2. Nutrition and NAFLD

Poor nutrition is the leading contributing factor to the development of NAFLD. This means that a healthy diet, combined with weight loss and increased physical activity, forms the most effective therapeutic approach for NAFLD management. In fact, since no standardised or specific medications have yet been approved to treat NAFLD, treatment currently centres around improving patients’ diets and exercise habits.

Dietary intervention (especially the Mediterranean diet) is effective in normalising aminotransferases and lowering intrahepatic fat, whilst exercise improves insulin sensitivity and decreases BMI ^[14][15]. NAFLD patients should be advised to maintain a low-calorie diet and stop alcohol and tobacco consumption ^[16]. A daily caloric restriction of 500–1000 kcal is an extremely effective intervention as primary or secondary prevention against NAFLD. A weight-loss reduction of 3% to 5%, meanwhile, is also associated with decreased NAFLD, but a greater reduction in weight (7–10%) is necessary to achieve NAFLD remission and fibrosis regression. The goal of calorie restriction should therefore be to achieve ≥10% overall body weight loss ^[17], though some authors advise to not exceed 1.6 kg/week weight reduction so as to avoid a worsening of fibrosis and hepatocyte necrosis ^[18].

Dietary intervention (especially the Mediterranean diet) is effective in normalising aminotransferases and lowering intrahepatic fat, whilst exercise improves insulin sensitivity and decreases BMI [64,65]. NAFLD patients should be advised to maintain a low-calorie diet and stop alcohol and tobacco consumption [66]. A daily caloric restriction of 500–1000 kcal is an extremely effective intervention as primary or secondary prevention against NAFLD. A weight-loss reduction of 3% to 5%, meanwhile, is also associated with decreased NAFLD, but a greater reduction in weight (7–10%) is necessary to achieve NAFLD remission and fibrosis regression. The goal of calorie restriction should therefore be to achieve ≥10% overall body weight loss [67], though some authors advise to not exceed 1.6 kg/week weight reduction so as to avoid a worsening of fibrosis and hepatocyte necrosis [68].

The overall goal is to find the best dietary pattern and macronutrient composition to prevent, attenuate, or reverse hepatic steatosis and its progression to steatohepatitis. Diets that can improve IR, oxidative stress, or inflammation are potentially good candidates to treat NAFLD. Below, we summarise what is currently known about the relationship between the core nutritional elements (carbohydrates, fats, and proteins), food groups, and dietary patterns, and the likelihood of NAFLD.

2.1. Carbohydrate Intake and NAFLD

Carbohydrates are classified as simple (fructose, glucose, galactose) and complex (starch). There is evidence suggesting that lower carbohydrate intake (≤40% of daily energy intake) may be beneficial for NAFLD patients ^[19] and that it is preferable to avoid consumption of carbohydrates with a high glycaemic index ^[20]. However, most of the conducted studies focus on fructose consumption and its relationship with NAFLD due to its unique hepatic metabolism and because the rise in the consumption of added sugars, particularly fructose, parallels the increasing incidence of NAFLD (

Figure 1).

Fructose is present in natural foods (fruits, vegetables, honey) and in processed foods (juices, nectars, other beverages) ^[26]. In addition, fructose is a primary component in the most widely used sweeteners (sucrose or high fructose corn syrup [HFCS]). Fructose consumption has increased by 30% in the last 40 years and by 500% over the last century due to the increased consumption of processed foods ^[27]. Sugar-sweetened beverage (SSB) intake increased by more than 40% from 1990 to 2016 ^[28]. Today, added sugar intake makes up 15% of total daily calories in the average Western diet ^[29]. In parallel, there has been a progressively higher incidence and prevalence of obesity, NAFLD, T2DM, and MetS. The increased consumption of added sugars, particularly fructose, is a major underlying cause of chronic metabolic diseases, including NAFLD, T2DM, obesity, hypertension, and CVD ^{[30][31][32][33]}. Numerous experimental and clinical studies have found that high fructose consumption is a major risk factor for NAFLD and its consequences ^[34][35].

Observational studies clearly reveal a close relationship between overconsumption of added sugars and the development of NAFLD in adults and children. A systematic review and meta-analysis of seven studies (six cross-sectional studies and one cohort study) involving 4639 subjects demonstrated that SSB consumers had a 53% increased risk of developing NAFLD compared with non-consumers ^[35]. Another systematic review and meta-analysis of 12 studies involving 35,705 participants showed that higher consumption of SSBs was positively associated with a 40% increase of NAFLD ^[11]. Notably, this study found that consumption of SSBs has a dose-dependent effect on the risk of NAFLD. Specifically, low doses (<1 cup/week), medium doses (1–6 cups/week), and high doses (≥7 cups/week) of SSBs significantly increased the relative risk of NAFLD by 14%, 26%, and 53%, respectively. This result was mainly drawn from a cross-sectional study involving 26,790 Chinese adults and showed how a small amount of soft-drink intake is associated with a 14–16% increase in the prevalence of NAFLD, even when adjusted for the presence of MetS ^[36]. The main limitation of these studies was that they only reflected NAFLD trends in China since almost all data was obtained from the Chinese population. However, other studies with different populations, such as the Framingham Heart Study (

n = 2600), also showed a dose–response association ^[37].

n = 9887) that analysed the influence of all foods on NAFLD development found that dietary intake of added fructose specifically (in the form of sucrose or HFCS) was positively correlated with the likelihood of NAFLD (OR 1.29, CI 95% 0.19–1.40) ^[38]. It is not yet known if there is a safety threshold for SSB or sugary food consumption for NAFLD prevention ^[39].

n = 47) resulted in the accumulation of liver fat as proven by magnetic resonance spectroscopy ^[40]. Even short-term interventions demonstrated that consumption of a high-fructose diet (HFruD) had a significant lipogenic effect on the liver ^[41]. This role was confirmed by a small study that found (

n = 16) administering a high-sugar-hypercaloric diet (>1000 kcal simple carbohydrates/day) to subjects for 3 weeks induced DNL, as measured by the lipogenic index ^[41].

n = 41) with habitual high fructose consumption showed reductions in liver fat and DNL when compared to controls fed an isocaloric diet ^[42]. Furthermore, in a subgroup of the same sample, there was an improvement in metabolic parameters such as diastolic blood pressure, serum triglycerides, and IR ^[43]. While there is consistent evidence for fructose’s causative role in liver fat deposition in a hypercaloric setting, there is conflicting evidence about the impact of carbohydrates on NAFLD in the context of hypocaloric or isocaloric diets. A systematic review and meta-analysis of six observational studies and 21 intervention studies concluded that there was insufficiently robust evidence available to demonstrate that high intake of fructose, HFCS, or sucrose is associated with a higher incidence of NAFLD and suggested that the relationship between fructose intake and NAFLD could be confounded by excessive energy intake ^[44]. In accordance with this, a recent short-term interventional study found that a high-fructose diet (150 g a day for 8 weeks) in an isocaloric context does not have negative health impacts on weight, liver health, cholesterol, insulin sensitivity, or glucose/blood sugar tolerance in healthy individuals ^[45]. In this trial with healthy individuals, no change in intramuscular or intrahepatic fat was observed, but there were higher levels of serum triglycerides and DNL. Additionally, a randomised trial that compared a hypocaloric low-carbohydrate diet with a classical hypocaloric low-fat diet showed comparable beneficial results from both interventions; this suggests that total energy deficit is the mediating factor for decreasing liver fat and that carbohydrate consumption has little impact ^[46].

However, these studies have some limitations ^[44][45][47]. The follow-up period in these trials is short (mostly 4 weeks or less), whereas we know from animal models that fatty liver develops after at least 8–24 weeks on a high-fructose diet ^[48]. Other important limitations are the small sample size, poor quality, heterogeneity of the studies, and the fact that most participants were healthy men, which limits the ability to apply the implications of these findings to other populations at high risk for developing NAFLD (T2DM and obese patients).

It is noteworthy that one cross-sectional study from Finland found an inverse relationship between fructose intake and NAFLD. This is because the population studied obtained fructose mostly through fruits and not through sugary drinks ^[49]. Although fruits contain considerable amounts of simple sugars such as fructose, they are unlikely to induce NAFLD and related diseases for various reasons ^[50]. Fruits have a lower fructose content per gram (compared to soft drinks) and contain beneficial phytochemicals, micronutrients, and fibre (which is a contributor to overall metabolic health).

2.2. Fat Intake and NAFLD

Dietary intervention studies performed in animal and human models have analysed the effect of fats on NAFLD ^[51]. Most studies suggest that high fat consumption plays a role in NAFLD’s pathogenesis ^[52][53]. However, monounsaturated, polyunsaturated, and saturated fatty acids (MUFAs, PUFAs, SFAs, respectively) or trans fatty acids (TFAs) do not exert the same effect on the liver, and the main source of each of them is different.

Interventional studies performed in mice show that a long-term high-fat diet promotes NAFLD development ^{[51][52][53][54][55]}. For example, an 80-week high-fat diet (60% fat, 20% protein, and 20% carbohydrate) in mice led to obesity and IR, while histological analysis demonstrated liver steatosis, cell injury, inflammation, and fibrosis ^[54].

Epidemiological studies show that NAFLD patients typically have greater intakes of SFAs and cholesterol and a lower PUFA intake, compared to matched controls without NAFLD ^[56][57]. The findings of some experimental studies with human beings confirm these results. In an 8-week double-blind randomised trial with obese individuals (

n = 61), SFAs increased liver fat content (50% relative increase) and serum ceramides, whereas PUFAs did not ^[58]. Additionally, a 10-week randomised trial (

n = 67) comparing two isocaloric diets containing Omega-6 PUFAs or SFAs showed that Omega-6 PUFAs supplementation caused a reduction in hepatic steatosis, insulin levels, and inflammatory markers when compared with SFAs, though no changes in weight were observed ^[59]. In another small clinical trial, 38 overweight subjects were overfed for 3 weeks with different macronutrients (unsaturated fats, SFAs, and simple carbohydrates) and showed intrahepatic triglyceride (IHTG) increases of 55%, 33%, and 15% in the SFA, carbohydrate, and unsaturated fat groups, respectively. Furthermore, SFAs induced IR and endotoxemia and significantly increased multiple plasma ceramides ^[60]. This effect was also confirmed in healthy adults taking a palm oil bolus, which resulted in a measurable increase in IHTG, energy metabolism, and IR ^[61].

n = 223) found that most children with NAFLD have insufficient intake of Omega-3 PUFAs (<200 mg/day) ^[62]. Indeed, as precursors of eicosanoids, Omega-3 fatty acids have an anti-inflammatory effect, regulate hepatic lipid composition, and improve IR ^[63][64], whereas Omega-6 fatty acids are proinflammatory and could have a negative effect on NAFLD ^[65].

In fact, many systematic reviews and meta-analyses of randomised controlled trials have addressed the topic of Omega-3 supplementation and its effect on patients suffering from NAFLD. These studies conclude that Omega 3 PUFAs supplementation (>3 g/day) is useful for the reduction of liver fat, hepatic enzymes, BMI, triglycerides, and cholesterol ^[66][67]. This potentially sets Omega-3 food supplementation as a safe, viable, and effective intervention to help treat NAFLD.

Regarding the role of MUFA intake in the prevention of NAFLD, there are heterogeneous results: observational studies suggest a neutral effect ^[68][69], while interventional studies such as the PREDIMED study show a beneficial effect with extra virgin olive oil (EVOO) supplementation (90% of fat in EVOO is MUFA) ^[70]. Some small, short-term randomised clinical trials have also shown clear positive results ^[71][72]. Thus, the consumption of a high-MUFA diet (28% of total caloric intake) for 8 weeks by T2DM patients reduced liver fat by 29% without body weight changes, in comparison to a baseline diet moderately rich in SFAs (13% of total energy). Recently, a double-blind randomised controlled clinical trial with NAFLD patients (

n = 66) found that consumption of 20 g/day of olive oil attenuated fatty liver grade and reduced body fat percentage ^[73]. It has been reported that MUFA may prevent the development of NAFLD by improving plasma lipid levels, reducing body fat accumulation, and decreasing postprandial adiponectin expression ^[74]. Based on these results, the EASL-EAS-EASO Clinical Guidelines recommend the Mediterranean diet for NAFLD subjects due to its high content of MUFAs ^[75].

Finally, TFA consumption has a pro-oxidative effect and is associated with an increased risk for CVD, IR, obesity, and systemic inflammation and may also damage the liver ^[76]. These data are mainly based on population studies such as the Rotterdam study (

n = 3882), which indicates that TFAs found in desserts and processed foods were associated with a higher prevalence of NAFLD ^[69]. Due to its unhealthy effects, there is no human clinical trial to assess the effect of TFA intake in the liver.

2.3. Protein Intake and NAFLD

Little is known about the influence of protein and amino acids on NAFLD since most of the intervention and mechanistic studies have focused on carbohydrates and fats. Furthermore, intervention and observational studies on the relationship between protein and NAFLD show conflicting results. Some studies found evidence that higher protein intakes might have a negative effect on NAFLD, while other studies showed a neutral or positive effect ^[77][78]. These inconsistencies might be explained by the type of protein consumed, as animal protein seems to have a negative effect on NAFLD, while plant protein has an inverse association ^[77]. It is well established that high meat intake, especially of red meats and processed meats, is associated with IR, T2DM, and CVD ^[79][80][81]. NAFLD has also been associated with red and processed meat consumption even when intakes are low ^[82][83]. However, this effect may be because of the saturated fat content, salt, additives, and cooking methods, more than the effect of the protein itself.

2.4. Fibre Intake and NAFLD

The relationship between fibre intake and NAFLD has been assessed in several observational studies and in a small number of small clinical trials with conflicting results. Although most of the studies suggest that high daily fibre consumption is associated with a preventive effect against NAFLD, other authors have not confirmed these results ^[84]. Recently, a study performed in a Dutch population found that subjects with a high fatty liver index (FLI) consumed less fibre daily, compared to those with a low FLI ^[68]. Similar results were also obtained in a subgroup of 6613 US adults participating in the National Health and Nutrition Examination Survey ^[84]. However, it is not fully known whether the protective effect of fibre against NAFLD is due to an indirect action through modulation of the microbiome or if it is a result of the direct anti-inflammatory properties of fibre ^[26].

2.5. Food Groups

Several epidemiological studies have attempted to assess the relationship between different food groups and the risk of NAFLD; however, the results obtained were quite heterogeneous. Recently, Chinese investigators performed a meta-analysis to study the association between eleven food groups (red meat, soft drinks, nuts, whole grains, refined grains, fish, fruits, vegetables, eggs, dairy products, and legumes) and the probability of developing NAFLD. They collected 24 studies (15 cross-sectional and 9 case–control studies) and made the following conclusions: (1) there is a positive association between red meat and soft drink consumption and the risk of NAFLD, (2) there is a negative association between nut consumption and the likelihood of NAFLD, and (3) no significant causal relationship was observed between the other food groups and NAFLD. Although this study had some limitations (reduced number of studies, risk of bias could not be tested, most studies did not stratify food intake), the results obtained were consistent with the current recommended guidelines for treating NAFLD ^[75][38].