A total of 7,275 participants from the Korea National Health and Nutrition Examination Survey (KNHANES) cohort, and 40,418 participants with NAFLD from the National Health Insurance Service (NHIS) cohort were included for the cross-sectional and longitudinal analyses, respectively. For the cross-sectional analysis, the primary outcome was prevalent CKD, defined as an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73m2. For the longitudinal analysis, the primary outcome was incident CKD, defined as the occurrence of eGFR <60 mL/min/1.73m2 or proteinuria (≥trace) on two consecutive measurements during follow-up.

In the KNHANES cohort, prevalent CKD was observed in 229 (6.1%), 48 (2.6%), and 36 (2.1%) participants in the 0, 1-2, and ≥3 exercise sessions/week groups, respectively. The likelihood of prevalent CKD was lowest in participants allocated to the ≥3 sessions/week group (adjusted OR 0.49; 95% CI, 0.33-0.71; P<0.001). During a median follow-up of 5.0 years in the NHIS cohort, incident CKD occurred in 1,047 (9.7/1,000 person-years), 188 (7.3/1,000 person-years), and 478 (7.4/1,000 person-years) participants in the 0, 1-2, and ≥3 sessions/week groups, respectively. The risk of incident CKD was lowest in participants allocated to the ≥3 sessions/week group (adjusted HR 0.85; 95% CI, 0.76-0.95; P=0.004).