Background: Guidance for the discontinuation of vasopressors in the recovery phase of septic shock is limited. Norepinephrine is more easily titrated; however, septic shock is a vasopressin deficient state, which exogenous vasopressin endeavors to resolve. Discontinuation of vasopressin before norepinephrine may result in clinically significant hypotension.
Methods: This retrospective, cohort study compared discontinuation of norepinephrine and vasopressin in medically, critically ill patients in the recovery phase of septic shock from May 2014 to June 2016. Difference in clinically significant hypotension after norepinephrine or vasopressin discontinuation was evaluated with χ2 test. Linear regression was performed, examining the effect of agent discontinuation on clinically significant hypotension. Baseline variables were examined for a bivariate relationship with clinically significant hypotension; those with P < .2 were included in the model.
Results: Vasopressin was discontinued first or last in 62 and 92 patients, respectively. Sequential Organ Failure Assessment scores at 72 hours (7.9 vs 7.6, P = .679) were similar. In unadjusted analysis, when vasopressin was discontinued first, more clinically significant hypotension developed (10.9% vs 67.8%, P < .001). There was no difference in intensive care unit (174 vs 216 hours, P = .178) or hospital duration (470 vs 473 hours, P = .977). In adjusted analyses, discontinuing vasopressin first was associated with increased clinically significant hypotension (odds ratio [OR]: 13.837, 95% confidence interval [CI]: 3.403-56.250, P < .001) but not in-hospital (OR: 0.659, 95% CI: 0.204-2.137, P = .488) or 28-day mortality (OR: 0.215, 95% CI: 0.037-1.246, P = .086).
Conclusion: Adult patients receiving norepinephrine and vasopressin in the resolving phase of septic shock may be less likely to develop clinically significant hypotension if vasopressin is the final vasopressor discontinued.
Keywords: hypotension; norepinephrine; septic shock; vasopressin.