Direct hemoperfusion with a polymyxin B column versus vasopressin for gram negative septic shock: a matched cohort study of the effect on survival.

Background: Vasopressin and direct hemoperfusion with an immobilized polymyxin B column (PMX) have emerged recently as treatments for septic shock. This study assessed the impact of these two treatments on the survival of patients with septic shock. Methods: A retrospective, matched cohort study was performed to compare patients who had septic shock and were treated with PMX or vasopressin. The primary endpoint was the 90-day mortality rate. 30 patients receiving PMX were matched to 30 treated with vasopressin. Results: The 90-day survival rate was significantly higher in the vasopressin group than the PMX group (83% vs. 53%; p = 0.008). In the PMX group, the subgroup with gastrointestinal surgery had a significantly higher survival rate than the subgroup without surgery (76.9% vs. 52.9%; p = 0.01). Cox multivariate analysis showed that vasopressin therapy associated with 73% risk reduction of 90-day mortality (versus PMX; HR = 0.27; p <0.01). The estimated hazard ratio for the SOFA score was 1.44 (p < 0.01), and total pressor dosage an baseline was 1.26 (p = 0.01). Conclusion: We found a significant reduction of the mortality rate in septic shock patients who received vasopressin compared with PMX. In the PMX group, the subgroup with gastrointestinal surgery had a significantly higher survival rate than the subgroup without such surgery.