperative transfusion in octogenarian patients continued to be significantly associated with operative mortality (P = 0.019; OR 1.97, 95 % CI 1.12-3.47). Perioperative transfusion is an independent predictor of postoperative mortality in surgical octogenarian patients.