Non-HDL Cholesterol: When—and How—to Treat
Non-HDL Cholesterol: When—and How—to Treat
Over the past 4 decades, our understanding of the role of elevated cholesterol in cardiovascular disease (CVD) has undergone radical change. During that time, we have moved from a belief that cholesterol does not matter and that atherosclerosis is an irreversible process to a strong conviction that treating elevated cholesterol, especially elevated low-density lipoprotein cholesterol (LDL-C), can slow and perhaps halt the progression of atherosclerosis. But it has been a slow process for several reasons.
In the 1960s, the Framingham investigators demonstrated that elevated serum cholesterol is a risk factor for CVD.1 However, there was no compelling evidence that treating elevated cholesterol reduces cardiovascular events until the Lipid Research Clinics Coronary Primary Prevention Trial in 1984.2 Cholestyramine resin was used in this trial. The results were not generally accepted by the medical community because the study included only middle-aged men; moreover, cholestyramine is difficult to tolerate.
In 1987 the first statin was introduced. This was a watershed moment because a better-tolerated drug was now available that lowered LDL-C more effectively than other pharmacological agents. But it was not until the late 1990s that studies clearly demonstrated a reduction in cardiovascular events across a broad range of patients. The results of the Heart Protection Study, one of the most influential trials, were not published until 2002.3 This and other landmark studies conclusively demonstrated that elevated cholesterol, specifically elevated LDL, poses significant risk and that lowering it will reduce cardiovascular events.
However, lowering LDL-C does not completely eliminate risk. Most patients obtain only a 25% to 35% risk reduction. A patient treated with a statin therefore has a residual risk of 65% to 75% of his or her pre-treatment risk.4 In many cases, additional drugs are required to reduce risk further.5 This creates problems with cost, additional adverse effects, drug interactions, and patient acceptance. Here I present a case that demonstrates the challenges facing clinicians, and I discuss how these challenges can best be addressed.
A PATIENT AT RISK: WOMAN WITH DIABETES, PREVIOUS MI, AND DYSLIPIDEMIA
Mrs S. is a 58-year-old postmenopausal Hispanic woman who comes to a primary care office because of increasing fatigue. She has a family history of diabetes. She has 3 children, and she thinks she had “a touch of diabetes” with her last pregnancy. One year earlier she had a myocardial infarction (MI), and her blood glucose level was found to be elevated.
Her examination is unremarkable except for a body mass index (BMI) of 28, blood pressure (BP) of 140/88 mm Hg, and decreased sensation in both feet. Laboratory studies reveal the following values: hemoglobin A1c (HbA1c), 7.8%; fasting blood glucose, 185 mg/dL; normal complete blood cell count, creatinine, and blood urea nitrogen; triglycerides, 350 mg/dL; total cholesterol, 200 mg/dL; LDL-C, 105 mg/dL; high-density lipoprotein cholesterol (HDL-C), 25 mg/dL; and non–HDL-C, 175 mg/dL. She is treated with lifestyle changes, metformin, and an angiotensin-converting enzyme inhibitor.
When she returns 3 months later, she has lost 5 lb and now walks 3 days a week. Her BMI is 27, and BP is 130/78 mm Hg. HbA1c level is 6.3%; triglycerides, 250 mg/dL; total cholesterol, 200 mg/dL; LDL-C, 120 mg/dL; HDL-C, 30 mg/dL; and non–HDL-C, 170 mg/dL.
BEYOND LDL-C LOWERING
This case raises several questions. Why did her LDL-C level increase instead of decrease? The LDLC level obtained in most lipid profiles is calculated using the Friedewald formula6: total cholesterol minus HDL-C minus triglycerides divided by 5. Thus, to calculate this patient’s initial LDL-C level: (200 — 25) — (350/5) = 105.
Although random variation (up to 19%) can influence numbers when retesting, the formula may create a false impression by underestimating the LDL-C level when the triglyceride level is high. As the level of triglycerides increases, the LDL-C level mathematically decreases, and as triglycerides decrease, LDL-C increases. So the change in LDL-C is a reflection of the change in triglycerides. This is why the National Cholesterol Education Program (NCEP) guidelines state that the LDL-C level is not a valid basis for therapeutic decisions when the triglyceride level is over 200 mg/dL.7
The next question is, How is the non–HDL-C level useful? Obtaining a non–HDL-C level does not require additional testing. If the total cholesterol and HDL-C values are in the patient’s chart, the non–HDL-C level can be calculated as follows: total cholesterol minus HDL-C. Thus, to calculate this patient’s initial non–HDL-C level: 200 — 25 = 175.
Patients with type 2 diabetes characteristically have a dyslipidemia that consists of decreased HDL-C levels, elevated triglyceride levels, and normal to elevated LDL-C levels. 8 The NCEP recognizes hypertriglyceridemia as a risk factor for coronary artery disease; in its guidelines, non–HDL-C is identified as the therapeutic target rather than LDL-C if the triglyceride level is greater than 200 mg/dL.7 Subsequent studies have demonstrated that the level of non–HDL-C predicts CVD in persons with diabetes.9 Moreover, non–HDL-C may be superior to LDLC in predicting CVD and should be used as the primary lipid target in persons with diabetes.10 Non–HDL-C contains the highly atherogenic, small, dense lipoproteins that are associated with a high incidence of CVD.11