B型肝炎及C型肝炎高盛行地區其非酒精性脂肪肝 盛行率、發生率及其相關決定因素之探討： 社區型前瞻性研究

Prevalence, Incidence, and Determinants of Non-alcoholic Fatty Liver Disease in Hyperendemic Hepatitis B/C Area: A Population-based Prospective Study

賴俊隆(Chun-Lung Lai)

指導教授 ： 陳秀熙

國立臺灣大學/公共衛生學院/預防醫學研究所/碩士(2005年)

https://doi.org/10.6342/NTU.2005.02144

摘要

非酒精性脂肪肝病變(Non-alcoholic fatty liver disease, NAFLD)是常見的肝臟病變。然而有關於非酒精性脂肪肝病變的盛行率及發生率卻較少研究提及，再者於B型肝炎及C型肝炎高盛行地區有關於非酒精性脂肪肝之相關決定因素之研究探討更闕如。研究目的：本研究主要目的將探討(1)非酒精性脂肪肝之盛行率及發生率估計；(2)探討除了腹部超音波對非酒精性脂肪肝作診斷外，進一步探究能早期偵測出非酒精性脂肪肝之指標及其決定因素為何；(3)探究並說明決定性相關因子對脂肪肝之影響，尤其是代謝症候群，而進一步探討非酒精性脂肪肝與B/C型病毒性及生活習慣之影響。研究方法: 67804位民眾參與基隆地區整合式篩檢，其中以兩階段方式進行肝癌篩檢，肝癌高危險群定義為:B型或C型病毒性肝炎血清學檢驗陽性或肝功能指數異常、胎兒蛋白指數異常者須接受第二階段腹部超音波檢查，其檢查結果包括脂肪肝、肝硬化及肝癌等。根據腹部超音波檢查結果受檢民眾可分為疾病世代以及健康世代，並可依此兩種結果進行盛行率及發生率估算。本研究以首次發生脂肪肝個案進行盛行率估算。此外，首次完全正常者為健康世代，再進行每年追蹤檢查，其後發生脂肪肝病變的個案即為新個案。以概似對比檢定或ROC曲線計算非酒精性肝病變的身體質量指數(BMI)、尿酸及ALT最佳切點值估計。以盛行率/發生率(P/I)比值方式進行非酒精性脂肪肝平均滯留期估算。以時間依存性Cox回歸模式進行非酒精性脂肪肝相關因子分析。結果: 以腹部超音波進行非酒精性脂肪肝診斷結果，20歲以上之高危險群民眾非酒精性肝病變的總盛行率為35.38%，其中男性為38.22% 女性為33.25%；其非酒精性脂肪肝發生率為14.92%，其中男性為15.11%及女性14.82%。同時利用log(ALT)， BMI及 log(serum uric acid)等項目進行所有研究對象的非酒精性脂肪肝預測，以-0.59為切點，結果非酒精性肝病變的盛行率為28%、發生率為7.5%；於高危險群，其非酒精性脂肪肝平均滯留時間為2.37年，一般民眾平均滯留時間為3.76年。控制年齡及性別之單變項分析，於肝功能指數偏高族群中，代謝症候群仍具統計顯著意義(aOR=2.57(2.41-2.75))，特別於非B型且非C型肝炎帶原者更具顯著(aOR=3.28(3.03-3.54))。另外於B型及C型肝炎皆為陽性者為(aOR=2.61(1.14-5.95))、B型肝炎陽性且C型肝炎為陰性族群為(aOR=1.53(1.29-1.83))。此外，除了代謝症候群之外，其他顯著相關因子有高總膽固醇(aOR=1.15(1.08-1.22))、高尿酸(aOR=1.70(1.61-1.81))及有嚼檳榔習慣(aOR=1.19(1.08-1.31))。控制年齡性別因子後，發現代謝症候群對非酒精性脂肪肝效應會隨著病毒帶原盛行的不同而不同，其中以B型及C型肝炎皆帶原者最顯著(aOR=7.76(1.31-45.90))，B型肝炎陽性且C型肝炎為陰性之族群則次之(aOR=1.89(1.31-2.72))，B型肝炎陰性且C型肝炎陽性族群具邊緣性顯著統計意義為(aOR=1.82(0.99-3.33))，而於B型肝炎陽性且C型肝炎陰性為(aOR=1.53(1.29-1.83))。除此之外，高尿酸為(aOR=1.30(1.03~1.63))，運動習慣為(aOR=0.76(0.67-0.87))。於肝功能指數ALT小於45 IU/dL者發現，控制其他相關因子後，代謝症候群對於非酒精性脂肪肝的效應相當。高尿酸對於非酒精性脂肪肝仍具其顯著獨立作用。結論：此篇社區型前瞻性研究，可於肝癌高危險群及一般族群中估算非酒精性脂肪肝病變發生率及盛行率。代謝症候群(直接或間接透過肝功能指數ALT)對於非酒精性肝病變的影響，在B型肝炎與C型肝炎帶原狀況不同而不同。高尿酸血症與高身體質量指數(BMI)對於非酒精性肝病變發生具顯著獨立性作用存在。

關鍵字

非酒精性脂肪肝病變 ； 盛行率 ； 發生率 ； 代謝症候群 ； 整合式篩檢 ； B型肝炎 ； C型肝炎 ； 時間依存性Cox回歸模式

並列摘要

Background Non-alcoholic fatty liver disease (NAFLD) is the most common liver disease. However, little is known about prevalence and incidence rate and its determinants in hyperendemic hepatitis B/C area.
Objectives We aimed to (1) estimate prevalence and incidence of NAFLD; (2) to develop early indicators other than abdominal sonography to identify suspected NAFLD; (3) and to elucidate the effect of determinants, particularly metabolic syndrome (MS), on NAFLD and fatty liver taking viral hepatitis and other life style factors into account.
Methods A total of 67804 study subjects were enrolled for analysis derived from a community-based screening program in Keelung, Taiwan. One of the diseases for screening in this program was liver cancer with two stage method where participants were assessed by positive results of five markers, HBsAg (+), Anti-HCV (+), abnormal aspartate aminotranferase(AST) and abnormal alanine aminotranferase(ALT), and high level of α-fetoprotein of whom were subsequently referred to undergo sonography to identify liver diseases including fatty liver, chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma. To estimate prevalence and incidence of fatty liver, we classified the memberships of screened population into two groups, case cohort who had already fatty liver disease and normal cohort who were free of NAFLD while participants attended first screen. To estimate the prevalence of NAFLD, we used data from the results of first screen to estimate prevalence of NAFLD. To estimate the incidence of FLD, the normal cohort was followed up at regular interval of one year to ascertain newly diagnosed NAFLD. Likelihood ratio test or Receiver operating characteristics (ROC) was applied to determine the optimal cutoff point based on ALT, serum uric acid and BMI. Prevalence/Incidence Ratio method was used to estimate average duration staying in NAFLD. Cox regression time-dependent model was used to assess the influence of relevant determinants on NAFLD.
Results By using abdominal sonography as the method for detecting NAFLD, the overall prevalence rate for the high risk group with positive results participants aged 20 years or older was 35.38%, with being 38.22% for male and 33.25% for female. The overall incidence rate of NAFLD in this high-risk group was 14.92% with 15.11% for male and 14.82% for female. By the application of cut-off score, -0.59, with the combination of three variables (log(ALT), BMI, and log(serum)), overall prevalence and incidence of NAFLD for the underlying general population were 28% and 7.5%. The application of the concept of the ratio of prevalence to incidence yielded 2.37 years of average duration of NAFLD among this high-risk group. For the general population of NAFLD with the application of three variables, the application of the concept of the ratio of prevalence to incidence rate yielded 3.76 years of average duration of NAFLD.
After controlling for these factors in the univariate analysis in each other, the effect of MS on elevated ALT still remained statistically significant (aOR=2.57 (2.41-2.75)). The most remarkable effect was seen in subject without HBV and HCV infection (aOR=3.28 (3.03-3.54)), followed by HBV (+) and HCV (+) (aOR=2.61 (1.14-5.95)) , and HBV (+) and HCV (-) (aOR=1.53 (1.29-1.83)). In addition to MS, other significant risk factors included hypercholesterolemia (aOR=1.15 (1.08-1.22)), hyperurecemia (aOR=1.70 (1.61-1.81)), and betel quids chewing (aOR=1.19 (1.08-1.31)).
After controlling for these factors in the univariate analysis in each other, the effect of MS on the risk for NAFLD was dependent on the presence of viral hepatitis B/C. The most remarkable effect was seen in subject with HBV and HCV infection (aOR=7.76 (1.31-45.90)), followed by HBV (+) and HCV (-) (aOR=1.89 (1.31-2.72)). The effect of MS on the risk for NAFLD was statistically significant in subjects in the absence of HCV and absence of HBV (1.71 (1.36-2.15)).The effect of MS on the risk for NAFLD was marginally statistically significant in subjects in the presence of HCV but absence of HBV (1.82 (0.99-3.33)). , and HBV (+) and HCV (-) (aOR=1.53 (1.29-1.83)). In addition to MS, only Hyperuricemia (aOR=1.30 (1.03-1.63)), and exercise (aOR=0.76 (0.67-0.87)). The equivalent association in subjects with ALT < 45 mg/dL found that the effect of MS on NAFLD was statistically significant after controlling other significant factors. Hyperuricemia was another independent risk factor for NAFLD after controlling for other significant risk factors.
Conclusions The present study used a population-based prospective study to estimate incidence and prevalence of NAFLD for high-risk group as well as general population. The effect of relevant correlates, particularly MS (direct and indirect effect through elevated ALT), on NAFLD was heterogeneous with the presence of hepatitis B/C infection. Hyperuricemia and high BMI may be considered as another two independent factors for NAFLD.

並列關鍵字

Non-alcoholic fatty liver disease ； prevalence ； incidence ； metabolic syndrome ； community-based screening program ； hepatitis ； Cox regression time-dependent model

參考文獻

Angulo, P. 2002. Nonalcoholic fatty liver disease. N.Engl.J.Med., 346(16): 1221-1231.

Choudhury, J. & Sanyal, A. J. 2004. Clinical aspects of fatty liver disease. Semin.Liver Dis., 24(4): 349-362.

Falck-Ytter, Y., Younossi, Z. M., Marchesini, G., & McCullough, A. J. 2001. Clinical features and natural history of nonalcoholic steatosis syndromes. Semin.Liver Dis., 21(1): 17-26.

Hultcrantz, R. & Gabrielsson, N. 1993. Patients with persistent elevation of aminotransferases: investigation with ultrasonography, radionuclide imaging and liver biopsy. J.Intern.Med., 233(1): 7-12.

Matteoni, C. A., Younossi, Z. M., Gramlich, T., Boparai, N., Liu, Y. C., & McCullough, A. J. 1999. Nonalcoholic fatty liver disease: a spectrum of clinical and pathological severity. Gastroenterology, 116(6): 1413-1419.