Gamma-glutamyl transferase and risk of type II diabetes: an updated systematic review and dose-response meta-analysis

Highlights

• There is a nonlinear relationship between GGT and risk of T2DM

• In males, there is a steeper increase in T2DM risk at GGT levels 4.5 to 23 U/L followed by a gradual increase in risk

• In females, risk of T2DM starts to increase at GGT levels of 7 U/L

• In a comparison of extreme thirds of baseline levels of GGT, there is a 30% higher risk of T2DM.

• Assays for GGT have the potential to aid in the identification of individuals at high risk of T2DM

Abstract

Purpose

We assessed the nature of the dose-response relationship between gamma-glutamyl transferase (GGT) levels and risk of incident type II diabetes mellitus (T2DM) in the general population.

Methods

Systematic review and dose-response meta-analysis of published prospective studies. Relevant studies were identified in a literature search of MEDLINE, EMBASE, and Web of Science databases up to June 2014. We examined a potential nonlinear relationship using restricted cubic splines.

Results

Of the 300 titles reviewed, we included 24 cohort studies with data on 177,307 participants and 11,155 T2DM cases. In pooled analysis of 16 studies with relevant data, there was evidence of a nonlinear association between GGT and T2DM risk in both males (P for nonlinearity = .02) and females (P for nonlinearity = .0005). In a comparison of extreme thirds of baseline levels of GGT, relative risk for T2DM in pooled analysis of all 24 studies was 1.34 (95% confidence interval, 1.27–1.42). There was heterogeneity among the studies (P < .001), which was to a large part explained by blood sample used, study size, degree of confounder adjustment, and quality of studies.

Conclusions

Circulating level of GGT contributes to an increased risk of T2DM in the general population in a nonlinear dose-response pattern.

Introduction

Type II diabetes mellitus (T2DM) and its complications have risen to pandemic proportions. It is expected that the number of people living with T2DM will reach 552 million by 2030 according to the most recent estimates released by the International Diabetes Federation [1]. In line with current guidelines from the International Diabetes Federation [2], much attention is now being paid to early identification of individuals at high risk and potential prevention of T2DM. To date, established risk factors for T2DM include age, gender, hypertension, family history of diabetes, and anthropometric data (body mass index or waist circumference), and several risk scores incorporating majority of these risk factors exist [3], [4], [5]. Participants at high risk of developing T2DM according to thresholds of these risk scores are amenable to preventive measures. The pathogenesis of T2DM is however still not fully established and appears to involve multiple factors.

Gamma-glutamyl transferase (GGT), a sensitive but nonspecific index of liver dysfunction and a biological clue of excessive alcohol intake, has been strongly linked to the development of T2DM [6], [7], [8]. Although interest in GGT as a risk marker for T2DM goes back at least several decades, this was further stimulated with the systematic review and meta-analysis by Fraser et al. [9], who quantified the prospective association of GGT with risk of incident T2DM, but assumed a log–linear relationship. Although the evidence suggests that there is a strong association between elevated baseline circulating GGT and high risk of incident T2DM, which indicates a dose-response relationship, there is uncertainty about the nature of the dose-response relationship and specifically, whether it is linear or there is a threshold effect. With available evidence suggesting that the GGT–T2DM association is apparent even within the normal range of GGT level [10], [11], it is important to establish if a threshold exists, which could potentially better identify individuals at increased risk of T2DM. In addition, since the publication of the previous review, a large number of studies have been published on the topic. Against this background, we aimed to characterize in detail the nature of the dose-response relationship between baseline GGT level and risk of T2DM in the general population by conducting a dose-response meta-analysis of available prospective data. We also conducted an updated systematic review and meta-analysis to obtain a precise estimate of the magnitude of the association by including all relevant studies published since the previous review [9] and transformed reported risk estimates from all studies to a consistent comparison (top vs. bottom thirds of the baseline levels of GGT) to enhance comparability of the findings.