NAFLD as a Sexual Dimorphic Disease: Role of Gender and Reproductive Status in the Development and Progression of Nonalcoholic Fatty Liver Disease and Inherent Cardiovascular Risk

One large sample study based on an electronic health file database reported that the risk of “recorded” NAFLD/NASH diagnosis increased linearly with BMI, was higher in males than females and in those with T2D [87]. However, the proportion of true NASH cases among the “recorded” diagnosis of NAFLD/NASH was unclear.

A longitudinal study in subjects with biopsy-proven NAFLD found that sex was not an independent risk factor for the progression of fibrosis [88]. In agreement, a systematic review has shown that age and hepatic necroinflammation are the only independent predictors of the development of advanced fibrosis in NASH patients, while other parameters such as MetS features and sex are not [89].

However, conflicting with the above studies, data from cross-sectional studies (Table 2) [29, 57, 65, 90‐97], which are based, in the majority of cases, on a histological diagnosis of NASH, tend to suggest that the risk of NASH and advanced fibrosis is indeed higher in females than males independent of metabolic factors [65, 90, 93, 94], and only a few studies conflict with the above findings [91, 92, 95].

As is easily foreseeable, obese and postmenopausal women, compared to premenopausal and non-obese women, suffer from a remarkably higher prevalence of NAFLD and NASH as a result of a worse metabolic profile [82]. At variance, a recent study reported that, compared to men and postmenopausal women, the risk of lobular inflammation and hepatocyte injury was significantly increased both in premenopausal women and in those taking synthetic hormones such as oral contraceptives and HRT [97]. Given the supposed effects of estrogens on metabolic health and liver injury, the findings of this study appear counterintuitive. It is noteworthy that the study by Yang et al. has several limitations, including some important sources of potential bias, such as the restricted enrollment at tertiary academic centers, menopausal category and synthetic hormone use were self-reported, and information on cumulative estrogen and/or progesterone exposure and serum hormonal levels were lacking. Of note, the authors highlight that, despite increased liver injury and inflammation, premenopausal women were at decreased risk of liver fibrosis compared with men and postmenopausal women. Moreover, sensitivity analyses separately assessing the impact of progesterone use and estrogen use clearly suggested that only the former was associated with liver damage. Collectively these findings provide novel hints regarding a potential multifaceted impact of sex hormones in NAFLD.

Consistent with the hypothesis that estrogens do exert a beneficial effect on NAFLD, menopause has been independently associated with significant fibrosis both in women with NAFLD and in an experimental zebrafish steatosis model [29]. A recent study has shown that men display a higher risk of advanced fibrosis compared to premenopausal women, while after menopause both sexes show a similar severity of liver fibrosis, suggesting that estrogens protect from the development of fibrosis [95]. A subsequent study limited to non-obese women with biopsy-proven NAFLD confirmed that, even in non-obese NAFLD patients, postmenopausal women still had more severe fibrosis, when compared to premenopausal subjects [98]. Accordingly, a study conducted in 488 postmenopausal women with biopsy-proven NAFLD has shown that, independently of age and metabolic factors, the longer the estrogen deficiency in postmenopausal status is (i.e., premature menopause and time from menopause), the higher the risk of fibrosis is [96].

The role of gender in influencing liver-related mortality in NAFLD is still uncertain, although some studies have reported that men are at an increased risk [99‐101].

A universal feature of HCC is the striking male prevalence, with a male/female ratio averaging 2:1 to 7:1; the latter proportion is more often found in HBV-related HCC [102]. Nevertheless, compared to viral etiologies of HCC, NAFLD-related HCC was found to be associated with the lowest male/female ratios in one study [103]. NAFLD-related HCC may also occur in non-cirrhotic livers, but this seems to be more likely in men [104]. The sexual dimorphism in HCC is also maintained regarding prognosis, with women showing better survival rates [102, 105]. However, menopause seems to attenuate these advantages [102, 105].

In summary, the incidence of NAFLD is higher in men than in women, and some longitudinal studies indicate that male gender is an independent predictor of NAFLD development. The prevalence of NAFLD is globally higher in men than women, but after menopause women display a similar or even a higher prevalence compared to men, a finding supporting a protective effect of estrogens. In cross-sectional studies, male gender and menopausal status have often been associated with the risk of NAFLD, independent of age and metabolic factors. The prevalence of NASH and advanced fibrosis has been found to be higher in postmenopausal women than in men; however, longitudinal studies have failed to support a role for gender in influencing the progression of liver fibrosis. Finally, HCC is definitely more common in men than in women in cases due to viral etiology; NAFLD may probably lower the male/female ratio in the risk of developing HCC and it is possible that gender affects the prognosis of HCC.

Several longitudinal studies investigating the association between CVD and NAFLD have reported multivariate analysis models, which have been adjusted for multiple confounders, including sex; as a result of this, the influence of gender in the risk of CVD cannot be evaluated in such studies (reviewed in [106, 107]). A large population-based American study (NHANES III) found that NAFLD assessed with ultrasound, together with male sex, age, race, and metabolic factors independently predicted incident CVD [109]. Another study addressing CHD as a prespecified outcome found that patients with NAFLD had a higher 10-year risk for CHD, as calculated by the Framingham risk score, than the matched control population. This estimated risk was higher in men than in women and more CVD events were reported among men than women at the end of 10 year follow-up, although this finding was not statistically significant [110].

Many cross-sectional studies in which the diagnosis of NAFLD was based on liver imaging studies (either ultrasound or CT scanning) showed an independent association of male sex with coronary artery calcifications [111] as well as significant CHD [112‐115].

Studies in which the diagnosis of NAFLD was based on surrogate indices, such as otherwise unexplained raised liver enzymes, found that sex modulates the association of NAFLD with incident CVD/mortality. For example, most population-based cohort and meta-analytic studies reported an independent association between raised GGT and incident CVD in both sexes [116]; conversely, a population-based cohort study from Germany found that increased GGT was associated with higher risk of all-cause and CVD mortality only in men but not in women, and this association was stronger in men who also had ultrasound scanning findings compatible with steatosis [101]. Increased ALT has been variably associated with incident CVD either in both sexes [117, 118] or in men only [119]. A recent study found that ALT levels independently predicted insulin sensitivity only in women, suggesting that this gender-specific association might explain the sex difference in the predictive role of increased ALT for CVD [120]. Finally, a recent large Korean cross-sectional study reported that men had more prevalent ultrasonographic fatty liver disease, carotid plaques, and increased carotid intima-media thickness (IMT) values than women, but ultrasonographic fatty liver disease independently predicted subclinical carotid atherosclerosis (IMT and plaques) in women only [121].

No gender difference has been reported in the association between NAFLD assessed with liver ultrasound and incident CVD/mortality. A study based on a national Danish registry showed that patients with a hospital discharge diagnosis of NAFLD had a higher all-cause mortality, including liver and CVD related, which was similar among sexes [122]. A study conducted in a community-based Japanese cohort of 1637 apparently healthy subjects found that the diagnosis of NAFLD based on ultrasound findings was a predictor of CVD in both men and women [123].

The only study carried out in postmenopausal women found a significant correlation between NAFLD (based on CT scanning findings) and prevalence of coronary artery calcification (CAC); however, NAFLD was not independently associated with CAC in these postmenopausal women [124].

In summary, male patients carry an increased risk of CVD; moreover, NAFLD seems to be associated with CVD independently of metabolic factors in both sexes. Few data are available in postmenopausal women and studies should specifically be conducted to ascertain whether NAFLD is a specific/independent cardiovascular risk factor in this population of patients.

A pioneering study supporting the notion that gender-dimorphic risk factors are associated with NAFLD was published in 2000. This study, by evaluating 199 individuals, found that elevated BMI was an independent predictor of fatty liver in either sex. Glucose area under the curve and a central-type body fat distribution predicted fatty liver only in women [126]. Similarly, insulin sensitivity has also been reported to be strongly associated with gender-dimorphic risk factors, i.e., fasting insulin and leptin levels (but none of the liver enzymes) in men versus BMI and ALT in women [120]. Of note, Suzuki et al. demonstrated that the associations between anthropometric measures (regional adiposity) and degree of fibrosis clearly differ between premenopausal women and postmenopausal women/men [127].

In women, a complex interaction including genetic polymophisms, dietary habits, endogenous sex hormones, age at menarche, menopausal status, dysmetabolic traits, and HRT modulates the risk of developing NAFLD, NASH, and fibrosis [84, 128‐130].

Early menarche has been associated with higher alanine aminotransferase, C-reactive protein, triglyceride levels, BMI, waist circumference, adult diabetes, cardiovascular morbidity and mortality, advanced liver disease, and HCC [131]. A recent Chinese study conducted in postmenopausal women has identified early menarche as a potential risk factor for NAFLD later in life; consistently, late menarche protects from NAFLD [130]. These associations were significantly attenuated after adjustment for current BMI or HOMA-IR, suggesting that obesity and insulin resistance may partly mediate the association between age at menarche and NAFLD [130]. The biological mechanism linking early menarche with increased risk of NAFLD is far from being clearly elucidated. It has been suggested that early maturation may determine a longer duration of positive energy balance and a greater accumulation of body fat [132]. Consistently, a large cross-sectional study among middle-aged Korean women confirmed that the inverse association between age at menarche and NAFLD was partially mediated by adiposity [133]. Again, a recent study from the American CARDIA cohort showed that early menarche was associated with NAFLD and visceral and subcutaneous abdominal ectopic fat depots in middle adulthood; these associations were attenuated after adjustment for weight gain between young and middle adulthood [131]. Finally, a Chinese study suggested that the presence of central obesity and MetS, but not NAFLD, after menopause was predicted by longer duration of menstruation and early menarche [134].

Compared to the fertile age, menopause increases the risk of NAFLD and liver fibrosis [135] via long-standing estrogen deficiency associated with ovarian senescence and dysmetabolic features such as T2D, hypertriglyceridemia, and central obesity [29, 81]. Consistently, HRT protects from NAFLD development [84], and oophorectomy in young women with endometrial cancer independently increases the risk of NAFLD together with the development of T2D and hypercholesterolemia [136]. These findings are in agreement with a study suggesting that, in HCV infection, increasing severity of fibrosis is associated with a higher BMI, advanced steatosis, and the menopause and that, conversely, menopausal women receiving HRT exhibit a lower stage of fibrosis [137]. Collectively, these data support the notion that estrogens have antifibrogenic properties in humans. This antifibrotic activity may occur by triggering anti-inflammatory, antioxidant, and antiapoptotic molecular pathways [138], which are mimicked by exercise training [139].

Of concern, however, young women in their reproductive age and those exposed to female synthetic hormones (oral contraceptive or HRT) are not completely spared the risk of developing NAFLD and, indeed, they tend to have more severe hepatocyte injury and inflammation. Notably, despite the possibility of enhanced hepatocellular damage, premenopausal women have been consistently reported at decreased risk of liver fibrosis compared to men and postmenopausal women. Moreover, sex hormones exert complex and variable effects on human NAFLD; indeed, the detrimental pro-inflammatory impact may be conveyed by progesterone, but not estrogen [97].

Table 3 summarizes the physiological role of estrogen, progesterone, and androgens according to gender, age, reproductive status, and obesity [140‐157].

Obesity, which is closely linked with NAFLD, mimics hypercortisolism. It is of interest that despite the prevalence of obesity being higher among women than men, the former are somewhat protected from the associated cardiometabolic consequences; however, this wanes after menopause, suggesting a role for estrogens. Mouse models suggest that sexually dimorphic expression and activity of glucocorticoid metabolizing enzymes may have a role in the differential metabolic responses to obesity in males and females [158].

Biochemical, genetic, and therapeutic studies have provided robust evidence for 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) being key in the pathogenesis of NAFLD [159‐161]. It is of interest, therefore, that 11β-HSD1 gene expression is regulated in a tissue-specific and sexually dimorphic manner. In particular, intact rats exhibit hepatic 11β-HSD1 mRNA levels 18-fold lower in the female than the male [162].

Regional adiposity displays a typical sexual dimorphism in humans. Women have a larger capacity to store fat in the subcutaneous compartment [163]. Men generally have twice as much visceral fat compared to women for any given fat mass value [164]. This is relevant given that, compared to subcutaneous adipose tissue depots, visceral adipose tissue depots in general display greater secretory capacity and a more pro-inflammatory profile [165]. Moreover, visceral adipose depots release free fatty acids (FFAs) directly into the portal blood and thus potentially overload the liver [166]. Not surprisingly, those phenotypes which occur whenever, due to hormonal effect, peripheral adiposity is relatively restricted and visceral adiposity is expanded (i.e., male obesity and postmenopausal women) have all been associated with NASH, and fibrosing NASH [127, 167‐169].

Steatosis will invariably occur as a result of an imbalance among enhanced steatogenesis and decreased capacity of oxidation of fatty acids [170]. Moreover, qualitative changes in the hepatic lipidomics may concur with lipotoxicity. Data suggest that all these mechanisms are under the control of sex hormones. For example ovariectomy in rats is associated with increased intrahepatic steatogenesis which occurs through a decreased synthesis of peroxisome proliferator-activated receptor (PPAR), and an increased transcription of genes encoding sterol regulatory element-binding protein 1 [(SREBP-1), a nuclear transcription factor and master regulator of the endogenous synthesis of cholesterol, fatty acids, triglycerides, and phospholipids] and stearoyl coenzyme A desaturase 1 [(SCD1), the rate-limiting enzyme for generating monounsaturated fatty acids (MUFA) from saturated fatty acids protects from hepatocyte lipotoxicity]; all these effects of ovariectomy are prevented by 177β-estradiol replacement, indicating a role for estrogens in the prevention of hepatic fat accumulation [171]. Consistently, hypoestrogenemia is associated with hepatic steatosis through changes in gene expression of molecules related to fat oxidation and lipogenesis; resistance and endurance training prevent this both in rats and human models [172‐175]. Moreover, a normal activity of stearoyl-CoA desaturase (SCD), the rate-limiting enzyme for generating monounsaturated fatty acids from saturated fatty acids, protects from hepatocyte lipotoxicity. Indeed, in mouse models of NAFLD, either genetic manipulation or dietary changes that inhibit the activity of SCD promote fibrosing NASH via hepatocyte lipotoxicity, despite inhibiting obesity and improving insulin resistance [176, 177]. Interestingly, ovarian hormones are also involved in MUFA biosynthesis, via SCD1 [178].

Excess 16:0 fatty acids associated with de novo lipogenesis from high carbohydrate diet inhibits the synthesis of highly unsaturated fatty acids; this may potentially account for the improved metabolic profile which results from supplementing a hypercaloric diet with preformed eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) [179]. For example, findings from a genetically engineered NASH mouse model fed a Western diet have shown that dietary DHA was superior to EPA in attenuating Western diet-induced hyperlipidemia and hepatic injury and at reversing those deleterious metabolic and histological effects induced in the liver by a Western diet [180]. Interestingly, human (women using the contraceptive pill or HRT and transsexual subjects) and experimental data in rats consistently indicate that sex hormones act to modify plasma and tissue n-3 PUFA content, possibly by altering the expression of desaturase and elongase enzymes in the liver [181]. Collectively, these findings suggest that enzyme activities which are key in the development of a potentially hepatotoxic lipidomic signature are critically controlled by sex hormones. Finally, it is of clinical importance that the therapeutic activity of PPAR-α agonists in obesity and fatty acid oxidation is modulated by sex and estrogens [182].

Reactive oxygen species (ROS) result from the oxidation of fatty acids within mitochondria and peroxisomes. The antioxidant defenses which physiologically constrain oxidative stress are hindered by nutritional deficiencies or changes in the intestinal microbiome that limit availability of choline [183]; by aging and the content of cysteine in diet, which, in turn, will affect the intrahepatic synthesis of glutathione [184]; and by sex and menopause which both affect the normal metabolism of choline [185].

Quali-quantitative changes in intestinal microbiota (dysbiosis), which are often associated with dietary indiscretions, have consistently been associated with increased gut permeability which could lead to increased translocation of bacterial products from the intestinal lumen into the portal circulation, thereby triggering chronic inflammation [170]. It this context, it is relevant that in an experimental model of liver failure due to endotoxemia after hepatectomy, sexually mature female rats are more exposed than males to endotoxin-induced liver injury and that ovariectomy abrogates this sexual dimorphism [186]. Whether this also applies to human NAFLD, however, remains to be proven.

Whether metabolic inflammation is a gender-dimorphic phenomenon has not been explored in a systematic and organized manner. Obesity, however, will typically exhibit changes in the innate and adaptive immune mechanisms [187]. However, how mechanistically such obesity-related dysregulation of immune defenses will impact on the pathogenesis of NASH is not fully elucidated. Experimental data obtained in the ob/ob mouse model suggest a role for natural killer T cells (NKT cells) which may at least partly account for the interindividual variation in immune responses being a key modifier in the development and progression of NAFLD [125]. In this respect, it should be highlighted that sex is a major determinant in the immune response. Supporting this notion, one study in young children reported that immune responses to vaccines were consistently higher or equivalent in girls compared with boys [188].

Liver fibrosis is the end-stage result of various liver injuries. In recent times, importance has been given to the Hedgehog signaling pathway on the grounds that activation of this pathway will stimulate the proliferation and differentiation of hepatic stellate cells (HSCs) as myofibroblasts (MF-HSCs), and, conversely, inhibiting Hedgehog activity in myofibroblasts derived from HSCs causes them to revert to a more quiescent (namely less fibrogenic) phenotype [125]. Accordingly, Hedgehog ligands and other factors that control fate decisions in HSCs are critical determinants of the development of cirrhosis due to various causes as well of the course of NASH [125]. For example, hepatic expression of Hedgehog ligands and Hedgehog pathway activity progressively increase from simple steatosis, to NASH, and reach highest levels in NASH-cirrhosis [189]. Of major significance to the pathogenesis of NAFLD, the Hedgehog pathway is strongly regulated by lipids [190], and conversely, Hedgehog signaling is a master regulator of body fat distribution and glucose metabolism [191, 192]. These findings suggest that interindividual variation in Hedgehog signaling might contribute to variability in both hepatic and extrahepatic outcomes of the metabolic syndrome [125].

Experimental overexpression of Hedgehog ligand in hepatocytes is able to induce a fibrogenic liver response and to promote hepatocarcinogenesis in a transgenic mouse model [193]. At least in part by modulating intermediary metabolism [194, 195], Hedgehog also interacts with pathways which regulate growth and differentiation [194, 196‐198], such as Wnt/β-catenin signaling, which are of potential significance for the development of hepatocellular carcinoma [199].

Treatment with compounds, e.g., vitamin E, that suppress Hedgehog ligand production and reduce the hepatic accumulation of Hedgehog-responsive myofibroblasts has proven beneficial in human NASH [200].

Further studies are necessary to establish whether and to what extent sex hormones affect Hedgehog signaling and how manipulation of cellular energy homeostasis might be exploited to prevent and manage fibrosing NASH, cirrhosis, and HCC in individuals with NAFLD.

As detailed in Table 3, estrogens exert several beneficial metabolic effects. Experimental studies suggest that estrogens promote the accumulation of peripheral gluteofemoral subcutaneous adipose tissue and, within the liver, promote FFA beta-oxidation and prevent the accumulation of triglycerides; moreover, they regulate energy homeostasis, enhance insulin sensitivity, and exert a protective role on the function of pancreatic beta-cells [201]. Finally, estrogens seem to have antisteatotic, antioxidant, and antifibrogenic properties in the liver [204, 205]. Estrogens may protect from liver steatosis and fibrosis in female mice via upregulation of miRNA-125b and miRNA-29, respectively [206, 207]. A recent experimental study showed that the estrogen/estrogen receptor alpha signaling plays essential roles in ROS detoxification and in the reduction of oxidative damage in the liver via partnering with hepatic PPARG coactivator 1 alpha, thus halting the transition from simple steatosis to steatohepatitis [205]. Estrogens are also known to inhibit stellate cell activation and fibrogenesis in experimental models [204]. Of note, a study showed that the natural estrogen 17β-estradiol prevents deoxycholic acid-induced hepatocellular damage in several cell lines. This hepatoprotective effect of estrogen was sustained by mechanisms which were unlikely to be mediated by nuclear estrogen receptor alpha [208]. Interestingly, it has recently been demonstrated that a non-nuclear estrogen receptor plays a key role in reducing hepatic steatosis in female mice [209]. Similarly, estrogen deficiency leads to fat redistribution toward visceral fat accumulation and its inherently unfavorable metabolic derangements, and is thus able to induce the development and progression of NAFLD/NASH both in mice and humans. Of note, both aromatase knockout mice, which are unable to synthesize endogenous estrogen, and estrogen receptor alpha knockout mice develop hepatic steatosis [210‐212]. Patients with breast cancer treated with tamoxifen, a potent estrogen antagonist, develop progressive liver steatosis and NASH [213]. Surgical and physiological menopausal status have been invariably associated with excess risk of NAFLD progression and liver fibrosis, and duration of estrogen deficiency has been directly related to increased fibrosis risk in postmenopausal women with NAFLD [95, 96, 98, 136].

Despite the above premises, the role of HRT in reverting the metabolic alterations associated with low levels of estrogens is a matter of dispute. In animal models, both estrogen and raloxifene, a second-generation selective estrogen receptor modulator, have improved inflammation and ballooning so halting the progression of liver fibrosis in diet-induced NASH in female ovariectomized mice [30, 214]. However, the theoretical benefits of HRT in liver disease remain to be proven in humans. A seminal randomized placebo-controlled study suggested that HRT containing low-dose estradiol and norethisterone was able to reduce liver enzyme concentrations in women with T2D, potentially through the reduction of accumulation of fat in the liver [45]. Accordingly, a north American study using data of the NHANES III survey showed that postmenopausal women who received HRT had a significantly lower risk of NAFLD than postmenopausal women who did not (OR 0.69, 95% CI 0.48–0.99) [215]. However, whether HRT reduces the risk of NASH and/or fibrosis among postmenopausal women remains uncertain. Two studies showed no, or borderline, beneficial effects of HRT on fibrosis among postmenopausal women with NAFLD [95, 96]. Worryingly, a recent study reported that synthetic hormone use was associated with more severe hepatocellular injury and lobular inflammation. Further analysis showed that progesterone, but not estrogen use, was associated with worse liver histological lesions, suggesting a multifaceted impact of sex hormones on NASH features [97].

Several lines of evidence support a key role of choline as an essential nutrient and cellular component. Choline is the major source of methyl groups in the diet and is a precursor of phosphatidylcholine, which is an important component of cell membranes and VLDL, required for the export of lipids from the liver into the bloodstream. Thus, depletion of choline inhibits hepatic triglyceride export and induces fatty liver in both experimental models and humans [185, 203, 216, 217]. Of note, postmenopausal women are more prone than premenopausal women and men to develop NAFLD as a consequence of choline deficiency [128, 185, 217], and decreased dietary choline intake has been significantly associated with increased fibrosis in postmenopausal women with NAFLD [218]. Indeed, menopausal status determines a differential susceptibility to choline deficiency arising from the estrogen-inducible expression and activity of phosphatidylethanolamine-N-methyltransferase (PEMT), a key enzyme in de novo choline biosynthetic pathway. Following menopause, the endogenous supply of choline decreases as a result of estrogen deficiency; consequently, the dietary requirement of choline is increased [185, 203, 216, 217]. Consistently, choline supplementation has been reported to improve or revert liver steatosis in patients receiving long-term total parenteral nutrition [219]. However, no interventional studies have addressed whether choline supplementation is able to prevent NAFLD or reduce NAFLD progression; moreover, future studies investigating the role of preventive or therapeutic choline supplementation in postmenopausal women are eagerly awaited.

Lifestyle changes, i.e., physical activity and balanced diet, are the milestone of intervention for NAFLD treatment. In general, weight loss induced by lifestyle changes improves IR and features of the MetS, including NAFLD. Of note, a modest weight loss of as little as 2–3 kg is associated with NAFLD reversal [39], and weight loss of greater than 7–10% significantly improves histological features of NASH, including fibrosis [220]. However, significant weight reduction achieved through dietary restrictions results in negative effects on lean muscle and bone mass in postmenopausal women. Therefore, while weight reduction in postmenopausal women with metabolic abnormalities and risk factors for NAFLD should be strongly recommended, special attention should be paid to how weight loss is obtained in older women to prevent the worsening of loss of lean mass [139, 203]. An integrated approach consisting of dietary changes along with regular exercise is mandatory to prevent the untoward effect of losing lean mass. Both resistance training and aerobic exercise are effective in preventing muscle and bone loss during weight loss [203], and equally reduce liver fat content [221, 222]. Postmenopausal women with higher levels of physical activity seem to have lower total body and visceral fat and to be less likely to gain fat mass during menopause [139, 223]. However, only few studies have specifically assessed the role of physical activity in NAFLD prevention/reversal in postmenopausal women. Data have shown that exercise training effectively reduces liver enzymes in overweight postmenopausal women, probably as a reflection of liver fat content [224], and aerobic physical activity reduces cardiovascular risk factors in postmenopausal women with NAFLD [225]. Further studies are needed to define the ideal structure and duration of exercise-based interventions in postmenopausal women with NAFLD or at risk of developing it.

In summary, there are no codified therapeutic interventions approved for tackling NAFLD in women. Given the theoretical beneficial metabolic and hepatic effects of estrogens, it is intriguing to speculate that HRT might possibly play a role in the prevention and treatment of metabolic alterations associated with menopause, including NAFLD. However, studies in humans are lacking and the potential metabolic benefits should be weighed, in clinical practice, against the detrimental cardiovascular and neoplastic risk associated with long-term HRT [226, 227]. Moreover, the multifaceted and variable impact of different sex hormones in NAFLD should be carefully considered. Dietary changes associated with either aerobic or resistance physical exercise should be considered as the milestone of treatment of NAFLD. Future studies will have to evaluate the effects on NAFLD prevention/treatment of supplementing choline to a balanced diet in postmenopausal women. Finally, the physiopathological peculiarities of NAFLD in women and the gender-based differences in the kinetics and dynamics of drugs and xenobiotics should be taken into account when developing new treatment strategies and proposing interventional trials for NAFLD.