Valproic acid for agitation in the intensive care unit: an observational study of psychiatric consults

Background Agitation is a common clinical problem encountered in the intensive care unit (ICU). Treatment options are based on clinical experience and sparse quality literature. Aim The aim of this study was to describe the effect of valproic acid (VPA) as adjuvant treatment for agitation in the ICU, identify predictors of response to VPA and evaluate the independent effect of VPA on agitation compared to standard of care (SOC). Method This retrospective single center observational study evaluated adult patients admitted to the ICU for whom a psychiatric consultation was requested for agitation management, with agitation defined as a Richmond Agitation Sedation Score of 2 or greater. A descriptive analysis of the proportion of agitation-free patients per day of follow-up, the incidence of agitation-related-events, as well as the evolution of co-medications use over time are presented. A logistic regression model was used to assess predictors of VPA response, defined as being agitation-free on Day 7 and generalized estimating equations were used to evaluate the independent effect of VPA as adjuvant therapy for agitation in the critically ill. Results One hundred seventy-five patients were included in the study with 78 receiving VPA. The percentage of agitation-free patients on VPA was 6.5% (5/77) on Day 1, 14.1% (11/78) on Day 3 and 39.5% (30/76) on Day 7. Multivariate regression model for clinical and demographic variables identified female gender as predictor of response on Day 7 (OR 6.10 [1.18–31.64], p = 0.03). The independent effect of VPA was non-significant when compared to SOC. Conclusion Although VPA used as adjuvant treatment was associated with a decrease in agitation, its effect when compared to SOC did not yield significant results.