Elevated Liver Enzymes and Incident Type Two Diabetes Mellitus Risk in Yemeni Patients

Liver disease is a major cause of morbidity and mortality in patients with type two diabetes mellitus (T2DM). This case-control study was aimed to assess the association between liver enzymes and incident T2D in Yemeni patients. The present study comprising 142 T2D patients and 142 healthy control subjects were recruited from the diabetic outpatient clinic of Ibn-Sina hospital in Mukalla during the period from 1 st January to 30 th May 2020. Serum fasting blood glucose (FBG), total cholesterol, triglyceride, high-density lipoprotein cholesterol (HDL-C), alanine aminotransferase (ALT), aspartate aminotransferase (AST), and gamma-glutamyl transferase (GGT) were analyzed using the Cobas Integra Plus 400 autoanalyzer. Also, anthropometric and blood pressure measurements were taken from each participant. T2D patients had signicantly higher FBG (P= <0.0001), total cholesterol (P= <0.0001), LDL-C (P= <0.0001), and GGT (P= <0.0001) while, HDL-C was signicantly lower in T2D patients (P= 0.021). In correlation analysis, serum GGT levels were positively associated with systolic BP (r= 0.134; P= 0.025), diastolic BP (r= 0.218; P= <0.001), FBG (r= 0.216; P= <0.0001), total cholesterol (r= 0.196; P= 0.0001), triglyceride (r= 0.123; P= 0.038), and LDL-C (r= 0.209; P= <0.0001). Also, serum ALT and GGT levels were signicantly associated with increased incident T2D risk (P= 0.006 for ALT and 0.022 for GGT) and the odds ratio at 95% CI comparing highest versus lower tertiles of ALT and GGT were 2.75(2.01-3.48) Data presented as mean±SD for normal continuous variables and median (interquartile range) for continuous non-normal variables. Independent sample T-test for normally distributed continuous variables and Mann-Whitney U test for skewed continuous variables. P-value <0.05 was considered statistically signicant.


Introduction
Diabetes mellitus is a metabolic disorder characterized by chronic hyperglycemia which results from defective insulin action and secretion or both [1]. World Health Organization projects that the number of diabetic patients will exceed 350 million by 2030 [1]. Previous data have documented liver disease is a major cause of morbidity and mortality of type 2 diabetes patients [2][3]. It is well known that the liver is a vital organ in the metabolism of carbohydrates and in maintaining glucose homeostasis during fasting and postprandial period [2,4].
Research indicates that diabetes is associated with a number of liver diseases. Non-alcoholic fatty liver disease (NAFLD) is the scope of chronic liver disease in patients T2D [5], which is characterized by excess deposition of fat in the liver and associated with hepatic insulin resistance (IR) [3] and T2D risk [5]. Serum alanine aminotransferase (ALT) and gamma-glutamyl transferase (GGT) has been shown to be good biomarkers of NAFLD. ALT has been considered the speci c marker of liver injury, as found in high concentrations in hepatocytes [6], while GGT is present on the surface of most cell types and highly active in the liver, pancreas, and kidneys [7]. Besides, GGT is responsible for the extracellular glutathione catabolism and may be linked to oxidative stress [8] and chronic in ammation [9]; both oxidative stress and chronic in ammation are important pathways for hepatic IR and subsequently T2D development [10].
Many studies on liver enzymes and incident T2D risk were conducted in Europe [11][12][13][14] and Asian population [15][16][17][18][19][20][21]. However, inconsistent ndings are reported: some studies shown that GGT and ALT are signi cantly associated with improved T2D prediction [15,21], while others did not [20]. Therefore, we hypothesized that liver is the primary organ are more susceptible to the effects of hyperglycemia-induced oxidative stress, which may lead to liver injury and subsequently chronic in ammation. To our knowledge, there are no previous studies addressing the association between liver enzymes and T2D risk in Yemen Hence, this present study was aimed to assess the association between these liver enzymes and T2D risk in a sample of Yemeni patients.

Study Design and Subjects Selection
This case-control study was carried out at the College of Medicine and Health Sciences, Hadhramout University, and the subjects were recruited from the diabetic outpatient clinic of Ibn-Sina hospital, Mukalla during the period from 1 st January to 30 th May 2020. A total of 284 Yemeni adult subjects, randomly selected, and recruited into this study. At recruitment, an in-person interview was conducted using a structured questionnaire to collect health-related information. The study group was subdivided into two groups: 142 healthy control subjects composed of 51 males and 91 females (age: 46.0±7.94 yr.), and 142 T2D patients composed of 64 males and 78 females (age: 54.0±8.29 yr.). T2D patients were those who reported being diagnosed with T2D. Healthy control subjects were selected from the remaining participants who were free of T2D and were matched for age, sex, and dialect group with cases on a 1:1 ratio. Moreover, the selected healthy control subjects were screened for the presence of undiagnosed T2D at the time of blood donation by measuring fasting blood glucose (FBG). Healthy control subjects with FBG ≥7.0mmol/L were excluded from the study. Written consent was obtained from each participant entered into the study. The study was approved by the Ethics Committee of the Medicine and Health Sciences College, Hadhramout University, Mukalla, Yemen. Patients with co-morbidities such as chronic liver disease, chronic renal disease, cardiovascular disease, and malignancy were excluded.

Data Collection
A questionnaire form focusing on demographic information and diabetes history was given to all subjects. The patient's demographic information, clinical presentation, medical history, and physical ndings were taken from each subject. This included the patient's age, sex, smoking status (never, current or past), hypertension status (yes or no), diabetes status (yes or no) diabetes duration (years), diabetes medication, and diabetes complications. Patients were diagnosed with diabetes based on medical history, present intake of diabetes medications, or according to the American Diabetes Association (ADA) criteria [22]. Classi cation of Body Mass Index (BMI) was based on the World Health Organization (WHO) [23].

Anthropometric and Blood Pressure Measurements
Weight and height were measured following measured according to WHO guidelines [23]. Body mass index (BMI) was calculated as weight/height 2 (Kg/m 2 ). Obese subjects were de ned as BMI ≥30 kg/m 2 and normal-weight subjects having a BMI of 18-25. Patients who had a blood pressure of ≥140/90 mmHg or were taking antihypertensive medications were diagnosed with hypertension [24]. A true healthy normal ALT level ranges from 29 to 33 IU/l for males, and 19 to 25 IU/l for females and levels above this should be assessed as described by the American College of Gastroenterology (ACG) [25].

Biochemical Investigations
Ten millilitres of the venous blood sample was obtained from consenting subjects. The blood samples were collected by vein puncture in tubes without anticoagulant. The blood samples were then transported to the laboratory immediately. The serum was separated and stored at −20°C freezers till analyses. The serum samples of matched case-control pairs were randomly placed next to each other with the case/control status blinded to the laboratory personnel and were processed, and tested in the same batch. All laboratory equipment was calibrated. Thawing freezing was avoided by dividing the samples into aliquots. Plasma fasting blood glucose (FBG), total cholesterol, triglycerides, and HDL-cholesterol (HDL-C) were determined enzymatically using a chemical autoanalyzer (Cobas Integra 400 Plus, Roche diagnostic GmbH, Mannheim, Switzerland), following the standard procedures as described by the manufacturer. Concentrations of LDL-cholesterol (LDL-C) were calculated using Friedwald's formula [26].
All biochemical investigations were analyzed in the National Center for Public Health Labs-Mukalla, Yemen.

Statistical Analysis
Data were analyzed using the Statistical Package for the Social Sciences for Windows (version 24) and are expressed by mean ± standard deviation (SD) for continuous variables (normally distributed). Noncontinuous variables are expressed by median (inter-quartile range) and n (percentage) for categorical variables. Independent Student's t-test used for normally distributed continuous variables and Wilcoxon signed-rank test for skewed continuous variables. The Pearson correlation test was performed with ALT, AST, and GGT as the dependent variables. ALT, AST and GGT were divided into tertiles according to their distribution in the healthy control samples and the lowest tertile was served as the reference group. We used conditional logistic regression to model their associations with T2D adjusting for age (continuous), smoking (never, current and past smoker) and BMI (continuous) (model 1). Model 1 plus adjusted total cholesterol (mmol/L), triglyceride (mmol/L), HDL-C (mmol/L), and LDL-C (mmol/L) (all in tertiles). We repeated the same analysis in 176 cases and control pairs with baseline FBG <7.0 mmol/L. Conditional regression analyse was done to estimate the odds ratios (ORs) and 95% CIs for T2D patients. The statistical analysis was conducted at a 95% con dence level and a P -value <0.05 was considered statistically signi cant.

Results
Descriptive statistics of anthropometric and biochemical data of the study population are presented in

Discussion
Despite the incidence of diabetes is increasing worldwide and its prevalence is higher in developing countries, no studies have examined the relationship between elevated liver enzymes and T2D risk in Yemeni patients. Our study therefore was focused on hepatocytes as the vital organ contributing in glucose homeostasis during fasting and postprandial stage. Serum ALT, AST, and GGT were taken from each participant and used for this work. Additionally, most people aged ≥45 years in devel oping countries suffer from diabetes [27]. These ndings were convenient with our study showed that T2D patients had signi cantly higher mean age compared to healthy control subjects (Table 1).
Besides, our present ndings also observed signi cantly increased BMI, systolic BP, and diastolic BP in T2D patients than healthy control subjects. The present ndings also showed that serum FBG, total cholesterol, and LDL-C were signi cantly higher in T2D patients than healthy control subjects, while, no signi cantly difference was found among both groups for serum triglyceride. In contrast, HDL-C was signi cantly lower in T2D patients. Our study further revealed higher levels of GGT in T2D patients. While, AST was signi cantly lower in T2D patients. Besides, no signi cantly difference was found among both groups for ALT.
Our study also revealed positive correlations between GGT with FBG, total cholesterol, triglyceride, and LDL-C across the combined group before and after adjustment for age and BMI, whilst, the association between ALT with FBG and triglyceride was no longer signi cant after adjustment for age and BMI. Such a positive relationship between liver enzymes and blood lipids pro le in T2D patients has been observed in previous studies [4,[28][29][30][31]. This nding supports the role of hepatic IR in the pathogenesis of NAFLD in patients with T2D [6,32]. Moreover, Cho et al. reported a correlation between ALT activity and increased fatty liver [16]. The impairment of the normal process of syn thesis and elimination of triglycerides may progress to brosis, cirrhosis, and hepatocellu lar carcinoma [33][34]. Triglyceride is a major form of lipids stored in the liver of patients with NAFLD.
In addition to its effect on lipid metabolism, insulin also contributes a proin ammatory effect to liver abrasion [35]. Thus, in ammation contributes to hepatic IR. Additionally, pro-in ammatory cytokines and transcription factors are highly expressed in white adipose tissue and liver. Obesity is de ned as a state of chronic low-grade in ammation and a highly risk factor for hepatic IR and NAFLD. Thus, it is a primary cause of decreased insulin sensitivity. Obesity leads to lipid accumulation and activates the c-Jun Nterminal kinase (JNK) and nuclear factor-kappa B (NF-κB) signaling pathways, which consequently increase production of pro-in ammatory cytokines, such as tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) [36]. In addition, various adipose tissue-derived proteins, such as adiponectin and leptin, are considered to be major links between obesity, hepatic IR and related in ammatory disorders [37].
GGT is known as a marker of hepatobiliary dis orders and is associated with other pathologi cal conditions like diabetes. Free radicals generated by diabetes consume glutathione which induces the increased expression of GGT in hepatocytes. Various studies have suggested the association of GGT concentrations with T2D [38-41] and hyperlipidemia [42]. These ndings are in agreement with our study; GGT was signi cantly associated with the hyperglycemic and hyperlipidemic pro le. We observed ALT and GGT together were positively correlated. Moreover, some data also reported ele vated GGT levels with ALT in T2D patients with dyslipidemia [39][40]43]. Although we did not con rm the presence of fatty liver by ultrasound techniques, we showed the relationship of ALT, AST, and GGT with the pre dictors of diabetes and lipid pro le parameters, presenting hepatocellular injury.
A study of male Korean workers found that AST was independently associated with diabetes [44], while in a study of male Japanese o ce work ers AST was not associated with T2D risk [40]. Some studies also reported that ALT is a signi cant predictor of diabetes while AST is not [45]. These ndings are in agreement with our ndings as AST does not show considerable relation ship with the studied parameters. Besides, Clark et al. also sug gested that mild or chronic elevations of these aminotransferases may be due to NAFLD [46][47]. However, our study is limited to the standard method of liver biopsy for the prediction of NAFLD but it goes with the analysis of the Third National Health and Nutritional Examination Survey where individuals with NAFLD are known to have elevated aminotransferases.
In addition, our study also found that increased ALT and GGT levels improve prediction of T2D risk, and this was supported by several previous studies: a meta-analysis reported a pooled relative risk of 1.34  [21]. A Mendelian randomization study further provided an evidence for the relationship between higher GGT levels and hepatic IR study [50]. In contrast, our study did not observe any relationship of AST incident T2D risk, which was consistent with previous studies [18,30,38,51], and this may be due to lack of speci city of AST for liver diseases [18]. However, one Korean study showed positive correlation between GGT/ALT and T2D risk among patients without fatty liver, suggesting alternative pathways exist [17]. Thus, increased GGT and ALT levels were linked to T2D development as surrogate measures of NAFLD [52]. NAFLD also, may indicate fat deposition in other organs such as skeletal muscle, the myocardium, and the pancreas, which predispose individuals to T2D risk [52]. Moreover, research evidence showed that the relations of GGT and ALT with T2D risk were also independent of other important pathologies in T2D development such as whole-body insulin resistance [3,45 ] and blood lipids pro le [3,[14][15][16].
The strength of the present study included adjustment for well-established diabetes risk factors including BMI, blood lipids, and hypertension, and using comprehensive statistical methods (Pearson correlation coe cient and regression analysis) to explore the predictive utility of liver enzymes with other risk factors. However, there are some limitations. First, our sample size may be small and thus underpowered to detect the interaction with ALT and GGT. Second, we measured liver enzymes only once and may not represent long-term pro le. Third, we did not measure hepatitis B and C infection, which could result in elevated liver enzymes. Fourth, we did not measure insulin, CRP, leptin, and adiponectin as predictive biomarkers links between obesity, hepatic IR and related in ammatory disorders in T2D patients. Thus, further large sample size with measurement of insulin, hs-CRP, leptin, adiponectin, and interleukins are required to con rm these correlations. In conclusion, increased levels of ALT and GGT are positively associated with higher risk of T2D in Yemeni patients. Thus, routine screening of liver enzymes in T2D patients is recommended for the early detection of liver abnormalities.

Conclusions
Increased levels of ALT and GGT are positively associated with higher risk of T2D in Yemeni patients, and may be used as the predictive biomarkers in developing T2D risk. Thus, routine screening of ALT and GGT in T2D patients is recommended for the early detection of liver abnormalities.

Declarations Data Availability
All requests for data access should be addressed to the corresponding author. Proposals requesting data access will have to specify how they plan to use the data.