Dyslipidemia treatment of patients with diabetes mellitus in a US managed care plan: a retrospective database analysis

Cardiovascular disease (CVD) is a primary cause of morbidity and mortality among patients with hyperglycemia or type 2 diabetes mellitus (DM) despite the availability of effective therapies to treat major risk factors such as elevated blood pressure and cholesterol levels [1, 2]. The most common lipid pattern in patients with DM, hypertension, and/or the metabolic syndrome includes hypertriglyceridemia, increased concentrations of small dense low-density lipoprotein particles, low levels of high-density lipoprotein cholesterol (HDL-C), increased remnant lipoproteins, and elevated apolipoprotein B concentrations [3, 4]. This profile of mixed dyslipidemia significantly increases risk for all forms of atherosclerotic disease, including coronary heart disease (CHD) [5–8].

The burgeoning prevalence of insulin resistance throughout the world is greatly increasing the incidence of mixed dyslipidemia. It is estimated that approximately 40% of patients with coronary artery disease have low-density lipoprotein cholesterol (LDL-C) levels below 130 mg/dL yet these patients also have low levels of HDL-C, with or without increased levels of triglycerides (TG) [9, 10]. Overall, isolated low HDL-C affects 20% to 30% of patients with CHD, representing several million people in the United States [9, 10].

Observational studies indicate that low HDL-C levels are strongly and independently associated with increased CHD risk [11–13]. The results of the first major clinical trial specifically focusing on the treatment of low HDL-C demonstrated that lipid treatment that raised HDL-C and lowered TG but had no effect on LDL-C substantially reduced the incidence of major cardiac and cerebrovascular events [14].

Elevated serum levels of TG are an independent risk factor for CHD even after adjustment for HDL-C [7]. In a subanalysis of the Pravastatin or Atorvastatin Evaluation and Infection Therapy-Thrombolysis In Myocardial Infarction 22 (PROVE-IT TIMI 22) trial to assess the impact of low on-treatment TG on CHD risk beyond LDL-C < 70 mg/dL, Miller et al. [15] demonstrated that on-treatment TG below 150 mg/dL was independently associated with a lower risk of recurrent CHD events. The Veterans Affairs HDL Intervention Trial (VA-HIT), which included patients with low HDL-C (mean 31 mg/dL) and LDL-C (mean 111 mg/dL), found that patients receiving gemfibrozil had a 22% relative risk reduction in the primary end point of time to first nonfatal myocardial infarction or CHD death (95% CI 7% to 35%, p = 0.006). There was also a significant reduction in cerebrovascular events. Among 627 patients with DM, there was a 24% relative risk reduction for the expanded end point (CHD-related death, nonfatal myocardial infarction, or definite stroke). Likewise, among 1,449 patients with ≥ 3 criteria that define the metabolic syndrome (impaired fasting glucose; hypertension; obesity; high TG; or low HDL-C), the relative risk reduction with gemfibrozil was an even more impressive 35% for the expanded end point [14]. Together with data from the Helsinki Heart Study [16] and a subgroup analysis from the Bezafibrate Infarction Prevention study [17], these data suggest that fibrates may be particularly effective in treating dyslipidemia in patients with the metabolic syndrome.

Multiple guidelines writing groups advocate combination therapy for the management of multiple lipid abnormalities (NCEP ATPIII, ADA, AHA/ACC) [18, 19]. Statin-fibrate combination therapy has been found to be more successful than statin monotherapy in achieving therapeutic targets in dyslipidemic patients with DM or the metabolic syndrome [20–22]. The ongoing Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial was designed to evaluate the effect of simvastatin-fenofibrate combination therapy on cardiovascular risk in patients with type 2 DM [23]. The effect of statin-niacin combination therapy on atherosclerosis progression in patients with high cardiovascular risk (CHD and low HDL-C levels) has been evaluated in both the Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol 2 (ARBITER 2) and the HDL Atherosclerosis Treatment (HATS) studies. These trials demonstrated that the use of statin-niacin combination therapy is associated with stabilization of atherosclerotic disease.

Despite the abundance of data confirming the role of suboptimal lipoprotein levels as a risk factor for CHD and the availability of guidelines enumerating therapeutic lifestyle modifications and pharmaceutical treatment options for the management of mixed dyslipidemia, a substantial proportion of dyslipidemic patients remain untreated or inappropriately treated [24, 25], or patients choose to discontinue treatment soon after it is initiated [26]. The primary objective of the present study was to achieve greater understanding of treatment patterns in "real-world" patients by evaluating pharmacologic treatment of DM patients with mixed dyslipidemia enrolled in a commercial health plan. Specifically, the purpose was to determine the proportion of DM patients with suboptimal LDL, HDL-C, and/or TG values not being appropriately treated with lipid-modifying medications.

Despite the availability of treatment guidelines for dyslipidemia management and the abundance of clinical trial evidence highlighting the benefits of lipid-modifying therapy to reduce the risk for CHD and its clinical sequelae, treatment of DM patients with abnormal LDL-C, HDL-C, and TG levels continues to be suboptimal. In our patient population, 16% of patients with suboptimal lipid values and 21% of our mixed dyslipidemia group had DM. Our DM patient population had a high (> 50%) proportion of patients with abnormal HDL-C and TG levels. This is in line with the results of a recent study by Grant et al. [27], who also found a high prevalence of suboptimal HDL-C among patients with type 2 diabetes, with nearly half (49.5%) of patients exhibiting low HDL-C. Another study, involving a high-risk population with documented CHD or CHD risk equivalents, also found a high prevalence of low HDL-C across various LDL-C levels, including among patients taking statins [28]. In that study, low HDL-C was most prevalent in patients with LDL-C of 70 mg/dL or lower, and was equally and highly prevalent in patients taking statins (67%) and those not taking statins (64%).

A large percentage of DM patients with suboptimal lipid levels in our study did not receive lipid-modifying therapy either prior to or after a suboptimal lipid value was obtained (53% and 42%, respectively). While the percentage of patients receiving lipid-modifying therapy increased in the post-index period, a substantial number of DM patients with suboptimal lipid levels did not receive any lipid-modifying therapy in the post-index period irrespective of abnormal lipid parameter, or the degree of the abnormal value. At least 30% of DM patients with suboptimal lipid levels across all mixed dyslipidemic groups were not treated with any lipid-modifying medication for at least 6 months post-index. Although lipid treatment rates increased post-index in all mixed dyslipidemic groups, and even though treatment rates were correlated with the number of total risk factors, abnormal lipid parameters and more severe lipid value quartiles, the incremental use of lipid-modifying medication was not as substantial as would be expected, given the compelling association between multiple lipid abnormalities and cardiovascular risk in patients with DM. Of particular note is that in DM patients with low HDL-C and elevated TG, the most common post-index lipid therapy used was statin monotherapy (in 37% of patients), and only 17% received any niacin or fibrate therapy to target suboptimal HDL-C or TG. Grant et al. [27] described similar findings, with the vast majority of their study patients receiving statin monotherapy and a very small percentage being prescribed treatment targeting the suboptimal HDL-C level.

These results are also consistent with those reported by Klingman and colleagues [29], who evaluated data from the 1999–2000 National Health and Nutrition Examination Survey to assess the state of dyslipidemia management in the US adult population and to determine whether treatment patterns were consistent with guideline recommendations. They demonstrated that 44% of treatment-eligible adults had never been told by a physician, or any other healthcare professional, that they had dyslipidemia. Among all treatment-eligible adults, only 30% were adopting therapeutic lifestyle changes (TLC) and only 10% were receiving lipid-modifying therapy. Very high percentages of treatment-eligible adults were receiving no treatment at all for dyslipidemia: 69% overall, 61% of high-risk adults, 70% of medium-risk adults, and 77% of low-risk adults. Unlike the present study, however, Klingman and colleagues focused on the achievement of LDL-C goals and did not evaluate HDL-C and TG levels.

Although the reasons for the suboptimal treatment of dyslipidemia are beyond the scope of this study, several explanations have been proposed to account for the discrepancy between guideline recommendations and real-world treatment patterns. These include suboptimal patient-follow-up, use of low-potency statins in high-risk patients, inaccurate pill counts and refill records, patients' inability to pay for medications, lack of patient motivation [30], physician time constraints, difficulty understanding and applying NCEP guidelines [19], and lingering concerns about the potential hepatic and skeletal muscle toxicity of combinations of lipid-modifying drugs [31].

NCEP ATPIII recommends adjunctive therapy with niacin or fibrates (to achieve non-HDL goal) for dyslipidemic patients with multiple CVD risk factors who have low HDL-C and/or high TG, after the LDL-C goal has been achieved with statin monotherapy [7, 19, 32]. Our study results suggest that the majority of patients with DM receiving lipid-modifying treatment are being treated with statin monotherapy. Although it is well documented that the different classes of lipid-modifying medications have different (complementary) effects on lipid parameters [19], statin monotherapy continued to be prescribed to 81% of patients with more than 1 suboptimal lipid value who received treatment during the post-index period of our study, even though many patients with multiple lipid abnormalities – i.e., those with suboptimal HDL-C and TG levels – may have gained greater benefit from adjunctive treatment with either niacin, a drug with the greatest capacity to increase HDL-C, or fibrates, which have the most pronounced capacity to decrease elevations in serum TG. This is despite the fact that NCEP ATPIII was published in 2001, 5 years prior to the conduct of this study [19].

Our findings regarding prescribing habits for mixed dyslipidemia are consistent with results reported by Stacy and colleagues [33], who evaluated 600 high-volume prescribers of lipid-modifying drugs in 6 metropolitan areas identified from the IMS Health prescription database. Their study also found that 40% of the study population potentially had mixed dyslipidemia. These investigators reported that a high percentage of patients (69%) were prescribed monotherapy with lipid-modifying medication, of which 82% was statin monotherapy; furthermore, 14% were not prescribed any lipid-modifying drug therapy.

In real-world clinical practice, pharmacologic treatment rates increased only slightly upon assessment of multiple lipid abnormalities and in patients with DM. Over 42% of DM patients with mixed dyslipidemia received no lipid-modifying therapy after suboptimal lipid levels were ascertained, and those patients who were treated, even those with low HDL-C and high TG levels who may benefit from adjunctive treatment with niacin or fibrates, were primarily prescribed statin monotherapy. These study results suggest DM patients are not being treated to ADA-suggested targets. Further research is warranted to investigate how physicians tailor lipid-modifying therapy in these patients so that solutions aimed at improving treatment rates can be devised in order to help DM patients better achieve guideline-defined LDL-C, HDL-C, and TG goals, thereby potentially reducing their risk for cardiovascular events.