REVIEW OF: Heikkinen M et al, Addiction 2021;116(8):1990–1998

STUDY TYPE: Cohort study

In the US, only 9% of patients diagnosed with alcohol use disorder (AUD) are prescribed AUD medication (Kranzler HR and Soyka M, JAMA 2018;320(8):815–824). A Swedish study provides evidence that further illuminates the potential harms from this lack of treatment. The study compares the real-world impact of four medications, three of which—disulfiram, naltrexone, and acamprosate—are FDA approved for AUD.

Researchers used national databases in Sweden to identify 125,556 people (62.5% men) ages 16–64 with a diagnosis of AUD and followed them prospectively for 10 years. They excluded individuals with schizophrenia or bipolar disorder. The researchers recorded whether participants picked up prescriptions for naltrexone, disulfiram, acamprosate, or nalmefene (an opioid antagonist similar to naltrexone that is not available in the US). The primary outcome researchers tracked was hospitalization due to AUD.

Data suggested that naltrexone reduced rates of hospitalization, both as a monotherapy and when combined with disulfiram or acamprosate. For hospitalizations due to alcohol-related causes, naltrexone monotherapy decreased rates by 11%, naltrexone + acamprosate decreased rates by 26%, and naltrexone + disulfiram decreased rates by 24%. The results were nearly identical for all-cause hospitalization.

At first glance, it may seem that naltrexone in combination with another medication is the best option. But the authors point out that even though combinations outperformed naltrexone monotherapy, patients who receive combination medication may also be more likely to be receiving specialist case, and that the lower hospitalization rates might result from the specialist care rather than the medications themselves.

Surprisingly, acamprosate was associated with increased risk of hospitalization. The authors speculate that acamprosate’s poor efficacy may partially be explained by the fact that it requires dosing three times daily, and as a result we can surmise that adherence was poorer in the acamprosate arm than for other medications. But it’s not clear why this would be associated with worse outcomes than no medication at all. Disulfiram and nalmefene monotherapy were not associated with significantly different rates of hospitalization.

Finally, researchers found that benzodiazepines were associated with both an 18% increased hospitalization rate due to AUD and an 11% higher mortality rate. While this result is not unexpected, it does emphasize the dangers of prescribing benzos to patients who are actively drinking or diagnosed with AUD.

CATR’S Take
While this study does not investigate the nuances of treatment planning, it does support that naltrexone, both alone and in combination with disulfiram or acamprosate, is one of the most promising medications for AUD. It also emphasizes that we should exercise great caution when prescribing benzodiazepines for these patients.