Non–High-Density Lipoprotein Cholesterol and Guidelines for Cholesterol Lowering in Recent History

Abstract

Background

The National Cholesterol Education Program (NCEP) released guidelines for treating cholesterol in 1988, 1994, and 2002. After a hiatus, the guidelines were released again in 2013, 2016, 2017, and 2018.

Methods

In this article, I review these guidelines, factors that affected their release, how they evolved, and why recommended treatment targets are reasonable. Also, to aid reader understanding, I briefly discuss biochemical mechanisms and the pathophysiology of beta-lipoproteins, focusing on the importance on non–high-density cholesterol (non-HDLC) in assessing risk and as a target for treatment. The concepts discussed are important to laboratory clinicians because those workers inscribe target values on the reports and may consult with medical staff members.

Conclusions

The newest recommendations, released in 2018, are an extension of the 2017 guidelines that defined non-HDLC as equivalent to low-density lipoprotein cholesterol (LDLC). For the reasons discussed herein, non-HDLC has advantages over LDLC. Laboratories reporting cholesterol results should include non-HDLC values and cutoffs in their reports.

cholesterol, apo B, non-HDL cholesterol, recommendations, LDL cholesterol, metabolic pathways

In 2013, the American College of Cardiology (ACC) and the American Heart Association (AHA) released new guidelines from an expert group¹ that recommended that therapy for lowering low-density lipoprotein cholesterol (LDLC) be focused on the intensity of drug treatment, rather than on a targeted concentration (target value), as previous expert groups had recommended. In 2016, an ACC Expert Consensus Committee² released new guidelines that restored target values and introduced non–high-density lipoprotein cholesterol (non-HDLC) as an equivalent target to LDLC in patients with diabetes mellitus and those with elevated triglycerides. In 2017, the American College of Cardiology Task Force on Expert Consensus Committee released still-newer guidelines on the role of nonstatin therapies for preventing coronary disease.³ These guidelines, which were an update of the 2016 ACC guidelines,² extended the use of non-HLDC as a target for all risk groups.

In this review, I discuss briefly how these recommendations evolved, how treatment targets reasonably meet the needs for reducing coronary disease, the biochemical mechanisms that support the recommendations, and why the specific target values selected are reasonable. To help readers understand dyslipidemias treatments, I also briefly discuss the pathophysiology of beta-lipoproteins, the relationship of those lipoproteins to atherosclerotic disease, the biochemical sites at which treatments act, and the growing importance of non-HDLC as a marker and target. These concepts are important to laboratory clinicians because those workers inscribe target values on the reports and may have to consult with medical staff members regarding the rationale for and use of these values.

The National Cholesterol Education Program (NCEP) and the Development of its Guidelines

The idea that cholesterol was a cause of coronary disease has been supported by a wealth of evidence from experimental animal models, epidemiological studies, genetic studies, and strongly suggestive human clinical trials. However, through the years, for various reasons, this idea was rejected by many scientists.⁴ An article published by Anitschkow and Chalatow⁵ in 1913 proposed a central role of hypercholesterolemia in atherogenesis. Yet, only in 2017 did a consensus statement from the European Atherosclerosis Society note that LDLC was not only a biomarker but a definitive cause of coronary disease. This statement was based on evidence from inherited disorders, prospective epidemiological studies, Mendelian randomization studies, and randomized control trials (RCTs).⁶

Anitschkow was a careful experimentalist.⁴ He described lipid droplets, cholesterol accumulation, and white blood cell accumulation in the artery tissue, as well as conversion of the fatty streak to fibrous plaque. One reason that the findings of Antischow were not taken seriously was because he worked with rabbits, which have low-density lipoprotein (LDL) metabolism similar to that of humans, whereas later investigators worked with rodents, which have different LDL metabolism compared with humans. Also, the prevalent view of atherosclerosis was generally accepted to be that it accompanies aging, and there was a rigid preconceived notion that only aging and not cholesterol was responsible (for a brief discussion of the work of Antischow, see reference⁴). Even today, some investigators⁷ doubt the cholesterol hypothesis and claim that most, if not all, evidence implicating cholesterol as a cause of atherosclerosis is faulty.

The first guidelines recommending treatments for lowering cholesterol, called the Adult Treatment Panel (ATP), were released by the NCEP in 1988.⁸ These defined target values for LDLC range from less than 130 mg per dL to 160 mg per dL or less, according to estimated coronary risk. The many events in the history of cholesterol are described by Steinberg in a series of reviews.^9–12 The discovery of a defective gene in familial hypercholesterolemia and the identification of the LDL receptor in the 1970s by Goldstein and Brown^13–15 were among many milestones in the history of cholesterol and were major finding in its understanding. However, the development of the first ATP guidelines was probably dependent on 2 other milestones.

The first such milestone is the Coronary Primary Prevention Trial (CPPT).¹⁶ This trial, conceived in the early 1970s and completed in the early 1980s, was a double-blinded RCT that enrolled 3806 men who were followed up for an average of 7.4 years. These men had to consume packets of sandy cholestyramine—a bile-acid sequestrant that lowers LDLC—or a similar gritty placebo 3 times per day. Although the compliance was poor overall, causing the lowering of cholesterol to be much less than was expected, the treated group showed a statically significant 19% reduction in coronary events. Thus, it was apparent that heart disease could be reduced by lowering cholesterol. However, there was no simple way to lower it until Akira Endo identified a substance in mold that could inhibit cholesterol synthesis—the second milestone.

Week after week for 2 years, Endo and his associates assayed more than 6000 substances to finally find a substance called compactin that inhibited cholesterol synthesis at the HMG-CoA reductase step—this was the first statin.¹² In 1978, Endo, along with Goldstein, Brown, and colleagues,¹⁷ published an article showing the large effect of statins on cells. At the time, there was great concern that intensive cholesterol lowering could lead to cancer. Despite this, the drug company Merck & Co, Inc.¹² went on to identify and support clinical trials for another statin—lovastatin—that was to become the first commercially available statin. Thus, by 1988, when the ATP I was released, it seemed clear that lowering LDLC could reduce coronary disease and that taking statins were an effective, innocuous way to lower LDLC.

The ATP I guidelines were followed by ATP II and ATP III in 1994 and 2002. These guidelines affirmed the approach of ATP I and set a new LDLC goal of 100 mg per dL or less as optimal and the target for the highest risk groups.^18,19 It designated persons with diabetes (mellitus) as among the highest-risk group, identified persons with multiple risk factors for more-intense treatment, and introduced the idea that persons with metabolic syndrome should be considered for intensified therapeutic lifestyle changes. The 2002 guidelines,¹⁹ which were endorsed by the AHA and the American Diabetes Association,²⁰ indicated that for patients with triglycerides between 200 mg per dL and 500 mg per dL, non-HDLC could be a secondary target, whereas for patients with triglyceride levels greater than 500 mg per dL, triglycerides should be lowered and then LDLC should be treated until the patient reaches the goal value, with non-HDLC as a secondary target. Likewise, the guidelines from the European Atherosclerosis Society^21,22 indicated that non-HDLC may be substituted for LDLC in assessing risk in patients with elevated triglycerides, especially when associated with diabetes mellitus, kidney disease, or metabolic syndrome.

Replacement of the NCEP with Consensus Committees in 2013

In 2013, the ACC and AHA released new guidelines from an expert group¹; these guidelines replace the ATP. The ATP IV Committee transitioned to joining the members of the ATP IV expert group, who were working on, but had not yet completed, ATP IV. The term coronary heart disease, used in the ATP guidelines, was replaced with the term atherosclerotic cardiovascular disease (ASCVD). The Committee defined 4 treatment groups, as shown in Table 1. These guidelines were based largely on evidence obtained from RCTs. The guidelines did not speculate on evidence from other types of trials. The Expert Panel was unable to find evidence from RCTs that lowering of LDLC or non-HDLC concentrations to target values reduces risk of ASCVD, as had been suggested in the ATP reports.¹ Nor did the panel find sufficient evidence from RCTs that drugs other than statins should be used to reduce risk. Therefore, the 2013 Committee focused on the intensity of statin treatment, as shown in Table 1 and Table 2.

Table 1.

Intensity of  Treatment for the 4 Studied Groups^a

Group	Treatment Type(s) and Intensity
Adults aged ≥21 y with clinical ASCVD	For patients aged ≤75 years, high intensity, or moderate intensity if high intensity is intolerable
	For patients aged >75 y, moderate to high intensity
Adults aged ≥21 years with LDLC ≥190 mg/dL (familial hypercholesterolemia)	High-intensity statin to achieve >50% reduction in LDLC
	May consider nonstatin for further reduction
	Screening of close relatives
Adults aged 40–75 years without ASCVD who have diabetes mellitus and LDLC of 70–190 mg/dL	Moderate intensity
	High intensity if ≥7.5% 10-y risk
Adults without ASCVD or diabetes mellitus, with >7.5% 10-y risk	Estimate with pooled cohort
	Equation; moderate- to high-intensity statin

ASCVD, atherosclerotic cardiovascular disease; LDLC, low-density lipoprotein cholesterol.^aAdapted from Lloyd -Jones DM et al².

Table 2.

Statin Intensity and Average Percentage Lowering of LDLC by Various Statins^a

High-Intensity Statin Therapy	Moderate-Intensity Statin Therapy	Low-Intensity Statin Therapy
Daily dose lowers LDLC by ≥50%, on average	Daily dose lowers LDLC by 30%–50% on average	Daily dose lowers LDLC by <30%, on average
Atorvastatin, 40–80 mg	Atorvastatin, 10–20 mg	Lovastatin, 10–20 mg
Rosuvastatin, 20–40 mg	Fluvastatin, 20 mg	Fluvastatin, 40 mg 2× day
	Fluvastatin XL, 80 mg	Pitavastatin, 1 mg
	Lovastatin, 40 mg	Pravastatin, 10–20 mg
	Pitavastatin, 2–4 mg	Simvastatin, 10 mg
	Pravastatin, 40–80 mg
	Rosuvastatin, 5–10 mg
	Simvastatin, 20–40 mg

LDLC, low-density lipoprotein cholesterol.^aAdapted from Stone NJ et al.¹

The committee also developed a risk calculator (Omnibus Risk Estimator) that is available at http://tools.acc.org/ascvd-risk-estimator-plus/#!/calculate/therapy or can be downloaded as an MS Excel (Microsoft Corporation) program at https://professional.heart.org /professional/index.jsp under the heading ASCV risk calculator. This calculator provides the 10-year risk for persons aged 20 to 79 years and the lifetime risk for persons aged 40 to 59 years, compared with optimal risk.

The characteristics entered into the calculator are sex; age; race; total cholesterol; HDLC; and systolic blood pressure, on treatment for high blood pressure, diabetes mellitus, and smoking. Although the ATP Panels relied largely on the Framingham Study 10-year risk scores, the Omnibus estimator is based on multivariate equations that were developed from 9 longstanding population-based cohort studies funded by the National Heart Lung and Blood Institute. The study participants included apparently healthy white, African American, and Hispanic women and men, aged 40 to 79 years.²³

The Return to Target Values and the Introduction of Non-HDLC as a Primary Target

In 2016, new guidelines were released² by an ACC Expert Consensus Committee; these guidelines were further revised in 2017.³ The guidelines maintained the 4-treatment groups and the Omnibus Risk Estimator but reintroduced target values, as shown in Table 3. Moreover, the 2016 Committee indicated that non-HDLC was an equivalent target value to LDLC for select groups—namely, those with elevated triglycerides and those with diabetes mellitus. As shown in Table 3, the 2017 Committee defined non-HDLC as equivalent to LDLC for all groups.

Table 3.

Patient Risk Groups and Targets Concentrations^a

Risk Groups	Target	Additional Drug Treatment to Consider to Reach Target After Statin
Patients aged 40–75 years without clinical ASCVD and without diabetes mellitus, with baseline LDLC 70–189 mg/dL, and with 10-y risk >7.5% for primary prevention	30%–49% reduction while taking statin (may consider LDLC <100 mg/dL or non-HDLC <130 mg/dL)	Ezetimibe (first) Bile-acid sequestrant (second)
Stable clinical ASCVD with no comorbidities on statins for secondary prevention	≥50% reduction while taking statin (may consider LDLC <70 mg/dL or non-HDLC <100)	Ezetimibe (first) PCSK9 inhibitor (second)
Clinical ASCVD with comorbidities on statins For secondary prevention	≥50% reduction while taking statin (may consider LDLC <70 mg/dL or non-HDLC <100 mg/dL)	Ezetimibe (first) PCK9 inhibitor (second)
Clinical ASCVD with LDLC ≥190 mg/dL due to familial hypercholesterolemia for secondary prevention	≥50% reduction while taking statin (may consider LDLC <70 mg/dL or non-HDLC <100 mg/dL)	Ezetimibe (first) or bile-acid sequestrant or PCSK9 inhibitor (second)
Without ASCVD with LDLC ≥190 mg/dL due to familial hypercholesterolemia	≥50% reduction on statin (may consider LDLC <100 mg/dL or non-HDLC <130 mg/dL)	Ezetimibe (first) or bile acid sequestrant or PCSK9 inhibitor (second).
Patients aged 40–75 without clinical ASCVD, with diabetes mellitus, and with baseline LDLC 70–189 mg/dL for primary prevention	≥50% reduction while taking statin (may consider LDLC <100 mg/dL or may consider non-HDLC <130 mg/dL)	Ezetimibe (first) or bile-acid sequestrant or PCSK9 inhibitor (second)

ASCVD, atherosclerotic cardiovascular disease; LDLC, low-density lipoprotein cholesterol; PCSK9, proprotein convertase subtilisin/kexin type 9.^aAdapted from Levinson SS.³²

The Beta-Lipoprotein Metabolic Pathways and Dyslipidemias

When lipids are ingested, they are converted to chylomicrons in the intestine, and the chylomicrons are transported via the lymphatics and blood to the liver, where the lipids are removed.²⁴ During transportation, some chylomicrons are broken down into remnant lipoproteins by lipoprotein lipase (LPL), releasing fatty acids for tissue metabolism. This is a rapid process that takes 2 to 3 hours. Much of the lipid is then transferred to the beta-lipoproteins by incorporation into very-low-density protein (VLDL).

The major beta-lipoprotein metabolic pathways have been known for many years.²⁴ Knowledge of these pathways is helpful in understanding where drugs act and which target values might be best for which dyslipidemias. A simple illustration of the pathways is shown in Figure 1. Apolipoprotein B-100 (apo B) is the major structural protein in beta-lipoproteins and is needed for VLDL synthesis.²⁵ VLDLs are synthesized in the liver and released in to the bloodstream. VLDLs are triglyceride-rich and, like chylomicrons, are broken down by LPL to form intermediate-density lipoprotein (IDL) remnants that have approximately equal amounts of cholesterol and triglycerides. These remnants are removed from the bloodstream by receptors that bind apolipoprotein E (apo E)²⁶ or are converted to LDL by hepatic lipase (HL). LDLs have little or no apo E and are rich in cholesterol. LDL particles containing apo B are removed from the bloodstream by the LDL receptor via endocytosis.²⁶

Figure 1

Beta-lipoprotein pathways. HL indicates hepatic lipase; LPL, lipoprotein lipase; sdLDL, small dense lipoprotein lipase; and R, reaction. Major apolipoproteins (apo) contained in each type lipoprotein particle are indicated in block below the type of particle.

Normally, the removal of LDL particles is the rate-limiting step. The conversion from VLDL to LDL takes 5 to 7 hours, so that generally after a meal, the lipids in chylomicrons have been transferred to LDL in fewer than 10 hours. Thus, an overnight fast is usually appropriate when seeking a fasting specimen in which VLDL-triglyceride is minimal and LDLC is at its steady-state concentration.

Various dyslipidemias can be explained by alterations of the pathway (Figure 1). These include:

Deficiency of LPL: Hypertriglyceridemia may occur if there is a deficiency of LPL or abnormalities in apolipoprotein C2 or C3 (apo C) that inhibit LPL (reaction 1 [R1] in Figure 1). In such a case, VLDL cannot progress to IDL or LDL, and LDLC is usually very low. The triglyceride level usually rises beyond 500 mg per dL.

Dysbetalipoproteinemia: This is a rare familial dyslipidemia in which a person inherits 2 apo E2 genes. In this case, the process cannot progress to form LDL (reaction 2 [R2] in Figure 1), and cholesterol and triglycerides are elevated, from approximately 300 mg per dL to approximately 500 mg per dL. This process is called dysbetalipoproteinemia, type III hyperlipidemia, or broad beta-hyperlipidemia because on electrophoresis, the IDL migrates in a wide (broad) band between the places where the VLDL and LDL should normally migrate. In this case, the process again cannot progress to LDL, and LDLC is usually low.

Hypercholesterolemia: LDLC may be elevated because the LDL receptor may be slow in removing LDL or reduced in number (reaction 3 [R3] in Figure 1), as in familial hypercholesterolemia, so that LDLC may become elevated higher than 160 mg per dL or even 190 mg per dL.

Combined hyperlipidemia (also known as the atherogenic phenotype): This is a type of dyslipidemia in which there is a net overproduction of VLDL. In this case, no specific step is inhibited, so neither cholesterol nor triglyceride levels get extremely high but triglycerides may be moderately increased, in the range of approximately 200 mg per dL to 400 mg per dL. LDLC may or may not be elevated; the HDLC is moderately decreased. The reason the LDLC may not be increased is because in this type of lipidemia, much LDL is synthesized as small dense LDL (sdLDL; Figure 1).²⁷ sdLDL contains more protein relative to cholesterol, so that although the LDLC may not appear to be elevated, there are nevertheless more LDL particles and there is more apo B.

Mechanism of ASCVD

Lindgren et al²⁸ first separated lipoprotein particles using analytical ultracentrifugation. These investigators found that the most buoyant lipoproteins were the chylomicrons, which float to the top during ultracentrifugation and the most-dense HDL, which sinks to the bottom, with the beta-lipoproteins floating in between. Although there are now other ways to separate lipoproteins, classification by density remains the preferred method and seems to have functional importance.²⁴

Substantial evidence^10,29 indicates that LDL infiltrates the artery wall, where it is modified and rapidly taken up by macrophages that become lipid-laden foam cells—apparently, the first event in a sequence that leads to atherosclerotic plaque. Thus, high concentrations of LDL cause more penetration. Besides, smaller lipoproteins, such as sdLDL, and remnant particles, such as chylomicron remnants and IDL, are considered especially atherogenic because it is thought that they can more easily penetrate the artery wall.^10,27,30

Measurement of LDLC

LDLC is usually measure by calculation, most commonly using the Friedwald equation (LDLC = Total C – HDLC – triglyceride/5; all in mg/dL).^31,32 This measurement is generally made on fasting specimens because triglyceride levels greater than 400 mg per dL may cause unacceptable clinical alterations. Triglyceride-rich VLDL and chylomicrons contain a different ratio of cholesterol to triglycerides than LDL, so that as the chylomicron triglycerides increase, the division by 5 becomes increasingly inaccurate. After a meal, chylomicrons may be substantially increased.

Generally, when the triglycerides level is greater than 400 mg per dL, direct LDL (dLDL) measurement is recommended. Otherwise, calculated LDLC (cLDLC) is preferred because it comes with the routine lipid profile and is as accurate^31,32 for at least 2 reasons. First, different dLDLC commercial methods use different chemicals to inhibit the reaction of cholesterol in beta-lipoproteins other than LDL, so it is unclear that these methods all yield similar results. Second, because of overlap of fractions, separation of beta-lipoproteins from one another by chemical inhibition is very difficult, so that one might not only expect variability between methods but also from lot to lot.³²

Besides, some of these methods measure the potentially atherogenic lipoprotein Lp(a) but others do not. Lp(a) is always included in the cLDLC concentration. Lp(a) is not affected by the standard cholesterol treatment drugs, namely, statins and ezetimibe. Very high levels of Lp(a) may be atherogenic. Therefore, in a patient with a family history of ASCVD, if a very high cLDLC level is identified that is hardly lowered by these standard drugs, it may be worthwhile to measure the Lp(a).

Apo B and Non-HDLC

Apo B is usually measured by immunological techniques.³³ As seen in Figure 1, apo B is found in all species of beta-lipoproteins. Moreover, there is 1 molecule of apo B in each particle. As such, much evidence indicates it is the best lipoprotein marker for ASCVD.¹⁹ Besides, unlike cLDLC, its measurement is not altered by elevated triglycerides, so that randomly collected specimens are acceptable. Generally, apo B targets are approximately 30 mg per dL less than LDLC targets. Thus, an LDLC level of 160 mg per dL would be equivalent to apo B levels of approximately 130 mg per dL.

Non-HDLC is also a calculated value (nonHDLC = Total C – HDLC) but it is a simpler, lesser variable and more accurate calculation than LDLC.³² It is a measure of all of the cholesterol in the beta-lipoproteins and thus, like apo B, it is a measure of all particles. Studies have shown non-HDLC correlates better with apo B than LDLC,^34,35 and evidence¹⁹ indicates it is a better marker for ASCVD than LDLC. According to receiver operating characteristic (ROC) curve analysis, non-HDLC, in conjunction with traditional risk factors, provides the same coronary risk information as apo B more than 97% of the time³⁶ and, like apo B, non-HDLC is unaffected by elevated triglyceride levels.

With the epidemic of obesity that is occurring worldwide, combined hyperlipidemia has become the most common type of dyslipidemia. Patients with combined hyperlipidemia tend to have borderline low HDLC and moderately elevated triglyceride levels and sdLDL so that, although the LDLC may not appear elevated, there are more LDL particles and approximately a 3-fold increase in ASCVD risk.³⁰ Many patients with metabolic syndrome express this phenotype, in which the overall coronary risk may be multiplied as a result of hypertension and insulin resistance. apo B and non-HDLC are invariably increased in these persons and are better markers of risk than LDLC.

Mechanisms by Which Recommended Lipid-Lowering Drugs Act

Mechanisms by which the various lipid-lowering agents act help medical professionals to understand how to best reach target values. Statins, bile-acid sequestrants, and ezetimibe all act by reducing intracellular cholesterol levels. Those levels are controlled by the sterol regulatory element-binding proteins (SREBPs),³⁷ which are transcription factors that bind to the sterol regulatory element DNA sequence TCACNCCAC. As shown in Figure 1, when intracellular cholesterol is low, the SREBP acts to upregulate the LDL receptor. As a result, more receptor is synthesized and more blood cholesterol is transported into the cell, lowering blood cholesterol.

Statins inhibit the 3-Hydroxy 3-methylglutarate coenzyme A (HMG-CoA) reductase step in cholesterol synthesis. This process lowers intracellular cholesterol and upregulates the LDL receptor. Statins are powerful lipid-lowering drugs because they are competitive inhibitors of HMG-CoA reductase, with synthesis inversely proportional to statin dose. Statins are more powerful agents for reducing blood cholesterol (Table 2) than ezetimibe (lower LDLC by 15%–20% or bile-acid sequestrants [lower LDLC by 15%–30%]), as is apparent from their mechanism of action.

Ezetimibe selectively inhibits intestinal cholesterol uptake, reducing intracellular cholesterol.³⁸ Ezetimibe and its active glucuronide metabolite circulate interhepatically and therefore, there is very little systemic exposure. Ezetimibe added to a moderate-intensity statin (40 mg of simvastatin; IMPROVE IT-Trial)³⁹ demonstrated a 6.4% statistically significant reduction (P = .02) in ASCVD events. Simvastatin lowered LDLC to 69.5 mg per dL, and 10 mg of ezetimibe further lowered LDLC to 53.7 mg per dL. Nevertheless, a 6.4% reduction was considered too modest, and the Endocrinologic and Metabolic Drugs Advisory Committee of the U.S. Food and Drug Administration (FDA) voted against recommending approval of ezetimibe as an add-on to a statin. Still, patients in that study who were at particularly high risk of recurrent ASCVD events showed a substantial risk reduction (relative risk [RR] reduction, 19%)⁴⁰ which is why the most recent guidelines recommend ezetimibe as the first primary second-line therapy after first-line statins (Table 3).^3,40

Bile-acid sequestrants are a group of ion-exchange resins that bind bile acids in the intestine, thereby disrupting the enterohepatic circulation of bile acids. Because cholesterol is the precursor of bile acids, liver cells synthesize more bile acids, reducing intracellular cholesterol stores. Bile-acid sequestrants are generally recommended as a second-line treatment after ezetimibe (Table 3).

It has been known for some time that besides acting on the LDL receptor, SREBPs act to upregulate protease proprotein convertase subtilisin/kexin type 9 (PCSK9).³⁷ The protease initiates a process whereby the LDL receptor is degraded (Figure 1). Thus, when low concentrations of intracellular cholesterol cause an increase in the LDL receptor, PCSK9 tries to counteract the effect. More recently, antibodies that inhibit PCSK9 have been developed. These antibodies are named evolocumab and alirocumab; they are subcutaneously injected⁴¹ and reduce LDLC levels approximately 50% beyond those yielded by standard statin therapy. These agents were shown to statistically significantly reduce ASCVD beyond that yielded by a statin by approximately 12% to 15% (RR, 0.85–0.92) and reduce death from any cause (RR, 0.85)⁴² over a medium duration of 2.8 years. These agents were approved by the FDA.² The combination of a statin and a PCSK9 inhibitor is a powerful tool for reducing blood LDLC because the statin directly inhibits the intracellular cholesterol that increases the LDL receptor and the PCSK9 inhibitor reduces the counteraction that degrades the receptor. Moreover, data from these studies have made it even clearer that the lower the cholesterol, the lower the risk of ASCVD—the lower the better. Because of expense and possible poor compliance in some patient conditions, these agents are recommended as third-line agents except for familial hypercholesterolemia, in which they may be used as second-line agents (Table 3).

Risk-assessment studies are derived from primary populations with a normal incidence of ASCVD, with the possible exception of the CPPT. However, most, if not all, treatment studies have been conducted as secondary prevention studies with high-risk populations, such as patients with type 2 diabetes mellitus, those who have known ASCVD, or those with familial hypercholesterolemia. Moreover, in secondary prevention trials, many subjects have multiple comorbidities, so they are apt to die from factors other than ASCVD, which causes cholesterol lowering to appear to be less effective than it really is. Therefore, although it is clear that the lower the LDLC the better, the exact effect of cholesterol-lowering treatments on persons without ASCVD in the general population remains unclear. Also, from the results of secondary prevention studies, it is clear that the higher the baseline LDLC concentration, the greater the relative effect, so that those with LDLC concentrations less than 70 mg per dL will derive less risk reduction from a 50% lowering than those with cholesterol levels of 160 mg per dL and higher.

Thus, it seems reasonable to define less than 70 mg per dL as a target value for the highest-risk groups and less than 100 mg per dL for those at risk but who do not have ASCVD. Moreover, although PCSK9 inhibitors can reduce LDLC to much lower levels, the added effect on risk reduction is a modest 0.08% to 0.15% (1.08-fold–1.15-fold).

PCSK9 inhibitors are very expensive. Also, many patients undergoing secondary prevention have numerous comorbidities and are already taking many other drugs. Thus, adding a PCSK9 inhibitor would mean the introduction of another drug that is to be taken by injection, with only modest risk reduction. As such, it seems that compliance would be poor. Therefore, it is reasonable to use these inhibitors as third-line drugs for these patients. However, in familiar hypercholesterolemia, in which LDLC is greater than 160 mg per dL, the risk of ASCVD is approximately 5- to 10-fold. Here, there is an expected large risk reduction by reducing LDLC. So, PCSK9 inhibitors are reasonable additives to statins because these patients may have no other comorbidities but a long life expectancy, during which one could reasonably expect good compliance.

One area of weakness in the recent guidelines relates to persons with characteristics of metabolic syndrome. Body mass index (BMI), weight circumference, triglyceride levels, and glucose level are all not included in the Omibus Risk Estimator. Patients with metabolic syndrome and low LDLC may not be identified as being at higher risk, especially those at a young age, because age is by far the most sensitive variable in the risk score. Nevertheless, elevated non-HDLC levels in these patients will usually be recognized as high total cholesterol in the risk calculation. Besides, many will have low HDLC levels and hypertension, so that their lifetime risk will be very elevated compared with the optimal risk. Lifetime risk can be used as a sign to encourage lifestyle changes or even to initiate statin treatment.

Discussion

Non-HDLC has the advantage that it correlates most strongly with apo B, does not require fasting and, like apo B, identifies persons whose LDLC may be close to desirable levels but who are at increased risk because they exhibit sdLDL and remnant lipoproteins. This has become the most common type of lipidemia as obesity has become more rampant. Moreover, unlike apo B, it is derived from the routine lipid profile at no extra cost. One problem with using non-HDLC as a primary risk-assessment instrument is that most previous studies used LDLC as the eligibility criteria, which means if the studies are reexamined and non-HDLC is substituted for LDLC, the result would be statistically suspect because it would become a retrospective indicator.

I believe that the 2017 committee³ should be commended for recommending non-HDLC as being equivalent to LDLC as a target because most clinical-outcome studies have targeted LDLC. In coming to this conclusion, the Committee considered more than RCT results but all of the evidence. This appraisal seems to have been confirmed by the results of recent studies in which PCSK9 inhibitor trials use LDLC and non-HDLC as thresholds for eligibility despite LDLC being below the eligibility cutoff, creating a prospective view.^43–45 Also, non-HDLC has been shown^45,46 to have a more robust relationship with ASCVD death (and all-cause mortality) than LDLC.

When triglycerides are elevated and the calculated cLDLC values are suspect, they can be compared against the non-HDLC values to assess accuracy, with the knowledge that there should be a 30 mg per dL difference between the 2 values. Concurrently, it is important to recognize that patients with metabolic syndrome and elevated triglycerides may exhibit borderline LDLC, but their non-HDLC may be elevated because they have increased numbers of sdLDL and remnants. According to the guidelines (Table 3), many of these patients can experience a 50% or greater drop in LDLC/non-HDLC. If the cLDLC and non-HDLC values are in disagreement, laboratory professionals should be able to consult with one another to determine how non-HDLC may be telling or, in rare instances, suggest measurement of dLDLC or apo B.

In November of 2018, the AHA and ACC released updated guidelines,⁴⁰ and concurrently, a document on risk-assessment tools to guide medical professionals in primary prevention.⁴⁷ A major focus of the risk-assessment article⁴⁷ was the addition of coronary-artery calcium measurement (CAC) in persons who might be at borderline risk but only if it is thought CAC measurement can provide sufficient information to modify the decision.

As previously, these guidelines pointed out that the main protein embedded in LDL and VLDL is apo B and like non-HDLC, apo B is a stronger indicator of atherogenicity than LDLC by itself.⁴⁰ The guidelines stated that in most cases, nonfasting specimens were sufficient for routine measurement of LDLC by calculation, which was a departure from previous guidelines, but if there was reason to suspect triglyceride interference, direct LDLC or apo B should be considered. However, these guidelines seemed to focus on LDLC, seemingly placing non-HDLC in a more subservient position. Nevertheless, these guidelines were described as an extension of the 2017 guidelines, with the goal of making the guidelines “shorter and enhancing user friendliness.” Therefore, the guidelines did not appear to change the target values from those defined in Table 3. For this reason, clinical laboratories should include non-HDLC values and cutoffs in their reports.

Abbreviations

Abbreviations
 
• ACC

  American College of Cardiology;

 
• AHA

  American Heart Association;

 
• LDLC

  low-density lipoprotein cholesterol;

 
• non-HDLC

  non–high-density lipoprotein cholesterol;

 
• NCEP

  National Cholesterol Education Program;

 
• RCTs

  randomized control trials;

 
• LDL

  low-density lipoprotein;

 
• ATP

  Adult Treatment Panel;

 
• CPPT

  Coronary Primary Prevention Trial;

 
• ASCVD

  atherosclerotic cardiovascular disease;

 
• LPL

  lipoprotein lipase;

 
• VLDL

  very-low-density lipoprotein;

 
• apo B

  apolipoprotein B-100;

 
• IDL

  , intermediate-density lipoprotein;

 
• apo E

  apolipoprotein E;

 
• HL

  hepatic lipase;

 
• apo C

  apolipoprotein C;

 
• R1

  reaction 1;

 
• R2

  reaction 2;

 
• R3

  reaction 3;

 
• sdLDL

  small dense low-density lipoprotein;

 
• dLDL

  direct LDL;

 
• cLDLC

  calculated low-density lipoprotein cholesterol;

 
• ROC

  receiver operating characteristic;

 
• SREBPs

  sterol regulatory element-binding proteins;

 
• HMG-CoA

  3-Hydroxy 3-methylglutarate coenzyme A;

 
• FDA

  U.S. Food and Drug Administration;

 
• RR

  , relative risk;

 
• PCSK9

  proprotein convertase subtilisin/kexin type 9;

 
• BMI

  , body mass index;

 
• CAC

  coronary-artery calcium measurement;

 
• ASCVD

  atherosclerotic cardiovascular disease.