A Study on the Association between Body Mass Index and Diabetes Mellitus Incidence according to the Gamma-Glutamyl Transferase Level

Abstract

Background: Gamma-Glutamyl Transferase (GGT) has been widely used as a marker for excessive alcohol intake and various hepatobiliary diseases such as alcoholic hepatitis, active chronic infection, fatty liver, and liver cancer. Accordingly, a number of studies are being conducted on the rise of GGT levels and incidence of chronic diseases. However, studies analyzing the association between body mass index (obesity) and diabetes according to GGT levels remain limited. This study sought to analyze how the association between body mass index (BMI) and diabetes prevalence changes according to serum GGT levels within the normal range among the Korean people. Methods: This study was conducted on participants of the Korean Cancer Prevention Study-II. As a general population cohort developed by the Seoul Metabolic Syndrome Research Project in 2005, 121,053 participants were included in the final analysis among the total 160,407 participants whose blood samples had been obtained. Those with prediabetes at the time of registration and those with missing values on blood tests or questionnaires were excluded. Among 121,053 participants, there were 18,116 cases of diabetes, which were analyzed by using a control group of 102,937 persons. After extracting SNPs having causality with GGT using the Genome Wide Association Study (GWAS) method, the weighted Genetic Risk Score (GRS) was calculated as follows. Results: A total of 121,053 participants were studied, consisting of 77,312 men and 43,741 women. By age, 13,305 persons were 20-30 years old, 55,515 persons 30-40 years old, 40,972 persons 40-50 years old, 20,007 persons 50-60 years old, 6,544 persons 60-70 years old, and 1,309 persons were 70 years or older, indicating that the persons in their 30s made up the majority. BMI levels were 21.9±3.36(kg/m2) for those aged 20-30 years, 23.38±3.35(kg/m2) for those aged 30-40 years, 23.90±2.90(kg/m2) for those aged 40-50 years, 24.23±2.73(kg/m2) for those aged 50-60 years, 24.45±2.75(kg/m2) for those aged 60-70 years, 24.11±2.83(kg/m2) for those aged 70-80 years, and 24.68±3.72(kg/m2) for those aged 80 year or older. The BMI level increased with age. The BMI levels were 24.41±2.90(kg/m2) for men and 22.15±3.09(kg/m2) for women, showing that the BMI level was higher in men than in women. The GGT levels were 24.85±24.88(mg/dl) for those aged 20-30 years, 35.32±42.79(mg/dl) for those aged 30-40 years, 39.83±55.55(mg/dl) for those aged 40-50 years, 41.76±67.98(mg/dl) for those aged 50-60 years, 37.11±72.37(mg/dl) for those aged 60-70 years, 31.59±34.41(mg/dl) for those aged 70-80 years, and 27.87±23.21(mg/dl) for those aged 80 years or older. BMI levels increase until the age of 50 but tend to decrease from the age of 60. The GGT levels were 47.14±60.93(mg/dl) for men and 18.80±19.8(mg/dl) for women. In both men and women, when the BMI level increased, the risk ratio for diabetes significantly increased (HR:1.97; 95% CI:1.69-2.31), and this tendency was stronger in the group with the higher GGT level (HR:2.45; 95% CI:2.02-2.98). However, no difference was observed in the association according to the GGT genetic risk score (GRS). For both men and women, the association between BMI level and diabetes tended to increase as the GGT level increased. Conclusion: In Korean adult men and women, the association between BMI and the risk of diabetes tended to increase in the group with higher GGT levels. However, the association according to the GGT genetic risk is unclear. This seems to suggest that an increase in GGT caused by factors influences the association between obesity and diabetes. 배경 및 목적: 글루타민전이효소(GGT)는 과도한 알코올 섭취, 알코올성 간염, 간경변증, 지방간, 간암 등의 각종 간담도계 질환의 표지자로 널리 사용되어져 왔다. 이에 GGT 상승과 당뇨병에 대한 다수의 연구가 진행되어 왔지만, GGT에 따라 체질량지수와 당뇨병의 연관성에 대해 분석한 연구는 아직 부족한 실정이다. 본 연구에서는 정상 범위내의 GGT 수준에 따라 체질량지수와 당뇨병 발생위험과의 관련성에 대해 알아보고자 하였다. 연구 방법: 연구는 2004-2013 전국 18개 종합검진센터에서 서면동의와 함께 연구에 참여한 160,407명으로 구축된 한국인 암 예방연구-Ⅱ를 대상으로 하였다. 대상자 중 당뇨병 유병자, 유전검사 결과를 이용할 수 없는 대상자, 주요 변수의 결측지를 가진 대상자를 제외한 121,053명을 최종 분석 대상자로 하였다. 체질량지수, GGT, 유전위험점수(Genetic Risk Score, GRS)로 환산한 GGT_GRS수치를 사분위수로 층화하여 분석하였다. GGT에 따른 체질량지수와 당뇨병 발생은 콕스 비례모형 회귀분석을 사용하였다. 연구 결과: 연구 대상자는 총 121,053명으로 남성 77,312명(63.87%), 여성 43,741명(36.13%) 이었다. 이들 중 추적기간 동안 당뇨병 발생자는 20,701명(15%)이었다. 남성의 체질량지수는 24.41 kg/m2(표준편차: 2.90)이며, 여성은 22.15 kg/m2(표준편차: 3.09)로 남성이 여성보다 높았다. GGT 수치는 남성은 47.14 mg/dl이며, 여자는 18.80 mg/dl이었다. 남성과 여성 모두 체질량지수가 증가할수록 당뇨병에 대한 위험비가 유의하게 증가하는 경향을 보였다. 특히, 이러한 관련성은 GGT 수치가 높은 군에서 강해지는 경향을 보였다. 그러나 이러한 관련성은 GGT_GRS 수준에 따라서 차이를 보이지 않았다. 결론 및 고찰: 한국 성인에서 체질량 지수와 당뇨병 발생과 관련성은 GGT 수치가 높은 군에서 증가하는 경향을 보였지만, GGT_GRS 수준에 따라서는 뚜렷한 경향성은 관찰되지 않았다. 따라서 체질량 지수와 당뇨병 발생과 관련성은 유전요인과 같은 내부적인 요인보다는 환경 등 외부요인에 의해 증가된 GGT 수준에 의해서 modify 되는 것으로 해석될 수 있다.