Keywords
Glycemic control - high-density lipoprotein - phenotype
Introduction
Diabetic dyslipidemia, consisting of any or a combination of hypertriglyceridemia, reduction in high-density lipoprotein cholesterol (HDL-C), and elevated levels of small dense low-density lipoprotein cholesterol, contributes to cardiovascular burden in type 2 diabetes mellitus (T2D) patients.[[1]] Triglycerides (TGs) in diet are digested by gastric and pancreatic lipases into monoglycerides and free fatty acids. Cholesterol esters are de-esterified by pancreatic esterase into cholesterol. TGs are transported by chylomicrons from within enterocytes through lymphatics into the bloodstream. Lipoprotein lipase converts chylomicron TG into fatty acids and glycerol for energy use by adipocytes and skeletal muscle. HDLs are produced in the enterocytes and the liver. The role of HDL is to transport cholesterol from tissues and other lipoproteins to other tissues and lipoproteins using cholesteryl ester transfer protein.
There are two phenotypes of HDL: HDL2 and HDL3, with densities of 1.063–1.125 kg/L and 1.125–1.210 kg/L, respectively. HDL2 contains apo A-I whereas HDL3 contains both apo A-I and A-II.[[2]] HDL2 particles are partially depleted of cholesteryl esters and enriched in TGs. Hepatic lipase hydrolyzes the TG-enriched HDL2 particles and regenerates HDL3, yielding particles that are again suited to accept cholesterol from peripheral cells. The apo A-I HDL particles are associated with the cholesterol efflux-promoting effects of HDL; particles containing both apolipoproteins (e.g., HDL3) are less effective at mobilizing cholesterol from peripheral stores and appear to have various other functions. The clinical significance of the different HDL subfractions is not fully understood, but studies done outside Africa have demonstrated that HDL2 cholesterol is a better predictor of coronary atherosclerosis and glycemia than HDL3 or total HDL cholesterol in persons with T2D.[[3]],[[4]] Abnormalities in HDL-C are common in African patients with a significant impact on glycemic status.[[5]],[[6]],[[7]],[[8]],[[9]],[[10]]
Information of such association between HDL-C phenotypes and glycemia is lacking in Nigerians with T2D. This study, therefore, targeted to describe this gap in scientific knowledge. The aim of this study was to determine the relationship between the two phenotypes of HDL-C and glycemia in Nigerian T2D patients.
Materials and Methods
This was a cross-sectional study done at the side laboratory of the medical ward complex of the Nnamdi Azikiwe University Teaching Hospital (NAUTH), Nnewi, South-East Nigeria, involving T2D patients aged 30–69 years who attended the diabetes mellitus (DM) clinic of the NAUTH, Nnewi. Study participants were consecutively recruited from the NAUTH DM clinic for this study after informed consent was sought and details of the study were explained to them. History taking and physical examination were done on each study participant, and data were entered in a researcher-administered study protocol. Weight, height, and waist circumference were determined using the WHO STEPS anthropometric tool.[[11]] Body mass index was calculated as weight in kilograms divided by the square of the height in meters. Blood pressure was measured using a mercury sphygmomanometer (Accoson, England), and hypertension was defined as systolic blood pressure of 140 mmHg and above, or diastolic blood pressure of 90 mmHg and above, or use of antihypertensive medications.[[12]]
Ten milliliters of venous blood was collected from the intercubital vein of each participant between 8 and 10 am after an overnight fast of 12–14 h. Two milliliters of the sample was put in a fluoride oxalate bottle for plasma glucose estimation using the Trinder glucose oxidase technique whereas 1 mL of blood was put in an ethylenediaminetetraacetic acid bottle for glycosylated hemoglobin (HbA1c) estimation using the boronate affinity chromatography method.[[13]],[[14]] The remaining 7 mL of blood was placed in a plain bottle for plasma HDL-C phenotype estimation (precipitation method), serum creatinine estimation (alkaline picrate method), serum alanine transferase estimation (dinitrophenylhydrazine method), hemoglobin estimation, and hemoglobin electrophoresis in the NAUTH chemical pathology laboratory.[[15]],[[16]],[[17]],[[18]],[[19]] Blood samples were stored in a freezer (Haier Thermocool Ltd., UK) at a temperature of −20°C before laboratory analysis using a spectrophometer (Spectronic ZOD, Milton Roy Company, England). Unknown blood levels of biochemical indices were determined by interpolation of absorbances on the calibration curves. Within-run and between-run assay coefficients of variation were within the acceptable limit of <10%.[[20]]
Poor glycemic control was defined as HbA1c >6.5%.[[21]] Exclusion criteria included presence of type 1 DM; use of medications such as statins, fibrates, nicotinic acid, glucocorticoids, androgens, hormonal contraception, and beta-blockers; pregnancy; renal disease (defined as estimated glomerular filtration rate of <60 mL/1.73 m2/min); necro-inflammatory liver disease (defined as serum alanine transferase >45 IU/L which was the upper reference limit for alanine transferase in the NAUTH chemical pathology laboratory); anemia (defined as hemoglobin <12 g/dL); familial forms of dyslipidemia; and presence of sickle cell disease.[[22]],[[23]] Individuals with presence of atherosclerotic cardiovascular diseases such as cerebrovascular disease, coronary artery disease, or peripheral artery disease were also excluded from this study. Details of medication use, history of familial dyslipidemias, and presence of atherosclerotic cardiovascular disease were obtained from the participants' clinic case files.
Collected data were transferred to Microsoft Office Excel® 2010 software for data management before transfer to the Statistical Package for the Social Sciences version 26 (IBM corporation, California, USA) for statistical analyses. Kolmogorov–Smirnov test was used to determine the normality of data distribution. Continuous variables were presented as median interquartile range (IQR) whereas categorical variables were presented as proportions, n (%). IQR was presented as (25th percentile, 75th percentile). Logarithmic transformation of skewed data was done before correlation of quantitative variables. Quantitative variables were compared among two groups using Mann–Whitney U-test whereas strength of association between continuous variables was done using Spearman rank correlation coefficient, rs. Statistical significance was defined as P < 0.05. Tables, figures, and text were used to summarize results.
Results
Four hundred participants with T2D were recruited for this study. [[Table 1]] shows the baseline characteristics of study participants. The study participants consisted of 174 (43.5%) males and 226 (56.5%) females with a median age of 60 years.
Table 1: Baseline characteristics of study participants
[[Table 2]] shows the comparison of demographic, clinical, and biochemical indices in participants with and without good glycemic control. Two hundred and thirty-five (58.8%) and 165 (41.2%) study participants had poor and good glycemic control, respectively. There were significantly lower total HDL-C, HDL2-C phenotype, and HDL2-C/HDL3-C ratio in participants with poor glycemic control compared to those with good glycemic control. The difference in HDL3-C among patients with and without good glycemic control was not statistically significant (P = 0.21). The median total HDL-C values in individuals with poor and good glycemic control were 1.20 mmol/L and 1.27 mmol/L, respectively.
Table 2: Demographic, clinical, and biochemical indices in study participants with and without good glycemic control
[[Figure 1]] shows the relationship between HbA1c and HDL2-C in study participants. There was a significant negative correlation between HbA1c and HDL2-C (rs = −0.12, P = 0.01). [[Figure 2]] shows the relationship between HbA1c and HDL3-C in the participants. There was a negative correlation between HbA1c and HDL3-C, but this association was not statistically significant (rs = −0.06, P = 0.21). [[Figure 3]] shows the relationship between HbA1c and HDL2-C/HDL3-C ratio in study participants. There was a significant negative correlation between HbA1c and HDL2-C/HDL3-C (rs = −0.11, P = 0.03).
Figure 1: Relationship between glycosylated hemoglobin and high-density lipoprotein 2 cholesterol phenotype. This relationship is described by the linear regression equation: Log glycosylated hemoglobin = 0.83-0.06 log high-density lipoprotein 2 cholesterol (rs = -0.12, P = 0.01)
Figure 2: Relationship between glycosylated hemoglobin and high-density lipoprotein 3 cholesterol phenotype. This relationship is described by the linear regression equation: Log glycosylated hemoglobin = 0.86-0.04 log high-density lipoprotein 3 cholesterol (rs = -0.06, P = 0.21)
Figure 3: Relationship between glycosylated hemoglobin and ratio of high-density lipoprotein 2 cholesterol to high-density lipoprotein 3 cholesterol. This relationship is described by the linear regression equation: Log glycosylated hemoglobin = 0.84-0.08 log high-density lipoprotein 2 cholesterol/high-density lipoprotein 3 cholesterol (rs = -0.11, P = 0.03)
Discussion
This study found significantly lower HDL2-C and HDL2-C/HDL3-C ratio in T2D patients with poor glycemic control compared to those with good glycemic control. There were low HDL3-C levels in persons with poor glycemic control compared to those with good glycemic control, though the difference was not statistically significant. There was a significant inverse relationship between HbA1c and each of HDL2-C and HDL2-C/HDL3-C with HDL2-C having a stronger association with HbA1c. There was no significant association between HDL3-C and HbA1c. The median values of total HDL-C in individuals with poor and good glycemic control were 1.20 mmol/L and 1.27 mmol/L, respectively. These values may be explained by the nonparametric nature of the population distribution and the fact that these patients were on treatment for DM. These have been compared with findings from previous research works.
Xian et al. reported a significant inverse relationship between glycemic control and HDL2-C in Chinese T2D patients.[[24]] This is similar to the finding from the present study which also found a similar relationship between the HDL2 phenotype and glycemic control. Even though the study by Xian et al. was carried out in 38 persons compared to the present study that involved 400 T2D patients and with different methodologies for HDL-C phenotype measurement in both studies, the findings were similar. Maeda et al. reported a stronger negative correlation between FPG and HDL2-C compared to that between FPG and HDL3-C in 418 Japanese-Americans studied.[[25]] The present study also found a stronger negative correlation of HbA1c with HDL2-C than with HDL3-cholesterol.
Some studies differed in their findings regarding the association of glycemic control and HDL-C phenotypes in diabetic patients when compared to findings from the present study. Paniagua et al. found a significant relationship between glycemia and the HDL3-C phenotype in T2D patients although this study recruited only twenty participants compared to the present study that involved 400 T2D patients.[[26]] Fukui et al. found a significant association between HDL3-C and HbA1c in 69 T2D patients studied although some of these individuals were on statin therapy.[[27]] Individuals on lipid-lowering agents (statins inclusive) were excluded in the present study.
The main strength of this study was the characterization of HDL-C phenotypes and their relationship with glycemic control in Nigerian T2D patients which has not been reported in previous Nigerian data. Individuals on lipid-lowering medications were also excluded from the present study because these agents play a role in affecting HDL-C levels. The main limitation of this study was the use of the precipitation technique in measuring HDL-C phenotypes and not the gold standard ultracentrifugation method. The ultracentrifugation method is expensive, cumbersome, and requires large volumes of blood for HDL-C phenotype analysis compared to the simple precipitation method. The precipitation method has gained widespread use because of its high accuracy in measuring HDL-C phenotypes when compared to the ultracentrifugation method.[[28]]
Conclusion
The present study showed a stronger association between HDL2-C phenotype and glycemic status compared to that between each of total HDL-C and HDL3-C and glycemia in Nigerian T2D patients. The clinical implication of this study finding is that the HDL2-C phenotype should be measured in T2D patients because of its strong association with glycemia in these individuals. Total HDL-C is routinely measured because of its proven role in the pathogenesis of the atherosclerotic process in T2D patients and its strong association with glycemic status in these persons. The stronger association of HDL2-C with glycemic control implies that this molecule should be a better variable to measure in T2D patients than total HDL-C.
Authors contribution:
All authors contributed substantially to the conception, conduct and reporting of the study. They all reviewed the manuscript for intellectual content and approved its final version.
Compliance with ethical principles
This study was approved by the NAUTH research ethics committee, Nnewi, Nigeria, before the study was commenced. All participants provided a written informed consent.
Data availability
Data are available by reasonable requests to the corresponding author.
