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


The applied search strategy led to a total number of 131 published works, which were obtained from the above described databases. Of these, 14 reports were found to be duplicates and another five did not meet the language criteria, and thus were immediately removed. In the next step, 59 records were selected due to being of potential interest for the present analysis based on article titles and abstracts (53 records were excluded as their content fell outside the scope of the current study). Of the 59 full articles, which were carefully assessed, 22 were considered for inclusion in our systematic review. The 37 omitted articles did not include data on human ibuprofen-induced liver injury useful for performing analyses of phenotypic characterisation. The 22 selected articles consisted of 17 case reports and two case series on idiosyncratic ibuprofen-induced liver injury and three additional case reports on ibuprofen overdose-related liver injury, which were all included and thoroughly analysed in the present work (Figure 1).

Figure 1.

Preferred reporting items for systematic reviews and meta-analyses (PRISMA) flow diagram of the report selection in the current study

Demographic Characteristics of Idiosyncratic Ibuprofen-induced Hepatotoxicity

Seventeen published case reports and two case series of idiosyncratic ibuprofen-derived liver injury from 1976 to 2018 were retrieved, carefully analysed and summarised in Table 2 and Table S1.[16–34] Of the 22 identified cases, 12 involved females (55%) and the mean patient age was 31 years (range 7 months—59 years). Thirteen patients (59%) had underlying chronic conditions, which were mainly related to rheumatic disorders (three patients with systemic lupus erythematous, one with juvenile rheumatic arthritis and one with polyarthritis) and hepatic disorders (four patients with hepatitis C virus infection). Ibuprofen was the only administered medication in six cases (27%), whereas it was administered simultaneously with other medications in twelve additional cases. In the remaining four cases, the authors did not provide information on concomitant treatments. The cumulative ibuprofen doses ranged from 0.4 to 180 g (mean 30 g) over a time period of 1–42 days with a mean ibuprofen treatment duration of 14 days. The mean time to onset of the DILI episode was 12 days (range 1–42 days; Table S1).

Clinical, Biochemical and Histological Profile of Idiosyncratic Ibuprofen-induced Liver Injury

Eighteen patients presented clinical manifestations at onset. The most prevalent symptoms were rash (56%), fever (56%), jaundice (50%), choluria (39%), vomiting (39%) and abdominal pain (22%). In addition, four patients were asymptomatic, and the diagnosis of liver injury was based on routine blood tests (Table S1).

The mean values for peak liver tests were as follows: total bilirubin (TBL) 7.6 mg/dL, aspartate aminotransferase (AST) 986 IU/L, alanine aminotransferase (ALT) 968 IU/L and alkaline phosphatase (ALP) 610 IU/L (Table 2). To depict an overview of potential severity of the ibuprofen cases, peak ALT and TBL values were graphed for 13 of the 22 cases (cases 2–5, 9, 11–16, 20 and 22) with available information (Figure 2). This figure also includes 25 ibuprofen-induced hepatotoxicity cases from the Spanish DILI Registry and Latin-American DILI Network, for comparative purposes. These cases, which do not form part of the current systematic review, have been published previously as a cohort study, but not as case reports.[12] Detailed information on these cases can therefore be found in Table S2. Figure 2 shows that a large proportion of cases from both groups had a TBL and ALT level >2 and >5 times the upper limit of normal, respectively, indicating a higher risk of severe outcome (Hy's law). The cases included in the current study, however, demonstrate a higher proportion of worst outcome cases compared to the Spanish/Latin-American cases (31% vs 12%).

Figure 2.

Idiosyncratic ibuprofen-induced liver injury events depicted according to peak alanine aminotransferase (ALT) and total bilirubin (TBL) levels. The hepatotoxicity cases include 13 cases identified from the literature (part of the current study) that are compared with 25 cases from the Spanish and Latin-American DILI Registries that are not included in the current study (case details can be found in Table S2). Worst outcome: death or liver transplantation; ULN: upper limit of normal

Fourteen of the 22 patients (cases 1–3, 5, 10–13, 15, 17–18, 20–22) had hypersensitivity features (fever, rash and/or eosinophilia). However, hypersensitivity features are not specific to DILI and must therefore be considered in the comorbid context of each patient. Only eight of the patients (cases 3, 5, 10–12, 15, 20 and 22) had hypersensitivity features most likely related to the DILI episode.

Liver histology information was available for 15 of the analysed patients (Table S1). These patients had their initial liver biopsies performed 10–63 days after DILI recognition. The primary findings were necrosis in three cases (cases 9, 10 and 22), cholestasis in five cases (cases 3, 5, 10, 11 and 14) and fatty changes were present in three cases (cases 1, 10 and 17). A total of seven cases had bile duct injury with significant bile duct loss in five cases (cases 5, 11, 12, 16 and 20). Mixed inflammatory infiltrate was detected in six cases (cases 5, 10, 11, 15, 20 and 22), while lymphocytic infiltrate predominated in cases 12, 20 and 22, and eosinophilic infiltrate was observed in case 13.

Hepatocellular pattern of liver injury was the most frequently observed injury pattern, with 11 cases presenting biochemical and/or histopathological criteria of hepatocellular injury, while three cases presented cholestatic and three cases mixed liver injury. The remaining five cases provided insufficient data to assess type of liver injury (Table 2). Six of the patients were diagnosed with VBDS after confirmation of compatible hepatic histology (cases 4, 5, 11, 12, 16, 20). In addition, several patients developed clinical manifestations that were associated with drug reaction with eosinophilia and systemic symptoms (DRESS; case 22), Stevens-Johnson syndrome (SJS; cases 3 and 5), or toxic epidermal necrolysis (TEN; cases 12 and 20).

Outcome and Follow-up Information on Idiosyncratic Ibuprofen-induced Liver Injury

Eleven of the analysed ibuprofen-induced liver injury patients fully recovered from the DILI episode (50%) and the mean time to resolution was 15 weeks (range 2–32 weeks). In addition, four patients with underlying HCV infections (cases 6–8 and 19) also recovered from the DILI episode, and liver injury markers returned to baseline values (ALT 105–119 IU/L, elevations due to the underlying chronic condition present prior to the DILI episode) within a mean time of 10 weeks (range 8–12 weeks). Patient follow-up visits ceased prior to complete normalisation for case 16 and no outcome information was available for case 18. One patient had a fatal outcome after suffering massive hepatic fatty metamorphosis and pleural effusion, and died 1 week after onset (case 1), whilst two patients (cases 9 and 22) had ALF and underwent liver transplantation within 10 weeks from onset. Two patients, who developed VBDS and remained deeply jaundiced 12 months after onset despite pharmacological therapy with immunosuppressive agents, were finally referred for liver transplantation (patients 4 and 5). Three cases (patients 6, 17 and 21) had an inadvertent rechallenge to ibuprofen, which triggered a new flare of aminotransferase elevations. These additional episodes subsided after ibuprofen dechallenge and the patients recovered (complete liver profile normalisation for patients 17 and 21, while patient 6 returned to baseline values; Table S1).

Intrinsic Ibuprofen-induced Hepatotoxicity

Three case reports were also found describing liver damage following ibuprofen overdose (cases 23–25).[35–37] Two of the events (case 23 and 25) were suicide attempts with a single ibuprofen intake of 20 and 60 g, while the reason for a single intake of 9.6 g of ibuprofen in case 24 remains unknown. In terms of outcome, one case developed ALF and underwent liver transplantation 4 weeks after onset, one case fully recovered (time to resolution unknown) and the outcome of the third case is unknown as follow-up was lost after 2 weeks (Table 2 and Table S1).