探討男性睪固酮缺乏與第二型糖尿病之關係

The Association of Testosterone Deficiency with Type 2 Diabetes Mellitus in Men

何承勳(Chen-Hsun Ho)

指導教授 ： 余宏政 ； 劉詩彬 ； 趙福杉

國立臺灣大學/醫學院/臨床醫學研究所/博士(2015年)

https://doi.org/10.6342/NTU.2015.10891

摘要

背景

許多研究已經證實男性睪固酮缺乏與第二型糖尿病之間有密切的關係。臨床上可以觀察到兩者常有共病的關係，例如睪固酮低下或攝護腺癌患者接受雄激素剝奪治療。睪固酮缺乏與第二型糖尿病都對男性的健康有顯著的負面影響：它們都會造成勃起功能障礙，也都被證明是心血管疾病和死亡的獨立危險因子。然而目前對於他們共病的機轉仍然沒有完全了解。目前的證據顯示，兩者可能互為因果，或者一些其他的因素，例如肥胖或性激素結合球蛋白，也能在兩者的關係中扮演重要的角色。
過去相關的臨床研究，主要研究的對象是已經診斷為第二型糖尿病的患者，其中大部分正在接受糖尿病的治療。糖尿病前期是一種血清葡萄糖濃度已經上升，卻未達到糖尿病的診斷標準的狀態。在許多研究中，它已被認為是心血管疾病和腎臟疾病的風險因子，這讓我們也想研究糖尿病前期的男性是否也有較高的睪固酮低下的風險，釐清這個問題有助於了解：(一）澄清睪固酮缺乏在第二型糖尿病的發展中的角色; (二）辨識適合睪固酮替代療法的潛在對象，並進行下研究針對這些患者給予睪固酮補充是否可以減少其演進至第二型糖尿病的機會，或預防心血管疾病。
另一方面，文獻上極少討論治療第二型糖尿病是否可逆轉睪固酮缺乏的狀況。根據統計，約有四分之一的糖尿病患者是屬於未診斷與未治療的，這個族群的血清睪固酮濃度很少被報導過，研究這群病人的血清睪固酮濃度有助於進一步闡明睪固酮缺乏與胰島素抵抗之間的因果關係。
最後，用於研究睪固酮缺乏與葡萄糖耐受不良之間的關係的動物模型尚未被建立。雖然大鼠或小鼠是最廣泛被應用於各個醫學研究領域的動物模型，但是以往關於睪固酮缺乏之大小鼠的代謝特徵的研究結果是相互矛盾的。因此，我們的研究也評估閹割雄性大鼠是否適合作為研究睪固酮缺乏與葡萄糖耐受不良和代謝症候群相關研究的動物模型。

研究方法

臨床研究：不同階段第二型糖尿病男性患者睪固酮缺乏的盛行率及危險因子

這是一個橫斷面的研究。我們從臺灣大學附設醫院健康管理中心的資料庫收集資料，收集1,306位於2009年接受性激素檢測的男性。所有研究對象均完成一份自填的問卷，用來收集他們的基本資料和病史，所有研究對象均經內科醫師看診，且做一次詳細的身體檢查。每個研究對象均接受兩次血液採集：第一個樣本是禁食一夜後在早上8點和10點之間收集，用來測定空腹血糖、男性賀爾蒙和其他血清生化資料；第二個樣本是收集午餐後兩小時的血液，用來測量的餐後血糖值。總睪固酮與性激素結合球蛋白的濃度由化學發光微粒子免疫分析進行測定，並利用Vermeulen公式計算出游離睪固酮。總睪固酮低下的定義為總睪固酮水準小於300毫微克/分升，和游離睪固酮低下的定義為小於6毫微克/分升。糖尿病的標準為：(一)病人有之前已診斷出糖尿病(已知糖尿病患者)，或(二)血糖變數達到糖尿病標準(新診斷糖尿病患者)：空腹血糖大於等於126毫克/分升，餐後2小時的血糖大於等於200毫克/分升，或糖化血紅蛋白大於等於6.5% (Association 2012)。糖尿病前期的診斷標準需滿足以下條件：(一)空腹血糖100-125毫克/分升，(二)餐後2小時的血糖140-199毫克/分升，或(三)糖化血紅蛋白5.7% - 6.4%。藉由上述資料，我們將所有1,306按血糖及過去疾病史將其分類為正常血糖耐受性、糖尿病前期、新診斷糖尿病、已知糖尿病等各組，我們比較各組的血清睪固酮濃度及睪固酮低下之盛行率，並使用邏輯回歸分析，校正年齡、身高體重指數、腰圍、代謝症候群等可能的干擾因子。

基礎研究：評估去勢大鼠是否可以用於睪固酮缺乏與葡萄糖耐受不良相關的研究

本實驗採用成年雄性Sprague Dawley（SD）大鼠(週齡10-12週；體重300到350公克）。將大鼠隨機分組（每組約6-8隻）：假手術對照組、去勢組、去勢後注射低劑量睪固酮組 (每週注射丙酸睪固酮，每公斤0.5毫克)、中劑量組 (每週注射，每公斤2毫克)和高劑量組 (每週注射，每公斤8毫克)。所有的大鼠餵食標準飲食。每隻大鼠的葡萄糖耐受性試驗共進行兩次，分別在第8週和16週。首先大鼠隔夜禁食16個小時，試驗從上午8時開始葡萄糖耐受性試驗，在時間0分鐘以及在15，30，60，90，和120分鐘分別採血進行測量。繪製每個時間點的血糖濃度圖，以及計算各組大鼠血糖濃度曲線下的面積（AUC）。另外在0和15分鐘採集兩個血液樣本，作為空腹和葡萄糖誘導之後的胰島素濃度分析。大鼠在第16週做完腹腔葡萄糖耐受性試驗後立即犧牲，切除胰臟和部分的內臟脂肪，將胰臟附近的脂肪去除後秤重，將胰臟進一步分為三個部分：頭段、中段、尾段。所有的組織（胰腺和內臟脂肪）進行石蠟包埋，且切下5 µm厚的切片用蘇木精和伊紅染色。部分的胰臟利用胰島素的抗體去染β細胞，這些切片用來進行脂肪細胞的尺寸的測量、胰臟形態檢查、β-測量細胞量。

結果

臨床研究：不同階段第二型糖尿病男性患者睪固酮缺乏的盛行率及危險因子

1306位的男性受試者中，有577（44.1%）位是血糖正常、543（41.5%）位是糖尿病前期、186（14.4%）位是糖尿病。糖尿病前期的男性，在調整年齡後總睪固酮低下的勝算比為1.87（95%信賴區間[CI]：1.38-2.54），而糖尿病患者的勝算比為2.38（95%信賴區間：1.57-3.60）。糖尿病前期之男性的勝算比，在進一步校正身體質量指數、腰圍、代謝症候群組成的因子和代謝症候群的多變數分析下仍然顯著，在調整代謝症候群之後，其總睪固酮低下的勝算比為1.49（95%信賴區間：1.08-2.06），幾乎等於糖尿病患者的勝算比1.50。
　在診斷為糖尿病的186位患者中，有81位為新診斷的糖尿病患者，105位為已知糖尿病患者。已知糖尿病患者的年齡較大，有較高的舒張壓和更高的空腹血糖濃度。兩組的總睪固酮的中間值為（358 [ 155 ] ng/dl vs 363 [ 154 ] ng/dl，P = 0.68），游離睪固酮的中間值為（7.2 [ 2.5 ] ng/dl vs 7.4 [ 2.4 ] ng/dl，P = 0.84），性激素結合球蛋白的中間值為（27.3 [ 22.3 ] nmol�L和28.7 [ 14.9 ] nmol�L，P = 0.46），兩組間沒有顯著的差異。在新診斷和已知第2型糖尿病組，其總睪固酮低下的患病率分別為28.4%和26.7%，游離睪固酮低下的患病率分別為21%和19%。新診斷和已知的第2型糖尿病總睪固酮低下的危險因子類似，包括糖化血紅蛋白≥7%、性激素結合球蛋白＜20 nmol / L、BMI≥27公斤/米2、腰圍≥90釐米、三酸甘油酯≥150毫克/分升、尿酸≥7毫克/分升，與代謝症候群。相較於PSA小於1毫微克/毫升，PSA≥2.5毫微克/毫升為保護因子。

基礎研究：評估去勢大鼠是否可以用於睪固酮缺乏與葡萄糖耐受不良相關的研究

睪丸切除術導致血清中睪固酮幾乎消失，大鼠在去勢兩周後，血清睪固酮都低於10毫微克/分升，與假手術對照組相比，低或高劑量睪固酮的補充組，在第8和16週，其大鼠有顯著較低或高睪固酮濃度。初始體重各組之間是相似的，但之後去勢大鼠相較於假手術對照組相比，體重明顯變輕。對去勢大鼠補充睪固酮後體重增加，超過去勢大鼠，但低於假手術對照組。去勢大鼠和補充睪固酮組的脂肪細胞大小顯著低於假手術對照組。第8周時，葡萄糖耐受性試驗的各組血糖曲線下面積表現無顯著差異。到第16周，去勢大鼠的血糖曲線下面積比較發現，去勢及去勢後補充低劑量睪固酮組的血糖曲線下面積顯著大於假手術對照組及補充中或高劑量的睪固酮組。在第8週與16週，無論是空腹或葡萄糖誘導的胰島素濃度皆與血清裡的睪固酮的趨勢相關，去勢大鼠比起假手術對照組有顯著較低的空腹和葡萄糖誘導的胰島素水準。無論是假手術對照組，去勢睪固酮或補充睪固酮狀態，皆不影響胰腺形態。

結論

糖尿病前期會增加其總睪固酮低下的風險。此風險在調整身體質量指數後、腰圍、代謝症候群的組成因子或代謝症候群後仍然顯著。調整代謝症候群後，糖尿病前期的男性族群，其總睪固酮低下的風險與糖尿病患者的風險幾乎相等，我們的研究顯示，糖尿病前期的男性也應要有常規的睪固酮檢測。新診斷第二型糖尿病的男性患者，其血清睪固酮濃度與睪固酮低下的盛行率，與已知第二型糖尿病的患者相似。在已診斷糖尿病的患者，糖化血色素與血清睪固酮呈現負相關。新診斷和已知第二型糖尿病男性患者總睪固酮低下之危險因子包括較高的糖化血色素、較低的性激素結合球蛋白、肥胖、高尿酸血症、高血糖症與代謝症候群；另外，上升的攝護腺特異抗原會降低睪固酮缺乏的風險。
基礎研究方面，我們的研究顯示，在經過十六週由去勢所引發睪固酮缺乏的狀態下，雄性大鼠的體重會降低，會有空腹血糖異常與葡萄糖耐受不良的情況，這些血糖的變化起因於胰島素分泌減少所導致。反觀人類男性在缺乏雄激素的影響下，會導致肥胖，且因為胰島素敏感性降低而導致葡萄糖耐受性不良。當雄性大鼠被應用於睪固酮和代謝之間的研究時，應該將這些物種間的生理差異列入考慮。

關鍵字

睪固酮缺乏 ； 糖尿病前期 ； 第二型糖尿病 ； 胰島素抗性 ； 大鼠模式

並列摘要

Background
The relationship between testosterone deficiency and type 2 diabetes mellitus has been well established in men. The connection is frequently observed in various clinical scenarios, such as late-onset hypogonadism and prostate cancer patients receiving androgen deprivation therapy. Both conditions have a significant negative impact on men’s health: they both cause lower urinary tract symptoms and erectile dysfunction; moreover, both of them are proved as an independent risk factor for cardiovascular disease and mortality. Nevertheless, the underlying mechanism of their coexistence remains not fully understood. Current evidence supports that testosterone deficiency can be either a cause or a consequence of insulin resistance. Besides, some other factors, such as obesity or sex hormone binding globulin, can also play a role within this relationship.
Previous related clinical studies were mainly based on subjects with overt type 2 diabetes, most of whom were receiving anti-diabetes treatment. Prediabetes is a state of increased serum glucose concentration, while it does not reach the criteria of overt diabetes mellitus. In many studies, it has been considered as a risk factor for cardiovascular disease and nephropathy. It raised the question of whether it is also associated testosterone deficiency. The purpose of defining the serum testosterone concentration of prediabetic subjects includes: 1) to clarify the role of testosterone deficiency in the development of type 2 diabetes; 2) to identify the potential subjects for testosterone replacement therapy.
On the other hand, it has been less addressed whether the treatment of type 2 diabetes can reverse the condition of testosterone deficiency. And the epidemiologic characteristics of untreated, newly-diagnosed type 2 diabetic men has been rarely addressed. To investigate the serum testosterone concentration of the untreated, newly-diagnosed type 2 diabetic men may help to further clarify whether testosterone deficiency is a cause or a consequence of insulin resistance.
Lastly, the optimal animal model for the study of the relationship between testosterone deficiency and glucose intolerance has not yet been established. While rats or mice are most widely used animal model in every field of medical research, previous studies evaluating the metabolic characteristics of rats or mice of testosterone deficiency were conflicting. We therefore aimed to evaluate the suitability of castrated male rats as an animal model for the related studies regarding testosterone deficiency, glucose intolerance, and metabolic syndrome.

Method
Clinical Studies: The prevalence and risk factors of testosterone deficiency in men at different stages of type 2 diabetes
This is a cross-sectional study. We obtained the data from the database of Health Management Center, National Taiwan University Hospital. In 2009, a total of 1,306 men receiving sex hormone measurement as part of their medical examination constituted the study subjects of the current study. The study protocol was approved by the institutional review board (IRB) of National Taiwan University Hospital (201207058RIC). All participants completed a self-administered questionnaire to collect their basic demographic data and medical histories. All subjects were then interviewed by an internal medicine physician, and a detailed physical examination was performed. Two blood samples were collected from each subject: the first sample was collected after an overnight fast between 8 am and 10 am, and was used to measure fasting blood glucose, sex hormones, and other biochemical data; the second blood sample was collected two hours after a standard lunch and was used to measure the postprandial glucose. Total testosterone and SHBG were measured by chemiluminescent microparticle immunoassay. Free testosterone was calculated by the formula proposed by Vermeulen. Low total testosterone was defined by total testosterone <300 ng/dL [16-18], and low free testosterone was defined by free testosterone <6 ng/dL. Prediabetes was diagnosed if any of the following criteria was met: 1) fasting glucose 100-125 mg/dL (IFG), 2) two-hour postprandial glucose 140-199 mg/dL (IPG), or 3) HbA1c 5.7%-6.4%. Diabetes was diagnosed if the patient had a prior history of diabetes or if the glycemic variables reached the criteria of diabetes: fasting glucose ≥126 mg/dL, two-hour postprandial glucose ≥200 mg/dL, or HbA1c ≥6.5%. Continuous data are presented as the mean ± standard deviation (SD), and categorical data are presented as count and percentage (%).Logistic regression was performed to obtain the odds ratios for TD in men with prediabetes and diabetes compared with those with normoglycemia. Five statistical models were used for multivariate analyses: Model 1, adjusted for age; Model 2, adjusted for age and body mass index (BMI); Model 3, adjusted for age and waist circumference; Model 4, adjusted for age and the number of MetS components; Model 5, adjusted for age and MetS. Multiple linear regression was performed to assess the association between total and free testosterone and prediabetes or diabetes.

Basic Studies: Evaluating castrated male rats as an animal model of glucose intolerance
Male Sprague-Dawley rats (10-12 weeks) were randomly divided into five groups (n=6-8 in each): sham-operated, castrated, and castrated with low- (intramuscular injection of testosterone propionate 0.5 mg/kg/week), intermediate- (2 mg/kg/week), and high-dose (8 mg/kg/week) of testosterone replacement. Each animal received intraperitoneal glucose tolerance test (IPGTT) twice on week 8 and 16 respectively; the area under the curve (AUC) of glucose concentrations were calculated to represent the glucose tolerance. Fasting and glucose-induced insulin concentrations were measured at 0 and 15 minutes during IPGTT respectively. After the rats were euthanized on week 16, visceral fat and pancreas were prepared for morphologic measurements.

Results
Clinical Studies: The prevalence and risk factors of testosterone deficiency in men at different stages of type 2 diabetes
Normoglycemia, prediabetes, and diabetes were diagnosed in 577 (44.2%), 543 (41.6%), and 186 (14.2%) men, respectively. Prediabetes was associated with an increased risk of subnormal total testosterone compared to normoglycemic individuals (age-adjusted OR=1.87; 95%CI=1.38-2.54). The risk remained significant in all multivariate analyses. After adjusting for MetS, the OR in prediabetic men equals that of diabetic patients (1.49 versus 1.50). IFG, IPG, and HbA1c 5.7%-6.4% were all associated with an increased risk of testosterone deficiency, with different levels of significance in multivariate analyses. However, neither prediabetes nor diabetes was associated with subnormal free testosterone in multivariate analyses. Men with previously known T2DM were older and had higher diastolic pressure and greater fasting glucose. There was no significant difference in total (358.0 [155.0] ng/dL vs 363.0 [154.0] ng/dL, P=0.68) and free (7.2 [2.5] ng/dL vs 7.4[2.4] ng/dL, P=0.84) testosterone and SHBG (27.3 [22.3] nmol/L vs 28.7 [14.9] nmol/L, P=0.46). The prevalence of low total and free testosterone were 28.4% and 21.0% in men with newly diagnosed T2DM, and were 26.7% and 19.0% in those with previously known T2DM. In men with previously known T2DM, better glycemic control (HbA1c<7%) was associated with a higher level of total testosterone and a lower risk of low total testosterone. Men with newly diagnosed and previously known T2DM shared similar risk factors of low total testosterone, including high HbA1c (≥7%), low SHBG (<20 nmol/L), obesity, hyperuricemia, hypertriglycemia, and metabolic syndrome. Elevated prostate-specific antigen (PSA) was a protective factor of low total testosterone. However, none of these factors was associated with low free testosterone.

Basic Studies: Evaluating castrated male rats as an animal model of glucose intolerance
Castration decreased the body weight and the adipocyte size, which were restored by testosterone replacement. Compared to controls, the castrated rats had a significantly greater AUC glucose and lower fasting and glucose-induced insulin concentrations on week 16. Testosterone replacement generally restored the insulin secretion and glucose tolerance in a dose-dependent fashion. The pancreatic islet morphology and the β-cell mass were not significantly different among groups.

Conclusion
Men with prediabetes are at an increased risk of subnormal total testosterone. The risk is reduced, but remains significant after adjustment for BMI, waist circumference, the number of MetS components, or MetS. After adjustment for MetS, the risk for TD in men with prediabetes is almost equal to that of men with diabetes. The substantially increased risk suggests that testosterone should be measured routinely in men with prediabetes. The prevalence and the risk factors of testosterone deficiency are similar between the newly diagnosed and previously known type 2 diabetic men. In men with previously known T2DM, the HbA1c is inversely associated with the serum testosterone. Men with newly diagnosed and previously known T2DM have common risk factors for low total testosterone, including high HbA1c, low SHBG, obesity, hyperuricemia, hypertriglycemia, and metabolic syndrome. Elevated PSA is associated with a lower risk of testosterone deficiency in type 2 diabetic men.
Castration-induced androgen deficiency in male rats reduces body weight and impairs glucose tolerance in part by attenuating fasting and glucose-induced insulin secretion. This is in contrast to the effect of androgen deficiency in humans, which frequently leads to obesity and impaired glucose tolerance through decreased insulin sensitivity. The interspecies difference should be carefully considered whenever male rats are applied in the research associated with the interaction between testosterone and metabolism.

並列關鍵字

testosterone deficiency ； prediabetes ； type 2 diabetes mellitus ； insulin resistance ； rat model

參考文獻

1 Association AD. Diagnosis and classification of diabetes mellitus. Diabetes Care 2012;35 Suppl 1:S64-71.

2 Garcia-Cruz E, Leibar-Tamayo A, Romero-Otero J, Asiain I, Carrion A, Castaneda R, Mateu L, Luque P, Cardenosa O, Alcaraz A. Marked Testosterone Deficiency-Related Symptoms May be Associated to Higher Metabolic Risk in Men with Low Testosterone Levels. J Sex Med 2014;11:2292-2301.

3 Martinez-Jabaloyas JM, group D-Ss. Testosterone deficiency in patients with erectile dysfunction: when should a higher cardiovascular risk be considered? J Sex Med 2014;11:2083-2091.

4 Corona G, Rastrelli G, Monami M, Guay A, Buvat J, Sforza A, Forti G, Mannucci E, Maggi M. Hypogonadism as a risk factor for cardiovascular mortality in men: a meta-analytic study. Eur J Endocrinol 2011;165:687-701.

5 Araujo AB, Dixon JM, Suarez EA, Murad MH, Guey LT, Wittert GA. Clinical review: Endogenous testosterone and mortality in men: a systematic review and meta-analysis. J Clin Endocrinol Metab 2011;96:3007-3019.