By Sebastian Gallego, MD, and Nancy Selfridge, MD
Dr. Gallego is Clinical Skills Facilitator, Clinical Foundations Department, Ross University School of Medicine, Barbados, West Indies.
Dr. Selfridge is Professor, Clinical Foundations Department at Ross University School of Medicine, Barbados, West Indies.
• Exercise with accompanying weight loss has been shown to improve histopathologic changes associated with metabolic associated fatty liver disease (MAFLD).
• Aerobic and resistance exercise without weight loss have been shown to reduce intrahepatic lipid content, assessed noninvasively.
• In this non-randomized controlled trial, 12 weeks of aerobic exercise reduced histopathologic changes characteristic of MAFLD, namely hepatocyte fibrosis and hepatocyte ballooning.
• Twelve weeks of aerobic exercise also increased cardiorespiratory fitness, an independent predictor of reduced all-cause mortality in patients.
SYNOPSIS: A small interventional study assessed the histological appearance of liver biopsies from patients with metabolic associated fatty liver disease (MAFLD) who completed 12-weeks of structured and supported aerobic exercise. Compared to biopsies from a nonexercising control group, the intervention arm demonstrated some reversal of histopathologic changes caused by MAFLD.
SOURCE: O’Gorman P, Naimimohasses S, Monaghan A, et al. Improvement in histological endpoints of MAFLD following a 12-week aerobic exercise intervention. Aliment Pharmocol Ther 2020;52:1387-1398.
Metabolic dysfunction caused by insulin resistance, type 2 diabetes mellitus, and obesity is associated with an increased risk of cardiovascular disease and metabolic (dysfunction) associated fatty liver disease (MAFLD), previously referred to as non-alcoholic fatty liver disease.
MAFLD has a global prevalence estimated at 25% and is a major contributor to morbidity and mortality from chronic liver disease worldwide.1
The natural history of MAFLD includes potential progression of this chronic liver inflammation to non-alcoholic steatohepatitis (NASH), liver cirrhosis, and hepatocellular carcinoma. There is no known pharmacologic treatment or cure for MAFLD, although both aerobic and resistance exercise have been shown to achieve histological benefit in MAFLD when associated with 7% to 10% weight loss. These strategies comprise the current evidence-based recommendations for management of this condition.
Exercise has been associated with a reduction in morbidity and mortality for several inflammatory conditions, such as cancer, type 2 diabetes mellitus, arthritis, and atherosclerotic cardiovascular disease. Therefore, it makes sense that it might be a helpful intervention for MAFLD. A recent meta-analysis of the effect of exercise interventions without weight loss on MAFLD demonstrated significant improvement in liver steatosis, measured noninvasively, after both aerobic and resistance exercise training interventions.2 However, other recent studies have shown no histologic improvement in MAFLD after exercise interventions.3,4 Thus, O’Gorman et al investigated the effect of a 12-week aerobic exercise program without dietary intervention or weight loss on improving MAFLD histological endpoints and explored the optimal dose, frequency, and type of exercise necessary for that outcome.
The authors further determined the effect of the prescribed exercise intervention on participant cardiorespiratory fitness, physical activity levels, and measures of cardiometabolic health, including body composition, vascular health, glucose metabolism, lipid metabolism, and circulating inflammatory markers. Sustainability of the exercise intervention was determined at 12 weeks and 52 weeks post-exercise intervention.
Twenty-eight participants with biopsy-confirmed MAFLD, all patients attending an outpatient hepatology clinic, were divided into a treatment group (n = 18) and a control group (n = 10), based upon participant preference. Of these, four participants dropped out before the completion of the first follow-up assessment in week 13, two in the treatment group and two in the control group, leaving 16 in the intervention group and eight in the control group for the week 13 data analysis. The age range of the participants was 46 to 77 years (mean 61), the male-to-female ratio was 7:17, and the average body mass index (BMI) was 35.7 kg/m2.
Inclusion criteria for participants included:
- age > 18 years;
- biopsy-proven MAFLD;
- ability to attend bi-weekly, clinic-based exercise classes for the full 12 weeks.
Exclusion criteria for the participants included:
- contraindications to exercise testing or prescription;
- significant orthopedic or neuromuscular limitations;
- unwillingness to participate;
- alcohol consumption > 40 g/day (males);
- Alcohol consumption > 20 g/day (females);
- coexisting liver disease.
All participants were assessed at baseline and at week 13 for the following: dietary intake using a four-day diet diary, hepatic elastography (a noninvasive assessment of hepatic steatosis and fibrosis), and cardio-respiratory fitness using a modified Bruce protocol and estimates of VO2 max. Cardiometabolic analysis measures included fat mass, skeletal muscle mass, waist and hip circumference, liver function tests (LFTs), lipid profile, fasting glucose, hemoglobin A1C, and circulating inflammatory markers (C-reactive protein [CRP], erythrocyte sedimentation rate [ESR], tumor necrosis factor alpha [TNF-alpha], interleukin 6 [IL-6], and interleukin 1 beta [IL-1 beta]). Additionally, the treatment group underwent follow-up liver biopsies at 13 weeks to assess changes in liver histological architecture. Statistical analysis using independent T tests and Mann-Whitney U tests found no differences in means for baseline measures for the control and treatment groups nor differences in baseline liver histology measures. Standard tests for multivariate analysis were applied to assess within-group differences in repeated measures for both continuous and categorical data. Effect size was calculated using eta squared, and statistical significance was set at P < 0.05.
The treatment intervention consisted of a 12-week moderate-to-intense aerobic exercise program, three to five sessions/week (two supervised by an exercise specialist and one to three unsupervised sessions). The exercise program was individualized and graduated in duration and intensity. Supervised group sessions consisted of five to seven minutes of warm-up, 21-42 minutes of moderate-intensity aerobic exercise (increasing over the 12-week study period), followed by a five- to seven-minute cool-down. For unsupervised sessions, participants were sent phone text messages to encourage them to repeat the same format of the supervised sessions and reminding them of the specific duration and intensity of the exercise prescription for the week. Unsupervised sessions were prompted once weekly for the first three weeks, and prompts increased to three per week in weeks 8-12. For the supervised sessions, participants were provided heart rate monitors to gauge exercise intensity, starting at 40% to 59% of heart rate reserve (HRR) and increasing to 55% to 75% HRR by week 9. Participants were trained to use the Borg scale to rate perceived exertion in order to duplicate the same intensity exercise for their unsupervised sessions. The control group was provided standard care. Diet was not changed during the study period in either group.
An assessment at week 13 showed statistically significant increased cardiorespiratory fitness in the treatment group, with the mean VO2 max increasing by 17% compared to the control group (P = 0.027). VO2 max also improved between baseline and week 13 within the treatment group. The treatment group also demonstrated improved cardiometabolic markers, including body mass (2.1% mean reduction, P = 0.038), waist circumference (4.0% mean reduction, P = 0.015), and fat mass (4.9% mean reduction, P = 0.007) compared to controls.
Additional improvements were noted within the treatment group compared to baseline measures for waist-to-hip ratio (2.4% mean reduction, P = 0.008) and increased skeletal muscle mass (3.8% mean increase, P = 0.034), weight loss, and BMI. No patient achieved 7% to 10% weight loss, although 19% of participants in the treatment group achieved 5% weight loss by the end of the intervention period. Histological changes in liver biopsy specimens were assessed for 12 of the 16 patients from the treatment group (four participants declined repeat liver biopsy). Those changes included improvement in liver fibrosis by one stage in 58% (P = 0.034) of patients and decreased hepatocyte ballooning (a characteristic histopathologic finding of steatohepatitis) by one stage in 67% (P = 0.020) of patients. Improvements in liver fibrosis and hepatocyte ballooning were associated with increases in estimated VO2 max by 25% (P = 0.020) and 26% (P = 0.010), respectively. The treatment group demonstrated no changes in hepatic steatosis (P = 1.000), lobular inflammation (P = 0.739), or non-alcoholic fatty liver disease (NAFLD) activity score (P = 0.172). Furthermore, there were no significant changes in LFTs, nor changes in measured circulatory inflammatory markers (CRP, ESR, TNF-alpha, IL-6, IL-1 beta), lipid profiles, or measures of glycemic control. Although participants were encouraged to continue exercising after the 12-week prescribed and supervised exercise intervention, none of these beneficial changes was sustained at a one-year follow-up.
The results from this study supported the existing evidence that exercise alone may improve pathologic liver changes characteristic of the metabolic dysfunction associated with obesity, insulin resistance, and type 2 diabetes. The limitations of the study methodology included a small sample size, nonrandomization of study participants, and an invasive and potentially risky biopsy procedure as part of the baseline and post-intervention assessment that likely limited recruitment. The control group was not subjected to liver biopsy after 12 weeks, and medication checks were not performed for either group after study participants were recruited. Thus, there was no control for changes in biopsy findings caused by non-exercise-related variables. Both conditions introduce a possibility of type II statistical error. Despite histologic improvement in hepatic fibrosis and hepatocyte ballooning noted in the treatment group, no other histologic changes met statistical significance; therefore, it is difficult to judge the clinical significance of the improvement without a longer study period.
The strengths of this study included intervention and control groups with no significant baseline differences despite nonrandomization and a superbly structured group exercise intervention that included a supervised component and had a high level of reported adherence by participants (93%) during the 12-week implementation period. In fact, it is likely that the structure of the intervention program (graduated moderate- to vigorous-intensity aerobic exercise, a supervised group setting, and text message prompts for unsupervised sessions) influenced the positive liver histology endpoints noted in this study that were not apparent in previous exercise intervention studies by Hickman et al and Eckard et al.3,4
Clinicians can capitalize on the findings from this study in clinical practice, citing an additional possible salubrious effect of a committed exercise program. First, this study suggests that adherence to exercise is easier and more likely in a set of patients with metabolic disorder and associated conditions when the exercise program includes a group setting, supervision, and structured encouragement. These are conditions for clinicians to promote when creating exercise prescriptions for patients. Although exercise with moderate weight loss remains the goal and foundation of lifestyle-change counseling for patients with metabolic disorder and MAFLD, patients who have repeatedly struggled with weight loss can be encouraged that exercise alone, performed with sufficient consistency and at least moderate intensity, can improve cardiometabolic risk profile and appears to improve associated fatty liver changes.
A complete list of references can be found online at http://bit.ly/3cMlV9h.