Which Antidepressants Cause Liver Damage?

Most antidepressants carry some risk of liver injury, but clinically significant damage is uncommon. A large German-speaking surveillance program reported that antidepressants caused drug-induced liver injury (DILI) in about 0.08% of patients, while mild, symptom-free liver enzyme elevations show up in roughly 0.5 to 3% of people taking these medications. The drugs most strongly tied to serious liver problems include nefazodone, certain older tricyclics like imipramine and amitriptyline, and a few others such as duloxetine, bupropion, and trazodone. SSRIs, as a class, appear to carry the lowest hepatic risk, though no antidepressant gets a completely clean bill of health.

How Often Do Antidepressants Actually Hurt the Liver

Two figures help frame the risk. A review of clinical data stretching back to 1965 concluded that somewhere between half a percent and 3% of patients on antidepressants develop mild, asymptomatic bumps in liver enzymes. These elevations often resolve on their own and never progress to anything a patient would feel. More concerning liver injury, the kind that shows up as jaundice, nausea, or abnormal lab results prompting a clinical workup, is far rarer. A multicenter drug-surveillance program tracking patients from 1993 to 2011 found that antidepressants caused clinically meaningful DILI in about 0.08% of cases.¹ That same surveillance data suggested SSRIs were less likely to contribute to DILI than other antidepressant classes.

A separate French cohort study involving over five million people registered in the national health insurance database compared the risk of serious liver injury in patients starting various non-SSRI antidepressants against those starting SSRIs. None of the drugs examined, including venlafaxine, duloxetine, mirtazapine, and agomelatine, showed a statistically significant increase in serious liver injury compared to SSRIs.² That finding sounds reassuring, but it does not mean no individual drug is a concern. Population-level studies can miss rare but severe events because so few people experience them. The drugs that worry hepatologists tend to surface through case reports and pharmacovigilance databases, not big cohort comparisons.

The Antidepressants Most Closely Linked to Liver Harm

A wide clinical review published in the American Journal of Psychiatry identified the antidepressants associated with the greatest risk of hepatotoxicity: iproniazid, nefazodone, phenelzine, imipramine, amitriptyline, duloxetine, bupropion, trazodone, tianeptine, and agomelatine.³ That list spans nearly every antidepressant class. A few of these deserve closer attention because they have caused the most dramatic problems or remain in widespread use.

Nefazodone

If one antidepressant stands out as a liver hazard, it is nefazodone. Case reports of severe hepatocellular injury, including liver failure requiring transplantation, began accumulating in the late 1990s. A report of three such cases in the Annals of Internal Medicine concluded that nefazodone causes liver damage in an unpredictable, idiosyncratic way, meaning it is not clearly dose-dependent and cannot be reliably predicted by baseline blood work.⁴ The authors recommended routine liver enzyme testing before and during therapy. In response to mounting evidence, the branded version of nefazodone was withdrawn from most markets, though generic versions remained available in some countries for a time.

Laboratory research has since illuminated part of the mechanism. Nefazodone impairs mitochondrial function by interfering with the enzymes responsible for oxidative phosphorylation, the process cells use to produce energy. It also ramps up production of damaging reactive oxygen species and weakens the cell’s own antioxidant defenses.⁵ In other words, liver cells exposed to nefazodone get starved of energy and hammered by oxidative stress at the same time.

Tricyclic Antidepressants

Tricyclics, the older class that includes amitriptyline, imipramine, and nortriptyline, have a well-documented if uncommon association with liver injury. The American Journal of Psychiatry review noted that cross-toxicity has been described among tricyclic and tetracyclic antidepressants, meaning a patient who reacts badly to one drug in this family may also react to another.³ That is an important clinical point: switching from amitriptyline to nortriptyline after a liver enzyme flare may not solve the problem.

Amitriptyline, the most widely prescribed tricyclic, has been linked to rare but severe hepatotoxicity. Research into the mechanism suggests the drug undergoes metabolic activation in the liver to form reactive arene oxide intermediates. These reactive molecules bind to cellular proteins and may trigger the immune-mediated injury characteristic of idiosyncratic drug reactions.⁶

A separate line of research has identified another mechanism for nortriptyline, a metabolite of amitriptyline and a commonly prescribed drug in its own right. In laboratory studies, nortriptyline triggered activation of the NLRP3 inflammasome, an immune sensor inside cells that, when improperly switched on, causes inflammation. The effect was especially pronounced when cells were already in a mildly inflammatory state, and other tricyclics triggered the same inflammasome pathway.⁷ This helps explain why tricyclic liver injury may be more likely in people who already have low-grade liver inflammation from another cause.

SSRIs and Their Relatively Lower Risk

SSRIs are generally regarded as the safest antidepressant class when it comes to the liver, and population data supports that view. But “safest” is relative, not absolute. Case reports have linked paroxetine, fluoxetine, fluvoxamine, and citalopram to reversible liver injury that resolves when the drug is stopped.⁸ Sertraline deserves specific mention because, despite being one of the most commonly prescribed antidepressants worldwide, patient cases have been linked to fatal outcomes.

Laboratory work has uncovered a plausible mechanism for sertraline-related liver toxicity. The drug uncouples mitochondrial energy production and inhibits key respiratory chain complexes in liver cells. It also triggers a process called mitochondrial permeability transition, essentially punching holes in the mitochondrial membrane and causing the cell to lose its energy supply rapidly.⁹ In practical terms, this means sertraline can damage liver cells by cutting off their power supply, though this appears to happen only in a very small fraction of patients. For most people, sertraline is well tolerated.

Duloxetine and Pre-Existing Liver Disease

Duloxetine, a serotonin-norepinephrine reuptake inhibitor (SNRI) used for both depression and chronic pain, occupies an unusual position in the liver-risk conversation. The large French cohort study found no significant increase in serious liver injury compared to SSRIs at a population level.² Yet duloxetine-induced liver injury in patients with pre-existing liver disease or chronic alcohol use is a recognized clinical phenomenon.¹⁰ The drug’s prescribing information includes warnings against use in patients with substantial alcohol intake or chronic liver conditions. For someone with a healthy liver, duloxetine is not considered high-risk. For someone who drinks heavily or has hepatitis, it may be.

Bupropion and Trazodone

Both bupropion and trazodone appear on the list of antidepressants associated with greater hepatotoxicity risk.³ Patient cases involving both drugs have been linked to fatal outcomes.⁸ Bupropion also surfaced in the AMSP drug surveillance program as a cause of drug-induced liver injury, though only as individual cases rather than a pattern suggesting high frequency.¹¹

Trazodone is widely prescribed at low doses as a sleep aid, which is worth bearing in mind. When clinicians think about liver risk, they tend to focus on full antidepressant doses, but trazodone’s occasional use for insomnia means many patients taking it may not even consider it an “antidepressant” or mention it when asked about their medications. That gap in reporting makes its true liver risk harder to pin down.

Agomelatine and Mandatory Liver Monitoring

Agomelatine, a melatonergic antidepressant available in Europe and some other regions but not in the United States, is unique in that regulators built liver enzyme monitoring directly into its prescribing requirements. A pooled analysis of over 7,600 treated patients found that liver enzymes rose to more than three times the upper limit of normal in 1.3% of patients on the 25-mg dose and 2.5% of patients on the 50-mg dose, compared to 0.5% for placebo. Most of these elevations appeared within the first 12 weeks. The median time to recovery after stopping the drug was just 14 days, and about a third of patients saw their enzymes normalize even while continuing agomelatine, suggesting the liver can sometimes adapt.¹² No cases of acute liver failure or death were reported in that dataset, but the enzyme elevation rate is high enough to justify regular blood tests.

The agomelatine story illustrates something important: a drug can cause frequent lab abnormalities without necessarily causing frequent serious liver disease. The monitoring requirement catches problems early enough that they resolve before real damage occurs. Whether this represents a genuinely safer drug or just a more carefully watched one is a matter of perspective.

What Antidepressant Liver Injury Looks Like

When antidepressants do damage the liver, the injury pattern is most often hepatocellular, meaning the liver cells themselves are the primary target, rather than the bile ducts. Cholestatic injury (where bile flow is blocked) and mixed patterns occur less frequently.¹ Symptoms, when they appear at all, can include fatigue, nausea, abdominal pain in the upper right side, dark urine, and yellowing of the skin or eyes. Many cases produce no symptoms and are caught only through routine blood work.

Timing matters for suspicion. Most antidepressant-related liver injury develops within the first several weeks to months of treatment. The agomelatine data showing 64% of enzyme elevations appearing before 12 weeks is broadly consistent with what is seen across the class.¹² If you have been on an antidepressant for a year without any enzyme changes, a new liver problem is more likely to have another explanation, though it is not impossible.

For drugs like paroxetine, fluoxetine, fluvoxamine, citalopram, mirtazapine, and venlafaxine, the liver injury is typically reversible once the drug is stopped.⁸ The more dangerous scenario is the idiosyncratic reaction seen with nefazodone, where the damage can progress rapidly to acute liver failure before anyone recognizes what is happening.

Who Faces Greater Risk

The single clearest risk factor for antidepressant-related liver injury is pre-existing liver disease. This is especially well documented for duloxetine, where the connection to injury in patients with chronic liver conditions or heavy alcohol use is explicitly acknowledged in prescribing information.¹⁰ But the principle extends broadly: a liver that is already under stress from alcohol, hepatitis, fatty liver disease, or another medication is more vulnerable to a second hit from an antidepressant.

The nortriptyline inflammasome research offers a biological explanation for why this happens. Tricyclics activated the NLRP3 inflammasome most aggressively in cells that were already in a mildly inflammatory state.⁷ If your liver already has some background inflammation, the drug’s effect on this immune sensor is amplified. This likely applies beyond just tricyclics and may help explain why some people develop liver injury on drugs that millions of others tolerate without incident.

Other risk factors include polypharmacy (taking multiple medications that compete for the same liver enzymes), older age, and possibly genetic variation in the liver enzymes that metabolize antidepressants. If you metabolize a drug slowly because of your genetic profile, you may accumulate higher drug levels than the average patient, and those higher levels can stress the liver.

St. John’s Wort Is Not Necessarily Safer

People who worry about prescription antidepressant liver effects sometimes turn to St. John’s Wort, the herbal supplement with mild antidepressant properties. The assumption that “natural” means “easier on the liver” does not hold up. Laboratory research comparing the two main active ingredients in St. John’s Wort, hypericin and hyperforin, against citalopram (a common SSRI) found that hypericin was a mitochondrial toxicant at low concentrations. It depleted cellular energy stores at concentrations far below those at which citalopram caused any measurable toxicity.¹³ Hyperforin, the other major active component, was not toxic at the tested concentrations. But the combination present in a whole-herb extract means you are getting both compounds.

St. John’s Wort also has potent effects on liver enzymes involved in drug metabolism, which can alter the blood levels of many other medications. This enzyme-induction property creates its own set of liver-related concerns, especially if you are combining the supplement with other drugs. The “natural alternative” framing obscures real biochemical risks that deserve the same caution you would give a prescription.

Practical Takeaways for Monitoring

Routine liver enzyme testing before starting an antidepressant is not universally recommended for SSRIs, but it is reasonable for anyone with existing liver disease, heavy alcohol use, or a plan to start a drug with known hepatotoxic potential like agomelatine or a tricyclic. For agomelatine, monitoring is mandatory in countries where the drug is available. For nefazodone, the original case reports recommended regular monitoring before and during therapy.⁴

If you develop unexplained fatigue, nausea, upper abdominal pain, or yellowing of the skin after starting a new antidepressant, those symptoms warrant prompt blood work. Most antidepressant liver injury is reversible when caught early and the drug is stopped. The rare catastrophic cases tend to involve either a delay in recognition or a drug like nefazodone where injury can progress rapidly. Keeping your prescriber informed about all medications and supplements you take, including St. John’s Wort, helps them account for the cumulative liver load when choosing a treatment.

1
PubMed Central. Antidepressants- and antipsychotics-induced hepatotoxicity
2
PubMed Central. Antidepressants and Hepatotoxicity: A Cohort Study among 5 Million Individuals Registered in the French National Health Insurance Database
3
PubMed. Antidepressant-induced liver injury: a review for clinicians
4
PubMed. Nefazodone-induced liver failure: report of three cases
5
PubMed. Involvement of mitochondrial dysfunction in nefazodone-induced hepatotoxicity
6
Chemico-Biological Interactions. Bioactivation of the tricyclic antidepressant amitriptyline and its metabolite nortriptyline to arene oxide intermediates in human liver microsomes and recombinant P450s
7
PubMed Central. Tricyclic antidepressants induce liver inflammation by targeting NLRP3 inflammasome activation
8
PubMed. Liver injury associated with antidepressants
9
PubMed Central. Mitochondrial dysfunction induced by sertraline, an antidepressant agent
10
PubMed Central. Duloxetine-induced liver injury in patients with major depressive disorder
11
International Journal of Neuropsychopharmacology. Drug-Induced Liver Injury during Antidepressant Treatment: Results of AMSP, a Drug Surveillance Program
12
PubMed. Characterisation of Agomelatine-Induced Increase in Liver Enzymes: Frequency and Risk Factors Determined from a Pooled Analysis of 7605 Treated Patients
13
Toxicology Letters. Mechanisms of hepatocellular toxicity associated with the components of St. John’s Wort extract hypericin and hyperforin in HepG2 and HepaRG cells