Why Naltrexone Matters
Naltrexone is the only first-line medication for a substance use disorder that is not itself a controlled substance, not an opioid, and carries no abuse or diversion potential. You can prescribe it on a plain prescription pad. There is no clinic requirement, no waiver, no urine-drug-screen gatekeeping, no risk of the patient selling it, and nothing to taper if they stop. For a large swath of patients and clinicians, that alone makes it the path of least resistance into medication-assisted treatment.
It has two solid FDA indications: alcohol use disorder (AUD) and opioid use disorder (OUD). In alcohol, it is the single best-evidenced pharmacotherapy we have, the "big winner" of the NIAAA-funded COMBINE trial, and it works by blunting the endogenous-opioid reward cascade that alcohol triggers, so the first drink is less rewarding and the slide into a heavy-drinking episode is less likely. In opioid use disorder it is the antagonist option: a full receptor blockade that, once on board, makes ordinary doses of opioids simply not work.
Its virtues and its central hazard are the same fact turned two ways. Because it is a pure mu-opioid antagonist, it asks nothing of the DEA and nothing of a methadone clinic, but it also demands that the patient be opioid-free before the first dose, or it will rip them into precipitated withdrawal. Get that one thing right and naltrexone is one of the safest, simplest tools in the addiction pharmacopeia. Get it wrong and you have a miserable, treatment-ending experience on your hands.
Naltrexone is safe, simple, and effective when you confirm the patient is opioid-free before starting, respect the hepatotoxicity warning with a baseline LFT and a recheck when the picture calls for one, plan ahead for pain and surgery, and counsel every patient that stopping the drug returns them to a state of dangerously low opioid tolerance. Do those things and you have access to the least encumbered medication in all of addiction treatment.
How to Use Naltrexone With Confidence
Naltrexone is prescribed less than the evidence warrants, not because clinicians decided it was wrong, but because a handful of specific fears make it feel riskier than it is. If any of these is why you reach for something else, this section is for you.
Unlike a narrow-therapeutic-window drug that needs perpetual monitoring, naltrexone's risk is almost entirely front-loaded into one decision at initiation. Handle that decision with a system, and the rest of the treatment is low-maintenance.
Fear #1: "I'm afraid I'll precipitate withdrawal."
This is the big one, and it is entirely preventable. Naltrexone throws an opioid-dependent patient into acute, forced withdrawal within minutes, far worse than natural withdrawal, and a guaranteed way to lose a patient from treatment. But it only happens if there is opioid still occupying the receptors. The entire prevention strategy is a clean opioid-free interval before the first dose.
The rule:
- Short-acting opioids (heroin, oxycodone, hydrocodone, fentanyl by history): at least 7 days opioid-free.
- Long-acting opioids (methadone, extended-release formulations, buprenorphine): 10 to 14 days opioid-free (methadone in particular needs the longer end).
- When in doubt, wait longer. There is no penalty for an extra day off opioids; there is a large penalty for a day too few.
When you're not sure the patient is truly clean, you have two tools:
- A urine drug screen, but remember it does not detect every opioid (fentanyl and some synthetics can be missed on standard panels), so a negative screen is reassuring, not definitive.
- A naloxone challenge. Naloxone is a short-acting antagonist; if you give it and no withdrawal appears, the patient is very likely opioid-free and safe for naltrexone. If withdrawal appears, you've triggered a short, self-limited (minutes-to-an-hour) reaction rather than the prolonged misery of a full naltrexone dose. Historically this was done with a small SC/IV naloxone dose under observation; a practical modern equivalent is to start with a low oral naltrexone test dose (e.g., 25 mg) and observe, rather than committing to the monthly Vivitrol injection you cannot take back.
Never start naltrexone, and especially never inject Vivitrol, unless the patient has been genuinely opioid-free for at least 7 to 10 days (longer for methadone and other long-acting agents). Starting too early precipitates a severe, forced withdrawal that is not typically life-threatening but is miserable, treatment-ending, and, with the irreversible Vivitrol depot, prolonged. When there is any doubt, wait longer or prove tolerability with a low oral test dose before committing to the injection.
Never lead with the Vivitrol injection when opioid-free status is uncertain. The oral tablet is your reversible probe. If 25 mg of oral naltrexone is tolerated without withdrawal, then you can transition to the monthly injection with confidence. You can wash out a bad tablet in hours; you cannot recall 380 mg of depot naltrexone from a gluteus.
Fear #2: "The opioid-free interval is a bottleneck, patients relapse while they wait."
This is a real and fair concern, and it is the honest structural weakness of the antagonist approach versus agonist therapy (buprenorphine can be started while a patient is still in withdrawal; naltrexone cannot). Acknowledge it rather than pretend it away.
The practical answers: use the waiting window actively (supportive comfort meds for withdrawal: clonidine, ondansetron, an antiemetic, hydroxyzine, trazodone for sleep); be honest with the patient that this drug requires them to get through the hump first; and if a patient can't reliably stay off opioids long enough to induct, that is a signal that buprenorphine or methadone may be the better fit for them. Naltrexone for OUD fits the motivated, already-detoxed patient, someone leaving a controlled setting (residential, incarceration, inpatient detox) opioid-free, or a patient who specifically refuses agonist therapy. Match the drug to the patient and the bottleneck stops being a problem.
Fear #3: "What happens when my patient needs surgery or pain control?"
Plan for it in advance, and it's manageable. A patient on naltrexone has their opioid receptors blocked: standard opioid analgesics won't work, and trying to "overpower" the blockade with escalating opioid doses is dangerous. So:
- For elective surgery, stop oral naltrexone at least 72 hours ahead (and account for the fact that the blockade fades over roughly that window). For Vivitrol, you need to plan around the injection: schedule elective procedures toward the end of the monthly cycle, understanding the depot's effect wanes over about 4 weeks but residual antagonism can persist.
- For acute or emergency pain, anesthesia and pain teams can manage with non-opioid analgesia (regional or neuraxial anesthesia, NSAIDs, acetaminophen, ketamine) and, when opioids are unavoidable, higher-monitored dosing in a setting equipped to watch for respiratory depression as the blockade recedes.
- Give every patient a wallet card or documented note stating they are on naltrexone, exactly as you would for an anticoagulant. This is the single most useful thing you can do to prevent a pain-control crisis.
The right time to have the surgery and pain conversation is at the first visit, not in a preoperative panic. Tell the patient: "While you're on this medication, ordinary opioid painkillers won't work on you. If you ever need surgery or land in an ER, tell them you're on naltrexone. For anything planned, call me first so we can time it."
Fear #4: "Vivitrol injections are logistically intimidating."
They're straightforward once you've set them up. Vivitrol is a deep intramuscular gluteal injection of 380 mg every 4 weeks, reconstituted from a kit and given with the manufacturer-supplied large-bore needle. Key logistics:
- It must go deep IM in the gluteal muscle, not subcutaneously: subcutaneous administration is what drives the serious injection-site reactions (induration, necrosis, abscess). Use the provided needle and proper technique; in patients with substantial subcutaneous tissue, ensure the needle is long enough to reach muscle.
- Confirm the opioid-free interval before every early injection, especially the first.
- Set up the cadence: same appointment slot every 4 weeks, ideally with a standing order so the medication and the nurse visit are pre-arranged.
- Cost and prior authorization are the real friction points (Vivitrol runs on the order of ~$1,000 to $1,300 per injection versus roughly ~$85 per month for oral), so reserve it for patients whose problem is adherence, the ones who won't reliably take a daily pill. For them, the monthly injection is transformative precisely because it removes the daily decision.
Naltrexone is not a high-maintenance drug, it is a high-stakes-at-one-moment drug. Almost all of its risk lives in a single question you must answer before the first dose: is this patient opioid-free? Answer that correctly, hand the patient a wallet card, check a baseline LFT, and you are left with one of the best-tolerated, lowest-hassle, non-controlled medications in psychiatry: no scheduled prescription, no diversion worry, no withdrawal on discontinuation. Confirm the interval, plan for pain, and prescribe it.
Part 1: Indications
FDA-Approved Uses
- Alcohol use disorder (oral, approved 1994; injectable Vivitrol, approved 2006)
- Opioid use disorder (as maintenance and relapse prevention after detoxification; Vivitrol FDA-approved for OUD in 2010)
Off-Label Uses
- Binge eating disorder: second-line (after lisdexamfetamine); the branded naltrexone/bupropion combination (Contrave) is FDA-approved for obesity and shows BED benefit.
- Obesity: naltrexone/bupropion (Contrave).
The Evidence-Based Clinical Picture
Alcohol use disorder. This is the indication where naltrexone earns its first-line status. In the COMBINE trial (2006, NIAAA-funded, no industry money, N=1,383, 9 arms), naltrexone was the clear winner: it beat placebo on percent-days-abstinent and on relapse to heavy drinking. Acamprosate, tested head-to-head in the same trial, was the "surprise loser," no better than placebo. Combining naltrexone with acamprosate added nothing over naltrexone alone.
Be honest about the size of the effect, because it shapes how you counsel patients. Naltrexone's benefit is modest but real:
- It reduces heavy drinking by roughly 17 to 25% and cuts relapse to heavy drinking meaningfully (on the order of 20 to 40% fewer relapses in the first 3 to 4 months versus placebo).
- It does not reliably increase total abstinence. Its sweet spot is harm reduction, fewer heavy-drinking episodes, rather than turning a drinker into a teetotaler.
That framing matters: naltrexone is a superb fit for the patient whose goal is to cut down or prevent the binge, and a reasonable tool even for the patient aiming at abstinence, but you should not sell it as a magic sobriety switch.
Because naltrexone works by making drinking less rewarding, some patients use it as targeted, as-needed dosing: 25 to 50 mg taken about 1 hour before an anticipated drinking situation, rather than daily. Over repeated pairings, alcohol becomes progressively less reinforcing (pharmacological extinction). It demands strict compliance (the pill before every drinking occasion) but is a legitimate, evidence-informed option for motivated moderate drinkers.
Opioid use disorder. Naltrexone for OUD is a different clinical animal. It is not the first thing most guidelines reach for: buprenorphine and methadone (agonist therapies) have the deeper evidence base and can be started without a long opioid-free wait. Naltrexone's niche in OUD is specific and valuable:
- The motivated, already-detoxed patient.
- Patients leaving a controlled, opioid-free setting: residential rehab, inpatient detox, incarceration, where the hard part (getting clean) is already done.
- Patients who refuse agonist therapy on principle or circumstance.
- Settings where agonist prescribing isn't available.
For OUD, the extended-release injectable (Vivitrol 380 mg IM monthly) is far more relevant than oral tablets, because daily oral adherence in active OUD is notoriously poor and a missed few days re-opens the door to relapse.
Binge eating disorder (off-label, second-line). Standard AUD doses (50 to 100 mg) appear insufficient for BED; an older pilot study in bulimia found reductions in bingeing and purging at 200 to 300 mg per day, but it was never replicated, and 300 mg a day is the dose that produced the transaminase elevations behind naltrexone's hepatotoxicity warning. The better-supported modern route is the naltrexone plus bupropion combination.
Predictors of a Good Naltrexone Response (AUD)
- A goal of reducing heavy drinking or preventing binges (rather than immediate total abstinence)
- Motivation and willingness to pair medication with brief counseling
- Ability to take it consistently (or a plan for the injection if not)
- Some data and clinical lore favor patients with a strong family history of alcoholism and prominent craving/reward-driven drinking, the endogenous-opioid-reward phenotype naltrexone directly targets
The most important adherence intervention in the COMBINE trial wasn't a drug, it was brief "medical management": nine 20-minute sessions over 16 weeks covering adherence, side effects, abstinence goals, and mutual-help attendance. This simple, low-intensity support outperformed structured CBT and is well within the reach of any prescriber's visit. Naltrexone plus five minutes of structured coaching beats naltrexone handed over silently.
Part 2: Before You Start
Naltrexone's pre-start checklist is short, but two items are non-negotiable.
The Two Things You Must Confirm
1. The patient is opioid-free. This is the single most important step.
- 7 days off short-acting opioids; 10 to 14 days off long-acting opioids (methadone, buprenorphine, ER products).
- Verify with history plus a urine drug screen, and use a naltrexone/naloxone challenge (or a cautious low oral test dose) when there is any doubt.
- Never start, especially never inject Vivitrol, into an uncertain opioid history.
2. The liver can tolerate it.
- Naltrexone carries a labeled warning for hepatotoxicity. The FDA removed the boxed warning in 2013, and the label now describes cases of hepatitis and clinically significant liver dysfunction under Warnings.
- Get baseline LFTs.
- Liver disease is no longer a contraindication; the acute hepatitis and liver failure entries left the label in 2013. Cirrhosis still raises exposure sharply, with naltrexone AUC about 5-fold higher when compensated and 10-fold higher when decompensated, so dose cautiously and involve hepatology when the picture is unclear.
- Mild-to-moderate transaminase elevations are generally acceptable: many patients with alcohol use disorder have some elevation from the alcohol itself. When uncertain, consult hepatology or addiction medicine rather than reflexively withholding treatment.
Baseline Workup
| Test | Why |
|---|---|
| LFTs (AST, ALT, bilirubin, albumin) | Labeled hepatotoxicity warning; establish a baseline and gauge any existing liver disease |
| Urine drug screen (opioids) | Confirm opioid-free status before starting; remember it can miss fentanyl and synthetics |
| Pregnancy test | In any woman of childbearing potential (limited safety data, see Special Populations) |
| Opioid history | Type, last use, short- vs long-acting; this determines the required washout |
Who Is a Poor Candidate?
- Anyone still taking opioids (illicit or prescribed): absolute contraindication until washed out
- Patients on chronic opioids for pain who cannot switch to non-opioid analgesia: naltrexone will block their pain control
- Decompensated cirrhosis, where naltrexone exposure rises about 10-fold
- Acute opioid intoxication or withdrawal
- Patients facing imminent surgery requiring opioid anesthesia or analgesia (time it, or defer)
Ask explicitly, at every initiation: "When did you last use any opioid, including any prescription painkiller, cough syrup with codeine, tramadol, or a friend's pill?" Patients routinely don't classify tramadol or codeine cough syrup as "opioids." The washout clock starts at the last dose of any of them.
Part 3: How to Start and Dose
| Parameter | Value |
|---|---|
| Formulations | Oral tablets 50 mg; Vivitrol (IM extended-release 380 mg monthly) |
| Dose | Oral 50 mg/day (some use 100 mg MWF); or Vivitrol 380 mg IM every 4 weeks |
| Must be opioid-free first | ≥7 days off short-acting / 10–14 days off long-acting opioids (or naloxone challenge) — otherwise precipitated withdrawal |
| Hepatotoxicity | A labeled warning rather than a boxed one since 2013; stop for signs of acute hepatitis; baseline LFTs |
| Analgesia | Blocks opioid pain relief — plan ahead for surgery/acute pain |
| Renal | Use caution (not studied in severe impairment) |
| Schedule | Not scheduled; no abuse potential |
Oral Naltrexone (ReVia and generics)
For alcohol use disorder:
- Start: 25 mg once daily for about 3 days to minimize nausea, then increase to the target. The 50 mg tablet is scored, so 25 mg is half a tablet.
- Target: 50 mg once daily. This is the standard, evidence-based dose.
- No benefit above 50 mg for AUD. Higher doses do not work better, and the transaminase elevations behind naltrexone's old boxed warning came from doses far above this range, 300 mg a day in the obesity study the label cites.
- Take with food to reduce nausea.
- Dose in the late afternoon, or about 1 hour before an anticipated drinking situation, not in the morning. Most drinking happens in the evening, and morning dosing wastes the drug's peak effect on the wrong part of the day.
As-needed / targeted dosing (Sinclair Method): 25 to 50 mg taken about 1 hour before a drinking occasion, for motivated moderate drinkers who will adhere before every such occasion.
For opioid use disorder (oral): 50 mg once daily after the opioid-free interval is confirmed, but recognize that daily oral adherence in OUD is poor, and the injectable is usually the better choice.
Injectable Naltrexone (Vivitrol)
- Dose: 380 mg deep intramuscular (gluteal), once every 4 weeks.
- Total naltrexone exposure across the month runs about 3 to 4 times that of oral 50 mg daily, and the depot produces much less 6-beta-naltrexol because it reduces first-pass hepatic metabolism. The reason to prefer it is the daily-adherence problem it removes: a randomized head-to-head in hospitalized patients with AUD found no significant difference from oral naltrexone in heavy drinking days at 3 months.
- Can be started de novo, no requirement to establish tolerance to oral naltrexone first, but only once opioid-free status is confirmed. When there is any doubt, prove tolerability with an oral test dose before committing to the irreversible injection.
- Injection technique is a safety issue: deep IM in the gluteal muscle with the supplied needle. Subcutaneous deposition causes the serious site reactions (see Side Effects).
In the pivotal alcohol trial, patients who had 4 or more days of abstinence before the first injection did better on every drinking outcome: longer time to first drink, higher rates of continuous abstinence, and fewer heavy drinking days per month. When feasible, get a few sober days on the board before you inject. It both improves outcomes and doubles as a safety buffer against residual opioid.
Choosing a Formulation
- Oral 50 mg/day: cheap (~$85 per month), reversible, easy to stop before surgery, good for motivated and adherent patients and for AUD harm reduction. Downside: daily adherence.
- Vivitrol 380 mg IM monthly: overcomes non-adherence, ideal for OUD relapse prevention and for AUD patients with documented compliance failure. Downside: cost (~$1,000 to $1,300 per injection), prior-auth hassle, injection-site reactions, and irreversibility, you can't take it back if you got the opioid-free interval wrong.
Part 4: Monitoring
Naltrexone monitoring is light, one of its great practical virtues, but the labeled hepatotoxicity warning means you don't skip the liver entirely.
Liver Function Tests
| Timepoint | Action |
|---|---|
| Baseline | LFTs before starting (also gauges any existing liver disease) |
| Early in treatment | A repeat check in the first few months is reasonable, more so with alcohol-associated liver disease or another hepatotoxic drug on board |
| Stable, dose ≤50 mg | The label sets no fixed schedule; individualize, and many clinicians check only if symptoms arise |
| Any hepatic symptom | Check LFTs promptly (RUQ pain, jaundice, dark urine, nausea, malaise) |
Across reported series, transaminase values above 3 times the upper limit of normal occur in about 1% of patients, and naltrexone has not been definitively linked to clinically apparent liver injury. Keep the dose at 50 mg for AUD and the liver risk stays small.
Vivitrol-Specific Monitoring
- Inspect the injection site at each monthly visit and educate the patient to report worsening pain, induration, swelling, blistering, or an open lesion, these can signal a serious site reaction needing prompt evaluation.
- LFTs at baseline; ongoing only as clinically indicated.
- Re-confirm opioid-free status before early injections.
What You Don't Have to Do
No serum "levels" to chase, no ECG requirement, no renal-dosing dance (minimal renal excretion), no scheduled-substance prescribing rules, no diversion surveillance. Compared with lithium, clozapine, or an agonist maintenance program, naltrexone monitoring is refreshingly minimal.
Part 5: Side Effects and How to Manage Them
Most naltrexone side effects are mild, early, and avoidable with slow titration and food. The two that demand respect are hepatotoxicity and, for Vivitrol, injection-site reactions.
Nausea, Anorexia, GI Upset (common, early, ~10 to 15%)
Usually mild and transient, and largely preventable.
- Titrate slowly: 25 mg × 3 days before 50 mg.
- Take with food.
- Dose in the late afternoon or evening, not the morning.
- Most nausea settles within the first days to weeks.
Hepatotoxicity
Naltrexone can raise transaminases, and the label records cases of hepatitis and clinically significant liver dysfunction. The signal that once earned a boxed warning came from an obesity study at 300 mg a day in which 5 of 26 subjects developed transaminase elevations after three to eight weeks. The FDA removed that boxed warning in 2013 and kept an ordinary Warning. Keeping AUD dosing at 50 mg is your primary prevention. Get baseline LFTs, recheck when the clinical picture warrants, and check promptly for any hepatic symptom. Stop the drug for signs or symptoms of acute hepatitis.
- Usually reversible on discontinuation.
- Remember that in AUD patients, alcohol itself is often the bigger hepatic insult; improving drinking may improve the LFTs.
Injection-Site Reactions (Vivitrol)
The characteristic risk of the depot. They range from common-and-benign to rare-and-serious:
- Common/mild: pain, tenderness, induration, a nodule at the site.
- Serious (uncommon): sterile abscess, cellulitis, tissue necrosis, some requiring surgical intervention.
- The dominant preventable cause is inadvertent subcutaneous injection. Use proper deep IM gluteal technique and an adequately long needle. Instruct patients to report worsening or non-healing site problems promptly.
Precipitated Opioid Withdrawal (if the opioid-free rule is broken)
Not a "side effect" so much as a mechanism-driven adverse event: giving naltrexone to an opioid-dependent patient causes rapid, severe, forced withdrawal. It is not typically life-threatening, but it is miserable, treatment-ending, and, with the irreversible Vivitrol depot, prolonged. Prevention is the whole game: confirm the interval (Part 2).
Mood and the "Dysphoria" Question
Because naltrexone blocks the endogenous opioid (endorphin) system, clinicians reasonably ask about depressed mood, dysphoria, or suicidality. The signal in the literature is weak and inconsistent, and untreated substance use disorders themselves carry high rates of depression and suicide. Practical stance: screen for depression and suicidality at baseline and at follow-ups (you should be doing this in any SUD patient regardless), don't attribute all low mood to the drug, and take a sustained worsening seriously. Do not let an overstated fear of "blocking endorphins" deprive a patient of an effective treatment.
Other
- Headache, dizziness, fatigue, insomnia, arthralgia/myalgia: generally mild, often transient.
- Appetite suppression: occasionally welcome.
Part 6: Overdose and Safety
Naltrexone Itself in Overdose
Naltrexone has a wide safety margin and is not a classic overdose agent: it is not sedating, not respiratory-depressing, and not lethal in the way opioids or lithium are. The dose-related worry is hepatotoxicity, not acute overdose death.
The Loss-of-Tolerance Overdose Danger
This is the most important safety fact in the entire guide, and it is counterintuitive because the danger comes not from taking naltrexone but from stopping it.
While a patient is on naltrexone, their opioid receptors are blocked and they are not using opioids, so their opioid tolerance falls back toward that of an opioid-naive person. If they then stop naltrexone and relapse to their previous opioid dose, that once-ordinary dose can now be fatal, because the tolerance that used to protect them is gone. The same physiology makes patients who try to override the blockade with escalating opioid doses acutely vulnerable to overdose as the blockade fades.
What to do about it:
- Counsel every OUD patient explicitly, at initiation and at discontinuation: "If you stop this medication and go back to using, the amount you used to take could kill you. Your tolerance is gone."
- Co-prescribe naloxone (Narcan) and train the patient and their people to use it.
- Frame discontinuation as a high-risk transition that warrants a plan, not a casual stop.
- Recognize that the highest-risk windows are the end of a Vivitrol month (as the depot wanes) and any lapse in adherence.
Naltrexone doesn't kill patients. Relapsing after naltrexone has lowered their tolerance can. Every OUD patient leaves your office with a naloxone kit and the tolerance conversation.
Part 7: Drug Interactions
Naltrexone is metabolically quiet, no significant CYP450 interactions, so its "interactions" are pharmacodynamic and center entirely on opioids.
Opioids
- Naltrexone blocks all mu-opioid effects. The blockade is total, not partial.
- It will precipitate acute withdrawal in anyone with opioids on board (absolute contraindication until washed out).
- It blocks opioid analgesia, ordinary opioid painkillers won't work. This is the surgical and pain-planning problem covered in "How to Use With Confidence" and Special Situations.
- The list of "opioids" patients forget includes tramadol, codeine (including in cough syrups), hydrocodone, oxycodone, morphine, and buprenorphine. All are blocked; all count for the washout.
Practical Interaction Management
- Elective procedures: stop oral naltrexone ≥72 hours ahead; plan Vivitrol timing around the monthly cycle.
- Emergency pain: lead with non-opioid strategies (regional anesthesia, NSAIDs, acetaminophen, ketamine); if opioids are unavoidable, manage in a monitored setting alert to respiratory depression as blockade recedes.
- Wallet card or documented allergy-style alert on every patient.
- Additive hepatotoxins: theoretical additive liver risk with other hepatotoxic drugs; monitor LFTs a bit more attentively if co-prescribing known hepatotoxins.
What Naltrexone Does Not Interact With
- No meaningful pharmacokinetic interactions with most psychotropics.
- No renal-clearance issues, no ECG concerns.
Part 8: Special Populations
Pregnancy
- Human data are limited and insufficient to establish a drug-associated risk. The concern is an animal one: daily naltrexone increased early fetal loss in rats and rabbits. Naltrexone is generally avoided in pregnancy unless the benefit clearly outweighs the risk.
- For a pregnant patient with opioid use disorder, the standard of care leans toward agonist therapy (buprenorphine or methadone), which has a far larger safety and outcomes database in pregnancy (e.g., the MOTHER data favoring buprenorphine on neonatal outcomes). For a patient already stable on naltrexone before conception and doing well, the decision is a nuanced, individualized one best made with maternal-fetal medicine.
- For alcohol use disorder in pregnancy, prioritize non-pharmacologic strategies and weigh risks case by case.
Lactation
Naltrexone is excreted in breast milk; data are limited. Most sources advise caution or avoidance during breastfeeding; individualize with pediatrics.
Hepatic Impairment
- Mild-to-moderate LFT elevation: generally acceptable, naltrexone is not contraindicated, though monitor.
- Cirrhosis (low albumin, elevated bilirubin): no longer a labeled contraindication, but naltrexone AUC rises about 5-fold in compensated and 10-fold in decompensated cirrhosis, so use caution and consult hepatology if the picture is ambiguous.
Renal Impairment
Minimal renal excretion; no formal dose adjustment is typically required. Use ordinary caution in severe renal disease.
Elderly
Reasonable to use; start at 25 mg, half of the scored 50 mg tablet, and titrate gently (nausea and dizziness are the main tolerability issues), and be attentive to age-related hepatic changes: keep the baseline and periodic LFTs.
Children and Adolescents
Not FDA-approved in pediatrics; off-label use is possible with appropriate counseling and consent. For adolescent OUD, weigh against agonist options, which have more supporting data.
Part 9: Discontinuation
Stopping naltrexone is pharmacologically simple; the care is in the clinical management.
- No physiologic withdrawal syndrome, no taper required. You can stop naltrexone abruptly without a rebound pharmacologic reaction, which is convenient.
- But the risk of relapse is real. In AUD, patients who stop are at higher risk of relapse, and the size of that risk has not been well quantified. The recommended course is a minimum of about 3 to 4 months, and many clinicians aim for at least 1 year, reassessing at least every 3 months.
- The "invisible benefit" trap: patients feel nothing from naltrexone, there's no subjective effect, so they conclude it isn't doing anything and stop. Preempt this at initiation: "You won't feel this medication. That's expected. Its whole job is to keep the next heavy episode from happening. If you're not drinking, it's working."
- For OUD, discontinuation is a high-risk safety event, not just an efficacy question, because of the loss-of-tolerance overdose danger (Part 6). Every OUD discontinuation should include the tolerance conversation and a naloxone kit.
Part 10: Naltrexone vs the Alternatives
For Alcohol Use Disorder
| Agent | Mechanism | Efficacy | Best for | Watch-outs |
|---|---|---|---|---|
| Naltrexone | Mu-opioid antagonist (blocks reward) | Strongest evidence; COMBINE winner; ~17 to 25% less heavy drinking | First-line; reward/craving-driven drinking; harm reduction | Nausea early; hepatotoxicity at high dose; needs opioid-free status |
| Acamprosate | GABA-glutamate modulator | Weak (COMBINE: no better than placebo; other data ~14%) | Hepatic disease (renally cleared, liver-safe); post-detox abstinence maintenance | High pill burden (TID); renal dosing; contraindicated CrCl <30 |
| Disulfiram | Aldehyde dehydrogenase inhibitor (aversive) | Moderate only if supervised; no effect unsupervised | Highly motivated patients committed to total abstinence, with a supervisor | Disulfiram-alcohol reaction; cardiac/hepatic cautions; adherence-dependent |
| Topiramate | GABA-glutamate rebalancer | ~23% less heavy drinking; roughly equivalent to naltrexone in head-to-head trials | Partial responders; patients who'd welcome weight loss | Cognitive slowing ("Dopamax"), paresthesias; teratogenic |
Bottom line for AUD: naltrexone first, especially for the reward-driven, cut-down-the-binge patient. Acamprosate is the liver-safe alternative when advanced liver disease makes naltrexone a poor fit. Disulfiram is for the specific, motivated, supervised abstinence-committed patient. Topiramate is a strong off-label second-line.
For Opioid Use Disorder
| Agent | Class | Prescribing setting | Key trade-off |
|---|---|---|---|
| Naltrexone (Vivitrol) | Full antagonist | Any prescriber; office-based; monthly injection | No abuse or diversion potential; no controlled-substance rules; but requires full detox first, and an injection given before the opioid-free interval is complete precipitates a withdrawal you cannot reverse. No agonist "cushion" if the patient relapses. |
| Buprenorphine | Partial agonist | Office-based (X-waiver now removed) | First-line; can start during withdrawal; respiratory-ceiling safety; reduces mortality; but is itself an opioid (diversion, stigma). |
| Methadone | Full agonist | Federally licensed clinics, daily dosing | Effective for severe/refractory OUD; but clinic-bound, full-agonist overdose risk, diversion potential. |
Bottom line for OUD: buprenorphine and methadone are the mortality-reducing agonist mainstays and can be started faster. Naltrexone's distinct advantages are that it is non-controlled, non-addictive, diversion-proof, and prescribable by anyone, making it the right tool for the already-detoxed, motivated patient, those leaving controlled settings, and those who refuse agonists. Match the patient to the mechanism.
Mechanism
Naltrexone is a pure, competitive mu-opioid receptor antagonist (with activity at kappa and delta receptors as well). It has no agonist activity, it simply occupies and blocks the receptor.
- In opioid use disorder, the mechanism is blockade: with receptors occupied, ordinary opioid doses produce no euphoria and no analgesia, removing the reinforcement that drives use.
- In alcohol use disorder, the mechanism is subtler and elegant. Alcohol triggers release of endogenous opioids (endorphins), which in turn drive dopamine release in the reward pathway (nucleus accumbens), a chunk of the pleasurable "buzz" and the craving that follows the first drink. By blocking the mu receptor, naltrexone attenuates that endogenous-opioid reward cascade: the first drink is less rewarding, the pull toward the next drinks is weaker, and a slip is less likely to snowball into a heavy-drinking episode. That is why naltrexone reduces heavy drinking more than it produces abstinence: it changes the reinforcement math of drinking rather than making the patient not want a drink at all.
Pharmacokinetics that matter clinically:
- Oral half-life ~4 to 10 hours (active metabolite 6-beta-naltrexol contributes to duration); dosed once daily.
- First-pass hepatic metabolism is significant for oral naltrexone, and the Vivitrol depot largely bypasses it, so the injection generates far less 6-beta-naltrexol.
- Vivitrol (380 mg IM) provides about 4 weeks of blockade, waning over the month, hence the end-of-cycle vulnerability window.
- Minimal renal excretion; hepatic handling is why the liver, not the kidney, is the organ you watch.
The Bedside Cheat Sheet
Before starting (non-negotiable)
- Confirm opioid-free: 7 days (short-acting), 10 to 14 days (long-acting/methadone/buprenorphine). UDS plus history; naloxone challenge or low oral test dose if unsure.
- Baseline LFTs. Liver disease is no longer a contraindication; mild-moderate elevation OK, caution in cirrhosis.
- Pregnancy test; full opioid history (include tramadol, codeine syrups).
Dosing
- Oral (AUD): 25 mg × 3 days → 50 mg/day. No benefit >50 mg; the 50 mg tablet is scored.
- Take with food; dose late afternoon or 1 hour pre-drinking, not morning.
- Vivitrol: 380 mg deep IM gluteal every 4 weeks. Aim for at least 4 sober days before the first injection (better outcomes plus safety buffer).
- Prove tolerance with an oral test dose before committing to Vivitrol when opioid-free status is at all uncertain.
Monitoring and side effects
- LFTs baseline, then as the clinical picture warrants rather than on a fixed schedule; check for any hepatic symptom.
- Vivitrol: inspect injection site monthly; use deep IM technique (subcutaneous means necrosis risk).
- Nausea (early, avoidable): slow titration plus food.
- Injection-site reactions: proper deep IM technique; report non-healing lesions.
Don't forget
- Blocks all opioid analgesia, plan surgery and pain ahead; stop oral ≥72 h pre-op; wallet card for every patient.
- Loss-of-tolerance overdose: stopping naltrexone plus relapse can be fatal. Counsel and co-prescribe naloxone for every OUD patient.
- No withdrawal, no taper, but relapse risk is real; treat ≥3 to 4 months, aim for ≥1 year.
- "If you're not drinking, it's working": patients feel nothing; educate to prevent premature discontinuation.
- Pair with brief medical management (it beat CBT in COMBINE).
Naltrexone asks very little of the prescriber compared with almost anything else in addiction medicine: no controlled-substance license, no clinic, no diversion worry, no serum levels, no taper. In exchange it offers the best-evidenced medication for alcohol use disorder and a clean, non-addictive antagonist option for opioid use disorder. Nearly all of its risk collapses into a single question you answer before the first dose, is this patient opioid-free, and a single fact you must teach every patient, stopping this drug and relapsing can be fatal because your tolerance is gone. Answer the question carefully, teach the fact plainly, and hand out a wallet card and a naloxone kit. With the opioid-free interval respected and a plan for pain in place, naltrexone is the default, uncomplicated way to put a patient on medication for their addiction, and one of the most useful and most underused tools in the whole pharmacopeia.