Systematic Review: Ibuprofen-induced Liver Injury

Miguel E. Zoubek; María Isabel Lucena; Raúl J. Andrade; Camilla Stephens


Aliment Pharmacol Ther. 2020;51(6):603-611. 

In This Article


Despite being considered as one of the safest NSAIDs in terms of the hepatic profile,[4,6,7] ibuprofen can cause hepatotoxicity. The prevalence of ibuprofen hepatotoxicity, however, appears to be relatively low considering the widespread use of this medication. In fact, vascular and gastrointestinal complications are probably more commonly associated with ibuprofen than hepatotoxicity.[39] The search for idiosyncratic ibuprofen-induced hepatotoxicity information in the literature resulted in 22 identified cases. Overall, ibuprofen-derived liver injury occurred after a relatively short time from treatment initiation with a mean time to onset of 12 days. With regard to sex, we noted a trend towards women more frequently developing ibuprofen hepatotoxicity than men. This was not the case in our previous report on ibuprofen hepatotoxicity cases in Spain and Latin-America, with similar frequency of male and female subjects, although a slightly higher proportion of women was observed in DILI due to other NSAIDs.[12] A possible explanation for this finding could be that 23% of the patients in the current study had underlying rheumatic disorders, which are more prevalent in women.[40] In addition, a recent study analysing the trends of NSAID use in US adults found that women are more likely to use NSAIDs.[41] Hence, a higher use of NSAIDs and a higher prevalence of rheumatic disorders requiring NSAID treatments may be the reasons behind the observed increase in females in published ibuprofen hepatotoxicity case reports rather than females being biologically more susceptible to this form of DILI than men.

We previously found a predominance of hepatocellular pattern of liver injury in Spanish and Latin-American DILI cases caused by ibuprofen.[12] The same observation holds for the cases obtained from the literature in the current study with 65% of the cases, with sufficient information to determine pattern of liver injury, presenting hepatocellular type of liver injury. Thus, the most common pattern of liver injury associated with ibuprofen-induced hepatotoxicity appears to have a hepatocellular character, although cholestatic/mixed liver injury can also occur.

Vanishing bile duct syndrome is characterised by bile duct injury and ductopenia, and occurred in six of the 22 idiosyncratic cases. Although rarely, VBDS can occur in DILI patients with progressive cholestasis potentially leading to liver failure and death or liver transplantation. It has been associated with causative drugs such as azathioprine, androgens, amoxicillin-clavulanate, carbamazepine, chlorpromazine, erythromycin, estradiol, flucloxacillin, phenytoin and co-trimoxazole.[42] A study of 363 DILI cases with biopsy data found that 7.2% of the cases had bile duct loss based on histopathological interpretations, of which 54% exhibited moderate to severe ductopenia with bile duct loss in more than 50% of the portal tracts.[43] However, it has been suggested that this incidence may be overestimated compared to observations in population-based studies due to the fact that the cases were recruited from tertiary referral centres, which are likely to see more severe cases.[44] Two of the case reports with VBDS in the current study had bile duct loss in more than 50% of portal tracts,[24,29] while the level of bile duct loss was not provided for the remaining four VBDS cases.

Four of the identified cases in the current study developed serious cutaneous reactions (progressive xanthomatosis, SJS or TEN) in addition to VBDS. The concurrence of VBDS and cutaneous reactions was similarly found in the aforementioned study of North American DILI cases, and suggests an aberrant hypersensitivity reaction affecting cholangiocytes and keratinocytes, potentially due to shared immunogenic proteins and cell surface presentation of drug-protein adducts or immunogenic drug metabolites.[43] Interestingly, one of the US cases with VBDS and TEN had taken ibuprofen prior to the liver reaction. The case was, however, adjudicated as DILI most likely caused by azithromycin, and only possibly due to ibuprofen.[43] Nevertheless, a role for ibuprofen cannot be completely ruled out in this case in terms of DILI development and clinical presentation.

Cutaneous reactions were not limited to those cases that developed VBDS. Our literature search also revealed three ibuprofen hepatotoxicity cases with cutaneous reactions (case 3, 15 and 22), but not VBDS. Cutaneous hypersensitivity reactions to ibuprofen are well known.[45] These reactions are often allergic in nature, mostly mild, occur rapidly after drug exposure and rarely contain hepatic involvement. In contrast, DRESS syndrome often presents with concurrent cutaneous and hepatic reactions.[46] However, ibuprofen does not appear to be a major cause of DRESS (with liver involvement) as we only identified one case in our literature search.

Our findings support that ibuprofen-induced liver injury has a wide clinical spectrum rather than a homogeneous signature. However, it should be noted that the analysed cases were all obtained from published case series and case reports and might therefore have been subjected to publication bias, as reports with severe or novel presentations tend to be preferred for publication compared to cases with mild and uncomplicated clinical courses. This may have contributed to the high proportion of identified ibuprofen hepatotoxicity cases with VBDS, SJS and TEN in the current study.

A severe clinical progression was identified in five idiosyncratic cases (one death, two liver transplantations and two liver transplantation referrals with unknown outcome) as well as one case involving a single ibuprofen overdose that led to liver transplantation. Similarly, the prospective ibuprofen-induced hepatotoxicity cohort previously reported from the Spanish DILI Registry demonstrated a concerning proportion of fatal/liver transplantation patients, which was higher than the proportion for DILI caused by other NSAIDs or non-NSAID agents.[12] These findings demonstrate that ibuprofen, although rarely, can be associated with worst outcome DILI. However, further studies involving a high number of carefully diagnosed ibuprofen hepatotoxicity cases are needed to confirm this and to determine the true incidence rate of worst outcome for this type of DILI.

Our study is based on a comprehensive database search to answer an unmet need for a better understanding of ibuprofen DILI, but it also has limitations. The availability of sufficient information to establish causality varied between cases. For example, presence of concomitant medications was not reported for some cases, which could reduce the diagnostic reliability of these cases. This highlights the importance of implementing and adhering to strict guidelines for DILI case reporting. The DILI criteria and diagnostic process may also have varied as the reporting period spanned across more than 40 years. Moreover, the distinction between hepatotoxicity and hypersensitivity with hepatic manifestation is not well defined. Furthermore, we cannot rule out publication bias and consequently under-representation of less striking cases. The low number of cases retrieved from the literature also implies that limited conclusions on the clinical presentation and outcome of ibuprofen hepatotoxicity can be drawn. Further evidence is required for more reliable conclusions.

In conclusion, ibuprofen-induced liver injury can occur, but the absolute risk of hepatotoxicity associated with ibuprofen is probably very low. It is in fact probably lower than the absolute risks of vascular and gastrointestinal complications. Ibuprofen-induced hepatotoxicity presents mainly as hepatocellular type of liver injury after a short latency period, but other presentations (including hypersensitivity features, cholestatic damage and VBDS) are known to occur. We found a relatively large proportion of patients in our study that died or required liver transplantation. However, the relatively high prevalence of underlying comorbidities including chronic liver disease causes uncertainties with regard to prognosis and causality of the liver failure/mortality cases. Additional studies including a substantial number of carefully diagnosed ibuprofen hepatotoxicity cases are therefore needed. Nevertheless, clinicians should not overlook ibuprofen intake when assessing a suspicion of hepatotoxicity, but be aware that ibuprofen has been associated with DILI in the literature. In line with other forms of DILI, careful follow-up and monitoring of patients suspected of having ibuprofen-induced liver injury is recommended until recovery. Although further studies are required to fully understand the role of ibuprofen in DILI, ibuprofen can be regarded as a safe and efficacious widely available NSAID.