Severe protein C deficiency predicts early death in severe sepsis

OBJECTIVE:To explore the relationship between measures of baseline disease severity and survival time in patients with severe sepsis. DATA SOURCE:A retrospective evaluation of the placebo group from a large placebo-controlled phase III clinical trial (PROWESS) comprising a total of 840 patients with severe sepsis from 164 medical centers. DATA EXTRACTION AND SYNTHESIS:Data collected included baseline demographics and disease severity measurements, baseline protein C and interleukin-6 levels, 28-day and in-hospital survival rates, and cause of death to 28 days. The survival curve for the placebo patients can be divided into three segments during which the rate of death seemed to be differentthe rapid alpha phase (day 0 to day 5), the beta phase (day 6 through day 15), and the gamma phase (day 16 to day 28). The risk of death during each phase was statistically significantly different. More patients died of refractory shock during the alpha phase than in the beta and gamma phases, whereas more patients died of respiratory failure during the beta and gamma phases than during the alpha phase. Multiple organ failure was a frequent cause of death during all phases. Protein C levels at the start of each time interval were highly predictive of outcome within that phase, with continued protein C deficiency being associated with mortality. Patients who died during either the alpha or beta phases had higher interleu-kin-6 levels at baseline than those who died later or who eventually survived. CONCLUSIONS:The rate and cause of death for patients with severe sepsis differs during the 28-day postdiagnosis period. Severe protein C deficiency (