ATT in Drug-Induced Hepatitis: 2024-2026 Guidelines

Overview

1. When to STOP ATT (Criteria for Withdrawal)

Note: A mild asymptomatic transaminase rise (<5x ULN without symptoms) represents hepatic adaptation and does NOT require stopping ATT - only increase monitoring frequency.

• Murray & Nadel's Textbook of Respiratory Medicine, p. 1216

2. What to Do While Waiting for LFTs to Normalize

• Severe, infectious TB (high bacillary load, early treatment): Start a liver-sparing (non-hepatotoxic) regimen: Ethambutol + a fluoroquinolone (levofloxacin) + an aminoglycoside (streptomycin) - minimum 3 drugs.
• Advanced treatment / low bacillary burden: It is reasonable to hold ALL TB treatment and simply wait for LFTs to recover.
• Exclude other causes before attributing DILI to ATT: viral hepatitis A/B/C/E, alcohol, biliary disease, other drugs.
• Follow serum ALT and INR/PT until they return to baseline.
• Murray & Nadel's Textbook of Respiratory Medicine, p. 1216
• Patel et al. 2025/2026 review - PMC12813275

3. When to RESTART ATT (Trigger for Reintroduction)

• Reintroduce only when ALT falls to <2x ULN (ideally normalized to normal range).
• In patients with pre-existing liver disease, restart when ALT returns to near baseline levels.

4. How to REINTRODUCE ATT - Guideline Comparison

WHO Protocol

• Reintroduce INH, RIF, and PZA simultaneously at full therapeutic doses from Day 1.
• If a second episode of DILI occurs, switch to sequential reintroduction.
• Advantage: Simplest, easiest to implement in resource-limited settings.

ATS (American Thoracic Society) Protocol

• Day 1: RIF full dose (least hepatotoxic of the three)
• Day 8: INH full dose
• Day 15: PZA - only if initial injury was mild (avoid if severe DILI/ALF)

BTS (British Thoracic Society) Protocol

• INH incremental dose first (start low, build up)
• Then RIF incremental dose
• Then PZA last

APASL Protocol

• RIF incremental dose first
• Then INH incremental dose
• Then PZA - only if initial injury was mild

RNTCP (India) Protocol

• Start with RIF first (any of the above sequential schemes); sequential full-dose regimen is most widely practiced in India.
• LFTs rechecked weekly during reintroduction to identify the culprit drug.

• Reintroduce one drug at a time (sequential preferred over simultaneous)
• If symptoms recur or ALT rises, stop the last drug added - this identifies the culprit
• PZA is the most hepatotoxic; avoid reintroduction if the initial injury was severe (ALF, coagulopathy), or if INH + RIF are tolerated alone
• RIF is less hepatotoxic than INH; most guidelines favor starting with RIF
• Murray & Nadel's Textbook of Respiratory Medicine, p. 1216
• Patel et al. 2025/2026, PMC12813275

5. Special Situations

6. Monitoring After Reintroduction

• Recheck LFTs weekly during the reintroduction phase.
• Once the full regimen is re-established and LFTs remain stable, resume standard monitoring frequency.
• If ALT rises again on rechallenge, the last added drug is the likely culprit - discontinue it and consider an alternative regimen without that drug.

Summary Flowchart

ATT started
    ↓
Mild ALT rise (<5x, asymptomatic)
    → Continue ATT, increase monitoring

ALT >5x ULN OR >3x + symptoms/jaundice
    → STOP INH, RIF, PZA (continue EMB)
    → Add liver-sparing regimen (EMB + LFX + aminoglycoside)
    → Exclude other causes of hepatitis
    ↓
ALT <2x ULN (normalized)
    → Reintroduce sequentially (preferred):
       Start RIF (full or incremental dose)
       → Wait 3-7 days, check LFT
       → Add INH (full or incremental dose)
       → Wait 3-7 days, check LFT
       → Add PZA ONLY if initial injury was mild
    → Check LFT weekly
    ↓
If DILI recurs → Stop last drug added (= culprit)
    → Design regimen without culprit drug

Role of ATT in Viral Encephalitis

The Core Concept

Why the Confusion Exists: TBM vs Viral Encephalitis

• Plum and Posner's Diagnosis and Treatment of Stupor and Coma, p. 420
• Murray & Nadel's Textbook of Respiratory Medicine, p. 1200

When Empirical ATT Is Justified (Key Clinical Principle)

"In view of the severity of TB meningitis, a presumptive diagnosis justifies empirical treatment if no other diagnosis can be established promptly."

• Murray & Nadel's Textbook of Respiratory Medicine, p. 1200

1. The Marais scoring system (uniform consensus case-definition) gives a score suggesting "possible" or "probable" TBM (score ≥6 out of 20 without imaging; ≥10 with imaging)
2. A viral cause has been excluded or is being actively ruled out (negative HSV PCR, negative viral serology, atypical temporal MRI)
3. The clinical course is subacute (>5 days) rather than hyperacute
4. The patient is from a high-risk group or endemic area (South/Southeast Asia, sub-Saharan Africa, HIV-positive, immunocompromised, malnourished)
5. Other diagnoses (bacterial meningitis, cryptococcal meningitis, autoimmune encephalitis) have been excluded

• Murray & Nadel's Textbook of Respiratory Medicine, p. 1200
• Index-TB Guidelines, MOHFW India: "The decision to give ATT should be based on clinical features and CSF profile. A negative Xpert result does not rule out TBM."

The Marais Diagnostic Scoring System (Used to Decide on Empirical ATT)

• Definite TBM: culture/NAAT positive from CSF
• Probable TBM: score ≥13 (without imaging) or ≥12 (with imaging)
• Possible TBM: score 6-12 (without imaging) or 6-11 (with imaging)
• Non-TBM: score <6

• Murray & Nadel's Textbook of Respiratory Medicine, p. 1200 (Table 53.1)

ATT Regimen for TB Meningoencephalitis

• Adams and Victor's Principles of Neurology, 12th Edition, p. 730

Role of Corticosteroids Alongside ATT in TBM

• Dexamethasone IV: 0.4 mg/kg/day for 1 week, then tapering over 3-6 weeks
• A landmark RCT (Thwaites et al., Vietnam) showed dexamethasone reduced mortality from 41% to 32% (without affecting residual disability)
• A systematic review/meta-analysis confirmed adjuvant corticosteroids reduced mortality by ~25% in TBM
• Used especially when there is threatened subarachnoid block, raised ICP, or rapidly declining consciousness
• Given only in conjunction with ATT, never alone
• Adams and Victor's Principles of Neurology, 12th Edition, p. 730

Summary: Decision Framework

Patient with encephalitis/meningoencephalitis
        ↓
Is this acute viral encephalitis? (HSV PCR positive, acute onset <5 days, temporal lobe MRI)
        ↓ YES → ATT not indicated. Treat with antivirals (acyclovir etc.)
        ↓ NO / UNCERTAIN
        ↓
Calculate Marais TBM score; review CSF, imaging, TB exposure history
        ↓
Score <6 → Non-TBM likely → Continue viral/autoimmune workup; hold ATT
Score 6-12 (Possible TBM) → Start empirical ATT + corticosteroids
Score ≥12 (Probable/Definite TBM) → Start ATT immediately + corticosteroids
        ↓
Regimen: 2HRZE (or HRZE/E*) / 10HR = 12 months total
Monitor LFTs; add pyridoxine with INH
Continue viral workup in parallel - if viral etiology confirmed, reassess ATT

Bottom Line

• ATT has no role in confirmed viral encephalitis.
• Its role arises when TB meningoencephalitis cannot be excluded - a common scenario in endemic countries where the two conditions mimic each other.
• A presumptive diagnosis of TBM justifies empirical ATT when no other diagnosis can be promptly established, because untreated TBM is uniformly fatal (median time to death ~19 days).
• The Marais scoring system is the standard tool to guide this decision.
• Regimen is 12 months (2HRZE + 10HR) with adjuvant dexamethasone.
• Diagnosis should continue in parallel with treatment using NAAT, culture, CSF ADA, and imaging.