Dietary alone interventions
The effect of dietary intervention alone for the management of NAFLD and NASH has been evaluated extensively over the past 2–3 decades [42,43]. These studies can be divided into two main types: those that have focused on weight reduction only, and clinical trials that have assessed the role of a particular dietary composition on hepatic steatosis.
Several small and uncontrolled studies have examined the effect of diet-induced weight reduction (without physical activity changes) on plasma aminotransferases levels and hepatic steatosis measured by either liver ultrasound [34,44] or computed tomography (CT) [35,36]. Taken together, a 4–5% weight loss by dietary intervention alone usually improves plasma aminotransferases levels and hepatic steatosis. Unfortunately, these surrogate markers of liver disease have been shown to have weak correlations with histologic findings , and do not allow to make firm conclusions regarding the long-term value of weight reduction by dietary intervention alone in NAFLD.
Table 20.1 summarizes the studies that have assessed the effect of weight reduction by diet alone on hepatic triglyceride accumulation by MRS in patients with NAFLD. As can be observed, independently of the type of diet used, weight reductions ranging from 6.5 to 11.0% have been associated with a significant reduction in liver fat content by MRS in the order of 38% to 81%.
Moreover, Tendler et al.  in a provocative small (n=5) pilot study reported that after 6 months of a low-carbohydrate, ketogenic diet, a reduction of ∼11% of body weight was associated with qualitative improvements in liver histology. Only a few studies have examined the role of dietary composition in addition to negative caloric balance. Ryan et al.  reported significant decrease in plasma aminotransferase levels with a low-carbohydrate diet when compared to a high-carbohydrate regime in patients with T2DM. However, interpretation is difficult as this improvement was associated with a larger weight reduction in the low-carbohydrate diet group. Kirk et al.  found similar reductions in liver fat content and weight after 11 weeks of a low-carbohydrate (high-fat) vs. high-carbohydrate diet. Of interest, short-term assessment after 48 hours of calorie restriction already showed a significant reduction in LFAT (∼20%) in the low-carbohydrate diet when compared to the high-carbohydrate arm, even before any meaningful weight change. In 170 overweight and obese patients with NAFLD treated over a 6-month period with a hypocaloric diet restricting either carbohydrates or fat, Haufe et al.  concluded that both interventions were equally effective in reducing weight and excessive hepatic triglyceride accumulation.
Based on the above findings, it may be concluded that the potential impact of dietary composition is less important than the overall weight reduction achieved. Unfortunately, studies have been of short duration (≤6 months) and few have tried to assess the effect of dietary composition alone during weight-stable diets (summarized in Table 20.2). As can be observed in Table 20.2, isocaloric diets that do not affect overall caloric intake and weight have only a minor impact on hepatic steatosis (∼13–20%) [22,23], suggesting that weight loss per se may be the most important factor in ameliorating hepatic Steatosis.
Physical activity only
Physical activity has also been suggested to be beneficial in patients with NAFLD and NASH . In a large cohort of 813 patients, 54% of patients with biopsy-proven NAFLD were considered to be physically inactive when screened using a self-reported physical activity questionnaire . Of note, the majority of patients with NAFLD reported that they did not engage at all in any physical activity. Though limited by the nature of the data collection (survey compared to direct observation), the above study suggests that increasing physical activity may play a role in the prevention or treatment of NAFLD.
Results among the many exercise studies in NAFLD are difficult to interpret due to heterogeneity in the type of exercise, baseline fitness levels, and amount of weight loss achieved. Moreover, few offer details about the changes in caloric content or dietary composition during the study. Like with dietary interventions, many of these trials have relied on weak surrogate biomarkers for the baseline diagnosis and follow-up of patients with NAFLD, such as liver aminotransferases and/or liver ultrasound [50–52].
Table 20.3 summarizes several studies that have used liver fat by MRS as the primary outcome of exercise-only interventions in patients with NAFLD. With some exceptions [53,54], all of them reported some reduction in hepatic steatosis after moderate exercise training [25–27,41,55]. Bacchi et al.  concluded that when aerobic and resistance exercise lead to a similar body weight reduction both are equally effective in reducing hepatic steatosis in patients with T2DM. In a recent study by Haus et al.  lack of weight reduction after 7 days of aerobic exercise resulted in no changes in total intrahepatic triglyceride content, emphasizing the importance of weight loss to achieve liver fat reduction. As can be appreciated in Table 20.3, most studies have been small and exercise alone (in the absence of weight reduction) was only moderately effective to reduce hepatic steatosis (∼10–20%) [25–27,54]. Patients who also lost weight had a greater reduction in liver fat content.
In a systematic review, Keating et al.  pooled data from 12 clinical trials involving 439 subjects. This work included an heterogeneous combination of studies that evaluated changes following physical activity on either liver aminotransferases, liver US, or hepatic fat content by MRS. Unfortunately, exercise intervention in most of the studies did not reach the minimal recommended exercise for the management of obesity and T2DM. While the effect on liver aminotransferases was inconsistent, the authors concluded that exercise alone had a small positive effect to decrease hepatic triglyceride accumulation in patients with NAFLD.
In summary, though the beneficial role of exercise per se has not been clearly demonstrated, there is no doubt that this intervention is beneficial in patients with NAFLD/NASH and/or T2DM, and that it should be recommended to patients in the management of NAFLD and NASH . More work is clearly needed to establish the type, quality, and intensity of long-term exercise prescriptions in this population.