| Summary: | 碩士 === 國防醫學院 === 公共衛生學研究所 === 91 === To examine determinants of elevated serum alanine aminotransferase (ALT) levels among health check-up populations, a follow-up study was conducted between 1997 and 2001. Study subjects were selected from the database maintained by a private health-testing institute. Subjects who were more than 20 years of age, had no history of liver cirrhosis or liver cancer, and had been followed up annually between 1997 and 2001 were eligible for this study. There were 1509 males and 1845 females included for the longitudinal analysis. The associations between sex, age, life-style habits, body composition, hepatitis virus infection, chronic disease status, liver sonography examination, and history of liver disease among relatives and elevated serum alanine aminotransferase levels (ALT＞35 IU/L) were assessed by means of generalized estimating equations (GEE) (for determinants of elevated ALT) and logistic regression model (for risk factors of intermittently elevated or persistently elevated ALT levels). The results indicated that the risk of abnormal ALT levels for males was significantly higher than that for females. In addition, the risk factors of elevated alanine aminotransferase levels were different between males and females. The risk of elevated ALT levels was higher in males who were at younger or middle age, had central obesity, whose percentage of body fat above the first quartile (by GEE analysis). The risk of intermittently elevated ALT levels was higher in males who had betel quid- chewing habit, had central obesity or whose percentage of body fat above the first quartile (by logistic regression analysis). Whereas, the risk of elevated ALT levels was higher in females who had hypertension, had diabetes mellitus, had history of hepatitis, had history of liver disease among relatives, had non-central obesity or central obesity (by GEE analysis). The risk of intermittently elevated ALT levels was higher in females who had diabetes mellitus, had hepatitis B virus infection, had non-central obesity, had central obesity or had history of liver disease among relatives (by logistic regression analysis). In summary, body composition was the major determinant of elevated ALT levels in this health check-up population.
|
|---|
