Validation of risk factors for surgical site infections and evaluation of hospital-acquired infections in other sites in patients undergoing cardiac surgery at Santa Casa de Marília: five years analysis

Abstract

Patients undergoing cardiac surgery are at risk of hospital infections, mainly the surgical site infections (SSI) which have important impact on morbidity and mortality. Although the incidence of mediastinitis is relatively low, ranging from 0.4 to 5%, the associated mortality is high, ranging from 14 to 47% and it results in longer length of stay and higher hospital costs besides of great human suffering. Several risk factors have been associated with the occurrence of mediastinitis, such as body mass index, diabetes, COPD, duration of extracorporeal circulation, blood glucose. In an attempt to identify which variables can predict more accurately the occurrence of surgical site infections, several predictive models have been proposed, with variable accuracy. Moreover, despite of ISSI be the most important, several studies have called attention to the other hospital infections which these patients are exposed. Published data suggest that about 27% of these patients develop nosocomial infections other than SSI, such as pneumonia, urinary tract infection or sepsis, which also impact the outcomes, contributing to increase the length of stay and higher mortality rates. Objective: To evaluate the risk factors associated with development of SSI in patients undergoing cardiac surgery at Santa Casa de Marilia in the last five years; to test the validity of predictive models from the American Society of Surgeons of Torax and the Brazilian model described by Magedanz in this population; evaluate the occurrence of non-surgical nosocomial infection in patients undergoing cardiac surgery, and the impact on mortality in hospital stay. Method: a retrospective cohort study will be done. Patients will be identified from the records of active surveillance and infection notification carried out by Infection control committee during five years. Additional information will be obtained from the medical record and laboratory. The various risk factors of patients with surgical site infection or the other sites of infection will be compared with those patients without nosocomial infection. The variables identified will be tested on predictive models. (AU)