ObjectiveTo investigate whether common infection foci (pulmonary, intra-abdominal and primary bacteraemia) are associated with variations in mortality risk in patients with sepsis.DesignProspective, observational cohort study.SettingThree surgical intensive care units (ICUs) at a university medical centre.ParticipantsA total of 327 adult Caucasian patients with sepsis originating from pulmonary, intra-abdominal and primary bacteraemia participated in this study.Primary and secondary outcome measuresThe patients were followed for 90 days and mortality risk was recorded as the primary outcome variable. To monitor organ failure, sepsis-related organ failure assessment (Sequential Organ Failure Assessment, SOFA) scores were evaluated at the onset of sepsis and throughout the observational period as secondary outcome variables.ResultsA total of 327 critically ill patients with sepsis were enrolled in this study. Kaplan-Meier survival analysis showed that the 90-day mortality risk was significantly higher among patients with primary bacteraemia than among those with pulmonary and intra-abdominal foci (58%, 35% and 32%, respectively; p=0.0208). To exclude the effects of several baseline variables, we performed multivariate Cox regression analysis. Primary bacteraemia remained a significant covariate for mortality in the multivariate analysis (HR 2.10; 95% CI 1.14 to 3.86; p=0.0166). During their stay in the ICU, the patients with primary bacteraemia presented significantly higher SOFA scores than those of the patients with pulmonary and intra-abdominal infection foci (8.5±4.7, 7.3±3.4 and 5.8±3.5, respectively). Patients with primary bacteraemia presented higher SOFA-renal score compared with the patients with other infection foci (1.6±1.4, 0.8±1.1 and 0.7±1.0, respectively); the patients with primary bacteraemia required significantly more renal replacement therapy than the patients in the other groups (29%, 11% and 12%, respectively).ConclusionsThese results indicate that patients with sepsis with primary bacteraemia present a higher mortality risk compared with patients with sepsis of pulmonary or intra-abdominal origins. These results should be assessed in patients with sepsis in larger, independent cohorts.

No abstract