Drug-induced liver injury in inflammatory bowel disease: Challenges in diagnosis and monitoring

Abstract

Drug-induced liver injury (DILI) is an important but often underrecognized complication in the management of inflammatory bowel disease (IBD), particularly in patients receiving long-term immunomodulatory or biologic therapies. Agents such as thiopurines, methotrexate, anti-tumor necrosis factor agents, and newer small molecules including tofacitinib and upadacitinib have all been implicated in hepatotoxicity, with clinical manifestations ranging from asymptomatic elevations in liver enzymes to severe hepatic injury. Differentiating DILI from hepatobiliary disorders commonly associated with IBD, such as primary sclerosing cholangitis, metabolic dysfunction-associated steatotic liver disease, and autoimmune hepatitis, remains a significant diagnostic challenge. The absence of standardized monitoring protocols, coupled with the variable latency and heterogeneous presentation of DILI, further complicates early recognition and management. In this narrative review, we synthesize current evidence on the epidemiology, pathophysiological mechanisms, and clinical spectrum of DILI in IBD. We also outline diagnostic strategies, including the role and limitations of causality assessment tools, and propose practical considerations for baseline evaluation, longitudinal monitoring, and therapeutic decision-making.

Key Words: Drug-induced liver injury; Hepatotoxicity; Immunosuppressive agents; Inflammatory bowel diseases; Monitoring

Core Tip: Drug-induced liver injury (DILI) in inflammatory bowel disease (IBD) is often underrecognized due to overlapping hepatobiliary disorders and nonspecific presentations. Distinguishing true DILI from IBD-related liver disease requires systematic evaluation, multidisciplinary input, and structured monitoring strategies. This review synthesizes current evidence on the epidemiology, mechanisms, and diagnostic challenges of DILI in IBD, highlighting practical approaches to optimize early detection and safe therapeutic decision-making.

INTRODUCTION

Inflammatory bowel disease (IBD), including ulcerative colitis (UC) and Crohn’s disease (CD), is a chronic relapsing-remitting disorder that requires sustained pharmacological interventions to induce and maintain remission and to prevent disease-related complications. In the past two decades, treatment options for IBD have expanded significantly, with immunomodulators, biologics, and small molecules targeting key inflammatory pathways now forming the cornerstone of standard care. Although these therapies have transformed disease outcomes, they are not without risk.

Drug-induced liver injury (DILI) represents an important yet often underrecognized complication in the management of IBD. The liver, as the principal organ of drug metabolism, is vulnerable to diverse hepatotoxic insults. In patients with IBD, this risk is amplified by several factors, including the frequent requirement for polypharmacy, the coexistence of hepatobiliary manifestations, and increased susceptibility to concomitant infections. Clinically, DILI exhibits a heterogeneous spectrum, manifesting as hepatocellular, cholestatic, or mixed patterns of liver injury. DILI can be broadly categorized into two types based on underlying mechanisms. Intrinsic DILI is dose-dependent, predictable, and usually develops gradually over weeks to months of therapy. In contrast, idiosyncratic DILI is dose-independent, unpredictable, and often manifests abruptly shortly after drug exposure[1].

Importantly, the clinical and biochemical features of DILI may overlap with other hepatobiliary disorders commonly associated with IBD, such as primary sclerosing cholangitis (PSC), autoimmune hepatitis, viral hepatitis, and metabolic dysfunction associated steatotic liver disease (MASLD). This overlap introduces significant diagnostic uncertainty and complicates the attribution of liver dysfunction to drug-related causes. Therefore, timely recognition and accurate differentiation of DILI from alternative pathologies remain a major challenge in clinical practice.

DILI also carries significant clinical consequences in the context of IBD. Hepatotoxicity may necessitate interruption or permanent discontinuation of essential therapies, potentially precipitating disease relapse or prompting a switch to alternative agents with different efficacy and safety profiles. In more severe cases, DILI can result in morbidity, hospitalization, and, albeit rarely, progression to acute liver failure or chronic liver disease. Thus, optimizing therapeutic efficacy while ensuring hepatic safety constitutes a critical aspect of IBD management.

The objective of this review is to summarize drug-specific risks of hepatotoxicity in IBD, delineate the diagnostic challenges in distinguishing DILI from competing hepatobiliary pathologies and propose strategies for monitoring for DILI in IBD.

LITERATURE REVIEW

A comprehensive electronic literature search was conducted in PubMed/MEDLINE, EMBASE, Scopus, and Cochrane Library/CENTRAL to identify studies published up to July 2025. The search strategy combined keywords and Medical Subject Headings (MeSH) terms, including: ‘Drug-induced liver injury’, ‘hepatotoxicity’, ‘5-aminosalicylates’, ‘mesalazine’, ‘thiopurine’, ‘anti-TNF’, ‘vedolizumab’, ‘ustekinumab’, ‘tofacitinib’, ‘upadacitinib’, ‘ozanimod’, ‘etrasimod’, ‘inflammatory bowel disease’, ‘ulcerative colitis’, and ‘Crohn’s disease’.

Studies were included if they reported on hepatotoxicity associated with IBD therapies in adult or pediatric populations. Exclusion criteria encompassed non-English articles, conference abstracts without full text, studies not involving IBD, and reports without clear liver-related outcomes.

EPIDEMIOLOGY AND RISK DETERMINANTS OF DILI IN IBD

The true incidence of DILI in patients with IBD remains difficult to determine[2]. This challenge largely arises from the frequent coexistence of hepatobiliary disorders in IBD which complicate the attribution of liver injury solely to pharmacological agents. Nevertheless, accumulating evidence indicates that individuals with IBD carry a higher risk of DILI compared with the general population. Although the overall risk appears broadly comparable between UC and CD, subtle differences are observed, reflecting variations in therapeutic practices, disease phenotype, and associated comorbidities.

Patient-specific risk factors significantly modulate the susceptibility to DILI[3]. Advanced age is associated with nearly a threefold increase in incidence, secondary to age-related changes in pharmacokinetics, diminished hepatic regenerative capacity, and a higher prevalence of comorbidities[4]. Polypharmacy is common in elderly patients with IBD, who often require treatment for cardiovascular, metabolic, and infectious comorbidities. This substantially increases the risk of hepatotoxicity by raising the likelihood of drug-drug interactions and adding to the cumulative hepatic burden. With improved survival and the rising prevalence of elderly-onset IBD, this subgroup is expected to contribute disproportionately to the overall burden of DILI, highlighting an important clinical concern.

Sex also appears to influence the risk of DILI. Although data specific to IBD remain scarce, studies in the general population demonstrate that females have a 1.5-1.7-fold higher risk of hepatotoxicity. This increased vulnerability is multifactorial, driven by sex-related differences in drug pharmacokinetics and pharmacodynamics, hormonal influences, and a greater propensity for aberrant immune responses following drug exposure[5,6].

The gut microbiota has recently emerged as a key determinant of pathogenesis of DILI[7,8]. Dysbiosis, a hallmark of IBD, alters microbial metabolism and affects drug biotransformation and toxicity. The interaction between host, microbiota, and drugs is bidirectional: Impaired gut barrier and dysbiosis increase translocation of bacterial products like lipopolysaccharides, triggering hepatic inflammation, while many IBD therapies-including antibiotics, immunomodulators, and biologics-can further disrupt the microbiome, exacerbating dysbiosis and hepatotoxicity.

Beyond microbiota-related mechanisms, genetic predisposition plays an important role. Polymorphisms in genes encoding drug-metabolizing enzymes and drug transporters have been implicated in interindividual variability in the risk of DILI[9]. Additional host factors such as MASLD, driven by systemic inflammation, metabolic comorbidities, and gut dysbiosis, further predispose IBD patients to hepatotoxicity while complicating diagnosis due to overlapping clinical and biochemical features[10]. Similarly, malnutrition, a frequent complication of active IBD, has been shown to impair hepatic resilience, thereby increasing susceptibility to DILI[11,12].

The interaction between host-related determinants, including age, sex, genetic susceptibility, comorbidities, and gut microbiota composition, and treatment-related exposures highlights the multifactorial nature of DILI. Nevertheless, the intrinsic hepatotoxicity of specific drugs remains a pivotal determinant of clinical outcomes. A detailed appraisal of these drug-specific risks is therefore essential to guide monitoring, facilitate timely recognition of hepatotoxicity, and inform therapeutic decision-making in clinical practice.

DRUGS IMPLICATED IN DILI IN IBD

Aminosalicylates (5-aminosalicylates)

5-aminosalicylates (5-ASA), including mesalazine and sulfasalazine, are first-line therapies in the management of mild-to-moderate UC and, to a lesser extent, in low-risk colonic CD[13]. Although hepatotoxicity with these agents is rare (0%-4%), it remains a clinically relevant concern. In a pharmacoepidemiologic analysis of 4.7 million sulfasalazine or mesalazine prescriptions, only seventeen cases of DILI were reported among patients with IBD[14].

The hepatotoxic potential of sulfasalazine is largely attributed to its sulfapyridine moiety, which can precipitate immune-mediated hypersensitivity reactions. These may manifest as cholestatic disease, granulomatous hepatitis, or, in rare instances, acute liver failure[15]. In contrast, mesalazine and its newer formulations, which avoid sulfapyridine exposure, exhibit a more favourable hepatic safety profile. Importantly, no significant differences in hepatotoxic risk have been observed among the various mesalazine formulations[16,17]. Reported patterns of mesalazine-associated DILI include mild, asymptomatic elevation of transaminases and idiosyncratic hepatocellular or cholestatic injury. Liver biopsy may reveal noncaseating granulomas with infiltration of epithelioid cells[18]. Unlike sulfasalazine associated DILI, mesalazine induced hepatotoxicity is infrequently associated with manifestations of systemic hypersensitivity such as fever, rash, or eosinophilia.

The majority of 5-ASA related hepatotoxic events resolve promptly upon discontinuation of the drug, with rapid normalization of liver biochemistry. In cases of persistent and significantly elevated transaminases or when features of systemic hypersensitivity are present, corticosteroids may be considered if liver injury does not regress after drug withdrawal.

Thiopurines

Thiopurines, including azathioprine (AZA), 6-mercaptopurine (6-MP), and 6-thioguanine (6-TG), are widely used immunomodulators in the management of IBD. AZA is a prodrug that undergoes conversion to 6-MP and subsequently to 6-TG nucleotides (6-TGN), the active metabolite that inhibits purine metabolism in activated T lymphocytes[19].

Hepatotoxicity has been reported in approximately 4%-14% of patients treated with thiopurines[20,21]. The development of hepatotoxicity is closely linked to thiopurine metabolism. 6-MP, either directly or as a metabolite of AZA, is methylated by thiopurine-S-methyltransferase (TPMT) to form 6-methylmercaptopurine (6-MMP), which is subsequently converted to 6-MMP ribonucleotides (6-MMPR). These metabolites are implicated in thiopurine-induced hepatotoxicity. The pattern of thiopurine toxicity depends on TPMT activity. Patients with low TPMT activity accumulate excessive levels of 6-TGN, predisposing to myelotoxicity. In contrast, patients with normal or high TPMT activity produce elevated levels of methylated metabolites, resulting in high 6-MMP/6-MMPR concentrations and an increased risk of hepatotoxicity[19,22]. The presence of pre-existing liver disease, including MASLD, further increases susceptibility to thiopurine-related DILI[23].

Thiopurine-related DILI encompasses a broad clinical spectrum, ranging from asymptomatic biochemical abnormalities to severe vascular complications. The most frequent presentation is transient, asymptomatic elevation of transaminases, which often resolves spontaneously and permits continuation of therapy under close monitoring. Persistent elevations may require dose reduction or drug discontinuation, and in selected cases, strategies such as split dosing or low-dose thiopurine-allopurinol co-therapy may mitigate hepatotoxicity[24-27]. Hypersensitivity reactions, typically occurring within the first three months of therapy, present as acute hepatitis and necessitate immediate drug withdrawal. Cholestatic liver injury, another early idiosyncratic manifestation, usually presents with fatigue and jaundice; pruritus is uncommon. Cholestatic liver injury generally resolves upon discontinuation, rare cases may persist or progress to vanishing bile duct syndrome[28].

Endothelial injury represents a dose-dependent pattern of hepatotoxicity and includes sinusoidal obstruction syndrome (SOS), peliosis hepatis, and nodular regenerative hyperplasia (NRH), typically arising between three months and three years of therapy[29,30]. Both AZA and 6-TG have been implicated in SOS and NRH. Clinical manifestations range from asymptomatic presentation to fatigue, hepatic dysfunction, and portal hypertension. SOS may present acutely with abdominal pain, ascites, hyperbilirubinemia, and elevation of transaminases. NRH is associated with older age, male sex, prior intestinal resection (≥ 50 cm), stricturing disease phenotype, cytopenias, elevated gamma-glutamyl transferase (GGT), elevated alkaline phosphatase (ALP), and macrocytosis[31]. In patient with endothelial injury, continuation of thiopurines is contraindicated, as ongoing therapy may lead to progression to end-stage liver disease.

Methotrexate

Methotrexate (MTX) is used for maintaining remission in corticosteroid dependent CD[32]. The mechanisms underlying MTX-induced hepatotoxicity are not fully elucidated. Proposed pathways include hepatocellular accumulation of MTX metabolites, impaired folate-dependent nucleotide synthesis, and activation of hepatic stellate cells leading to fibrosis. Genetic polymorphisms in folate-metabolizing enzymes, particularly MTHFR C677T and A1298C, have been investigated for their association with MTX toxicity. While some studies suggest an increased risk with the C677T variant, others report no consistent association[33,34].

The incidence of MTX-associated hepatotoxicity, as reflected by elevations in transaminases and treatment discontinuation due to such abnormalities, remains relatively low[35-37]. The majority of patients who develop elevated transaminases while receiving MTX have coexisting risk factors for liver injury, including obesity, dyslipidemia, type 2 diabetes, metabolic syndrome, or significant alcohol consumption[38,39]. Concerns have been raised regarding the potential risk of hepatic fibrosis with long-term MTX therapy, leading to the use of non-invasive modalities such as elastography and FibroScan for assessment of fibrosis in patients receiving MTX[40-42]. However, a meta-analysis demonstrated no significant association between cumulative MTX exposure and the development of liver injury[43]. Overall, the incidence of hepatic fibrosis or cirrhosis in IBD patients treated with MTX is low, and the occurrence of advanced fibrosis remains exceptional[44].

Despite low rates of MTX-related DILI, all patients receiving MTX should receive folic acid supplementation, either 5 mg weekly (ideally 1-2 days after the MTX dose) or 1 mg daily, as this has consistently been shown to reduce the risk of hepatotoxicity[45,46].

Anti-tumour necrosis factor agents

Anti-tumor necrosis factor (anti-TNF) agents, including infliximab (IFX), adalimumab, golimumab, and certolizumab pegol, constitute a cornerstone in the treatment of moderate-to-severe IBD. These biologics have significantly improved patient outcomes by inducing and maintaining remission, promoting mucosal healing, and reducing the need for hospitalization and surgery in both UC and CD. Despite their efficacy, anti-TNF agents have been associated with idiosyncratic and indirect hepatotoxicity.

The pathophysiology of anti-TNF-induced hepatotoxicity remains incompletely defined and is likely multifactorial[47]. Immune-mediated mechanisms are thought to be central, with genetic susceptibility influencing the risk. For example, carriage of the HLA-B39:01 allele has been linked to IFX-related hepatotoxicity[48]. A subset of patients develops an autoimmune hepatitis-like phenotype characterized by antinuclear antibody positivity, hypergammaglobulinemia, interface hepatitis, and lymphoplasmacytic infiltrates, suggesting that tumor necrosis factor-α (TNF-α) blockade may disrupt immune tolerance to hepatic autoantigens[47,49]. In addition, IFX can contribute to indirect hepatotoxicity by reactivating latent hepatitis B virus (HBV) infection due to impaired cytotoxic T-cell-mediated immunity[50]. Mechanistically, TNF-α plays a role in antiviral defence through pathways such as APOBEC-mediated degradation of HBV cccDNA; and blockade of this pathway by anti-TNFs facilitates viral replication and reactivation[51]. Pre-existing liver disease, metabolic syndrome, obesity, hepatic steatosis, and concomitant hepatotoxic medications may further enhance susceptibility to anti-TNF mediated DILI.

From an epidemiological perspective, clinically significant hepatotoxicity is uncommon, although mild biochemical abnormalities are frequently observed. Transient elevations of transaminases or ALP are the most common findings and often resolve spontaneously or after discontinuation of therapy. The usual latency period is between two weeks and six months after initiating treatment[49]. Some patients develop autoimmune hepatitis-like syndrome, which may require corticosteroid therapy if abnormalities persist after IFX withdrawal[52]. Importantly, recurrence is rare when patients are switched to other biologics such as adalimumab. Cholestatic hepatitis is less frequent but can manifest with jaundice and raised ALP, that usually resolve after drug discontinuation. Advanced fibrosis and cirrhosis remain rare[53].

A particularly important manifestation of DILI with anti TNFs is reactivation of HBV, which can potentially lead to acute liver failure, especially in hepatitis B surface antigen (HBsAg)-positive patients. The reactivation rates range from 5%-39%, and is more pronounced with IFX compared with other TNF-α inhibitors. Given the risk, universal HBV screening (HBsAg, anti-HBs, and anti-HBc) is recommended prior to initiating anti-TNF therapy. For both HBsAg-positive and HBsAg negative, anti-HBc positive patients, antiviral prophylaxis with entecavir or tenofovir is recommended throughout the duration of the therapy and continued for at least six months after discontinuation[54,55].

Monitoring during anti-TNF therapy should include liver function tests every 3-4 months, with more frequent assessments in patients with metabolic comorbidities or pre-existing liver disease. Persistent elevation of transaminases (> 3 × upper limited normal), after excluding alternative causes of liver injury, necessitates drug discontinuation.

Anti integrins

Anti-integrin agents, including vedolizumab (VDZ) and natalizumab, are biologic therapies that inhibit leukocyte trafficking to the gut, thereby attenuating intestinal inflammation while preserving systemic immune function. Natalizumab is associated with the risk of developing progressive multifocal leukoencephalopathy, which significantly limits its clinical application. VDZ, in contrast, has demonstrated a favourable safety profile and is widely used in both UC and CD.

With respect to hepatotoxicity, VDZ is generally considered safe, with rare reports of liver injury[56]. In the pivotal GEMINI phase 3 and subsequent open-label extension trials, approximately 4% of patients developed liver enzyme abnormalities, none of which required drug discontinuation[57]. Reported liver injuries with VDZ include mild elevations in transaminases, autoimmune-like hepatitis, and, less frequently, cholestatic injury[58-61]. Importantly, VDZ does not appear to worsen hepatic steatosis[62]. While the risk of HBV reactivation is considered minimal, screening for HBsAg and anti-HBc should be done prior to initiating VDZ[63].

Anti interleukins

Anti-interleukin (anti-IL) therapies represent an important advancement in the treatment of IBD. Ustekinumab, directed against the p40 subunit shared by IL-12 and IL-23, effectively inhibits both Th1- and Th17-mediated immune responses. More recently, selective anti-IL-23 agents, including risankizumab, guselkumab, and mirikizumab, have been approved for IBD, offering greater specificity. Ustekinumab has been associated with rare and typically mild elevations in liver enzymes or, occasionally, an autoimmune hepatitis-like presentation, while clinically significant liver injury remains uncommon[64,65]. Data on the newer anti-IL-23 agents remain limited due to their recent approval; however, preliminary evidence indicates a minimal and generally reversible risk of hepatotoxicity[66,67]. Nonetheless, regular monitoring of liver functions is recommended during therapy.

JAK inhibitors

JAK inhibitors, such as tofacitinib, upadacitinib, and filgotinib, represent an important class of small molecules used in the management of moderate-to-severe IBD. By targeting intracellular signalling pathways downstream of multiple cytokine receptors, these have broad immunomodulatory effects[68,69]. With regard to hepatotoxicity, mild elevations in liver enzymes have been observed, most often in patients with coexistent risk factors for liver disease. Neither registration trials nor real-world studies have demonstrated a significant risk of clinically meaningful or irreversible liver injury with JAK inhibitors[70-75]. It is also important to note that CYP3A4 is the primary enzyme responsible for the metabolism of tofacitinib. Careful monitoring for potential drug-drug interactions is recommended when tofacitinib is co-administered with agents that modulate CYP3A4 activity. Also, routine monitoring of liver function tests is advised during therapy.

S1P receptor modulators

S1P receptor modulators, including ozanimod and etrasimod, are emerging therapeutic options for UC, acting through modulation of lymphocyte trafficking to reduce intestinal inflammation. Data on their hepatic safety remain limited. Elevations in GGT have been reported in clinical trials, though without clear evidence of serious hepatotoxicity[76]. Given the paucity of long-term data and the potential for subclinical liver injury, serial monitoring of liver function tests is recommended for patients receiving S1P modulators.

Corticosteroids

Corticosteroids are frequently used for induction of remission in moderate-to-severe IBD, including acute severe UC. Although short term use of corticosteroids is generally considered hepatically safe, they can, in rare instances, precipitate hepatotoxicity. Reported manifestations include hepatic steatosis, steatohepatitis, and, less commonly, idiosyncratic liver injury[77]. These effects are often reversible upon dose reduction or discontinuation. However long-term corticosteroid exposure may contribute indirectly to metabolic risk factors such as obesity, insulin resistance, and dyslipidemia, thereby exacerbating the risk of MASLD.

Antibiotic-associated hepatotoxicity

Antibiotics are prescribed in IBD for perianal fistulizing disease, pouchitis, or infectious complications. Ciprofloxacin has been implicated in hepatotoxicity, with reported presentations ranging from cholestatic jaundice to fulminant hepatic failure, typically attributed to idiosyncratic hypersensitivity. A delayed and protracted cholestatic hepatitis with ductopenia has also been described in a patient with CD after prolonged therapy[78]. Levofloxacin has also been associated with ductopenia and vanishing bile duct syndrome[79,80]. Given their frequent use in IBD, monitoring for new-onset elevation of transaminases is suggested in patients receiving fluoroquinolones.

Hepatotoxicity of anti-tubercular therapy

The management of latent or active tuberculosis is a critical consideration in patients with IBD receiving biologic therapy, particularly anti-TNF agents, given the risk of reactivation of TB. Standard first-line anti-tubercular therapy (ATT) comprising isoniazid, rifampicin, pyrazinamide, and ethambutol is well recognized for its hepatotoxic potential, with isoniazid and pyrazinamide most frequently implicated. Hepatotoxicity may range from asymptomatic elevations in transaminases to acute hepatitis, and, in rare instances, fulminant hepatic failure[81]. The risk is further amplified in patients with underlying liver disease or concomitant use of other hepatotoxic agents. Systematic monitoring of liver function tests during ATT is essential to ensure timely detection and management of hepatotoxicity.

Table 1 summarizes the hepatotoxic profiles of drugs used in the management of IBD and provides a risk gradient for each category of drugs.

Table 1 Hepatotoxic profile of inflammatory bowel disease therapies.

Drug class	Agents	Reported hepatotoxicity	Risk profile/mechanism	Risk gradient
5-aminosalicylates	Mesalazine, Sulfasalazine	Rare hepatitis, cholestatic disease, granulomatous hepatitis, mild hepatocellular/cholestatic injury (mesalazine)	Sulfapyridine moiety associated with immune hypersensitivity	Low
Thiopurines	Azathioprine, 6-MP, 6-TG	Transient rise in transaminases, hypersensitivity hepatitis, cholestatic liver injury, vanishing bile duct syndrome; endothelial injury (sinusoidal obstruction syndrome, nodular regenerative hyperplasia, peliosis hepatis)	TPMT activity, 6-MMP accumulation; risk increases with concomitant MASLD	High
Methotrexate		Rise in transaminases, steatosis, steatohepatitis	Risk increases with concomitant obesity, diabetes mellitus, alcohol, metabolic syndrome	Moderate
Anti-TNF agents	Infliximab, adalimumab, golimumab, Certolizumab	Rise in transaminases, alkaline phosphatase, autoimmune hepatitis-like syndrome, cholestatic hepatitis, HBV reactivation	Immune-mediated, genetic (HLA-B39:01)	Moderate
Anti-integrins	Vedolizumab, Natalizumab	Rare; mild rise in transaminases, autoimmune-like hepatitis, cholestatic injury	Minimal risk	Very low
Anti-interleukins	Ustekinumab (anti-IL12/23), Risankizumab, Guselkumab, Mirikizumab (anti-IL23)	Rare; mild rise in transaminases, autoimmune-like hepatitis	Favourable hepatic safety profile, long term data needed	Verly low
JAK inhibitors	Tofacitinib, Upadacitinib, Filgotinib	Mild transient increase in transaminases, usually with other risk factors, autoimmune-like hepatitis, HBV reactivation, no serious DILI reported	CYP3A4 metabolism; caution with CYP3A4 modulators, long term data needed	Low
S1P modulators	Ozanimod, Etrasimod	Gamma-glutamyl transferase elevations	Unclear mechanism; clinical significance uncertain, long term data needed	Low
Corticosteroids	Prednisone, Budesonide, Methylprednisolone	Rare idiosyncratic reactions, HBV reactivation, long-term use associated with steatosis, MASLD	Indirect metabolic injury	Low
Antibiotics	Ciprofloxacin (fluoroquinolones)	Rare cholestatic jaundice, fulminant failure, delayed ductopenic cholestasis	Idiosyncratic hypersensitivity	Moderate
Anti-tubercular therapy	Isoniazid, Rifampicin, Pyrazinamide, Ethambutol	Rise in transaminases, acute hepatitis, fulminant failure (isoniazid and pyrazinamide)	High risk with pre-existing liver disease, polypharmacy	High

6-MMP: 6-methyl mercaptopurine; 6-MP: 6-mercaptopurine; 6-TG: 6-thioguanine; ALP: Alkaline phosphatase; HBV: Hepatitis B virus; IBD: Inflammatory bowel disease; MASLD: Metabolic dysfunction associated steatotic liver disease; TNF: Tumour necrosis factor; TPMT: Thiopurine-S-methyl transferase.

CHALLENGES IN DIAGNOSIS OF DILI IN IBD

The diagnosis of DILI in IBD is challenging owing to overlapping clinical manifestations with IBD-associated hepatobiliary disorders, the frequent absence of specific symptoms, and the influence of multiple potential confounders (Figure 1). The key diagnostic challenges are outlined below.

Figure 1 Challenges in the diagnosis of drug induced liver injury in patients with inflammatory bowel disease. AIH: Autoimmune hepatitis; LFT: Liver function test; MASLD: Metabolic dysfunction associated steatotic liver disease; PSC: Primary sclerosing cholangitis; RUCAM: Roussel Uclaf Causality Assessment Method; DILI: Drug induced liver injury; IBD: Inflammatory bowel disease.

Clinical overlap

The spectrum of hepatobiliary disorders associated with IBD, including PSC, autoimmune hepatitis, and MASLD, frequently mimics the clinical and biochemical presentation of DILI[82]. Autoimmune-like hepatitis triggered by biologics, such as IFX, may be indistinguishable from de novo autoimmune hepatitis. Similarly, cholestatic injury patterns can overlap with early PSC. Differentiating DILI from these intrinsic hepatobiliary conditions requires careful longitudinal evaluation and systematic exclusion of alternative etiologies.

Nonspecific symptoms

DILI in IBD is often asymptomatic and is most commonly detected incidentally during routine laboratory monitoring. When present, clinical manifestations such as fatigue, anorexia, or mild jaundice are nonspecific and can easily be misattributed to systemic disease activity, treatment-related adverse effects, or nutritional deficiencies.

Limitations of causality assessment tools

The Roussel Uclaf Causality Assessment Method (RUCAM) remains the most widely used instrument for the attribution of DILI, offering a structured approach that incorporates latency, dechallenge, alternative causes, and response to rechallenge[83]. However, its utility in IBD is constrained by several factors. Patients with IBD frequently present with concurrent hepatobiliary disorders, making exclusion of alternative causes challenging, particularly given the fluctuating and often subclinical course of these conditions. Furthermore, IBD management frequently involves polypharmacy, with simultaneous use of thiopurines/methotrexate and biologics, all with hepatotoxic potential. As RUCAM was designed to assess a single suspect drug, its reliability diminishes when multiple agents are plausible contributors. Additionally, baseline liver enzyme abnormalities due to PSC, cholestasis, or prior drug exposure complicate the timing and pattern recognition required by RUCAM, potentially obscuring atypical or delayed presentations. Importantly, RUCAM places significant emphasis on drug withdrawal and re-exposure, both of which may be clinically impractical or unsafe in patients reliant on continuous therapy for disease control.

Role of imaging and biopsy

Imaging modalities, particularly ultrasound and magnetic resonance cholangiopancreatography (MRCP), are essential for excluding structural or biliary pathology such as PSC but have limited specificity in establishing a diagnosis of DILI. Liver biopsy remains the reference standard in ambiguous cases, providing histological characterization such as autoimmune-like hepatitis, steatosis, or cholestatic injury[84]. However, biopsy findings are not pathognomonic, and their interpretation requires integration with clinical history and pharmacological data.

Confounding factors

Multiple comorbidities may further complicate the diagnosis of DILI in IBD. Chronic viral hepatitis (HBV, HCV), alcohol consumption, metabolic syndrome, and obesity-related MASLD can independently produce biochemical and histological abnormalities. In addition, polypharmacy remains a significant confounder, as patients are often treated with several concurrent medications with overlapping hepatotoxic potential.

MONITORING OF DILI IN IBD: PRACTICAL IMPLICATIONS AND SUGGESTED APPROACHES

Effective monitoring of DILI in IBD requires a systematic and individualized strategy, reflecting the heterogeneity of hepatotoxic risks across therapeutic classes and the frequent coexistence of alternative causes of liver dysfunction (Figure 2). The following practical considerations and structured approaches can optimize monitoring in clinical practice.

Figure 2 Approach to drug induced liver injury in a patient with inflammatory bowel disease. ¹Patients with symptomatic disease have poor outcomes. AST: Aspartate transaminase; ALT: Alanine transaminase; ALP: Alkaline phosphatase; ANA: Antinuclear antibody; ASMA: Anti-smooth muscle antibody; AMA: Anti-mitochondrial antibody; IgG: Immunoglobulin G; HBsAg: Hepatitis B surface antigen; Anti HBs: Antibody to hepatitis B surface antigen; HAV: Hepatitis A virus; HCV: Hepatitis C virus; HEV: Hepatitis E virus; LKM-1: Liver kidney microsome antibody; ULN: Upper limited normal; SOS: Sinusoidal obstruction syndrome; NRH: Nodular regenerative hyperplasia; DILI: Drug induced liver injury; IBD: Inflammatory bowel disease.

Role of causality assessment tools

The RUCAM remains a widely used tool for assessing suspected DILI, but its application in IBD is limited by polypharmacy and the presence of concomitant hepatobiliary disease. In this context, RUCAM should be regarded as an organized aid rather than a definitive arbiter. Its utility is maximized when integrated with expert clinical judgment and multidisciplinary input. Emerging digital or revised RUCAM tools may enhance reproducibility, although they require validation in IBD-specific cohorts.

Broadening the differential diagnosis

Given the clinical overlap between DILI and IBD-related hepatobiliary disorders, an expanded differential is essential. Conditions such as PSC, autoimmune hepatitis, MASLD, and cholestatic injury secondary to sepsis or ischemia should be systematically excluded. Imaging modalities, including MRCP, and liver biopsy may be necessary when uncertainty persists, particularly to differentiate DILI from chronic cholangiopathies or overlap syndromes.

Baseline assessment and serial monitoring

Accurate interpretation of potential hepatotoxicity requires robust baseline liver function tests prior to initiation of therapy initiation. Serial monitoring should follow drug-specific schedules, with closer intervals during the initial 3-6 months and longer intervals once stability is achieved. Evaluation of serial trends of liver enzymes and other biochemical markers is generally more informative than isolated elevations.

Polypharmacy and confounding agents

Many IBD patients are exposed to multiple potentially hepatotoxic agents simultaneously. In cases of suspected DILI, temporary discontinuation of non-essential drugs and therapeutic drug monitoring, when available, can help refine causality. Expert adjudication, akin to approaches used in the DILI network, is valuable when multiple plausible causative agents exist.

Non-invasive tools for assessment of fibrosis, such as transient elastography or shear-wave elastography, may be helpful in monitoring for chronic liver injury. Pharmacogenetic testing (e.g., TPMT and NUDT15 polymorphisms for thiopurines) may provide additional risk stratification, though its role in DILI is adjunctive.

Multidisciplinary management

Optimal monitoring necessitates collaboration between gastroenterologists, hepatologists, and clinical pharmacologists. Joint review of complex cases ensures accurate diagnosis, prevents unnecessary drug discontinuation, and facilitates timely intervention when true DILI occurs.

UNMET NEEDS AND FUTURE DIRECTIONS

Despite increasing recognition of DILI in IBD, substantial gaps remain in its diagnosis, monitoring, and management. No IBD-specific guidelines currently exist, and practice is often extrapolated from general hepatotoxicity recommendations, which may not account for polypharmacy, immunosuppressive regimens, or underlying hepatobiliary comorbidities. Future strategies should focus on IBD-specific algorithms integrating clinical, biochemical, and histological parameters. Artificial intelligence and machine learning models have potential to predict hepatotoxic risk using large datasets incorporating laboratory trends, pharmacogenomics, and comorbidities[85-87]. Novel biomarkers may enable earlier detection and improved risk stratification[88-90]. Large-scale registries and robust pharmacovigilance initiatives are essential for capturing real-world data across diverse therapeutic classes. Ultimately, personalized monitoring strategies, tailored to individual patient risk, are critical to optimize safety while maintaining effective IBD therapy.

CONCLUSION

DILI represents a clinically significant yet frequently under-recognized complication in patients with IBD. Its diagnosis is often challenging due to overlapping etiologies, including IBD-related hepatobiliary disorders, metabolic liver disease, viral hepatitis, and polypharmacy with potentially hepatotoxic agents. Consequently, distinguishing true DILI from alternative causes requires careful clinical assessment, integration of laboratory trends, imaging, and, when appropriate, histological evaluation. Vigilant monitoring is therefore essential, beginning with baseline liver function assessment and risk stratification, followed by serial evaluations tailored to the therapeutic agents used and individual patient risk profiles. Multidisciplinary collaboration between gastroenterologists, hepatologists, and clinical pharmacologists enhances diagnostic accuracy, guides safe continuation or discontinuation of therapy, and optimizes patient outcomes. Looking ahead, development of IBD-specific diagnostic algorithms, incorporation of novel biomarkers, and application of artificial intelligence and large-scale pharmacovigilance initiatives hold promise for advancing precision in risk prediction, early detection, and personalized monitoring strategies.