The Impact of Body Weight Change on Liver Histology in Metabolic Dysfunction-Associated Steatotic Liver Disease Across Various Histological Endpoints: A Systematic Review and Meta-Analysis.
Dujinthan Jayabalan, Sugam Dhakal, Jeffrey Chandra et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and random-effects meta-analysis assessed the relationship between body-weight change and liver histology in patients with biopsy-proven MASLD or MASH who had paired biopsies. Across 26 studies involving 1,963 participants, 12% achieved complete MASH resolution without worsening fibrosis overall. Among those with observed weight loss greater than 3%, the proportion was 25% (95% CI: 0.20-0.30; P < 0.001). At least one-stage fibrosis improvement without worsening MASH occurred in 17%. Outcomes did not differ significantly by baseline BMI, type 2 diabetes prevalence, or Hispanic/Latino ethnicity.
Key findings
- Complete MASH resolution without worsening fibrosis occurred in 0.12 overall and in 0.25 with observed weight loss >3% (95% CI: 0.20-0.30, P < 0.001).
- At least one-stage fibrosis improvement without worsening MASH occurred in 0.17 (95% CI: 0.13-0.22).
- Baseline BMI, type 2 diabetes prevalence, and Hispanic/Latino ethnicity were not associated with significant differences in the reported histological outcomes.
- Mean age contributed significantly to heterogeneity in MASH resolution estimates (P = 0.021, R 2 = 43%).
Why it matters
The findings quantify histological responses associated with weight change and suggest that MASH resolution may occur with weight loss exceeding 3%. The lack of significant differences across several reported patient characteristics supports the relevance of weight management across these groups, while the pooled estimates may help contextualize histological outcomes in future studies.
What to keep in mind
The analysis pooled studies with substantial heterogeneity for MASH resolution (I 2 = 76%) and fibrosis improvement (I 2 = 62%); the reported weight-loss association does not establish causality.