Evaluation of risk factors associated with phenobarbital dose escalation for the treatment of alcohol withdrawal in the critically ill

Aims While there are risk stratification tools for alcohol withdrawal syndrome (AWS), there are no well-established risk factors associated with phenobarbital escalation; the primary aim was to identify risk factors associated with needing phenobarbital dose escalation for AWS management. Methods This single-center, retrospective, cohort study evaluated adult patients with diagnosed AUD admitted to a surgical or medical intensive care unit of a large academic medical center within 24 hours of the initial phenobarbital dose for AWS management if admitted between 1 December 2021 and 1 December 2024. Results This study included 169 patients, 54 (32.0%) required dose escalation. The univariable analysis demonstrated that patients in the dose escalation group had a significantly higher percentage of preceding Clinical Institute Withdrawal Assessment of Alcohol Scale, Revised scores (CIWA-Ar) scores ≥15, required more concomitant continuous infusion sedation, higher body mass indexes, and shorter times between phenobarbital administration and detectable blood alcohol levels, mechanical ventilation, and hospital admission. In a final multivariable logistic regression model, CIWA-Ar score ≥ 15 (OR: 2.41, (95% CI: 1.25–4.69; P = 0.008) was the only variable helping predict phenobarbital dose escalation. Conclusion Based on our analysis, elevated CIWA-Ar scores may be the best predictor of phenobarbital dose escalation to treat severe AWS in the critically ill. These findings reinforce the importance of early intervention and dynamic clinical reassessment of phenobarbital dosing in the management of patients with AWS. Further prospective clinical trials are needed to assess risk factors in patients that require escalated treatment for AWS.