Free testosterone concentration is inversely associated with markers of liver fibrosis in men with type 2 diabetes mellitus

Transcription

1 2017, 64 (12), Original Free testosterone concentration is inversely associated with markers of liver fibrosis in men with type 2 diabetes mellitus Shozo Miyauchi 1), Teruki Miyake 2), Masumi Miyazaki 1), Toru Eguchi 3), Tetsuji Niiya 4), Shin Yamamoto 5), Hidenori Senba 2), Shinya Furukawa 6), Bunzo Matsuura 5) and Yoichi Hiasa 2) 1) Department of Internal Medicine, Uwajima City Hospital, Uwajima, Japan 2) Department of Gastroenterology and Metabology, Ehime University Graduate School of Medicine, Toon, Japan 3) Department of Diabetology, Endocrinology and Metabolism, Shimonoseki City Hospital, Shimonoseki, Japan 4) Department of Internal Medicine, Matsuyama Shimin Hospital, Matsuyama, Japan 5) Department of Lifestyle-related Medicine, Ehime University Graduate School of Medicine, Toon, Japan 6) Department of Epidemiology and Preventive Medicine, Ehime University Graduate School of Medicine, Toon, Japan Abstract. The association between serum testosterone level and liver fibrosis in patients with non-alcoholic fatty liver disease is unclear. To clarify this association, we investigated the relationship between serum free testosterone concentration and markers of liver fibrosis in men with type 2 diabetes mellitus but no obvious features of alcohol consumption. This retrospective observational cross-sectional study enrolled 248 men with type 2 diabetes mellitus. The FIB-4 index was measured as a marker of liver fibrosis, and multiple linear regression analysis was performed to examine its association with serum free testosterone concentration. In addition, the 7S domain of type IV collagen (IV-7S) was examined in 140 of the 248 patients. The mean free testosterone concentration was 10.6 ± 6.8 pg/ml and the means of the FIB-4 index and IV-7S were 1.64 ± 1.19 and 4.02 ± 1.11 ng/ml, respectively. After adjusting for all relevant variables, serum free testosterone concentrations were inversely associated with both the FIB-4 index and IV-7S (β; -0.28, p < , and β; -0.28, p = 0.002, respectively). Measuring serum free testosterone concentrations in men with type 2 diabetes mellitus may help to predict progression to advanced liver disease. Identifying patients at risk may help to prevent the development of cirrhosis and hepatocellular carcinoma. Key words: Free testosterone, FIB-4 index, Liver fibrosis, Non-alcoholic fatty liver disease, Type 2 diabetes mellitus THE PREVALENCE of non-alcoholic fatty liver disease (NAFLD), one of the most common chronic liver diseases, is increasing with changes in lifestyle [1, 2]. Unless appropriately treated, NAFLD can progress to steatohepatitis, cirrhosis, and liver failure. Moreover, hepatocellular carcinoma may complicate the course of NAFLD. It was recently reported that the prognosis of NAFLD is strongly associated with liver fibrosis [3]; therefore, it is desirable to identify high-risk patients with advanced NAFLD. NAFLD is a frequent complication of type 2 diabetes mellitus (T2DM), and T2DM is associated with the progression of non-alcoholic fatty liver (NAFL) to non-alcoholic steatohepatitis (NASH) and cirrhosis [4-9]. Moreover, T2DM is thought to be a Submitted May 25, 2017; Accepted Aug. 2, 2017 as EJ Released online in J-STAGE as advance publication Sep. 9, 2017 Correspondence to: Teruki Miyake, Department of Gastroenterology and Metabology, Ehime University Graduate School of Medicine, 454 Shitsukawa, Toon city, Ehime, , Japan. teruki-ygc@umin.ac.jp The Japan Endocrine Society risk factor for hepatocellular carcinoma [9]. Sex differences have also been observed in NAFLD and NASH. In Japan, the prevalence of NAFLD is higher in men than in women [10-12]. Consequently, the prevalence of NASH is also estimated to be considerably high in men with NAFLD [13, 14]. Testosterone, the main androgen in men, has important effects on bone, fat, and brain physiology and on the metabolism of glucose and lipids [15]. A low serum testosterone level is a risk factor for obesity, metabolic syndrome, and T2DM [16-24]. In the body, testosterone circulates in three forms: Tightly bound to sex-hormone binding globulin (SHBG) (60%), loosely bound to albumin (38%), and in a free form (2%) [15]. Free testosterone levels reflect the androgen status more accurately than do total testosterone levels. Several studies have shown a relationship between NAFLD and testosterone components: Low serum total testosterone levels were positively correlated with NAFLD in men [25, 26], and low serum SHBG levels were demonstrated

2 1138 Miyauchi et al. to be associated with NAFLD in patients with T2DM [27]. However, there are few reports on the association between free testosterone and NAFLD [27], and whether free testosterone is associated with liver fibrosis in patients with NAFLD remains unknown. In this study, we aimed to clarify the relationship between free testosterone and liver fibrosis in patients with NAFLD, by investigating the relationship between free testosterone concentration and both the Fibrosis-4 (FIB-4) index and the 7S domain of type IV collagen (IV-7S) in men with T2DM. Subjects and Methods This retrospective observational cross-sectional study enrolled 248 men with T2DM who were admitted to Uwajima City Hospital between May 1, 2013 and December 31, 2016 for treatment of their diabetes. We excluded patients based on the following exclusion criteria: obvious alcohol consumption (> 30 g/day); other types of liver diseases including viral or autoimmune hepatitis, and cholestatic or metabolic/genetic liver disease; and taking corticosteroids, antiretroviral drugs, or anticancer drugs. The enrolled patients were aged years. The questionnaire included exercise habits and the frequency and quantity of cigarettes smoked. Height and body weight, measured without wearing shoes or outer clothing, were used to calculate the body mass index (BMI) at this time of hospitalization. Blood pressure was measured with an automated sphygmomanometer after a 15-min rest the morning after admission. Blood samples were collected between 0700 and 0900 h after a 12-h fasting period. Fasting blood glucose, glycated hemoglobin (HbA1c), total cholesterol, and triglyceride levels; aspartate aminotransferase (AST), alanine aminotransferase (ALT), γ-glutamyl transpeptidase (γ-gtp), and free testosterone levels; and platelet count were measured. Serum free testosterone was measured using the free testosterone ELISA kit (Cosmic Corporation Co., Ltd., Tokyo, Japan). Moreover, IV-7S was checked in 140 of the 248 patients. All biochemical determinations were performed in the same laboratory. The visceral fat area was automatically measured with navel-level computed tomography images using a SYNAPSE VINCENT volume analyzer (FUJIFILM Corporation, Tokyo, Japan). The diagnosis of T2DM was based on the Japan Diabetes Society criteria [28]. The FIB-4 index was calculated according to the following formula: age (years) AST [U/L] / (platelets [ 10 9 /L] ALT[U/L]) [29]. This study was conducted in accordance with the revised guidelines of the World Medical Association s Declaration of Helsinki. Prior approval for this study was obtained from the Institutional Review Board of Uwajima City Hospital and the study fulfilled local ethical and legal regulatory requirements (Approval ID , University Hospital Medical Information Network ID: UMIN ). All procedures in this study were conducted in accordance with Good Clinical Practices. Informed consent was obtained from all patients. Statistical Analysis Spearman correlation coefficients were used to evaluate the relationships between FIB-4 or IV-7S and each of the clinical variables. If the p value was 0.2, the factor was selected as an independent variable. Then, multiple linear regression was performed to explore associations with FIB-4 or IV-7S. All statistical analyses were performed using JMP software, version 12.0 (SAS Institute Japan Inc., Tokyo, Japan). P values < 0.05 were considered significant. Results Baseline characteristics The baseline characteristics of the patients are shown in Table 1. The mean age was 60.3 years. The average BMI and HbA1c were 25.6 kg/m 2 and 9.9%, respectively, and the mean concentration of each of the measured liver enzymes was mildly raised. The mean free testosterone concentration was 10.6 pg/ml. The means (± standard deviations) of the FIB-4 index and IV-7S were 1.64 ± 1.19 and 4.02 ± 1.11 ng/ml, respectively. Clinical variables correlated with FIB-4 index and IV-7S The FIB-4 index was positively correlated with visceral fat area and the serum level of γ-gtp, and was inversely correlated with the HbA1c value; diastolic blood pressure (DBP) measurement; and the serum levels of total cholesterol, low-density lipoprotein cholesterol, free testosterone, and IV-7S (Table 2). As a result of these tests, the following variables were included as candidates in the multiple linear regression analysis: fasting blood glucose level; HbA1c; visceral fat area; DBP measurement; and serum concentrations of total

3 FT and liver fibrosis in men with T2DM 1139 cholesterol, γ-gtp, and free testosterone. IV-7S was positively correlated with age; BMI; visceral fat area; and the serum levels of AST, ALT, γ-gtp, and was inversely correlated with the platelet count and the serum level of free testosterone. From these results, 8 factors (age, BMI, visceral fat area, AST, ALT, γ-gtp, platelet count, free testosterone) were included in the multiple linear regression analysis. Analysis of the association between the FIB-4 index and free testosterone The results of the multiple linear regression analysis with the FIB-4 index as the dependent variable and the above-mentioned factors as independent variables are shown in Table 3; the FIB-4 index was inversely correlated with the concentration of free testosterone (β; -0.32, p < ). Moreover, the results of the multiple linear regression analysis using the FIB-4 index as the dependent variable and all clinical variables as independent variables showed significant inverse associations between the FIB-4 index and the concentration of free testosterone (β; -0.28, p < ). Analysis of the association between IV-7S and free testosterone The results of the multiple linear regression analysis with IV-7S as the dependent variable and the above candidates as independent variables are shown in Table 4; free testosterone was selected as an independent variable. After adjustment for age, IV-7S was inversely correlated with free testosterone concentration (β; -0.28, p = 0.001). The results of the multiple linear regression analysis using only selected clinical variables and that using all clinical variables as independent variables both showed a significant inverse association between IV-7S and free testosterone concentration (β; -0.29, p = and β; -0.28, p = 0.002, respectively). Discussion The findings of the present study show that serum concentrations of free testosterone are inversely associated with both the FIB-4 index and IV-7S, markers of liver fibrosis, in men with T2DM but without obvious Table 1 Baseline characteristics of patients (n=248) Variable Mean ± sd or n (%) Age (years) 60.3 ± 12.0 BMI (kg/m 2 ) 25.6 ± 4.7 Fasting glucose (mg/dl) ± 54.2 HbA1c (%) 9.9 ± 2.2 Regular exercise (yes) 52 (21.0%) Current smoker (yes) 57 (23.0%) Visceral fat area (cm 2 ) ± 75.9 SBP (mm Hg) ± 19.1 DBP (mm Hg) 75.4 ± 13.2 Total cholesterol (mg/dl) ± 45.1 Triglycerides (mg/dl) ± HDL-cholesterol (mg/dl) 47.4 ± 13.3 LDL-cholesterol (mg/dl) ± 40.1 AST (U/L) 27.3 ± 18.1 ALT (U/L) 31.6 ± 23.5 γ-gtp (IU/L) 68.3 ± Platelets ( 10 9 /L) 216 ± 73 FIB-4 index 1.64 ± 1.19 IV-7S (ng/ml) 4.02 ± 1.11 FT (pg/ml) 10.6 ± 6.8 sd, standard deviation; BMI, body mass index; HbA1c, hemoglobin A1c; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; AST, aspartate aminotransferase; ALT, alanine aminotransferase; γ-gtp, gamma glutamyl transpeptidase; IV-7S, 7S domain of type IV collagen; FT, free testosterone. Table 2 Correlations between the liver fibrotic markers and clinical variables FIB-4 index (n=248) IV-7S (n=140) r p value r p value Age NA NA BMI Fasting glucose HbA1c Visceral fat area SBP DBP Total cholesterol Triglycerides HDL-C LDL-C AST NA NA 0.43 < ALT NA NA γ-gtp 0.37 < < Platelet count NA NA FIB-4 index NA NA 0.47 < IV-7S 0.47 < NA NA Free testosterone IV-7S, 7S domain of type IV collagen; BMI, body mass index; HbA1c, hemoglobin A1c; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; AST, aspartate aminotransferase; ALT, alanine aminotransferase; γ-gtp, gamma glutamyl transpeptidase; NA, not applicable.

4 1140 Miyauchi et al. features of alcohol consumption. This significant association remained after adjusting for other possible confounding factors. Previously, Kim et al. conducted a single-center cross-sectional observational study of 495 healthy men to examine the relationship between a low level of serum total testosterone and NAFLD diagnosed by ultrasonography [25]. They reported that a low level of testosterone was independently associated with NAFLD Table 3 Associations between the FIB-4 index and variables including FT (n=248) Selected variables All variables Variable β p value β p value BMI (kg/m 2 ) NA NA Fasting glucose (mg/dl) NA NA HbA1c (%) Visceral fat area (cm 2 ) SBP (mm Hg) NA NA Triglycerides (mg/dl) NA NA HDL-C (mg/dl) NA NA LDL-C (mg/dl) γ-gtp (IU/L) 0.38 < < FT (pg/ml) < < FT, free testosterone; BMI, body mass index; HbA1c, hemoglobin A1c; SBP, systolic blood pressure; HDL-C, highdensity lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; γ-gtp, gamma glutamyl transpeptidase; NA, not applicable. [25]. Additionally, Seo et al. examined whether the total testosterone level was associated with NAFLD in 1,944 men who underwent ultrasonography at a health promotion center. The authors of that cross-sectional and longitudinal study showed the serum levels of total testosterone were significantly associated with ultrasonographic features of NAFLD, although the serum levels of total testosterone did not independently influence the development or regression of NAFLD [26]. Hua et al. investigated the association between SHBG, total testosterone, or free testosterone concentrations and NAFLD in a case-control study involving 240 patients with T2DM [27]. The multivariate analysis showed that serum SHBG levels were inversely associated with NAFLD, although free testosterone levels were not correlated with NAFLD [27]. However, neither of these studies assessed liver fibrosis. The mechanisms by which low serum testosterone concentrations are associated with NAFLD progression are not established. However, there are several possible explanations. In men, low serum testosterone concentrations induce the accumulation of visceral adipose tissue, with subsequent release of inflammatory cytokines and insulin resistance; this exacerbates steatosis [30]. Additionally, an increased concentration of circulating inflammatory cytokines is a risk factor for progression from NAFL to NASH and cirrhotic states [30]. In experimental studies, castration could render male Table 4 Associations between IV-7S and independent variables, including FT (n=140) Age-adjusted Selected variables All variables β p value β p value β p value Age (years) BMI (kg/m 2 ) NA NA Fasting glucose (mg/dl) NA NA NA NA HbA1c (%) NA NA NA NA Visceral fat area (cm 2 ) NA NA SBP (mm Hg) NA NA NA NA Triglycerides (mg/dl) NA NA NA NA HDL-C (mg/dl) NA NA NA NA LDL-C (mg/dl) NA NA NA NA AST (U/L) NA NA ALT (U/L) NA NA γ-gtp (IU/L) NA NA Platelet count ( 10 9 /L) NA NA FT (pg/ml) IV-7S, 7S domain of type IV collagen; FT, free testosterone; BMI, body mass index; HbA1c, hemoglobin A1c; SBP, systolic blood pressure; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; AST, aspartate aminotransferase; ALT, alanine aminotransferase; γ-gtp, gamma glutamyl transpeptidase; NA, not applicable.

5 FT and liver fibrosis in men with T2DM 1141 mice susceptible to liver inflammation [31]. Moreover, the serum testosterone levels of male and female mice correlated negatively with IL-17, CXCL-9, and CXCL- 10 expression in the liver, which was at least partly responsible for the liver inflammation observed. On the other hand, testosterone treatment was sufficient to completely suppress liver inflammation in female mice [31]. Additionally, testosterone treatment of female mice significantly reduced the expression of IL-17A, CXCL-9, and CXCL-10 within the liver. Therefore, we speculated that low levels of testosterone could be related to the inability to suppress inflammation, and could induce insulin resistance. These conditions may lead to the accumulation of visceral adipose tissues, which would exacerbate fat deposits in the liver by secreting free fatty acids. The combination of inflammation and fat deposits in the liver might lead to fibrosis, and eventually advanced NAFLD. Our study has some limitations. First, we included only men. There are several differences between men and women regarding the effects, balance, and normal ranges of hormones such as estrogen and testosterone. Future studies are needed to determine the effects of testosterone on NAFLD in women. Second, the ELISA kit (Cosmic Corporation Co., Ltd., Tokyo, Japan) was used to measure the free form of testosterone, because the equilibrium assay, which was recommended by the international guideline for late-onset hypogonadism, was unavailable. Additionally, only the free form of testosterone was measured to investigate testosterone levels; the levels of total testosterone and SHBG were not determined. The effects of these absent factors are unknown. Third, liver biopsies were not performed to confirm liver fibrosis. Fourth, subjects were enrolled at a single center, so there may be selection bias. A multicenter study is required. Finally, this was cross-sectional study. No cause-and-effect associations could thus be established in this analysis. It is necessary to consider the effect that improvements in the concentration of free testosterone may have in men with liver fibrosis. Although our study has several limitations, it did produce several notable results. In particular, our findings demonstrated a possible risk of liver fibrosis in men with T2DM. Moreover, serum free testosterone concentrations may help to predict which T2DM patients are at greater risk of developing advanced liver disease. This may help clinicians to prevent future cases of cirrhosis and hepatocellular carcinoma. Acknowledgements We thank Hisashi Yamashita for the visceral fat area measurements. Conflicts of Interest None of the authors have any potential conflicts of interest associated with this research. References 1. Younossi ZM, Stepanova M, Afendy M, Fang Y, Younossi Y, et al. (2011) Changes in the prevalence of the most common causes of chronic liver diseases in the United States from 1988 to Clin Gastroenterol Hepatol 9: e1. 2. Park SH, Jeon WK, Kim SH, Kim HJ, Park DI, et al. (2006) Prevalence and risk factors of non-alcoholic fatty liver disease among Korean adults. J Gastroenterol Hepatol 21: Ekstedt M, Hagstrom H, Nasr P, Fredrikson M, Stal P, et al. (2015) Fibrosis stage is the strongest predictor for disease-specific mortality in NAFLD after up to 33 years of follow-up. Hepatology 61: Cusi K (2016) Treatment of patients with type 2 diabetes and non-alcoholic fatty liver disease: current approaches and future directions. Diabetologia 59: Raff EJ, Kakati D, Bloomer JR, Shoreibah M, Rasheed K, et al. (2015) Diabetes mellitus predicts occurrence of cirrhosis and hepatocellular cancer in alcoholic liver and non-alcoholic fatty liver diseases. J Clin Transl Hepatol 3: Portillo-Sanchez P, Bril F, Maximos M, Lomonaco R, Biernacki D, et al. (2015) High prevalence of nonalcoholic fatty liver disease in patients With type 2 diabetes mellitus and normal plasma aminotransferase levels. J Clin Endocrinol Metab 100: El-Serag HB, Tran T, Everhart JE (2004) Diabetes increases the risk of chronic liver disease and hepatocellular carcinoma. Gastroenterology 126: Reid AE (2001) Nonalcoholic steatohepatitis. Gastroenterology 121: Tilg H, Moschen AR, Roden M (2017) NAFLD and diabetes mellitus. Nat Rev Gastroenterol Hepatol. 14:

6 1142 Miyauchi et al Suzuki A, Abdelmalek MF (2009) Nonalcoholic fatty liver disease in women. Womens Health (Lond) 5: Jimba S, Nakagami T, Takahashi M, Wakamatsu T, Hirota Y, et al. (2005) Prevalence of non-alcoholic fatty liver disease and its association with impaired glucose metabolism in Japanese adults. Diabet Med 22: Eguchi Y, Hyogo H, Ono M, Mizuta T, Ono N, et al. (2012) Prevalence and associated metabolic factors of nonalcoholic fatty liver disease in the general population from 2009 to 2010 in Japan: a multicenter large retrospective study. J Gastroenterol 47: Yatsuji S, Hashimoto E, Tobari M, Tokushige K, Shiratori K (2007) Influence of age and gender in Japanese patients with non-alcoholic steatohepatitis. Hepatol Res 37: Chitturi S, Farrell GC, Hashimoto E, Saibara T, Lau GK, et al. (2007) Non-alcoholic fatty liver disease in the Asia-Pacific region: definitions and overview of proposed guidelines. J Gastroenterol Hepatol 22: Anderson DC (1974) Sex-hormone-binding globulin. Clin Endocrinol (Oxf) 3: Selvin E, Feinleib M, Zhang L, Rohrmann S, Rifai N, et al. (2007) Androgens and diabetes in men: results from the Third National Health and Nutrition Examination Survey (NHANES III). Diabetes Care 30: Oh JY, Barrett-Connor E, Wedick NM, Wingard DL, Rancho Bernardo Study (2002) Endogenous sex hormones and the development of type 2 diabetes in older men and women: the Rancho Bernardo study. Diabetes Care 25: Stellato RK, Feldman HA, Hamdy O, Horton ES, McKinlay JB (2000) Testosterone, sex hormone-binding globulin, and the development of type 2 diabetes in middle-aged men: prospective results from the Massachusetts male aging study. Diabetes Care 23: Haffner SM, Shaten J, Stern MP, Smith GD, Kuller L (1996) Low levels of sex hormone-binding globulin and testosterone predict the development of non-insulindependent diabetes mellitus in men. MRFIT Research Group. Multiple Risk Factor Intervention Trial. Am J Epidemiol 143: Biswas M, Hampton D, Newcombe RG, Rees DA (2012) Total and free testosterone concentrations are strongly influenced by age and central obesity in men with type 1 and type 2 diabetes but correlate weakly with symptoms of androgen deficiency and diabetes-related quality of life. Clin Endocrinol (Oxf) 76: Hofstra J, Loves S, van Wageningen B, Ruinemans- Koerts J, Jansen I, et al. (2008) High prevalence of hypogonadotropic hypogonadism in men referred for obesity treatment. Neth J Med 66: Svartberg J, von Muhlen D, Sundsfjord J, Jorde R (2004) Waist circumference and testosterone levels in community dwelling men. The Tromso study. Eur J Epidemiol 19: Glass AR, Swerdloff RS, Bray GA, Dahms WT, Atkinson RL (1977) Low serum testosterone and sexhormone-binding-globulin in massively obese men. J Clin Endocrinol Metab 45: Fukui M, Soh J, Tanaka M, Kitagawa Y, Hasegawa G, et al. (2007) Low serum testosterone concentration in middle-aged men with type 2 diabetes. Endocr J 54: Kim S, Kwon H, Park JH, Cho B, Kim D, et al. (2012) A low level of serum total testosterone is independently associated with nonalcoholic fatty liver disease. BMC Gastroenterol 12: Seo NK, Koo HS, Haam JH, Kim HY, Kim MJ, et al. (2015) Prediction of prevalent but not incident nonalcoholic fatty liver disease by levels of serum testosterone. J Gastroenterol Hepatol 30: Hua X, Sun Y, Zhong Y, Feng W, Huang H, et al. (2014) Low serum sex hormone-binding globulin is associated with nonalcoholic fatty liver disease in type 2 diabetic patients. Clin Endocrinol (Oxf) 80: Committee of the Japan Diabetes Society on the Diagnostic Criteria of Diabetes Mellitus, Seino Y, Nanjo K, Tajima N, Kadowaki T, et al. (2010) Report of the committee on the classification and diagnostic criteria of diabetes mellitus. J Diabetes Investig 1: Sterling RK, Lissen E, Clumeck N, Sola R, Correa MC, et al. (2006) Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology 43: Tarantino G, Savastano S, Colao A (2010) Hepatic steatosis, low-grade chronic inflammation and hormone/growth factor/adipokine imbalance. World J Gastroenterol 16: Schwinge D, Carambia A, Quaas A, Krech T, Wegscheid C, et al. (2015) Testosterone suppresses hepatic inflammation by the downregulation of IL-17, CXCL-9, and CXCL-10 in a mouse model of experimental acute cholangitis. J Immunol 194: