Nonalcoholic fatty liver disease
Nonalcoholic fatty liver disease (NAFLD) refers to a condition where there is an excess buildup of triglycerides in the liver parenchyma in the absence of excess alcohol consumption 1. It ranges from simple steatosis to nonalcoholic steatohepatitis (NASH), which can progress to fibrosis, cirrhosis, and hepatocellular carcinoma (HCC). Simple steatosis is a relatively benign condition and is defined as the presence of fat comprising 5% of the liver weight in the absence of inflammation 2. In contrast, NASH is diagnosed by the additional presence of inflammation and hepatocyte injury, with the most defining features being hepatocyte ballooning and the formation of Mallory bodies 3: eosinophilic intermediate filament proteins found in the cytoplasm of damaged hepatocytes. NAFLD is considered to be the hepatic component of the metabolic syndrome (MetS) and is associated with increased visceral adipose tissue (VAT), insulin resistance, and dyslipidemia 4. NAFLD has been found to be present in 20–30% of the population of Western countries 5 and is the leading cause of chronic liver disease in the USA 6. The global prevalence of NAFLD continues to rise in conjunction with rapid increases in rates of obesity and diabetes in recent years 7.
In men, testosterone deficiency is associated with increased VAT and insulin resistance 8, and testosterone replacement in this setting has been found to reduce fat deposition 9; whereas hyperandrogenemia has been associated with increased risk of insulin resistance and VAT in women 10,11. Also, some reports suggest an association between testosterone and other sex hormones and severity of liver disease 12. Thus, the primary aims of this review are to discuss the current available data evaluating the association between testosterone levels and NAFLD, to discuss the potential clinical relevance of these data, and to identify gaps in the literature.
Pathogenesis of NAFLD
The pathogenesis of NAFLD, although still not completely understood, appears to be multifactorial. The traditional model describing the pathogenesis of NASH is the two-hit hypothesis 13. The first hit involves the accumulation of triglycerides and free fatty acids in the hepatic parenchyma due to an imbalance between the influx and synthesis of lipids and the export and β-oxidation of lipids. This makes the liver susceptible to injury caused by second hits such as inflammation, adipokines, gut-derived endotoxins, mitochondrial dysfunction, and oxidative stress. However, recently, there has been speculation that free fatty acids themselves can directly cause injury to the liver as they undergo esterification with glycerol to form triglycerides in hepatocytes and activate an inflammatory pathway 14. Furthermore, a third hit has been recently proposed in the pathogenesis of NAFLD 14. In a healthy liver, progenitor cells differentiate and replace dead hepatocytes. The third hit portrays the inability to regenerate dead hepatocytes in the presence of oxidative stress in NAFLD, thus worsening liver damage.
Insulin resistance, a component of the MetS, has been shown to promote NAFLD in epidemiological, experimental, and human studies 15. As insulin suppresses lipolysis in adipose tissue, in a state of insulin resistance, there is an increased influx of fatty acids to the liver 14. Hyperinsulinemia increases the expression of sterol regulatory element binding protein-1c, a transcription factor which up-regulates de-novo lipogenesis. It also inhibits β-oxidation of free fatty acids in the mitochondria and increases the degradation of very low-density lipoprotein in the liver, leading to an increase in hepatic fat accumulation 16. In addition to insulin playing a potential role in the pathogenesis of NAFLD, many factors including lipid metabolites can interfere with the insulin-signaling cascade and exacerbate insulin resistance, leading to a vicious cycle between insulin resistance and NAFLD 17.
Obesity and increased VAT may also play a role in the pathogenesis of NAFLD. In obesity, excess lipids can deposit ectopically into other organs such as the liver 18. Thus, both insulin resistance and excess VAT cause an increased influx of free fatty acids to the liver, leading to NAFLD 19. Increasing visceral obesity also increases proinflammatory cytokines such as tumor necrosis factor-α, interleukin 6, and C-reactive protein and decreased production of the anti-inflammatory adipokine, adiponectin, which can also play a role in the development of NAFLD 20.
Diagnosis of NAFLD
Most patients with NAFLD are asymptomatic. If symptoms present, they usually are nonspecific and include fatigue and abdominal discomfort. On physical exam, an enlarged, palpable liver can be present in a patient with NAFLD, but this sign is not universal and not specific. Hence, there should be evidence of hepatic steatosis by imaging or histology, with secondary causes of fat accumulation within hepatocytes such as significant alcohol consumption and viral infection excluded in order to establish a diagnosis of NAFLD 21. Liver biopsy remains the gold standard to diagnose NAFLD/NASH and associated fibrosis; however, it is rarely performed for this purpose as it is expensive and has a risk of morbidity and mortality because of its invasiveness. Further, there are also limits related to data obtained by liver biopsy given well-known variability in sampling and pathologist interpretation; these are significant considerations for missed diagnosis of NASH, NASH-cirrhosis, or NAFLD-related HCC. Although noninvasive imaging modalities such as ultrasound (US), computed tomography, and MRI can show the presence of a fatty liver, these do not always correlate with other histological features 22. In addition, patients with NAFLD may have elevated alanine aminotransferase, aspartate aminotransferase, and gamma-glutamyl transpeptidase levels. However, these lack sensitivity and specificity 23. Algorithms that combine liver enzymes with other widely available biomarkers or clinical data including the FIB-4 and NAFLD Fibrosis Score are also available and have high negative predictive values for determining presence of advanced NAFLD-associated fibrosis with similar positive predictive values 24. Acute phase proteins have also been evaluated for predicting NASH. Cytokeratin-18 fragments, markers of hepatocyte apoptosis, have shown to be useful in distinguishing steatosis from NASH 25. Nevertheless, the use of serum biomarkers has still not been approved in diagnosing NAFLD in routine clinical practice 21.
Testosterone circulates in the body either free (2%), loosely bound to albumin which is produced in the liver (38%), or tightly bound to its main carrier which is also synthesized in the liver: sex-hormone binding globulin (SHBG) (60%) 26. Bioavailable testosterone (BT), which consists of the sum of free testosterone (FT) and albumin-bound testosterone, can bind to and activate the androgen receptor (AR) in target tissues. It can also be converted to dihydrotestosterone (DHT), the strongest activator of the AR, or to estradiol (E2). On the contrary, testosterone bound to SHBG cannot bind to the AR and is unable to be taken up by target tissues. Plasma SHBG concentrations are increased by a variety of conditions including aging, hyperthyroidism, hypogonadism, and hepatic cirrhosis and decreased in obesity and diabetes mellitus 27. Because SHBG can be a potential confounding variable when evaluating androgen and estrogen levels in conditions such as obesity, aging, and liver disease, only FT and BT levels should be measured in these settings.
Equilibrium dialysis is the gold standard for measuring FT 28. However, it can be affected by sample dilution, pH, and changes in temperature 29. Immunoassays for quantifying FT are commonly available, but lack sensitivity 30. Liquid chromatography–tandem mass spectrometry (LC–MS/MS) has improved sensitivity and shows hope in becoming the reference technology to detect testosterone levels 31. In addition, FT and BT levels can be calculated by using the measurements of SHBG and serum total testosterone levels. The calculated FT and BT have shown to be in agreement with equilibrium dialysis 28. BT can also be measured by precipitating SHBG with ammonium sulfate. However, as ammonium sulfate can precipitate a small amount of albumin-bound testosterone, BT can be falsely underestimated 32.
Low testosterone, NAFLD, and the metabolic syndrome in men
In men, testosterone deficiency has been associated with an increased accumulation of VAT and insulin resistance, which are factors contributing to the MetS 19; and higher levels of testosterone are associated with decreased central obesity 33. Analysis from a cross-sectional study amongst elderly men showed that an increase of 1 SD of total testosterone level (∼153 ng/dl or 5.31 nmol/l) correlated with a 57% reduced risk of the diagnosis of MetS 34. In a population of middle-aged Finnish men, it was found that low total testosterone and SHBG can predict the development of MetS and diabetes 35. Moreover, testosterone replacement has been found to reduce insulin resistance in hypogonadal men with type 2 diabetes and/or MetS 36 and to reduce fat accumulation 37. Further support of the role of androgens in NAFLD is suggested by a case report of a man who developed NAFLD after treatment of prostate cancer with a gonadotrophin agonist that induced profound hypogonadism. 38. In spite of this strong association between testosterone and the MetS, causality has not been established 39.
Laboratory and clinical evidence showing association of testosterone to NAFLD
Male animal data
To determine the relative contribution of gonadal status to diet-induced fat accumulation in the liver, a study was conducted in young male rats placed in four different groups: intact eugonadal (I) rats fed a regular chow diet (I+RCD), intact eugonadal rats fed a high fat diet (I+HFD), castrated (C) rats on a high fat diet (C+HFD), and castrated rats with testosterone replacement on a high fat diet (C+HFD+T). Over a 15-week time period, the C+HFD had undetectable serum testosterone levels, whereas the C+HFD+T had serum testosterone levels that were significantly higher than those measured in the I+RCD and I+HFD groups. The hepatocytes of the I+HFD and C+HFD+T revealed only mild to moderate microvesicular steatosis. C+HFD had the largest macrovesicular accumulation of fat in hepatocytes, inflammation of the liver, and worst pathology score. These pathological features, which characterize NAFLD, decreased significantly with testosterone replacement in the C+HFD+T group in the absence of hyperglycemia, hyperinsulinemia, and insulin resistance. Consequently, this study suggests that testosterone deficiency enhances diet-induced accumulation of fat in the liver and the development of NAFLD, whereas testosterone replacement in frankly hypogonadal animals diminishes hepatic steatosis and inflammation, thereby possibly reducing the severity of NAFLD 40. Previous studies have shown that food consumption decreases by castration and increases by testosterone replacement in male rats 41. Although leptin levels were higher in HFD groups regardless of their testosterone status, other factors such as ghrelin could influence this association between testosterone and food consumption, thus possibly offering an alternate potential causal or modifying factor in the relationship between testosterone and diet-induced NAFLD. The timing of androgen exposure may also be important. For example, in sheep, maternal prenatal exposure to testosterone predisposes to fatty liver disease in the adult offspring of exposed ewes independently of central obesity 42. One limitation of these studies was the lack of measurement of DHT and inability to detect E2 levels lower than 2 pg/ml or 7.34 pmol/l as testosterone is converted to these active metabolites in the body.
Studies in men
Several epidemiological studies have investigated the association between testosterone and NAFLD in men such as a study in 1944 Korean men 43. This study showed that male individuals with US-confirmed NAFLD had a statistically significant lower median baseline total testosterone level, measured by radioimmunoassay, than the control group [∼418 ng/dl (14.5 nmol/l) vs. 496 ng/dl (17.21 nmol/l), P<0.05]. After adjusting for potential confounders such as age, smoking, exercise, history of hypertension and diabetes, waist circumference, systolic blood pressure, glucose, HDL-cholesterol, triglyceride, and erythrocyte sedimentation rate, a logistic regression model showed that as baseline total testosterone levels increased, the prevalence of NAFLD was lower at baseline and follow-up (odds ratio, 0.696 and 0.795 at baseline and follow-up, respectively). In a secondary analysis using Cox regression, higher baseline testosterone levels significantly hindered the development of NAFLD. However, this association was no longer significant after adjusting for confounders, suggesting that other factors may also play a role.
Another study was conducted in 55 neurologically stable male individuals admitted to a rehabilitation program in Sulmona, Italy because of chronic spinal cord injury 44. One blood sample was taken to measure total testosterone levels by chemiluminescence immunoassay, whereas NAFLD was diagnosed by abdominal US. Twenty-seven out of the 55 patients were diagnosed with NAFLD; all had a higher BMI, insulin resistance (HOMA-IR), triglyceride levels, and gamma-glutamyl transpeptidase values, and lower total testosterone and FT levels. However, multiple logistic regression analysis showed that only total testosterone and FT had an independent association with NAFLD, with males with lower total testosterone levels (<300 ng/dl or 10.4 nmol/l) having a 12-fold increase risk of NAFLD than those with higher total testosterone levels.
In a different retrospective cross-sectional study conducted on 495 healthy Korean adult men 45, total testosterone was measured in serum by radioimmunoassay, and abdominal US was performed to diagnose NAFLD. The 251 subjects that had NAFLD had lower testosterone levels [398 ng/dl (13.81 nmol/l) compared with 483 ng/dl (16.76 nmol/l)] and adverse metabolic profiles such as higher VAT, HOMA-IR, and low-grade inflammation. After adjusting for VAT, insulin resistance, low-grade inflammation, age, smoking, diabetes, exercise, BMI, and triglycerides, an inverse relationship was shown between testosterone and NAFLD.
Cross-sectional data from the Multi-Ethnic Study of Atherosclerosis evaluated the association between sex hormones and liver fat. Testosterone, dehydroepiandrosterone (DHEA), SHBG, and E2 were measured using various immunoassays and BT was calculated. Computed tomography scanning measured fatty liver. A total of 2899 males were included in the analysis. Men with fatty liver had lower total testosterone levels and SHBG levels and higher DHEA and E2 levels. These associations remained significant even after adjusting for age, ethnicity, sex, waist-to-hip ratio, hypertension, current smoking, and total cholesterol 46.
Studies in women
The association between testosterone and NAFLD was also evaluated in 2835 postmenopausal women in the Multi-Ethnic Study of Atherosclerosis 46. Women with fatty liver had lower SHBG levels and higher E2 levels. However, unlike in men, there was no association between fatty liver and total testosterone or DHEA levels, suggesting sex differences in the association of testosterone to NAFLD. After adjusting for confounding variables, it was found that women in the highest tertile of BT and E2 were significantly more likely to have fatty liver than in the lowest tertile of BT and E2, respectively. One limitation specific to this study is that estrone, the main postmenopausal estrogen, was not measured.
In Greece, a cross-sectional study was done in 40 postmenopausal women where 22 were diagnosed with NAFLD by liver biopsy. They underwent blood sampling to measure serum SHBG and total testosterone by immunochemiluminescence while FT and BT were calculated. The women with NAFLD had lower SHBG levels and higher FT and BT when compared with those women without NAFLD. After adjusting for age, BMI, and waist circumference, only SHBG and BT levels were independently associated with NAFLD, suggesting that the role of androgens may be the opposite in women compared with men. One limitation of this study is the small sample size 47. However, this is one of the first studies on testosterone and NAFLD that diagnoses NAFLD by liver biopsy, the gold standard to diagnose this disease.
A retrospective cross-sectional study conducted in 95 females with polycystic ovary syndrome (PCOS) to determine whether androgens play a role in the association between NAFLD and PCOS, a condition often accompanied by hyperandrogenism. NAFLD was diagnosed in 21 of these women by US, minimal alcohol consumption (no more than one drink per week), and the absence of positive hepatitis serologies. Total testosterone, dehydroepiandrosterone sulfate, and 17-hydroxyprogesterone levels measured by a chemiluminescent immunoassays were similar in women with and without NAFLD, suggesting that androgens may not play a role in the association of PCOS and NAFLD. However, confounding factors such as age, weight, and insulin resistance were not accounted for; for example, HOMA-IR was higher in the PCOS and NAFLD group, possibly influencing the results 48.
On the contrary, in the UK, a case–control study was conducted in 29 women with PCOS and 22 healthy controls. Total testosterone and SHBG were measured by a chemiluminescence method, and the free androgen index was calculated while hepatic steatosis was identified by proton magnetic resonance spectrometry. Women with hyperandrogenic PCOS, based on the free androgen index, had evidence of increased liver fat compared with women with normoandrogenic PCOS and healthy controls after adjusting for BMI, adipose tissue, and HOMA-IR. These data suggest that hyperandrogenic PCOS can increase the risk of hepatic steatosis independently of obesity and insulin resistance 49.
Summary of studies across sexes and their limitations
The studies are summarized in Table 1. In the studies in men described above, there was an association between lower levels of total testosterone and the development of NAFLD. FT, BT, SHBG, DHEA, and other sex hormones like E2 were not measured in all the studies in men so definite conclusions cannot be made about their relationship to NAFLD. In the first two studies mentioned in women, NAFLD was associated with higher levels of BT. In the two studies showing the association between PCOS and NAFLD, one found that androgens may not play a role, whereas the other found that increased androgens can predispose to hepatic steatosis. Other conclusions cannot be drawn at this time as different forms of testosterone and sex hormones were measured in these studies.
Immunoassays were used to measure the various sex hormones. This is a limitation as immunoassays lack sensitivity. Another limitation is that only one testosterone measurement was taken in some studies. Two samples are usually required to confirm the diagnosis of low testosterone. NAFLD and hepatic steatosis were diagnosed by abdominal US in many of the studies and US may not detect hepatic fatty infiltration less than 30% 50. Participants were excluded if they had consumed significant amounts of alcohol, which was defined differently across these studies. Further, recall bias could have occurred when asking participants if they had consumed significant alcohol or taken steatogenic drugs, excluding them from a clinical diagnosis of NAFLD. Although confounding variables were adjusted in most of the studies, all of them may not have been accounted for, causing limitations. Standardization of all these different factors would be important to take into consideration when future studies are conducted.
Possible mechanisms underlying the association of low testosterone to NAFLD
Several different mechanisms could explain the observed association between low testosterone and NAFLD in men. For instance, low testosterone leads to accumulation of VAT, which can cause insulin resistance and increased exposure of the liver to free fatty acids. Moreover, low serum testosterone is associated with increased inflammation 51. Testosterone may also influence microRNAs in the liver or the activity of hepatic lipase as seen in male rats 52. As testosterone can be converted to DHT and E2 by the enzymes 5α-reductase and aromatase, respectively, DHT and E2 may play a part in linking low testosterone and NAFLD in men. The deficiency or inhibition of 5α-reductase triggers the development of hepatic steatosis in male mice 53,54, suggesting low DHT levels may contribute to NAFLD. As DHT can initiate cell cycle arrest and apoptosis in androgen-sensitive liver cells by the PKR/eIF2a signaling cascade 55, low levels of DHT may increase HCC risk and possibly have a role in the pathogenesis of NAFLD-related HCC. Furthermore, low testosterone is associated with elevated E2 in men and this could be due to greater conversion of testosterone to E2. E2 has been found to reduce lipogenesis in male rats by decreasing fatty acid synthase and the phosphorylation of acetyl coenzyme A 56, and thus, could affect testosterone’s association with NAFLD.
Relevance of low testosterone and NAFLD association
A direct association between testosterone and NAFLD could help in the diagnosis of NAFLD. Testosterone levels could help predict the stage of NAFLD and prognosis of the disease or it may justify the screening of hypogonadal men for NAFLD. It is also possible that obesity causes both low testosterone and NAFLD as there are insufficient data to determine whether the association between low testosterone and NAFLD is causal. However, if the relationship is causal, testosterone may also protect the liver from developing hepatic steatosis and slow down the progression to HCC 53. As estrogen/estrogen receptor and DHT/AR signaling decrease fatty acid and cholesterol synthesis and increase β-oxidation 56, and E2 has suggested as a protective factor for NAFLD in healthy men 57, steroid hormones such as testosterone may have the ability to treat NAFLD in men. However, forms of oral testosterone that are 17-alpha-alkylated are hepatotoxic 58, and so these forms would not be appropriate to use in this setting. Further, there has been an association between anabolic steroid abuse and HCC 59, so caution must be taken when suggesting testosterone’s possible therapeutic effect for NAFLD.
Androgens and NAFLD in women
The liver is one of the most sexually dimorphic organs per gene expression and profound differences in circulating hormones exist between sexes. Also, there seem to be clear sex differences regarding the potential role of androgens in NAFLD. These differences could be driven by several factors including that in women, higher androgen levels are in general associated with increased visceral adiposity and insulin resistance 10. Though there may be a direct association between NAFLD and PCOS 60, it is still uncertain whether this association is because of the presence of high levels of testosterone or other metabolic factors such as obesity and insulin resistance 61. DHT, rather than testosterone, may play a role in women with PCOS; long-term DHT treatment in female mice displayed ovarian, endocrine, and metabolic characteristics of PCOS seen in humans 62. The contrasting role of androgens in men and women regarding its contribution to NAFLD is far from understood. It is possible that the differences in testosterone levels per se, where women still have comparatively much lower testosterone levels than men with testosterone deficiency, or a potential protective effect of estrogens could explain these differences.
Because of the growing obesity epidemic, NAFLD is the most common cause of chronic liver disease in the USA. This disease most often has an indolent course though a subset often develop more progressive liver diseases including NASH and NASH-cirrhosis, in whom there is an excess risk of HCC. There is experimental and clinical data suggesting that testosterone may help to prevent or ameliorate NAFLD in men. However, given limitations in studies, particularly the lack of randomized controlled trials, further investigations need to be conducted. The association between testosterone and NAFLD has been primarily evaluated in specific subgroups of females (e.g. with PCOS); additional research is needed to establish testosterone’s role in the development and progression of NAFLD in women across the age-spectrum within the general population. Sex-specific research is crucial given that the burgeoning obesity epidemic also impacts women; even with women having a lower risk of NAFLD/NASH progression to HCC than men 63, there are a large number of women with NAFLD. In addition, although experimental research suggests several mechanisms by which testosterone may influence NAFLD risk, additional epidemiological research in prospectively recruited and serially followed human populations is necessary to establish mechanisms in background of multiple common confounding factors in ‘real world’ populations including dietary intake, physical activity, and use of alcohol. Though physiological levels of testosterone were measured in the clinical studies, the impact of testosterone supplementation in the development of NAFLD should also be studied. These studies should also take into account long-term safety issues, particularly cardiovascular safety, as this has recently been questioned. 64
Dr Garcia receives research support from the Department of Veterans Affairs (BX000507 and I01 CX000174) and NIH (AG040583). Dr White receives salary support from the National Institute of Diabetes Digestive and Kidney Diseases (DK095082) and infrastructure and resource support from the Houston VA HSR&D Center for Innovations (CIN13-413) and the Texas Medical Center Digestive Disease Center (DK56338).
Conflicts of interest
Dr Garcia is a consultant for Viking Therapeutics. For the remaining authors there are no conflicts of interest.
1. Alba LM, Lindor K. Review article: non-alcoholic fatty liver disease. Aliment Pharmacol Ther 2003; 17:977–986.
2. Younossi ZM, Gramlich T, Liu YC, Matteoni C, Petrelli M, Goldblum J, et al.. Nonalcoholic fatty liver disease: assessment of variability in pathologic interpretations. Mod Pathol 1998; 11:560–565.
3. Hubscher SG. Histological assessment of non-alcoholic fatty liver disease. Histopathology 2006; 49:450–465.
4. Machado M, Cortez-Pinto H. Non-alcoholic steatohepatitis and metabolic syndrome. Curr Opin Clin Nutr Metab Care 2006; 9:637–642.
5. Masarone M, Federico A, Abenavoli L, Loguercio C, Persico M. Non alcoholic fatty liver: epidemiology and natural history. Rev Recent Clin Trials 2014; 9:126–133.
6. Adams LA, Lindor KD. Nonalcoholic fatty liver disease. Ann Epidemiol 2007; 17:863–869.
7. Mishra A, Younossi ZM. Epidemiology and natural history of non-alcoholic fatty liver disease. J Clin Exp Hepatol 2012; 2:135–144.
8. Zitzmann M. Testosterone deficiency, insulin resistance and the metabolic syndrome. Nat Rev Endocrinol 2009; 5:673–681.
9. Yassin A, Doros G. Testosterone therapy in hypogonadal men results in sustained and clinically meaningful weight loss. Clin Obes 2013; 3:73–83.
10. Phillips GB, Jing T, Heymsfield SB, Phillips GB, Jing T, Heymsfield SB, et al.. Does insulin resistance, visceral adiposity, or a sex hormone alteration underlie the metabolic syndrome? Studies in women. Metabolism 2008; 57:838–844.
11. Corbould A. Effects of androgens on insulin action in women: is androgen excess a component of female metabolic syndrome? Diabetes Metab Res Rev 2008; 24:520–532.
12. Kemp CJ, Leary CN, Drinkwater NR. Promotion of murine hepatocarcinogenesis by testosterone is androgen receptor-dependent but not cell autonomous. Proc Natl Acad Sci USA 1989; 86:7505–7509.
13. Day CP, James OF. Steatohepatitis: a tale of two ‘hits’? Gastroenterology 1998; 114:842–845.
14. Dowman JK, Tomlinson JW, Newsome PN. Pathogenesis of non-alcoholic fatty liver disease. QJM 2010; 103:71–83.
15. Utzschneider KM, Kahn SE. Review: the role of insulin resistance in nonalcoholic fatty liver disease. J Clin Endocrinol Metab 2006; 91:4753–4761.
16. Gusdon AM, Song KX, Qu S. Nonalcoholic fatty liver disease: pathogenesis and therapeutics from a mitochondria-centric perspective. Oxid Med Cell Longev 2014; 2014:637027.
17. Asrih M, Jornayvaz FR. Metabolic syndrome and nonalcoholic fatty liver disease: is insulin resistance the link? Mol Cell Endocrinol 2015. [Epub ahead of print]; doi: 10.1016/j.mce.2015.02.018.
18. Than NN, Newsome PN. A concise review of non-alcoholic fatty liver disease. Atherosclerosis 2015; 239:192–202.
19. Eguchi Y, Eguchi T, Mizuta T, Ide Y, Yasutake T, Iwakiri R, et al.. Visceral fat accumulation and insulin resistance are important factors in nonalcoholic fatty liver disease. J Gastroenterol 2006; 41:462–469.
20. Abenavoli L, Peta V. Role of adipokines and cytokines in non-alcoholic fatty liver disease. Rev Recent Clin Trials 2014; 9:134–140.
21. Chalasani N, Younossi Z, Lavine JE, Diehl AM, Brunt EM, Kenneth C, et al.. The diagnosis and management of non-alcoholic fatty liver disease: practice guideline by the American Gastroenterological Association, American Association for the Study of Liver Diseases
, and American College of Gastroenterology. Gastroenterology 2012; 142:1592–1609.
22. Saadeh S, Younossi ZM, Remer EM, Gramlich T, Ong JP, Hurley M, et al.. The utility of radiological imaging in nonalcoholic fatty liver disease. Gastroenterology 2002; 123:745–750.
23. Obika M, Noguchi H. Diagnosis and evaluation of nonalcoholic fatty liver disease. Exp Diabetes Res 2012; 2012:145754.
24. Kim D, Kim WR, Kim HJ, Therneau TM. Association between noninvasive fibrosis markers and mortality among adults with nonalcoholic fatty liver disease in the United States. Hepatology
25. Machado MV, Cortez-Pinto H. Non-invasive diagnosis of non-alcoholic fatty liver disease. A critical appraisal. J Hepatol 2013; 58:1007–1019.
26. Anderson DC. Sex-hormone-binding globulin. Clin Endocrinol (Oxf) 1974; 3:69–96.
27. Paduch DA, Brannigan RE, Fuchs EF, Kim ED, Marmar JL, Sandlow JI. The laboratory diagnosis of testosterone deficiency. Urology 2014; 83:980–988.
28. Wang C, Nieschlag E, Swerdloff RS, Behre H, Hellstrom WJ, Gooren LJ, et al.. ISA, ISSAM, EAU, EAA and ASA recommendations: investigation, treatment and monitoring of late-onset hypogonadism in males. Aging Male 2009; 12:5–12.
29. Sinha-Hikim I, Arver S, Beall G, Shen R, Guerrero M, Sattler F, et al.. The use of a sensitive equilibrium dialysis method for the measurement of free testosterone levels in healthy, cycling women and in human immunodeficiency virus-infected women. J Clin Endocrinol Metab 1998; 83:1312–1318.
30. Swerdloff RS, Wang C. Free testosterone measurement by the analog displacement direct assay: old concerns and new evidence. Clin Chem 2008; 54:458–460.
31. Shea JL, Wongt PY, Chen Y. Free testosterone: clinical utility and important analytical aspects of measurement. Adv Clin Chem 2014; 63:59–84.
32. Davies R, Collier C, Raymond M, Heaton J, Clark A. Indirect measurement of bioavailable testosterone with the Bayer Immuno 1 system. Clin Chem 2002; 48:388–390.
33. Wang C, Jackson G, Jones TH, Matsumoto AM, Nehra A, Perelman MA, et al.. Low testosterone associated with obesity
and the metabolic syndrome contributes to sexual dysfunction and cardiovascular disease risk in men with type 2 diabetes. Diabetes Care 2011; 34:1669–1675.
34. Muller M, Grobbee DE, den Tonkelaar I, Lamberts SW, van der Schouw YT. Endogenous sex hormones and metabolic syndrome in aging men. J Clin Endocrinol Metab 2005; 90:2618–2623.
35. Laaksonen DE, Niskanen L, Punnonen K, Nyyssonen K, Tuomainen TP, Valkonen VP, et al.. Testosterone and sex hormone-binding globulin predict the metabolic syndrome and diabetes in middle-aged men. Diabetes Care 2004; 27:1036–1041.
36. Jones TH, Arver S, Behre HM, Buvat J, Meuleman E, Moncada I, et al.. Testosterone replacement in hypogonadal men with type 2 diabetes and/or metabolic syndrome (the TIMES2 Study). Diabetes Care 2011; 34:828–837.
37. Saad F, Aversa A, Isidori AM, Gooren LJ. Testosterone as potential effective therapy in treatment of obesity
in men with testosterone deficiency: a review. Curr Diabetes Rev 2012; 8:131–143.
38. Gabbi C, Carubbi F, Losi L, Loria P, Costantini M, Bertolotti M, Carulli N. Nonalcoholic fatty liver disease induced by leuprorelin acetate. J Clin Gastroenterol 2008; 42:107–110.
39. Vanni E, Bugianesi E, Kotronen A, De Minicis S, Yki-Jarvinen H, Svegliati-Baroni G. From the metabolic syndrome to NAFLD or vice versa? Dig Liver Dis 2010; 42:320–330.
40. Nikolaenko L, Jia Y, Wang C, Diaz-Arjonilla M, Yee JK, French SW, et al.. Testosterone replacement ameliorates nonalcoholic fatty liver disease in castrated male rats. Endocrinology
41. Nunez AA. Dose-dependent effects of testosterone on feeding and body weight in male rats. Behav Neural Biol 1982; 34:445–449.
42. Hogg K, Wood C, McNeilly AS, Duncan WC. The in utero programming effect of increased maternal androgens and a direct fetal intervention on liver and metabolic function in adult sheep. PLoS One 2011; 6:e24877.
43. Seo NK, Koo HS, Haam JH, Kim HY, Kim MJ, Park KC, et al.. Levels of serum testosterone predict prevalent but not incident non-alcoholic fatty liver disease. J Gastroenterol Hepatol 2015; 30:1211–1216.
44. Barbonetti A, Caterina Vassallo MR, Cotugno M, Felzani G, Francavilla S, Francavilla F. Low testosterone and non-alcoholic fatty liver disease: evidence for their independent association in men with chronic spinal cord injury. J Spinal Cord Med 2015. [Epub ahead of print].
45. Kim S, Kwon H, Park JH, Cho B, Kim D, Oh SW, et al.. A low level of serum total testosterone is independently associated with nonalcoholic fatty liver disease. BMC Gastroenterol 2012; 12:69.
46. Lazo M, Zeb I, Nasir K, Tracy RP, Budoff MJ, Ouyang P, Vaidya D. Association between endogenous sex hormones and liver fat in a multiethnic study of atherosclerosis. Clin Gastroenterol Hepatol 2015. [Epub ahead of print]; doi:10.1016/j.cgh.2014.12.033.
47. Polyzos SA, Kountouras J, Tsatsoulis A, Zafeiriadou E, Katsiki E, Patsiaoura K, et al.. Sex steroids and sex hormone-binding globulin in postmenopausal women with nonalcoholic fatty liver disease. Hormones (Athens) 2013; 12:405–416.
48. Kauffman RP, Baker TE, Baker V, Kauffman MM, Castracane VD. Endocrine factors associated with non-alcoholic fatty liver disease in women with polycystic ovary syndrome: do androgens play a role? Gynecol Endocrinol 2010; 26:39–46.
49. Jones H, Sprung VS, Pugh CJ, Daousi C, Irwin A, Aziz N, et al.. Polycystic ovary syndrome with hyperandrogenism is characterized by an increased risk of hepatic steatosis compared to nonhyperandrogenic PCOS phenotypes and healthy controls, independent of obesity
and insulin resistance. J Clin Endocrinol Metab 2012; 97:3709–3716.
50. Dowman JK, Tomlinson JW, Newsome PN. Systematic review: the diagnosis and staging of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis. Aliment Pharmacol Ther 2011; 33:525–540.
51. Tarantino G, Savastano S, Colao A. Hepatic steatosis, low-grade chronic inflammation and hormone/growth factor/adipokine imbalance. World J Gastroenterol 2010; 16:4773–4783.
52. Langfort J, Jagsz S, Dobrzyn P, Brzezinska Z, Klapcinska B, Galbo H, Gorski J. Testosterone affects hormone-sensitive lipase (HSL) activity and lipid metabolism in the left ventricle. Biochem Biophys Res Commun 2010; 399:670–676.
53. Dowman JK, Hopkins LJ, Reynolds GM, Armstrong MJ, Nasiri M, Nikolaou N, et al.. Loss of 5α-reductase type 1 accelerates the development of hepatic steatosis but protects against hepatocellular carcinoma in male mice. Endocrinology
54. Livingstone DE, Barat P, Di Rollo EM, Rees GA, Weldin BA, Rog-Zielinska EA, et al.. 5α-Reductase type 1 deficiency or inhibition predisposes to insulin resistance, hepatic steatosis, and liver fibrosis in rodents. Diabetes 2015; 64:447–458.
55. Dai R, Yan D, Li J, Chen S, Liu Y, Chen R, et al.. Activation of PKR/eIF2alpha signaling cascade is associated with dihydrotestosterone-induced cell cycle arrest and apoptosis in human liver cells. J Cell Biochem 2012; 113:1800–1808.
56. Zhang H, Liu Y, Wang L, Li Z, Zhang H, Wu J, et al.. Differential effects of estrogen/androgen on the prevention of nonalcoholic fatty liver disease in the male rat. J Lipid Res 2013; 54:345–357.
57. Tian GX, Sun Y, Pang CJ, Tan AH, Gao Y, Zhang HY, et al.. Oestradiol is a protective factor for non-alcoholic fatty liver disease in healthy men. Obes Rev 2012; 13:381–387.
58. Bassil N, Alkaade S, Morley JE. The benefits and risks of testosterone replacement therapy: a review. Ther Clin Risk Manag 2009; 5:427–448.
59. Gorayski P, Thompson CH, Subhash HS, Thomas AC. Hepatocellular carcinoma associated with recreational anabolic steroid use. Br J Sports Med 2008; 42:74–75. (discussion 75).
60. Brzozowska MM, Ostapowicz G, Weltman MD. An association between non-alcoholic fatty liver disease and polycystic ovarian syndrome. J Gastroenterol Hepatol 2009; 24:243–247.
61. Kelley CE, Brown AJ, Diehl AM, Setji TL. Review of nonalcoholic fatty liver disease in women with polycystic ovary syndrome. World J Gastroenterol 2014; 20:14172–14184.
62. Caldwell AS, Middleton LJ, Jimenez M, Desai R, McMahon AC, Allan CM, et al.. Characterization of reproductive, metabolic, and endocrine features of polycystic ovary syndrome in female hyperandrogenic mouse models. Endocrinology
63. White DL, Kanwal F, El-Serag HB. Association between nonalcoholic fatty liver disease and risk for hepatocellular cancer, based on systematic review. Clin Gastroenterol Hepatol 2012; 10:1342–1359.
64. Hwang K, Miner M. Controversies in testosterone replacement therapy: testosterone and cardiovascular disease. Asian J Androl 2015; 17:187–191.