Gender-based differences in outcome in patients with sepsis

Hypothesis: Among factors postulated to affect outcome in sepsis is the gen der of the patient, with a suggestion that females may have lower mortality . This study tested the hypothesis that female patients admitted to the sur gical intensive care unit with a documented infection have a lower mortalit y rate. Design: Retrospective analysis of a prospectively collected data set. Setting: Surgical intensive care unit of a university hospital medical cent er. Methods: Analysis of a consecutive series of 1348 patients who had signs of systemic inflammatory response syndrome on admission to a surgical intensi ve care unit. A cohort of 443 patients (32.9%) admitted with documented inf ection-and who therefore had sepsis, severe sepsis, or septic shock-constit uted the study population. For each patient, APACHE (Acute Physiology and C hronic Health Evaluation) II and III scores, systemic inflammatory response syndrome score, gender, age, and hospital mortality were recorded. chi(2) With Fisher exact test was performed to compare mortality rates between mal es and females. Univariate analysis of variance tvas used to compare contin uous variables in discrete populations. Multivariate analysis of variance w as used to determine which factors independently predicted mortality. Primary Outcome Measures: Mortality, intensive care unit length of stay, ho spital length of stay, and maximal multiple organ dysfunction score. Outcom es stratified by gender. Results: Patients had mean +/- SEM age of 67 +/- 1 years; mean +/- SEM APAC HE II and III scores of 20.1 +/- 0.4 and 67.7 +/- 1.0 points, respectively. There were no demographic differences between genders. Overall, 104 (23.5% ) of 443 patients with sepsis died. The difference in mortality rates betwe en female and male patients was not significant, except in octogenarians (P = .05). Multivariate analysis of variance, APACHE III (P<.001), maximal mu ltiple organ dysfunction score (P<.001), and female gender (P = .02) predic ted mortality. In females, APACHE III (P = .03) and maximal multiple organ dysfunction score (P<.001) predicted mortality, but age did not. Conclusion: Female gender is an independent predictor of increased mortalit y in critically ill surgical patients with documented infection.