Accelerometer-derived moderate-to-vigorous physical activity and incident nonalcoholic fatty liver disease

Background The liver effects of concentrated vs. more evenly distributed moderate-to-vigorous physical activity (MVPA) patterns remain unclear. We aimed to examine the association of accelerometer-measured MVPA and different MVPA patterns with liver outcomes. Methods Eighty-eight thousand six hundred fifty-six participants without prior liver diseases from UK Biobank were included. MVPA was measured by a wrist-worn accelerometer. Based on the guideline-based threshold (>= 150 min/week), MVPA patterns were defined as inactive (< 150 min/week), active weekend warrior (WW; >= 150 min/week with >= 50% of total MVPA achieved within 1-2 days), and regularly active (>= 150 min/week but not active WW) patterns. The primary outcome was incident nonalcoholic fatty liver disease (NAFLD). Results During a median follow-up of 6.8 years, 562 participants developed NAFLD. Overall, there was a nonlinear inverse association of total MVPA with incident NAFLD (P for nonlinearity = 0.009): the risk of NAFLD rapidly decreased with the increment of MVPA (per 100 min/week increment: HR = 0.68; 95%CI, 0.57-0.81) when MVPA < 208 min/week, while moderately declined (HR = 0.91; 95%CI, 0.84-0.99) when MVPA >= 208 min/week. For MVPA patterns, compared with inactive group, both active WW (HR = 0.55, 95%CI, 0.44-0.67) and active regular (HR = 0.49, 95%CI, 0.38-0.63) group were associated with a similar lower risk of NAFLD. Similar results were observed for each secondary outcome, including incident severe liver diseases, incident liver cirrhosis, and liver magnetic resonance imaging-based liver steatosis and fibrosis. Conclusions Regardless of whether MVPA was concentrated within 1 to 2 days or spread over most days of the week, more MVPA was associated with a lower risk of incident liver outcomes, including NAFLD, liver cirrhosis, liver steatosis, and fibrosis, to MVPA more evenly distributed.