- Design
- Multinational retrospective cohort of biopsy-confirmed MASLD (Global MASLD project, 41 countries)
- Population
- 18,326 adults; 6.7% lean by body mass index
- Primary outcome
- All-cause mortality and clinical events (hepatocellular carcinoma, decompensation, transplant, death)
- Effect
- Lean vs overweight/obese: adjusted HR 0.90 (95% CI 0.69 to 1.16) for mortality; advanced fibrosis adjusted HR 2.10 (95% CI 1.71 to 2.57)
This multinational retrospective study drew on biopsy-confirmed metabolic dysfunction-associated steatotic liver disease from 41 countries, with 18,326 patients. Only 6.7% were lean by body mass index, and lean disease was most prevalent in Asia (11.3% by BMI).
Compared with overweight or obese patients, lean patients had less type 2 diabetes (38.2% vs 48.3%), less advanced fibrosis (29.3% vs 37.2%) and a lower activity score. Even so, outcomes were similar: all-cause mortality was 1.08 vs 0.91 per 100 person-years and clinical events 1.51 vs 1.43. After adjustment, lean status was not independently associated with mortality (HR 0.90, 95% CI 0.69 to 1.16) or clinical events (0.94, 0.76 to 1.17).
Advanced fibrosis (F3 to F4) was what mattered: adjusted HR 2.10 for mortality and 3.41 for clinical events. Transient elastography was more accurate for advanced fibrosis in lean patients (AUC 0.87 vs 0.83), whereas FIB-4 was slightly less accurate (0.76 vs 0.79). The data are retrospective and from biopsied patients, who are a selected group.
- Do not rule out fatty liver disease in a person of normal weight with metabolic risk factors.
- Stage fibrosis in lean patients too; fibrosis, not weight, tracked outcomes.
- Consider liver stiffness measurement where FIB-4 is borderline in a lean patient.
- Look for diabetes, raised triglycerides and central adiposity even when BMI is normal.
Why it matters
It challenges the assumption that fatty liver in a lean person is a lesser disease.
Don't overread it
The cohort was retrospective and biopsy-based, so it may not represent lean patients in the community who are never biopsied.
The statistics, in plain English
Adjusted hazard ratios near 1.0 (0.90 and 0.94, with intervals spanning 1) mean that once fibrosis and other factors were accounted for, being lean did not show a clear difference in outcome. The advanced fibrosis hazard ratios (2.10 and 3.41) are large and their intervals sit well above 1.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for top clinical updates, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free