Moderation of Prazosin's Efficacy by Alcohol Withdrawal Symptoms

Abstract

Objective: Alcohol use disorder (AUD) is a leading cause of global disease burden. Chronic, heavy use increases the likelihood of alcohol withdrawal symptoms and associated secondary outcomes of alcohol craving and mood, anxiety, and sleep disturbances, which are predictive of poor treatment outcomes. The authors examined whether alcohol withdrawal symptoms moderate the efficacy of prazosin in reducing alcohol intake and associated secondary outcomes.

Methods: A 12-week, double-blind, randomized, controlled proof-of-concept trial of prazosin (16 mg/day, with a 2-week titration) was conducted in community-recruited adults with current alcohol dependence (N=100) with varying levels of alcohol withdrawal symptoms assessed at treatment entry. Primary outcomes were daily self-reported drinking days and heavy drinking days, and secondary outcomes were average drinks/day and mood, anxiety, craving, and sleep quality ratings.

Results: Modified intent-to-treat analyses indicated a significant interaction of alcohol withdrawal symptom score by treatment by full-dose treatment period (weeks 3-12) for drinking days, heavy drinking days, and average drinks/day. By week 12, participants with high alcohol withdrawal symptoms on prazosin reported 7.07% heavy drinking days and 27.46% drinking days, while those on placebo had 35.58% heavy drinking days and 58.47% drinking days (heavy drinking days: odds ratio=0.14, 95% CI=0.058, 0.333; drinking days: odds ratio=0.265, 95% CI=0.146, 0.481). No such benefit of prazosin was observed in those reporting low or no alcohol withdrawal symptoms. Individuals with high alcohol withdrawal symptoms on prazosin compared with placebo also showed significantly improved anxiety, depression, and alcohol craving over the course of the trial.

Conclusions: The findings indicate that alcohol withdrawal symptoms are a significant moderator of prazosin treatment response for alcohol use outcomes and for associated symptoms of alcohol craving, anxiety, and mood symptoms. These data support further evaluation of alcohol withdrawal symptoms as a prognostic indicator of prazosin's efficacy in the treatment of AUD.

Keywords: Alcohol Use Disorder; Alcohol Withdrawal Symptoms; Prazosin.

FIGURE 1.

CONSORT diagram for a placebo-controlled study of prazosin’s efficacy in relation to alcohol withdrawal symptoms

FIGURE 2.

Interactions of alcohol withdrawal symptoms by treatment group in a study of prazosin’s efficacy in relation to alcohol withdrawal symptoms^{a a} The graphs show significant interactions of the continuous score for alcohol withdrawal symptoms with treatment (prazosin or placebo) during the full-dose period (weeks 3–12) for percent drinking days (p<0.002) (panel A) and percent heavy drinking days (p<0.01) (panel B). For the placebo group, the higher the alcohol withdrawal symptom scores, the greater the number of drinking days (p<0.0002) and heavy drinking days (p<0.003), but for the prazosin group, number of drinking days (p<0.86) and heavy drinking days (p<0.77) did not increase with alcohol withdrawal symptom score (also shown for week 12, at end of treatment, for comparison). No differences were observed between the prazosin and placebo groups for participants with low alcohol withdrawal symptom scores. A median split of alcohol withdrawal symptom continuous score to assess the high alcohol withdrawal symptom group by treatment indicates (panel C) that in individuals with high alcohol withdrawal symptom scores, for average drinking across the full-dose period (weeks 3–12), the percentage of drinking days was 41.2% for the placebo group and 26.9% for the prazosin group (z=2.24, p<0.01; odds ratio=0.50, 95% CI=0.28, 0.92), and the percentage of heavy drinking days was 27.11% for the placebo group and 8.83% for the prazosin group (z=3.36, p<0.0004; odds ratio=0.23, 95% CI=0.1, 0.55). As shown in panel D, by week 12, individuals with high alcohol withdrawal symptom scores showed the greatest treatment effect, with percentage of drinking days at 58.47% in the placebo group and 27.46% in the prazosin group (z=4.36, p<0.00006; odds ratio=0.265, 95% CI = 0.146, 0.481), and percentage of heavy drinking days at 35.58% for the placebo group and 7.07% for the prazosin group (z=4.45, p<0.00005; odds ratio=0.14, 95% CI=0.058, 0.333). Drinking days and heavy drinking days are binary outcomes assessed daily for the overall analyses. For depiction purposes and simple-effects analyses, the percentages of drinking days and heavy drinking days were calculated for each participant in each week and then averaged across time periods. Error bars indicate standard error of the mean. (See also Figure S2 in the online supplement.) **p<0.01. ***p<0.001. ****p<0.0001.

FIGURE 3.

Treatment effects on baseline-corrected anxiety, depression, and craving ratings during the 12-week trial^{a a} Panel A illustrates a three-way interaction of treatment (prazosin orplacebo) with alcohol withdrawal symptom continuous score and week for baseline-corrected weekly depression ratings (Center for Epidemiologic Studies-Depression Scale) (p=0.007). Panel B illustrates a three-way interaction of treatment, alcohol withdrawal symptom continuous score, and week for baseline-corrected weekly anxiety ratings (Hamilton Anxiety Rating Scale) (p=0.002). Panel C illustrates a two-way interaction of treatment and alcohol withdrawal symptom continuous score on alcohol craving, measured at weeks 4, 8, and 12 (Obsessive Compulsive Drinking Scale) (p=0.003). Alcohol withdrawal symptom continuous scores are shown here in a median split for low symptom levels (Clinical Institute Withdrawal Assessment for Alcohol-Revised [CIWA-Ar] score ≤2; placebo group, N=22; prazosin group, N=34]) and high symptom levels (CIWA-Ar score ≥3; placebo group, N=23; prazosin group, N=21]) for illustration of simple effects. Error bars indicate standard error of the mean. *p<0.05.