Objective: To identify predictors of 28-day mortality among patients with refractory septic shock treated with norepinephrine with or without vasopressin.
Design: Prospective observational cohort study.
Setting: A 1,200-bed academic medical center.
Patients: One hundred thirty-seven patients with septic shock treated with norepinephrine with or without vasopressin.
Interventions: None.
Measurements and main results: The 28-day mortality rate was 37.2% (n = 51). By multivariate analysis, significant predictors of death were norepinephrine plus vasopressin administration (adjusted odds ratio [AOR], 13.96; 95% confidence interval [CI] 6.47, 30.08; p = 0.001), lack of goal-directed fluid administration during initial resuscitation (AOR 15.82; 95% CI 6.16, 40.61; p = 0.003), inappropriate initial antimicrobial therapy (AOR 8.95; 95% CI 2.93, 27.33; p = 0.05), and higher Acute Physiology and Chronic Health Evaluation (APACHE) II score (AOR 1.14; 95% CI 1.07, 1.21; p = 0.033). Patients who received norepinephrine plus vasopressin (n = 68) had a significantly higher mortality rate than patients managed with norepinephrine alone (n = 69) 28 days after the initiation of vasopressors (54.4% vs. 20.3%; p < 0.001). This finding was confirmed in patients matched optimally across treatment groups.
Conclusions: Our study found an association between the use of norepinephrine plus vasopressin and 28-day mortality in refractory septic shock. In view of its known mechanism of action, vasopressin contributed to this excess mortality. Further recommendations regarding the use of vasopressin await the results of large randomized trials evaluating its efficacy and safety for septic shock.