The effects of plasma chromium on lipid profile, glucose metabolism and cardiovascular risk in type 2 diabetes mellitus. A case - control study

Background The study was aimed at determining the effect of plasma chromium concentration on the metabolism of glucose, and lipids and their subsequent cardiovascular risk in patients with type 2 diabetes in the Bolgatanga district of Ghana. Material and methods Fasting blood glucose and lipids profile were determined by enzymatic assay using the BT 5000® Random Access Chemistry Analyzer. Fasting serum insulin and High sensitive C-reactive protein were determined by ELISA, a solid phase direct sandwich immunoassay method. HOMA-IR, which is based on fasting blood sample for insulin and glucose concentrations measured in a single blood sample, was used to calculate insulin resistance. Plasma chromium was measured using an atomic Absorption Spectrometer. Results Patientswith diabeteshad significantly (p<0.0001) increased LDL, TC, TG, VLDL, insulin, CRP and HOMAIR and a significantly reduced plasma chromium (p<0.0001) (0.53± 0.02μg/l and 0.11±0.01μg/l control and case respectively). Low Cr (p ≤0.001) was associated with high blood pressure, obesity and lipid dysregulation. Plasma Cr significantly correlated negatively with blood pressure and LDL. Conclusion Lower plasma Cr level was associated with hyperglycaemia, hyperinsulinemia, hypertension, insulin resistance and high inflammation marker HsCRP.


Introduction
Excessive ingestion of high fat and carbohydrate diets are the major macronutrients that have been associated with the development of diabetes and obesity. However, micronutrients, including trace elements particularly chromium (Cr), are also known to play significant roles in lipid and carbohydrate metabolism [1][2][3] and thereby contributing to the etiology of diabetes, obesity and cardiovascular diseases (CVD) [4][5][6] Chromium is a transition metal and trivalent chromium has been referred to as the 'glucose tolerance factor' and implicated in the regulation of glucose and lipid metabolism [7]. The exact mechanism of action remains unclear; however, several mechanisms have been proposed to explain the signaling process. Chromium acts as a cofactor or secondary messenger to insulin, improves insulin sensitivity and facilitate glucose utilization by insulin target tissues and [8][9]. Chromium also improves insulin affinity to its receptors, and also activates insulin receptor kinases and at the same time inhibiting insulin receptor phosphatases [10]. Chromodulin, the Cr-binding protein, promotes tyrosine kinase activity of insulin receptor during response to insulin [11]. This hypothesis is supported by the observation that Cr deficiency leads to elevated blood glucose, total cholesterol and triglycerides and decreased in high density lipoproteins (HDL) and insulin sensitivity even in humans on normal diets [12,13]. Epidemiological evidence for the incidence of Cr deficiency is limited; however, several studies in humans and animals experiments support the beneficial effects of Cr supplementation on glucose metabolism and insulin sensitivity [14,15]. The relationship between chromium and insulin resistance has been further demonstrated in patients with high insulin resistance, having a corresponding high Cr excretion in urine in comparison to those patients with diabetes who have a lesser degree of insulin resistance [16]. However, some studies reported minimal beneficial effects of Cr [17,18] in patients with diabetes.
This study attempts to evaluate the correlations between plasma Cr levels, insulin resistance, glucose utilization, dyslipidemia and cardiovascular risk in type 2 diabetics attending the Bolgatanga Regional Hospital-Ghana.

Subjects
A case-control study was carried out at the outpatient department (OPD) and theDiabetes Clinic in the Upper East Regional Hospital (Bolgatanga-Ghana), among subjects visiting the facility as patients with diabetesor qualified blood donors, between the period August 2014 to July 2015. The study population was made up of 163 enrolled diagnosedpatients with type 2 diabetes, consisting of 68 males and 95 femalesaged between 35-65 years who reported at the diabetes clinic and 168consisting 70 males and 98 females of healthy non-diabetic volunteers from the same locality aged matched with the patients with diabeteswere used as the control. A total of 331 subjects were recruited for this study.Patients with diabeteswhose life style had not changed, pertaining to exercise and dietary habit in the past four weeks were enrolled, otherwise excluded from the study,also excluded from the study were pregnant women. This information was obtained by simple interview through a designed questionnaire.

Ethical clearance approval
All procedures were approved by the Committee on Human Research Publication and Ethics of School of Medical Sciences, KNUST Kumasi, Ghana (CHRPE/Student/113/09) and the committee on human research publication and ethic of the Navrongo Health Research Centre. A written consent form was completed and signed/thumb-printed by all the participants who were recruited into the study after the study was explained to them in a language they understand.

Anthropometric parameter measurements
Body weight was measured (to the nearest 0.5 kilogram). Height was measured (to the nearest 1.0 millimeter), Waist circumference (WC), Hip circumference (HC) and blood pressure were measured as recommended by the World Health Organization (WHO, 2006, WHO, 1998.

Sample collection
About 10.0 ml of venous blood samples from overnight fasting subjects was aseptically collected from the median antecubital or cephalic veins. 4.0ml of the blood was then dispensed into labelled plain BD vacutainer1 tubes for the lipid profile and other biochemical parameters measurements and. 1.0 ml in to fluoride oxalate coated tubes (Becton Dickenson, Plymouth, UK) for fasting blood glucose determination. 5.0 ml of blood was also dispensed into another labelled trace metal free evacuated tube containing EDTA (BD, Plymouth, UK, Royal blue top Vacutainer 1) for Cr assay. Samples for blood glucose assay were immediately analysed. After clotting, blood sample in the plain tubes were centrifuged and the serum stored at -20˚C until ready for analysis for chromium (Cr), lipid profile and other biochemical parameters.

Biochemistry analysis
Lipid profile and blood glucose were measured using enzymatic methods, whilst insulin and HsCRP were assayed using ELISA method at the Chemical Pathology laboratory at the Kwame Nkrumah University of Science and Technology, Kumasi, Ghana and HOMA-IR calculated.
Chromium was measured with an atomic Absorption Spectrometer (VARIAN AA 240FS-Atomic Absorption Spectrometer) at the Atomic Energy, Chemistry laboratory, Accra,Ghana.

Statistical analysis
All statistical analyses were performed using GraphPad prism 5.0 (Graph PadSoftware, San Diego California USA, www.graphpad.com) and Microsoft Excel 2007. Continuous variables were expressed as mean ± SEM, while categorical variables expressed as proportion. Comparisons of the subjects and control group, were performed using unpaired t tests, χ2 tests, or Fisher exact tests, where appropriate. One-sample t-test was used in comparison between means Cr levels within group variable. Odds ratio and their 95% confidence intervals were used to quantify the risk of subjects in highly risk population. A level of p<0.05 was acceptable as statistically significant unless otherwise stated. Comparison of clinical variables, biochemical markers and anthropometrics between diabetics and control groups was by Pearson Correlation Coefficient. Correlation was significant at the 0.05, 0.01, 0.001 levels (2-tailed) Table 1 above depicts a case-control study with a higher proportion of the participants being females (58.3%) as compared to males (41.7%). The total participants enrolled into the study were 331, of which diabetics were 163 and 168 non-diabetics (control group). There was no significant difference in age between the case and control groups (52.43 ± 0.69 vs. 51.96 ± 1.23, p = 0.7232). In the case group, subjects had a significant increase in blood pressure (SBP and DBP) compared to the controls group (p<0.0001). Patients with diabeteshad higher values in all of the anthropometric measurements; weight, BMI, HC, WC, WHR, WHtR, WTR and mid upper arm circumference as compared to non-diabetics (control group).

Results
The diabetic group had significantly (p<0.0001) increased LDL, TC, TG, VLDL. HDL level did not show any statistically significantly change (p = 0.3420). Mean FBG level was about two times higher in patients with diabetes (p< 0.0001). The mean plasma insulin, HsCRP and HOMAIR were also significantly higher in diabetic group than in the control group (p<0.05) but about four times lower level of Cr (p<0.05) ( Table 2).
Obesity, (measurement defined by BMI !30Kg/M 2 , WC (male ! 102 cm and female! 88 cm)WHR (men ! 1.0 and women and ! 0.85)WHtR(!0.5), WTR (!1.95) was significantly associated with lower Cr level than the non-obese (p<0.05). Plasma Cr was markedly decreased in patients with diabetes compared to normoglycemic control group (p<0.0001). However, Cr level was not statistically significantly changed between the dyslipidaemic and the control (p>0.05) when using both the WHO and NCEP III criteria for defining metabolic syndrome (Table 3).
Patients with diabetesand significantly low plasma Cr level(p 0.001) had increased blood pressures (SBP and DBP) and had a tendency to obesity as indicated by high values of their anthropometric measurements (BMI = 27.8 ± 0.53, WHR = 0.95 ± 0.01, WHtR, = 0.58 ± 0.01 and WTR = 1.89 ± 0.01). Also plasma lipid levels (TC, TG, and LDL-C), fasting blood sugar levels, hyperinsulinemia (insulin levels) and high insulin resistant (high HOMA-IR values) were significantly higher in subjects who had significantly low plasma Cr level (p 0.001). However, the mean level of HDL-cholesterol was increased in those with high plasma Cr level though not statistically significant (p>0.05) ( Table 4). There was no significant correlation between the lipids profiles and Cr of the patients with diabetes. However, plasma Cr was significantly but negative correlated with systolic and diastolic blood pressure and LDL (r = -0.200, r = -0.189 r = -0.180) respectively among the controls. HDL was positively correlated with TC in the diabetics (r = 0.324) and HOMA-IR in the controls (r = 0.210), while insulin was significantly negatively correlated with hsCRP (r = -0.214) and positively correlated with HOMA-IR, (r = 0.942) in the controls (Table 5).

Discussion
The metabolism of chromium is affected by several factors including, stress, diet, exercise, and diabetes [19] and increased intake of simple sugars results in increased chromium loss [20][21][22] . Therefore, subjects whose life style had not changed, pertaining to exercise and dietary habit in the past four weeks were enrolled for the study. The ability to covert chromium into a more useful form may be the key difference between chromium metabolisms in the diabetics compared to the non-diabetes patients. Patients with diabetes have been shown to have higher chromium absorption but also greater chromium excretion [23]. Individuals with diabetes tend to lose the ability to convert inorganic chromium to a useable form [24,25]. Perhaps, similar to the observation that diabetic mice lose the ability to convert inorganic chromium to a useable organic form (organic chromium) that activates insulin [22] Plasma lipids except HDL were significantly elevated in the patients with diabetes (Table 2). Lipoprotein abnormalities are usually present in type 2 diabetes, which includes hypertriglyceridemia, increased LDL and reduced plasma HDL-cholesterol and also LDL's are converted to smaller, and more atherogenic, lipoproteins [26,27]. These abnormalities are linked to the increased metabolism of apolipoprotein B (apoB). Experimental evidence suggests that regulation of apoB production, increase lipolysis in adipocytes due to poor insulin activity results in increased fatty acid release from adipose cells. Insulin has also been shown to directly increase the degradation of apoB which ameliorates dyslipidaemia [28][29]. Therefore, insulin deficiency or hepatic insulin resistance may increase the secretion of apoB, and upregulate VLDL and LDL and increase cardiovascular risk. In this study even though plasma lipids were elevated, they were still within the physiological ranges and the atherogenic or cardiovascular risk (total cholesterol / HDL) was low. The significantly low plasma chromium in patients with type 2 diabetes agrees with the derailed lipid metabolism. The low cardiovascular risk may be attributed to the life style of the subjects. Increased exercise or energy expenditure has been shown to improve lipid profile [30]. Indeed many of the subjects are subsistent farmers or petty traders and these types of occupations involves a lot of movement hence there is very minimal sedentary life style in this region compared to the other urban areas in the country. The staple diets in this population are rich in vegetable fibre. However, apart from the lipid profile, the metabolism of the biomarkers which are affected by chromium metabolism may be predictors of cardiovascular risk. CRP, insulin and HOMAIR were all significantly elevated ( Table 2) in patients with diabetes. CRP significantly correlated positively with diastolic blood pressure, triglycerides and HOMAIR and negatively correlated with insulin (Table 5). CRP is an acute phase reactant and a member of the pentraxin family of innate response proteins   Table 5 Values represent correlation coefficient (r). Ã Correlation is significant at P < 0.05 level (2-tailed) ÃÃ Correlation is significant at P < 0.01level (2 - [31]. Baseline levels of CRP are predictive of risk of myocardial infarction, and stroke. Evidence suggests that CRP may be directly linked to atherogenesis, and that arterial plaque can produce CRP, independent of the hepatic pathways [32]. Recent data suggest that CRP may play a direct role in atherogenesis by inducing the expression of intercellular adhesion molecule-1 (ICAM-1) and vascular adhesion molecule-1 (VCAM-1) by endothelial cells [33,34] and mediates the monocyte chemoattractant protein-1(MCP-1) induction. CRP activates complement [35], as well as mediating the uptake of LDL by macrophages [36]. The plasma level of the CRP of both the control and patients with diabetes though was more than twice the value (6.36±0.41 and 7.76±0.47 mg/l respectively) reported by the American Heart foundation as cardiovascular risk (greater than 3.0mg/l) [37], this may not necessarily imply the subjects have increased cardiovascular risk. These high values may be due to an immune response. Indeed it has been shown that in the Garu-Tempani district in the Upper East region of Ghana, CRP values as high as 24.6mg/l were recorded and this was attributed to immune response to P. Falciparumparasitaemia [38]. This work is sited around similar setting in the Upper East region and therefore the high CRP is not surprising because samples were collected throughout the year during some point when malaria or P. Falciparum parasitaemia is high.
Besides, it has also been shown that CRP levels abnormalities are changed by the ethnicity factor [39], possibly also accounting for the high level in this work. Since CRP in the diabetes patients was significantly higher than in the control, the results of CRP, triglycerides and HOMAIR may therefore be predictive of an increased cardiovascular risk independent of the lipid profile. Plasma insulin has been shown to be predictive of cardiovascular risk. High insulin levels may constitute a more sensitive predictor of CHD than the degree of glucose intolerance, it may be useful to avoid excessive plasma insulin concentration, and even to lower its level [35] The metabolism of CRP and insulin seems to be regulated by the plasma levels of chromium (Table 4). When Cr levels are significantly lower (diabetics) insulin is upregulated (coefficient (r);-0.072) whereas when plasma levels of Cr are significantly higher (in control subjects) insulin and CRP levels are downregulated (coefficient (r);-0.054 and -0.045 respectively). The mechanisms by which Cr metabolism affects plasma insulin and CRP levels are unclear, even though some studies have shown it improves insulin sensitivity [17,18,40,41]. In this study, it is suggested, it may improve insulin and CRP secretion. Similarly, significantly decreased plasma Cr (p 0.001) concentration was associated with significantly increased plasma total cholesterol, LDL and VLDL. Low Cr levels were also associated with both an increased systolic and diastolic pressure (p<0.0001). Indeed, normal Cr levels in the control subjects, were significantly correlated negatively both with systolic and diastolic blood pressure whereas low Cr level was contrary indicated in patients with diabetes as well (Table 5). Hence, high plasma Cr seems to ameliorate hypertension. Cr also seems to play a role in the anthropometry [42] of the subjects (Table 3). High plasma Cr was associated with fat distribution; WHtR, WC and WHR were significantly lower at significantly greater (p> 0.001) plasma Cr levels (Table 3 and  Table 4). However, Cr effect was not significant on dyslipidaemia as determined by the WHO and NCEP III Criteria (Table 3)

Conclusion
Lower plasma Cr level was associated with hyperglycaemia, hyperinsulinemia, hypertension, insulin resistance and CRP. High plasma Cr was also associated with favourable fat distribution: WHtR, WC and WHR were significantly lower at high plasma Cr levels. Plasma chromium concentration therefore ameliorate glucose and lipid metabolism. High plasma chromium there reduces cardiovascular risk.

Limitation
Plasma chromium levels are also affected by several factors including, stress, diet, exercise. It was not possible to evaluate these confounding factors on chromium levels in the patients with diabetes.