Things We Do for No Reason: Avoiding Naltrexone for Alcohol Use Disorder in Liver Disease

Abstract

Hospitalizations related to alcohol use disorder (AUD) are common. Yet, few patients receive pharmacotherapy consistent with guideline recommendations. Previous concerns over potential hepatotoxicity of naltrexone have been disproven and recent studies have shown its safety and efficacy in patients with cirrhosis. Naltrexone is an effective therapy to reduce heavy alcohol consumption, however lack of knowledge among prescribers inhibits greater uptake. Hospitalization is an opportune time for change --naltrexone can promote reduction or cessation of unhealthy alcohol consumption, as well as subsequent readmissions or progression of alcohol-related liver disease. Hospitalists should stop avoiding naltrexone in the treatment of AUD.

Background

Pharmacologic therapy for AUD is highly effective in reducing unhealthy alcohol use. A 2023 systematic review of 118 clinical trials including 20,976 participants showed oral naltrexone (50mg/day) had a number needed to treat (NNT) of 18 for return to any drinking and a NNT of 11 for a return to heavy drinking.¹ Despite its efficacy and guideline recommendations, only 1% of people with AUD in the United States receive evidence-based medications for AUD (mAUD).² Hospitalizations for AUD-related withdrawal or complications are common and patients may otherwise not be engaged in health care. However, rates of mAUD initiation at hospital discharge remain low. A study of Medicare patients from 2015 to 2017, excluding those who were already on mAUD, found that only 0.7% of patients with AUD overall and 2.3% with a primary discharge diagnosis of AUD were prescribed mAUD on discharge.³

Three medications (naltrexone, acamprosate, and disulfiram) are currently approved by the U.S. Food and Drug Administration (FDA) to treat AUD. Naltrexone and acamprosate are considered first line therapy, while pooled analyses have shown disulfiram does not significantly reduce alcohol consumption (Table 1).¹ A meta-analysis of 4 RCTS showed no significant difference between naltrexone and acamprosate in return to any drinking or heavy drinking.¹ But, in practice, clinicians often prefer naltrexone. The pill burden is substantially less—naltrexone is once daily, while acamprosate is two pills three times a day. Naltrexone has a long-acting injectable option every 28 days, which can be beneficial in some patients. Notably, acamprosate’s evidence is largely among individuals who have had a period of sobriety before initiation, whereas naltrexone has proven benefit without a lead-in period, as well as a greater effect on craving reduction.^1,4 Furthermore, acamprosate is typically more expensive than oral naltrexone, is less widely covered by insurance, and is often not Medicaid preferred therapy or on state formularies. These issues can be further barriers to treatment.⁵

Table 1:

Summary of Findings Assessing Efficacy of Medication for AUD and Associated Contraindications (Table adopted from McPheeters et. al)¹

FDA Approved Medications for AUD
	Naltrexone PO	Naltrexone IM	Acamprosate	Disulfiram
Dosing	50mg (1 pill) daily	380mg IM injection every 28 days	666mg (2 pills) 3 times daily	250mg (1 pill) daily
Return to any drinking (RR1, 95% CI2)	0.93 (0.87-0.99)	0.96 (0.90-1.03)	0.88 (0.83-0.93)	1.03 (0.90-1.17)
Return to heavy drinking (RR, 95% CI)	0.81 (0.72-0.90)	1.00 (0.82-1.21)	0.99 (0.94-1.05)	Not reported
Evidence Strength	Moderate		Moderate	Low (No significant effect)
Relative contraindications	• Decompensated cirrhosis • Acute hepatitis		• Creatinine Clearance 30-50	• Recent use of metronidazole • Liver disease
Absolute contraindications	• Opioid use for pain • Methadone or buprenorphine for opioid use disorder treatment		• Creatinine Clearance <30	• Coronary artery disease • Heart failure

¹RR, Risk Ratio;

²Confidence Interval

Heavy drinking days: ≥5 drinks/day for men, ≥4 drinks/day for women

Why you might think you should delay initiating naltrexone

Naltrexone concerns arose following several studies published in the 1980s where its use was associated with hepatocellular injury when administered in high doses (up to 300mg/day). This prompted the FDA to issue a “black box” warning regarding the risk of hepatotoxicity when given in high doses and among people with acute hepatitis or active liver disease.⁶ This concern remains in some guidelines. The American Association for the Study of Liver Diseases (AASLD)’s 2019 Alcohol-Associated Liver Disease (ALD) Practice Guidelines only list acamprosate and baclofen under their AUD treatment recommendations in patients with liver disease.⁷

Barriers to naltrexone receipt also come from the prescriber side with misunderstandings about the medication. A survey of inpatient clinicians at an urban academic medical center showed low institutional AUD prioritization, limited knowledge of naltrexone effectiveness, concerns of side effects and liver toxicity, and a belief that naltrexone initiation should be contingent upon a patient’s commitment to complete abstinence from alcohol inhibited initiation.⁸ Among general internists and family medicine physicians who practice inpatient and outpatient, 27% believed naltrexone should only be prescribed by psychiatry or addiction specialists and half thought it is dangerous to drink simultaneously while on the medication.⁹ Two-thirds believed prescriber and patient registration with an FDA Risk Evaluation and Mitigation Strategy (REMS) program was required for naltrexone.⁹

Why withholding naltrexone is unnecessary and potentially harmful

The FDA removed the warning label for hepatoxicity with naltrexone in 2013, as there had been no known cases of hepatic failure, and minimal evidence for liver disease progression attributable to its use.¹⁰ Furthermore, in the COMBINE (Combined Pharmacotherapies and Behavioral Interventions for Alcohol Dependence) clinical trial that compared naltrexone to acamprosate, only twelve of 1,383 participants (0.9%) had significant liver enzyme elevation (defined as >5x the upper limit of normal) with naltrexone 100mg/day and none had significant liver disease progression or liver failure.¹¹

Recent data support the efficacy and safety of naltrexone among people with alcohol-associated liver disease and cirrhosis. In a retrospective cohort study of 9,131 Veterans Affairs (VA) patients with alcohol-associated cirrhosis and high-risk AUD, naltrexone or acamprosate exposure was associated with improved survival relative to no medication (hazard ratio [HR] 0.80, 95% Confidence Interval 0.67–0.97, p = 0.02). Longer exposure to either medication was associated with further survival improvement.¹²

A retrospective study of 2,940 VA patients with cirrhosis who started naltrexone found that the median AST and ALT decreased and only 2% had liver enzyme elevations at 3-months, none of which were attributed to naltrexone. Importantly, there were no episodes of liver decompensation or death linked to the medication. Notably, 411 patients (14%) in this cohort had Child Pugh B cirrhosis, with mean MELD (model for end-stage liver disease)-NA 11 and 915 patients (31%) had history of at least 1 decompensation (ascites, variceal bleed, hepatic encephalopathy). Thus, this study provides the largest evidence base for the safety of naltrexone use in patients with compensated and decompensated (Child Pugh B) cirrhosis.¹³

Finally, hospitalization for AUD complications can be a pivotal moment to address unhealthy alcohol use, and the hospitalist plays a critical role. In a retrospective cohort study of AUD-related hospitalizations, receipt of naltrexone, acamprosate, or disulfiram on discharge was associated with a 42% decreased incidence and 18% absolute reduction in 30-day all-cause mortality, rehospitalizations, or emergency department encounters.¹⁴ During inpatient stays, patients undergo managed withdrawal with detoxification and may feel motivated to change their alcohol use behaviors given heightened concerns related to the health impacts from drinking. Naltrexone initiation following detoxification and alcohol abstinence is associated with reductions in heavy drinking and continued abstinence. Thus, hospitalization is an opportune time to offer treatment.⁴

When delaying or holding naltrexone initiation might be helpful

Naltrexone is a mu-opioid receptor antagonist. An alternative mAUD should be chosen in individuals on buprenorphine or methadone maintenance therapy for opioid use disorder or opioids for chronic pain. To prevent precipitated opioid withdrawal, there should be a seven-to-ten-day opioid-free washout period before naltrexone initiation.¹⁵

Data are lacking on the safety of naltrexone in severe decompensated cirrhosis or acute alcoholic hepatitis. While the aforementioned study of naltrexone safety in cirrhosis included 915 (31%) patients with prior cirrhosis decompensation, it only included 22 individuals with Child Pugh C cirrhosis.¹³ To date, there are no studies examining naltrexone’s safety among patients with acute alcoholic hepatitis. Therefore, it is reasonable to use acamprosate in these cases.

What you should do instead

Offer and prescribe naltrexone to hospitalized patients with AUD, including those with Child Pugh A and B cirrhosis. Side effects of naltrexone are uncommon but include abdominal cramping, nausea, restlessness, and decreased appetite, which usually present transiently after initiation and self-resolve.¹⁵

An AST and/or ALT >5 times the upper limit of normal is often referenced as a naltrexone contraindication. However, this cutoff emerged from prior hepatotoxicity concerns and clinical trial inclusion criteria,¹⁵ without specific evidence. In cases of >5x elevation, non-alcohol etiologies should be investigated and naltrexone can still be cautiously provided, especially if liver enzymes and liver synthetic function are improving.

If the patient has severe decompensated cirrhosis (Child Pugh C) and/or severe acute alcoholic hepatitis, discuss acamprosate. However, if there are cost or insurance barriers, challenges with the acamprosate dosing, or past treatment failure, the benefits of starting naltrexone to reduce heavy drinking likely outweigh the potential risk of hepatotoxic effects of naltrexone. This must be discussed with the patient so they can make an informed decision.

Prescribe naltrexone with referral to an outpatient clinician for treatment continuation. Whenever possible, refer the patient to a peer support group (such as Alcoholics Anonymous), outpatient counseling, or residential treatment for AUD. Medications for AUD should never be contingent on patient engagement in these activities.¹¹

Recommendations

1. Offer naltrexone for alcohol use disorder. Prescribe at least a 30-day supply with referral to ongoing care.

2. Do not avoid naltrexone in patients with Child Pugh A and B cirrhosis or chronic liver disease.

3. Offer acamprosate or naltrexone via shared decision making in cases of severe decompensated cirrhosis (Child Pugh C) or acute alcoholic hepatitis.

Conclusion

Unfortunately, the patient in this clinical scenario returned to drinking shortly after his hospital discharge and did not follow up with his primary care doctor. He was not offered naltrexone until his next hospitalization, and with naltrexone was able to stop drinking.

Hospitalists should always offer medications for AUD before hospital discharge unless there is a clear contraindication. Naltrexone reduces the risk of heavy alcohol use and return to drinking and hospitalization is an optimal time to initiate. Naltrexone is safe and effective, even among patients with compensated and early decompensated cirrhosis.

Clinical scenario.

A 54-year-old man with alcohol use disorder (AUD) and Child Pugh B cirrhosis is hospitalized for alcohol withdrawal. Labs are significant for an aspartate transaminase (AST) of 200 IU/L and an alanine transaminase (ALT) of 120 IU/L. The hospitalist treats the alcohol withdrawal and discusses medications for alcohol use disorder (AUD) with the patient but defers initiation given the patient’s cirrhosis and transaminase abnormalities. The hospitalist recommends the patient follow-up with a primary care doctor and recheck transaminases before starting naltrexone.