Changing pattern of organ dysfunction in early human sepsis is related to mortality

Objective: We examined the pattern of organ system dysfunction, the evoluti on of this pattern over time, and the relationship of these features to mor tality in patients who had sepsis syndrome. Design: Prospective, multicenter, observational study. Setting: Intensive care units in tertiary referral teaching hospitals. Patients: A total of 287 patients who had sepsis syndrome were prospectivel y identified in intensive care units. Materials and Measurements: Cardiovascular, pulmonary, neurologic, coagulat ion, renal, and hepatic dysfunction were assessed at onset and on day 3 of sepsis syndrome. Organ dysfunction was classified as normal, mild, moderate , severe, and extreme dysfunction. We calculated the occurrence rate and as sociated 30-day mortality rate of organ dysfunction at the onset of sepsis syndrome. We then measured the change in organ dysfunction from onset to da y 3 of sepsis syndrome and determined, for individual organ systems, the as sociated 30-day mortality rate. Results: At the onset of sepsis syndrome, clinically significant pulmonary dysfunction was the most common organ failure, but was not related to 30-da y mortality. Clinically significant cardiovascular, neurologic, coagulation , renal, and hepatic dysfunction were less common at the onset of sepsis sy ndrome but were significantly associated with the 30-day mortality rate. Wo rsening neurologic, coagulation, and renal dysfunction from onset to day 3 of sepsis syndrome were associated with significantly higher 30-day mortali ty than with improvement or no change in organ dysfunction. Conclusions: Increased mortality rate in sepsis syndrome is associated with a pattern characterized by failure of nonpulmonary organ systems and, in p articular, worsening neurologic, coagulation, and renal dysfunction over th e first 3 days. Although initial pulmonary dysfunction is common in patient s with sepsis syndrome, it is not associated with an increased mortality ra te.