Cryptogenic steatotic liver disease: a lean phenotype associated with increased liver-related mortality.

BACKGROUND

Cryptogenic steatotic liver disease (SLD) remains undefined.

OBJECTIVE

We investigated the clinical features and outcomes of cryptogenic SLD, operationalised as lean SLD without recorded cardiometabolic risk factors (CMRFs).

DESIGN

Clinical characteristics were evaluated in the UK Biobank (UKB) magnetic resonance proton density fat fraction (MR-PDFF) cohort (SLD defined as PDFF ≥5%). Findings were assessed in a hepatic steatosis index-based UKB cohort and externally validated in three cross-sectional cohorts, including the National Health and Nutrition Examination Survey and Korean National Health Insurance Service (KNHIS) cohorts. Outcomes were examined using Cox regression in two longitudinal cohorts: the UKB hepatic steatosis index cohort and the KNHIS cohort.

RESULTS

Among 30 847 MR-PDFF participants, 1195 (3.4%) had non-obese SLD (body mass index <25 kg/m²); 13.7% had cryptogenic SLD and 86.3% had lean metabolic dysfunction-associated SLD. Cryptogenic SLD showed less favourable metabolic and liver-related profiles than non-SLD/no CMRF across cohorts. In cohort-specific analyses, cryptogenic SLD exhibited higher contrast-enhanced T1-weighted image values and an increased prevalence of PNPLA3 and TM6SF2 risk variants in the UKB MR-PDFF cohort. Higher fibrosis rates were observed in the magnetic resonance elastography cohort. In longitudinal analyses, cryptogenic SLD was associated with liver-related death in KNHIS (HR 2.5, 95% CI 1.4 to 4.3). A similar but imprecise association was observed in the UKB hepatic steatosis index cohort (HR 13.2, 95% CI 1.9 to 92.4). The KNHIS association persisted after stricter alcohol exclusion.

CONCLUSION

Operationally defined cryptogenic SLD accounted for a substantial proportion of lean SLD. Despite no recorded CMRFs, this phenotype was associated with liver injury markers and liver-related mortality.