1. INTRODUCTION
Diabetes and hyperlipidemia are two major factors involved in the development of cardiovascular disease. The medical treatment and reduction of the effects of these conditions are key modalities in the prevention of heart disease [1-4]. The term diabetic dyslipidemia has been used to describe the pathophysiology surrounding the effects of insulin resistance on abnormal lipid levels. This concept states that defects in insulin action and increases in glucose can lead to higher amounts of lipoproteins in the blood. The subsequent increase in lipids, secondary to a state of glucose intolerance, adds to the progression of atherosclerosis and cardiovascular disease. As well, even slight increases in lipid levels in such diabetic patients are associated with a substantial increase in cardiovascular disease, more so than the general population. Dyslipidemia affects a diabetic patient to a greater extent than a non-diabetic patient, further reinforcing the severity of diabetic dyslipidemia. Therefore, diabetic dyslipidemia is a definite concern in the management and treatment of cardiovascular disease [1-3,5]. Very few studies to date have studied the direct relationship between changing levels of insulin resistance and dyslipidemias in a clinical setting.
The Adult Treatment Panel III, or ATP III, Guidelines discusses the importance of diabetic dyslipidemia and directly states that diabetic dyslipidemia is seen more frequently in patients with premature coronary heart disease. The ATP III also states that treatment of this condition should begin promptly and must take into account the many risk factors of this condition. It separates the risk factors into categories of either modifiable or non-modifiable and places diabetes into the category of a modifiable risk factor while also mentioning that diabetic control in relation to diabetic dyslipidemia has been shown to lead to a decrease in coronary heart disease [5].
Diabetic dyslipidemia also has a strong relationship with metabolic syndrome. Metabolic syndrome, a combination of risk factors including dyslipidemia, obesity, and insulin resistance, increases the risk of cardiovascular disease even more. Each factor of the metabolic syndrome perpetuates the other. For instance, obesity puts a patient at risk for diabetes or glucose intolerance, which further puts the patient at risk for dyslipidemias. To the same extent, a diabetic state can increase the risk of obesity, which can perpetuate and worsen a state of dyslipidemia even more. The underlying cause of metabolic syndrome is related to either the insulin resistant state of the patient or of the obesity. Much controversy exists regarding which of the two is the ultimate cause, yet, nevertheless, both are strongly involved with the pathogenesis of the resulting dyslipidemia. Metabolic syndrome is widely prevalent in patients with diabetic dyslipidemias, which ultimately puts these patients at an even greater risk for cardiovascular disease [1-3,5].
Patients with diabetes are predicted to have a 2 times - 3 times higher rate of coronary artery disease, 4 times higher rate of death during an acute myocardial infarction, and 2 times higher rate of post myocardial infarction morbidity when compared to the general population [1]. The link between diabetes and the development of cardiac disease revolves around such lipid abnormalities caused by insulin resistance, since atherogenesis is strongly linked to these abnormalities. The concept of diabetic atherosclerosis has been used to describe this observation [4].
Hypertriglyceridemia, increased low density lipoprotein (LDL), decreased high density lipoprotein (HDL), and post-prandial lipemia are said to be the four main processes that help to maintain an atherogenic environment in diabetic patients. Various biochemical pathways are involved, with most taking advantage of the dysfunction of the lipoprotein lipase enzyme seen with insulin resistance. As a result of this decreased lipoprotein lipase function, hypertriglyceridemia results. The hypertriglyceridemia subsequently creates an increase in LDL and decrease in HDL, further predisposing a patient to coronary artery disease [1,2].
Other possible mechanisms that influence the resulting dyslipidemia include insulin effects on the liver, apoprotein production, peripheral actions of insulin on other tissues such as adipose, as well as changes in regulation of enzymes responsible for maintaining adequate levels of lipoproteins in the blood. The liver is responsible for manufacturing apolipoprotein B, a major component of both very low density lipoprotein (VLDL) and LDL. In patients with resistance to insulin, an increase in lipolysis occurs, which will cause the liver to produce a greater amount of apolipoprotein B. This will ultimately create an increase in the amount of lipids in the blood. To the same effect, the actions of insulin are said to directly stimulate the liver to degrade apolipoprotein B. Therefore, in patients with insulin resistance, there will be an even higher amount of this substance in the blood secondary to the increase in production as well as the decrease in its degradation [1,2,5].
In patients with metabolic syndrome, there is further risk of dyslipidemias. For instance, the increased visceral fat seen in patients with metabolic syndrome secretes increased amounts of tumor necrosis factor alpha (TNFalpha) when compared to the general population. This increase in TNF-alpha has been shown to induce both insulin resistance, which will further increase one’s risk for obesity, as well as increase levels of triglycerides, further increasing one’s risk for dyslipidemia. Therefore, a state of metabolic syndrome compounded with a diabetic dyslipidemia ultimately creates even more difficulty in preventing and managing cardiovascular disease [1-3,5].
While using the basic science literature to form the hypothesis, our study attempts to quantify the direct relationship between various levels of insulin resistance and dyslipidemias in a clinical setting.
2. MATERIALS AND METHODS
Data was collected by a retrospective paper chart review of all active patients with diagnoses of either glucose intolerance or type 2 diabetes and any type of hyperlipidemia in a specialized lipid clinic that focuses on treating patients with hard-to-treat dyslipidemias. Fasting blood glucose levels in excess of 125 mg/dL on two consecutive measurements were used to classify patients as being diagnosed with type 2 diabetes while fasting blood glucose levels from 100 to 125 mg/dL on two consecutive measurements were used to classify patients as being glucose intolerant. Patients who were considered to have a diagnosis of a dyslipidemia had values of total cholesterol greater than 199 mg/dL, LDL greater than 159 mg/ dL, or triglycerides greater than 150 mg/dL. HDL levels were not used to determine whether or not a patient was considered to be dyslipidemic in our study.
Lipid levels and either glucose, in the glucose intolerant patient group, or hemoglobin A1c values, in the diabetic patient group, were recorded. The data points used in our analyses compared changes over a 6 month period in either glucose or hemoglobin A1c with changes in total cholesterol, LDL, HDL, and triglycerides during that same time period. A total of 34 data points from 21 glucose intolerant patients and 100 data points from 42 diabetic patients were used. If possible, multiple data points were extracted from each individual patient.
Strict exclusion criteria were followed. Data points were excluded if any changes to medication were made within a 6 month time period of that specific laboratory data point. By doing so, our study attempted to exclude any changes in laboratory values that could have been directly manipulated by changes in medication regimen. Data lacking complete medication regimens were also excluded for that same reason. If a patient’s diabetic condition became better controlled and the data points fell into the glucose intolerant or normal range those data points were excluded. To the same extent, if a glucose intolerant patient’s glucose levels fell within normal limits those data points were excluded as well. The same was applicable to changes in diagnosis in regard to the patient’s dyslipidemia. Only patients with stable diagnoses of either diabetes or glucose intolerance and hyperlipidemia were included.
The data points were subsequently analyzed by regression analyses comparing either changes in glucose, in the glucose intolerant patients, or hemoglobin A1c, in the diabetic patients, with that of all components of a lipid panel, including total cholesterol, triglyceride, LDL and HDL levels, during that same time period.
3. RESULTS
A positive relationship was seen with both change over time in glucose in the glucose intolerant patient group and change over time in hemoglobin A1c levels in the diabetic patient group with that of all components of the lipid panel except for HDL. The strongest relationship was seen with comparisons involving triglycerides. Our results take into account the R values for each comparison as well as the independent variable coefficient value generated from the regression analyses when considering strength of relationship.
Clinical characteristics of the patients used in this study are fully represented in Table 1. It can be noted that the data used in this study are comparable in respect to both the diabetic and glucose intolerant groups as well as male and female genders. Differences in medication regimens, age, and gender did not show any significant changes in our calculations.
When comparing changes in glucose and HgbA1c with changes in triglycerides, an R value of 0.722 was observed in the glucose intolerant patient group (Figure 1) and an R value of 0.787 was observed in the diabetic patient group (Figure 2).
When comparing changes in glucose and HgbA1c with changes in total cholesterol, an R value of 0.635 was observed in the glucose intolerant patient group (Figure 3) and an R value of 0.645 was observed in the diabetic patient group (Figure 4).
Table 1. Patient demographics in both groups.
Figure 1. Data from regression analysis showing change in glucose vs change in triglycerides (n = 34) with an R value of 0.722.
Figure 2. Data from regression analysis showing change in HgbA1c vs change in triglycerides (n = 77) with an R value of 0.787. †Hemoglobin A1c.
Figure 3. Data from regression analysis showing change in glucose vs change in total cholesterol (n = 33) with an R value of 0.635.
Figure 4. Data from regression analysis showing change in HgbA1c vs change in total cholesterol (n = 77) with an R value of 0.645. †Hemoglobin A1c.
When comparing changes in glucose and HgbA1c with changes in LDL, an R value of 0.484 was observed in the glucose intolerant group (Figure 5) and an R value of 0.527 was observed in the diabetic group (Figure 6).
When comparing changes in glucose and HgbA1c with changes in HDL, an R value of –0.567 was observed in the glucose intolerant patient group (Figure 7) and an R value of –0.135 was observed in the diabetic patient group (Figure 8).
Table 2 shows the predicted changes in the lipid modalities that can be expected with an increase in either glucose by 1 mg/dL or hemoglobin A1c by one percentage point. These values were calculated using the independent variable coefficient that resulted from the regression analyses. Note that direct comparisons should not