Evaluating Isoniazid Hepatotoxicity in Rheumatic Disease: Risks and Alternatives

In patients with autoimmune rheumatic disease and prior DMARD‑related liver injury, isoniazid for latent tuberculosis infection (LTBI) was associated with a 27.6% incidence of hepatotoxicity in the reported cohort, with particularly high recurrence after methotrexate exposure. That magnitude of risk shifts the immediate calculus for LTBI prophylaxis in rheumatology patients by increasing the likelihood of clinically relevant transaminase elevations during standard INH courses.

This 27.6% rate markedly exceeds expectations from unselected LTBI populations and underscores the amplifying effect of prior DMARD hepatotoxicity. Additive drug–drug hepatotoxicity and baseline hepatic vulnerability concentrate risk in previously injured livers; the first three months after starting prophylaxis are the critical window for early detection.

Baseline and follow‑up strategies should prioritize objective laboratory and clinical surveillance. Obtain a full LFT panel (ALT, AST, alkaline phosphatase, bilirubin) and perform viral hepatitis screening when clinically relevant, and provide clear symptom education about jaundice, dark urine, anorexia, nausea, and fatigue. For this higher‑risk group adopt a practical cadence: baseline testing, an early check at 2–4 weeks, then monthly monitoring through month three, with immediate evaluation for any new hepatic symptoms. This schedule aims to detect transaminase elevations early enough to prevent progression to severe injury.

Common thresholds to hold or stop treatment include symptomatic transaminase elevation ≥3× upper limit of normal (ULN) or asymptomatic elevation ≥5× ULN; assessment should actively consider recent exposure to other hepatotoxins such as prior DMARDs and alcohol. Prior methotrexate and other hepatotoxic DMARD exposure increases recurrence risk and may lower the practical threshold for selecting a non‑INH regimen in individual patients. In this cohort, rifampicin was associated with fewer hepatic events, but the subgroup was very small (n = 8), preventing reliable comparative safety conclusions; interpret this cautiously and weigh interactions and enzyme‑induction effects when choosing an LTBI regimen.

These data support individualized risk stratification, intensified early monitoring during the first three months, and consideration of non‑INH regimens for patients with substantial prior DMARD‑related liver injury. Shared decision‑making that balances the benefits of LTBI prevention with heightened hepatic surveillance offers an actionable path when treatment proceeds in this high‑risk group.

Key Takeaways:

• Observed INH hepatotoxicity incidence was 27.6% in rheumatic disease patients with prior DMARD‑related liver injury, with higher recurrence after methotrexate exposure.
• Risk in this cohort exceeds historical unselected LTBI expectations and concentrates in the first three months after initiation.
• Monitoring: baseline LFTs and viral hepatitis screening where relevant; recheck at 2–4 weeks and then monthly through month three; immediate testing for hepatic symptoms.
• Stop/hold thresholds: symptomatic ≥3× ULN or asymptomatic ≥5× ULN; assess other hepatotoxins including recent DMARDs and alcohol.
• Rifampicin appeared to have fewer hepatic events in this small subgroup, but interactions and induction effects must be considered.
• Management: individualized risk stratification, shared decision‑making, and close hepatic surveillance during LTBI prophylaxis.