Surgical site infections (SSIs) are among the most important complications of perioperative care and can culminate in in-hospital death, especially in more vulnerable patients or those undergoing more complex procedures. Understanding factors associated with mortality in SSI cases may contribute to improving prevention strategies and surgical care. This study aimed to analyze in-hospital deaths associated with SSI and identify related clinical and care variables in a 14-year retrospective cohort. Retrospective cohort formed from cumulative active surgical surveillance data from the infection control committees of two hospitals of different complexity levels, during patient hospitalization and up to 30 days after the procedure. The primary outcome was in-hospital death. Multivariable logistic regression was performed, adjusting for age, sex, hospital type, and contamination potential of the procedure. The variable “surgical specialty” was tested but excluded from the final model because it produced unstable estimates. Among 105,854 patients evaluated, 7.3% developed SSI. Presence of SSI was strongly associated with risk of in-hospital death (χ² = 330.6; p < 0.001), with OR = 2.42 (95% CI: 2.20–2.67). Age was an independent risk factor, with progressive increase per year (OR = 1.04; 95% CI: 1.04–1.04), as was male sex (OR = 1.42; 95% CI: 1.32–1.53). Surgeries with higher contamination potential were also associated with higher mortality (OR = 1.46; 95% CI: 1.35–1.57). On the other hand, patients operated on in the secondary hospital had a significantly lower risk of death (OR = 0.009; 95% CI: 0.005–0.017). Our study highlights SSI as a clinical marker of elevated risk of hospital mortality. Identification of factors associated with this outcome—such as age, sex, procedure type, and hospital complexity—can support targeted surveillance and prevention actions. Risk stratification and systematic SSI monitoring are essential to reduce mortality associated with infectious surgical complications.
Silva et al. (Sun,) studied this question.