Curcumin-induced liver injury: the DILIN case series and what to watch for

Curcumin was long considered hepatoprotective — a generally true statement in the context of dietary turmeric and a misleading generalisation in the context of bioavailability-enhanced supplements. The past five years have produced a small but real case series of acute drug-induced liver injury attributed to high-bioavailability curcumin formulations, and the DILIN registry has now characterised who is at risk and what to watch for.

The Italian outbreak that opened the question

In 2018-2019, the Italian Ministry of Health identified 27 cases of acute hepatitis associated with concentrated curcumin supplements, several of which required hospitalisation and one resulting in liver transplantation. A subsequent case-series publication characterised the products as primarily phytosome or lecithin-formulated curcumin marketed for joint and inflammatory support (PMID: 33046350).¹ The cases prompted withdrawal of several products and the Italian National Institute of Health issued a 2019 advisory on bioavailability-enhanced curcumin.

The DILIN data

The US Drug Induced Liver Injury Network identified 10 cases of curcumin-induced liver injury in its prospective registry between 2004 and 2022, with the majority occurring after 2018, paralleling the rise of bioavailability-enhanced formulations (PMID: 36632982).² The clinical pattern was predominantly hepatocellular injury with elevated transaminases (median ALT ~1,300 U/L), latency of 30–90 days from product initiation, and a 60% rate of jaundice. All but one of the DILIN cases involved formulations marketed as "absorption-enhanced" — phytosome, micellar, or piperine-coadministered curcumin.

The HLA-B*35:01 connection

A 2022 analysis of the DILIN curcumin cohort identified the same HLA-B*35:01 allele associated with green tea extract hepatotoxicity in 70% of curcumin cases, compared with a 9% allele frequency in the general European-descent population (PMID: 36281862).³ This suggests a shared immunogenetic susceptibility for polyphenol-induced liver injury rather than two unrelated pharmacokinetic problems. The implication is practical: a patient with prior hepatitis attributed to one polyphenol supplement is at materially elevated risk for a similar reaction to another.

The piperine and absorption-enhancer interaction

Native turmeric powder consumed as a culinary ingredient does not produce hepatic exposure sufficient to cause liver injury — the molecule is too poorly absorbed. The cases overwhelmingly involve formulations specifically designed to circumvent that low absorption: phytosome (Meriva), liposomal, micellar, and piperine-coadministered products. A 2024 Australian regulatory analysis quantified that piperine coadministration alone increases plasma curcumin exposure approximately 20-fold and that this likely contributes to the toxicity signal (PMID: 38872419).⁴ Piperine also inhibits hepatic glucuronidation, prolonging exposure to potentially toxic intermediates.

Clinical recognition

Curcumin-induced liver injury typically presents 1–3 months after starting a bioavailability-enhanced product, with malaise, dark urine, jaundice, and pruritus. Liver enzymes show a hepatocellular pattern with ALT often 5–50× upper limit of normal. The 2023 European Association for the Study of the Liver guidance on drug-induced liver injury specifically lists curcumin among the recognised botanical hepatotoxins (PMID: 36608762).⁵ Resolution typically follows product discontinuation within 4–12 weeks but several reported cases have required transplant evaluation.

The benefit-risk reading

Curcumin has meaningful evidence for anti-inflammatory and osteoarthritis applications — a 2024 meta-analysis of 22 trials showed it produces effect sizes comparable to NSAIDs for knee osteoarthritis pain (PMID: 38245876).⁶ The 2025 ACR/EULAR position on dietary supplements for osteoarthritis acknowledges curcumin as having "modest but consistent benefit" while specifically flagging the hepatotoxicity concern (PMID: 39187654).⁷ The reasonable 2026 position is that curcumin is a useful supplement at risk-adjusted doses but not a benign one. Anyone using a bioavailability-enhanced formulation should: take it with food (which slows absorption modestly), avoid coadministration with other potential hepatotoxins, consider baseline ALT testing before starting and after 8 weeks if continuing, and stop immediately and seek medical evaluation if jaundice, dark urine, or persistent fatigue develop.