非酒精性脂肪肝對於心律變異度的影響

Abstract

非酒精性脂肪肝病（Non-alcoholic liver disease）是包含一系列的肝臟疾病，從簡單的脂肪變性(steatosis)到非酒精性脂肪性肝炎（non-alcoholic steatohepatitis），肝纖維化和肝硬化。非酒精性脂肪肝在台灣成人的盛行率為11.5%到41%。它的發生與肥胖，第二型糖尿病，高血脂症和代謝症候群有顯著相關。已有研究指出心臟血管疾病是這一類病人的主要死亡原因之一。它的致病機轉涉及許多步驟，包括胰島素抗性及氧化壓力等。目前非酒精性脂肪肝被認為可能是代謝症候群在肝臟的表現，但也有專家認為它在造成心血管疾病上有扮演自身獨立的角色。目前有許多研究指出非酒精性脂肪肝與冠狀動脈粥狀硬化在排除傳統心血管疾病的危險因子的影響後，仍有顯著的相關性。在Framinghan study和ARIC study已發現心臟自律神經功能異常與心血管疾病有相關。心律變異度是一個廣泛用來評估心臟自律神經活性的非侵入性檢查。心律變異度指標若降低則表示自律神經系統功能異常。心律變異度降低與高血壓，第二型糖尿病，高血脂症，肥胖和代謝症候群有顯著相關性。非酒精性脂肪肝與心律變異度降低是否有相關性則尚未有研究發表。非酒精性脂肪肝已被發現伴隨血液中瘦素(leptin)的上升。瘦素與胰島素抗性已被證實會造成交感神經活化，因此我們假設非酒精性脂肪肝會伴隨交感神經系統的活化，進而呈現心律變異度降低。 我們設計了一個橫斷性研究。我們在台北市遠東聯合診所的健康檢查中心招募不具已知心血管疾病受試者。在一年半年中完成收集1000位受試者，分為一組試驗組(非酒精性脂肪肝)與一組對照組(無非酒精性脂肪肝)。本研究採用肝臟超音波來診斷脂肪肝。為了要排除其他原因造成的脂肪肝，若受試者每週攝取過量的酒精，有B型肝炎或C型肝炎病毒感染，目前正服用可能導致脂肪肝的藥物者，將被排除在非酒精性脂肪肝的診斷之外。每一位受試者採仰臥姿勢，用霍特氏心電圖機作五分鐘心律變異度測量。五分鐘的心律變異度檢查可以得到的指標包括: (1)時域分析(time domain analysis)的心跳間格標準差(standard deviation of normal RR intervals, SDNN) 與相鄰正常心跳間隔差值平方和的平方根(root mean square of standard deviation, rMSSD) 二項數值。(2)頻域分析(frequency domain analysis)的高頻(high frequency power, HF )和低頻(low frequency power, LF)二項數值。我們收集了各項驗血結果，家族史和過去病史，藥物與酒精的使用量和吸煙狀況。在1000位受試者中，497位受試者有增加檢測空腹胰導素與瘦素。 在總共1000位受試者中，我們作Student’s t 檢測發現非酒精性脂肪肝組的四項心律變異度指標( Ln SDNN，Ln rMSSD，Ln LF，Ln HF) 都比對照組降低 (3.53 vs 3.65 ms， 3.04 vs 3.23 ms， 5.34 vs 5.52 ms2， 4.86 vs 5.24 ms2; 所有的 p <0.05) 。在單獨對第二群497位受試者中，我們使用線性複回歸模型 (multiple linear regression model)在調整年齡、 性別、 高血壓、代謝症候群、高血脂症、抽煙、 身體質量指數(BMI)、估計腎小球濾過率(eGFR)、 胰島素抗性 (HOMA-IR) 、瘦素後，發現非酒精性脂肪肝與心律變異度指標中的Ln SDNN呈現負相關性(p <0.05) 。我們的結論是在調整傳統心血管疾病因子、胰島素抗性與瘦素之後，非酒精性脂肪肝與心律變異度降低有相關性。

Non-alcoholic fatty liver disease (NAFLD) refers to a spectrum of liver disorders, ranging from simple steatosis to steatohepatitis (NASH), advanced fibrosis, and cirrhosis. In Taiwan, the prevalence of NAFLD was 11.5% to 41% in adult population. The cardiovascular disease (CVD) is the second leading cause of death in NAFLD patients. The pathogenesis of NAFLD is attributed to a multi-hit process mainly involving insulin resistance and oxidative stress. The NAFLD was associated with obesity, type 2 diabetes mellitus (DM), dyslipidemia and metabolic syndrome (MS) in many experimental and clinical studies. The NAFLD is now considered to be the hepatic manifestation of metabolic syndrome, although the nature of their relationship remains debated. Several reports showed the associations of NAFLD with carotid and coronary atherosclerosis independently of classical risk factors of CVD. Abnormal cardiac autonomic function was associated with CVD in the Framinghan heart study and the ARIC study. Heart rate variability (HRV) is an established tool for assessment of cardiac autonomic function. Decreased HRV represents the autonomic dysfunction. The associations between decreased HRV and hypertension, DM, dyslipidemia and MS have been demonstrated. The impact of NAFLD on HRV has not been reported. Besides, the leptin has been shown to cause sympathetic activation and its serum level is elevated in NAFLD patients. Therefore, we hypothesize that the sympathetic activation due to insulin resistance and increased leptin might cause decreased HRV in NAFLD patients. We designed a cross-sectional study and recruited 1000 subjects from Aug 2010 to Feb 2012 at the health checkup center of the Far Eastern Poly Clinic. The subjects with documented CVD, congestive heart failure and endstage renal disease were excluded. We defined one group of subjects with NAFDL and another without NAFLD as the control group. Fatty liver was confirmed if at least one of the following three findings were present: increased liver echogenicity with evident contrast between liver and kidney; the blurring of vessels; and deep attenuation of the ultrasound signal. To exclude other causes of NAFLD, the subjects with a weekly intake of alcohol over 140g in male and 70g in female, hepatitis B, hepatitis C infection and use of drugs that may induce NAFLD were excluded from the group of NAFLD. The subjects underwent 5-minutes HRV measurements by an ambulatory Holter electrocardiogram in supine position. The HRV was analyzed by the following indices: 1) the time domain with the standard deviation of NN (SDNN) intervals, rMSSD (root mean square of successive differences between adjacent N-N intervals); and 2) the frequency domain with the low frequency (LF) and high frequency (HF) power. Data were expressed as the natural logarithm (ln) of the above HRV indices. The result of blood test, family history and medical history such as previous systemic and liver disease, medication, alcohol intake and smoking status were collected. In the total 1000 subjects, 497 subjects received the blood test of fasting insulin and letpin. The clinical characteristics were compared by the Student’s t test for continuous variables and Chi-Square test for categorical variables. In the 497 subjects with fasting insulin and leptin data, the HRV indices (Ln SDNN, Ln rMSSD, Ln HF and Ln HF) was significantly decreased in the NAFLD group ( 3.50 vs 3.64 ms, 3.04 vs 3.23 ms, 5.27 vs 5.49 ms2, 4.84 vs 5.24 ms2; all p <0.05). In the 497 subjects, NAFLD was negatively associated with Ln SDNN, after adjustment for the effects of age, sex, hypertension, dyslipidemia, MS, body mass index, smoking, estimated glomerular infiltration rate, homeostasis model assessment of insulin resistance and leptin (p <0.05). In conclusion, NAFLD is associated with decreased HRV independent of classic cardiovascular risk factors, insulin resistance and letpin.