After 2 years of follow-up in this randomized controlled trial that aimed to compare the effect of two personalized energy-restricted dietary strategies in overweight or obese participants with NAFLD, significant improvements were observed in both intervention groups. Body weight, body composition, glycemic profile, leptin, adiponectin and hepatic status parameters displayed significant improvements at all the follow-up visits of the study, compared to baseline values within both intervention groups. However, the FLiO group showed a greater and more persistent effect in the reduction of some metabolic and hepatic parameters including ALT, liver stiffness and FLI, and a greater increase in adiponectin after 24 months compared to the AHA group.
The current treatment of NAFLD relies on lifestyle modifications, mainly focused on diet and physical activity. Moreover, the achievement of weight loss is considered essential in NAFLD management, especially for subjects with overweight or obesity.[34,35] For instance, the AASLD recommends a weight reduction of at least 3%-5% of body weight to reduce liver steatosis and propose a weight loss of 7%-10% to improve fibrosis and other histological features of NASH. A recent systematic review and meta-analysis evaluated the effect of interventions aiming at weight loss in subjects with NAFLD, including behavioural programs with diet alone or accompanied by exercise, pharmacotherapy or bariatric surgery. The studies had a median duration of 6 months and researchers found a consistent association between weight loss interventions and the improvement in a variety of hepatic biomarkers. However, although clear evidence showed that weight loss interventions were associated with improvement in liver steatosis, no statistically significant change was observed in histological liver fibrosis or NAFLD fibrosis score. This aspect becomes relevant since the degree of fibrosis is considered the strongest predictor correlating NAFLD progression with life-threatening complications. In the current study, both groups achieved the weight loss recommendations of the AASLD for the reduction of liver steatosis at all the timepoints of the study without significant differences between them, but the average weight loss percentage was numerically greater in the FLiO group at 12 and 24 months. Thus, only the FLiO group maintained the weight loss percentage recommended for the improvement of liver fibrosis along the trial. Accordingly, when adjusting the comparisons between groups by weight loss percentage at each timepoint of the study, these numerical differences in terms of weight loss seemed to explain the greater and more persistent improvement of certain parameters in the FLiO group, including ALT levels, liver stiffness and FLI. In addition, dietary factors beyond caloric restriction and weight loss may have influence on the amelioration of liver fibrosis and longer periods of treatment and follow-up may be needed to observe consistent changes induced by the diet in this outcome.
The long-term adherence to dietary treatments and the maintenance of body weight and lifestyle modifications is a considerable challenge for many individuals. Previous studies have found that weight loss is largely unsuccessful in NAFLD patients in ambulatory care settings and that frequent clinical encounters are associated with a successful weight reduction. In the present research, although both groups significantly reduced body weight at all the time points of the study compared to baseline values, a tendency to regain weight was observed, especially in the AHA group. Thus, periodical reinforcement of behavioural changes in lifestyle interventions may be needed to maintain the beneficial metabolic and hepatic effects evidenced in short and medium-term trials. Moreover, the optimal duration of follow-up in patients with NAFLD is still undetermined.
In this context, few studies about the effect of long-term interventions on hepatic outcomes have been published. For instance, a research in postmenopausal women with obesity compared the effects of an ad libitum Paleolithic diet with a conventional low-fat diet on liver fat determined by proton resonance spectroscopy within a period of 2 years and found that body weight and liver fat content were significantly reduced in both groups at 6 and 24 months, with a more pronounced effect of Paleolithic diet at 6 months. Another study evaluated the 2-year effects of a lifestyle intervention with diet and exercise counselling on liver status of middle age and elderly men with NAFLD, compared with usual care. The results showed a greater improvement in steatosis degree evaluated by ultrasonography, ALT and FLI in the intervention group. Our results contribute to this growing body of evidence, showing a variety of beneficial metabolic and hepatic long-term effects in individuals following the AHA or the FLiO diet (both energy-restricted strategies) and evaluated by different noninvasive techniques, including ultrasonography, MRI, elastography and fatty liver indexes. However, the FLiO strategy may provide more persistent benefits in the evaluated outcomes.
Previous studies have shown the benefits of dietary interventions with high adherence to the MedDiet, and high antioxidant capacity in the context of energy-restricted diets for the management of NAFLD in subjects with overweight or obesity. Moreover, dietary patterns such as the MedDiet, the Dietary Approaches to Stop Hypertension (DASH) diet, and the Paleolithic diet have shown favourable effects for the management of some features of NAFLD, with a larger body of scientific evidence for the first pattern.[11,41,42] A sub-study of the CENTRAL 18-month trial showed that a Mediterranean/low-carbohydrate diet induced a greater reduction in hepatic fat content determined by MRI compared to a low-fat diet, while a recent systematic review of randomized controlled trials concluded that the MedDiet could improve NAFLD severity parameters, although inconsistencies among the included studies were found. According to our findings, both groups significantly improved their adherence to MedDiet during the whole duration of the intervention, which may have contributed to produce the beneficial effects observed in the two arms of the study.
To the date, no optimal protein:carbohydrate:fat ratio has been established for the management of NAFLD, although some research has suggested that high-protein diets are more effective to reduce hepatic fat than low-protein diets. Thus, further studies should explore this issue by means of the innovative "nutritional geometry" approach or "Geometric Framework for Nutrition", which aims to integrate nutrients and foods in a model to understand how food components interact to regulate the properties of diets having influence on health and disease.[8,44] In addition to the dietary approach, moderate exercise performed 3–4 times per week with an expenditure of about 400 calories each time by means of a combination of aerobic and restrictive exercise may be recommended to individuals with NAFLD. Moreover, the advances in technology and the use of algorithms developed by artificial intelligence would allow to create personalized lifestyle treatments for patients with NAFLD.
On the other hand, insulin resistance is implicated in both the pathogenesis of NAFLD and the progression from simple steatosis to NASH. Accordingly, the modulation of insulin resistance by means of dietary changes and/or use of medication has been proposed as a potential strategy for the treatment of NAFLD. In this sense, leptin seems to promote insulin resistance and hepatic fibrogenesis, while adiponectin has shown to reduce insulin resistance and to have anti-steatoic and anti-inflammatory effects by increasing free fatty acid oxidation and decreasing gluconeogenesis and de novo lipogenesis, among other mechanisms. Notably, both groups in this research maintained the beneficial significant effects on glycemic profile and insulin resistance evaluated by HOMA-IR at all the follow-up visits. Moreover, leptin evidenced a decrease within both dietary groups along the study when compared to baseline values, while adiponectin significantly increased in the two groups with a significantly greater improvement in FLiO at 6 and 24 months. Hence, both strategies may represent feasible options to improve glucose metabolism, insulin resistance and related adipokines in individuals with NAFLD, although the FLiO diet seems to promote a higher increase in adiponectin, which may entail greater benefits for metabolic and hepatic status.
Interestingly, a bidirectional association between NAFLD and hypertension has been observed and it has been suggested that the change in fatty liver status overtime might influence the incidence of hypertension. In this regard, the AHA group showed a more pronounced and persistent decrease in systolic blood pressure, although a significant difference was not reached when the changes in the two groups were compared. This finding might be expected since the AHA recommendations on diet and lifestyle are intended to reduce cardiovascular risk and aim for normal blood pressure. Thus, the AHA diet might be a suitable option for individuals with coexisting NAFLD and hypertension.
Finally, some limitations of the present study should be acknowledged. First, liver status was evaluated using only noninvasive techniques instead of liver biopsy, which is considered the most reliable approach for detecting NASH and staging fibrosis. However, we carried out a compressive evaluation of the liver by means of validated and widely used techniques for the assessment of liver steatosis (ultrasonography and MRI) and liver stiffness was evaluated as a surrogate marker of liver fibrosis (TE and ARFI elastography). In addition, blood biomarkers and hepatic indexes were used, which facilitates the translation of these findings to clinical and public health settings. Second, the proportion of patients lost to follow-up was high, thus, the sample size was considerably reduced especially after 24 months, which may have produced some bias in the results. Third, the dietary data were obtained using self-reported information of the participants by means of an FFQ, which may produce some bias in the assessment of the diets. On the other hand, the major strengths of the study include the design as a randomized trial, the long-term follow-up for 24 months, and that both strategies included personalized diets and promoted the adoption of behavioural changes and a healthy lifestyle with individual follow-up of the participants.
Liver International. 2021;41(7):1532-1544. © 2021 Blackwell Publishing