Role of gamma glutamyl transferase as a diagnostic marker of metabolic syndrome

Introduction

The metabolic syndrome (syndrome X, insulin resistance syndrome) comprises of cluster of meta bolic abnormalities that promote risk of acquiring cardiovascular disease (CVD) and cerebrovascular disease.1 Since the original definition by the World Health Organization in 1998, the criteria for the metabolic syndrome have evolved and are reflecting growing clinical evidence and analysis by a variety of consensus conferences and professional organizations. Central obesity, hypertriglyceridemia, low high-density lipoprotein (HDL) cholesterol, hyperglycaemia, and hypertension are the major features of the metabolic syndrome.1

The prevalence of metabolic syndrome worldwide is 20-25% (IDF) . In India the overall prevalence is around 31.4%.2 The upsurge in the prevalence of diabetes, obesity etc. is seen concomitantly with the rise in the number of patients with metabolic syndrome. In this sense, metabolic syndrome can considered as a cardiovascular risk factor. This syndrome not only includes, but is also strongly associated with, other complications of obesity, fatty liver, cholesterol gallstones, obstructive sleep apnoea, and polycystic ovarian syndrome.3

Several markers like adiponectin have been studied as a measure of increased adipose tissue but have not proven to be cost effective and easily available. Currently, the focus is on the pocket friendly and feasible biochemical marker is required to predict an early onset of this syndrome. Gamma Glutamyl Transferase (GGT) is one such marker which is cost effective, easily available and is routinely done as part of liver function tests.4

High levels of GGT have been associated in populations with increased risk of atherosclerotic cardiovascular disease (ASCVD).5 Several prospective studies reported baseline serum GGT concentration as an independent risk factor for the development of non-communicable diseases like coronary artery disease (CAD), hypertension, diabetes mellitus and stroke.6

The purpose of this study is to evaluate the utility of GGT as a diagnostic marker of metabolic syndrome.

Materials and Methods

Method of data collection

Ethical clearance was obtained from the M S Ramaiah Medical College ethics committee prior to the start of study was taken. Data was collected by pretested semi structured questionnaire; clinical examination and necessary investigations (involved in IDF criteria) after taking written informed consent from the patients. An estimation of GGT was done for all the study subjects including cases and controls. The two groups were compared to fulfill the study objective. Reference range for normal GGT values at M S Ramaiah Hospital laboratories:

Males < 55 IU/L

Females<38 IU/L

Statistics

Data were entered serially in MS Excel and were analyzed using SPSS software version 20. Descriptive statistics, unpaired t test and fischer exact test were applied wherever necessary. p<0.05 was considered as statistically significant. Results were expressed in the form of text, figure and tables.

Table 1

	Group	N	Mean	SD	Min	Max	‘t’ value	‘p’ value
GGT(IU/L)	Case	90	52.49	6.004	37	61	279.420	<0.001
	Control	90	34.58	8.202	23	51
FBS (mg/dl)	Case	90	142.25	59.792	64	315	69.709	<0.001
	Control	90	88.93	9.730	68	108
HBA1C (%)	Case	90	8.414	2.2171	5.2	13.8	83.467	<0.001
	Control	90	6.204	.5923	5.2	7.8
Triglyceride	Case	90	194.54	83.023	116	449	109.134	<0.001
	Control	90	100.52	19.936	92	144
HDL Cholesterol	Case	90	37.89	10.358	17	54	58.717	<0.001
	Control	90	47.46	5.752	38	63

Biochemical parameters of the cases and controls

On comparison of the biochemical parameters among cases and control higher/raised values were seen among cases. Table 1

Table 2

Variable		Case (N=90)		Control (N=90)		X 2 value #	‘p’ value
		n	%	n	%
GGT(IU/L)	<55/38	28	31.1%	90	100%	94.576	<0.001
	≥55/38	62	68.9%	0	0%

Serum GGT levels in cases and controls (above normal values)

[i] # fischer exact test

Among cases, 68.9% had GGT values above the reference range and also statistically significant. Table 2

Table 3

GGT result	Patients with metabolic syndrome (IDF criteria)	Patients without metabolic syndrome	Total
Positive(≥55/38 IU/L)	62	0	62
Negative(<55/38 IU/L)	28	90	118
Total	90	90	180

Sensitivity and specificity of GGT in diagnosis of metabolic syndrome

The sensitivity was 69% and specificity was 100%. The positive predictive value of the test was 100% and negative predictive value was 76.2 %. There were 31% false negatives and nil false positives. Table 3

Figure 1

ROC curve analysis for sensitivity and specificity of GGT in males

To calculate the best possible value for validity measure of GGT in males ROC curve analysis was done. The sensitivity and specificity through this analysis was computed as 90.7% and 97.6% respectively. The cut-off value of GGT for normal, through this analysis was 50 IU/L. Figure 1

Figure 2

ROC curve analysis for sensitivity and specificity of GGT in females

To calculate the best possible value for validity measure of GGT in females ROC curve analysis was done. The sensitivity and specificity through this analysis was computed as 100% and 93.9% respectively. The cut-off value of GGT for normal, through this analysis was 36 IU/L.Figure 2

Results

A total of 180 patients (90 cases and 90 controls) were recruited in the study. Of the enrolled patients, the mean age of cases 52.8±5.89 years and controls was 54.1±12.484 years. Out of cases, 54(60%) were males and 36(40%) females and in control group, 49(54.4%) were females, 41(45.6%) were males. Mean BMI and waist/hip ratio of cases was 29.96± 2.56 kg/m² and 1.04±0.70 and control group was 24.17±0.76 kg/m² and 0.7 4±0.51 respectively. The SBP, DBP of cases were 135±15.17 mm of Hg and 86.44±6.30 mm of Hg respectively and control group were 120.22±7.74 mm of Hg and 73.31±4.78 mm of Hg.

Discussion

In our study, 180 subjects were recruited comprising 90 cases of metabolic syndrome and 90 age and sex matched controls. The mean age in the study group was 54.1±5.895 years and 52.8±12.4 years in the control group. In a similar study done by B Kasapgolu et al4 and Harini G L et al,8 the mean age was 51.3±3.2 years and 50.76±10.36 in cases and 50.78± 10.58 in controls respectively

Out of cases, 54(60%) were males and 36(40%) females and in control group, 49(54.4%) were females, 41(45.6%) were males. In a similar study done by B Kasapgolu et al,4 the gender distribution showed 62% females and 38% males in the study group. Another study done by Vijayalakshmi Masalmani et al² included 42% males and 58% females in cases and 47% males and 53% females in the control group.

The mean BMI was 29.9±2.5 kg/m² in the study group and 24.17±0.76 kg/m² in the control group. The mean waist-hip ratio was 1.04±0.07 in the study group and 0.74±0.05 in controls. The reference study⁵ showed almost similar results; the mean waist circumference and BMI were 104.1±9.8 and 30.8±4.1 kg/m² respectively. Comparable findings were also seen in the study by R Shanmuga Priya et al9 where the mean BMI was 26±4.47 in the study group and 23.46±3.06 in the control group. The above observation indicates that obesity and increased central adiposity are pivotal to the pathogenesis of metabolic syndrome.

The SBP, DBP of cases were 135±15.17 mm of Hg and 86.44±6.30 mm of Hg respectively and control group were 120.22±7.74 mm of Hg and 73.31±4.78 mm of Hg. In another study done by A O Rantala et al10 higher values were observed with SBP being 160.2±20.3 and DBP being 98.2±10.2 which was significantly different from our study.

The mean HbA1C was 8.41±2.21, with 26% in the IGT group (HbA1C 5.7-6.4) and 74% in the diabetes group (HbA1C>6.5). Similar results were seen in a study done by Kishor phepale et al ¹³ where 73.6% of cases had HbA1C more than 6.5 whereas among controls only 24% had HbA1C more than 6.5 These observations suggest a high prevalence of type 2 diabetes in patients with metabolic syndrome.

The mean total cholesterol was 173.6±45.6, triglyceride was 194.5±83, HDL was 37.89±10.35, and LDL was 105.86±38.3. A total of 68 out of 90 cases had TG>150 including 40 males and 28 females. A total of 56 out of 90 cases had HDL <40mg/dl for males and <50mg/dl for females. The values in our study with respect to lipid profile were lower than the reference study wherein mean TG was 273.8±25.2, LDL was 131.4±8.9 and HDL was 42.1±9.7. A similar finding was noted in the reference study4 as well.

In the evaluation of liver function tests, GGT which is the biomarker being evaluated in this study had the following results. The mean GGT in the study group was 52.49±6.04 and that in the control group was 34.58±8.20. A total of 62 out of 90 cases had values of GGT (≥55 (males)/≥38 (females) IU/L) including 29 males and 33 females, comprising 69% of the study group (P<0.001). In the control group none of the subjects had GGT levels above the normal range. In a similar study done by B.Kasapgolu et al4 the mean GGT among cases was 40.9±10.2 and it was 21.0±7.1 in the controls. In a study done by Gurjar BS¹⁴ GGT values were 53.39 ± 19.12 among cases and 32.78± 10.71 in control group and similarly a study done by Vijayalakshmi Masalmani et al⁴ showed 74.77± 15.36 and 40.91± 8.3

The sensitivity and specificity of GGT to diagnose metabolic syndrome was found to be 67% and 100% in males and 94% and 98% in females respectively in our study. Comparable findings were seen in the study done by Harini GL et al8 where the overall sensitivity was 92% and specificity was 88% and in study done by Kishor phepale et al11 the overall sensitivity was 84% and specificity was 91.2%.

Thus, study showed that elevated level of GGT among people with metabolic syndrome. The sensitivity of the test to diagnose metabolic syndrome was better in females; but specificity had no gender bias. Considering the CVD risk primary prevention may be emphasized in patients of metabolic syndrome with high GGT values. Hence GGT as a marker can be proposed in the algorithm used for the evaluation of metabolic syndrome.

Source of Funding

None.

Conflict of Interest

None.