The global burden of nonalcoholic fatty liver disease (NAFLD) is rapidly increasing.
This study aimed to evaluate the effect of exercise on intrahepatic lipid (IHL), serum alanine aminotransferase (ALT), body mass index (BMI), and insulin resistance in NAFLD patients.
We searched MEDLINE, Embase, Cochrane CENTRAL, KMbase, and the Korean Studies Information Service System through April 2022. The included studies were randomised control trials (RCTs) of exercise, in which IHL was measured using magnetic resonance imaging in adult NAFLD patients.
Eleven RCTs with 577 participants were included in this meta-analysis. Exercise was significantly associated with a reduction in IHL (mean difference (MD), -2.03; 95% CI, -3.26 to -0.79; P = 0.001) and a decrease in ALT (MD, -4.17; 95% CI, -6.60 to -1.73; P = 0.0008). Regarding the duration of exercise, maintaining exercise for more than 3 months significantly improved IHL (MD, -3.62; 95% CI, -5.76 to -1.48; P = 0.0009), while exercise for less than 3 months did not (MD, -1.23; 95% CI, -2.74 to 0.29; P = 0.11). BMI and insulin resistance did not improve significantly with exercise.
We found that exercise improved IHL and ALT levels in NAFLD patients. The effect of exercise is particularly increased when one engages in exercises that last longer than 3 months.

OBJECTIVE: In populations of black African ancestry (BA), a paradox exists whereby lower visceral adipose tissue is found despite their high risk for type 2 diabetes (T2D). This systematic review investigates ethnic differences in other ectopic fat depots (intrahepatic lipid: IHL; intramyocellular lipid: IMCL and intrapancreatic lipid; IPL) to help contextualise their potential contribution to T2D risk.
METHODS: A systematic literature search was performed in December 2020 to identify studies reporting at least one ectopic fat comparison between BA and one/more other ethnicity. For IHL, a meta-analysis was carried out with studies considered comparable based on the method of measurement.
RESULTS: Twenty-eight studies were included (IHL: n = 20; IMCL: n = 8; IPL: n = 4). Meta-analysis of 11 studies investigating IHL revealed that it was lower in BA populations vs pooled ethnic comparators (MD -1.35%, 95% CI -1.55 to -1.16, I2 = 85%, P < 0.00001), white European ancestry (MD -0.94%, 95% CI -1.17 to -0.70, I2 = 79%, P < 0.00001), Hispanic ancestry (MD -2.06%, 95% CI -2.49 to -1.63, I2 = 81%, P < 0.00001) and South Asian ancestry comparators (MD -1.92%, 95% CI -3.26 to -0.57, I2 = 78%, P = 0.005). However, heterogeneity was high in all analyses. Most studies found no significant differences in IMCL between BA and WE. Few studies investigated IPL, however, indicated that IPL is lower in BA compared to WE and HIS.
CONCLUSIONS: The discordance between ectopic fat and greater risk for T2D in BA populations raises questions around its contribution to T2D pathophysiology in BA.