(63) AUD Readmission Risk Lower With Pre-Discharge Ketamine Infusion or Naltrexone Injection

Introduction:
Alcohol use disorder (AUD) impacts 15 million Americans [1], leading to 5 million annual emergency department (ED) visits and 2 million inpatient admissions [2]. Hospitalization is a vital opportunity to initiate addiction recovery, and two single-dose, adherence-independent, pharmacologic interventions – intravenous (IV) ketamine and intramuscular (IM) naltrexone – are well-suited to administer before discharge. However, their feasibility and effectiveness at reducing re-admissions among AUD inpatients are not well established.

Methods:
Our open-label pilot trial randomized participants to 1) IM naltrexone injection, 2) IV ketamine infusion, or 3) enhanced-linkage alone. Adult (age 18-65), hospitalized participants with severe AUD and at least one past-year hospital encounter were recruited through an addiction consultation service at an urban, safety-net hospital. Exclusion criteria included psychiatric instability, severe hepatic or cardiac disease, chronic or planned opioid use, and other severe substance use disorders. The primary aim was to test feasibility, and the primary clinical outcome was 30-day, all-cause re-admission rate. Secondary clinical outcomes included 30-day emergency department (ED) visit rate and 14-day outpatient addiction clinic follow-up rate.

Results:
Participants (n=44) were mean 45.1 years-old, 20.5% female, 56.8% white/Caucasian, 34.1% Hispanic, 61.4% stably housed. All interventions were feasible, with approximately 3.8 participants recruited per month, no serious adverse events reported, and 90.9% with complete inpatient protocol adherence. For the linkage-alone (LA), naltrexone (NTX), and ketamine (KET) arms, the thirty-day readmission rates were 41.2, 21.4, and 15.4% respectively. Thirty-day emergency department rates were 70.6, 57.1, and 53.9% respectively. Fourteen-day follow up rates were 41.2, 50.0, and 61.5% respectively. Comparing the KET arm to the LA arm, the relative risk of readmission was 0.37 (p=0.17). Comparing the NTX arm to the LA arm, the relative risk of readmission 0.52 (p=0.27)

Discussion:
A single ketamine infusion or naltrexone injection given pre-discharge, combined with enhanced treatment linkage, are feasible and may reduce readmission rates among medically hospitalized patients with AUD, though the differences did not reach statistical significance. Our results are consistent with Dakwar et al. [3] who demonstrated beneficial effects with a single ketamine infusion compared to their active control, midazolam, among a lower-acuity AUD population. IM naltrexone – representing an aspirational standard of care – also demonstrated favorable outcomes compared to the LA arm. A lack of blinding may have introduced expectancy effects for the two pharmacologic interventions. Moreover, despite randomization, the linkage alone arm appeared to be more housing insecure, male, and predisposed to care utilization than the other arms. The ketamine arm had a lower baseline daily quantity of alcohol than the other arms. Next steps include further dissemination of results and planning a larger, blinded trial. If effective in recovery initiation, IV ketamine and/or IM naltrexone could represent a novel addition to the inpatient standard of care for AUD.

Conclusion:
In this randomized pilot trial among hospitalized AUD patients, both the ketamine infusion and naltrexone injection arms showed lower readmission rates and higher outpatient follow-up rates compared to the linkage alone arm, though results were not statistically significant.