Antidepressant-Induced Acute Liver Injury: A Case-Control Study in an Italian Inpatient Population

Pre-marketing clinical trials show that antidepressant-induced liver injury seems to be a rare adverse event. Because of short follow-up trial duration, the incidence of liver injury due to antidepressant use could be underestimated.

We aimed to quantify the risk of acute liver injury associated with antidepressant use through a case-control analysis among an inpatient population.

A multicenter study was carried out in nine Italian hospitals from October 2010 to January 2014, within the DILI-IT (Drug-Induced Liver Injury in Italy) study project. After exclusion of all patients with a clear competing cause of liver injury, cases were defined as adults admitted to the hospital with a diagnosis of acute liver injury, while controls had any other acute clinical condition not related to the liver. Antidepressant exposure was evaluated within 90 days prior to the date of the first sign/symptom of liver injury. Odds ratio (OR) with 95% confidence interval (95% CI) was calculated as a measure of risk estimates for liver injury.

We included 17 cases exposed to antidepressants matched to 99 controls. According to the features of liver injury, all cases showed symptomatic liver function test abnormalities at hospital admission, with the main signs/symptoms represented by fatigue, nausea, asthenia, or dark urine. Citalopram was the antidepressant mostly involved in the increase of liver enzymes, mainly alanine aminotransferase. Compared with non-use, current use of antidepressants was associated with a significantly increased risk of liver injury (adjusted OR, OR_ADJ, 1.84; 95% CI 1.02-3.32). Specifically, an increased, but not significant, risk of developing liver injury was observed for citalopram, a selective serotonin-reuptake inhibitor (OR_ADJ 1.82; 95% CI 0.60-5.53).

The use of antidepressants is not as safe in terms of liver injury as expected; instead, the risk of antidepressant-induced liver injury is likely underestimated. The lack of significance does not reflect the absence of risk, but rather suggests the need to evaluate it in a wider setting of antidepressant users.