(Circulation. 1997;95:1651-1653.)
© 1997 American Heart Association, Inc.
Articles |
Cholesterol Screening Guidelines
Consensus, Evidence, and the Departure From Common Sense
From Tulane University Medical Center (J.C.L.), New Orleans, La, and Mary Imogene Bassett Research Institute (T.A.P.), Columbia University, New York, NY.
Correspondence to John C. LaRosa, MD, Chancellor, Tulane University Medical Center, 1430 Tulane Ave, SL 76, New Orleans, LA 70112.
Key Words: cholesterol • coronary disease • screening • lipoproteins
 |
Introduction |
|---|
 Top Introduction References |
|---|
In this issue of Circulation, Garber and Browner1 present a defense of the American College of Physicians (ACP) guidelines for cholesterol screening, which were originally published in Annals of Internal Medicine in May 1996.2 The bases of these guidelines are that widespread cholesterol screening, such as that recommended by the National Cholesterol Education Program (NCEP),3 will result in the inappropriate use of cholesterol-lowering drugs in persons at low risk of developing coronary disease. The authors contend that the "evidence-based approach," which the ACP allegedly used to reach these conclusions, is fundamentally different from the "consensus-based approach" used by the NCEP. Surprisingly, they then conclude that the NCEP "used logic similar to the ACP in development of its treatment recommendations." Garber and Browner point out correctly that the two groups are drawing conclusions on the basis of the same database.
The authors then discuss the process of developing such guidelines by citing the "extensive internal and external reviews" that the ACP guidelines underwent, without describing the content of those reviews or the background of the individuals asked to make them. This is, of course, in contrast to the NCEP process, which has been and continues to be a completely public endeavor. Contrary to what is implied by Garber and Browner, the NCEP process has been highly inclusive, drawing not only on individuals with expertise in lipids and lipoproteins but also on including individuals from other specialties and from the practicing community and even the most vociferous critics of cholesterol intervention.
It is also important to recognize, as Garber and Browner state, that there is no disagreement that cholesterol lowering in individuals with clinically apparent vascular disease prevents recurrent events and saves lives. It is an important intervention central to the medical management of such patients; no debate about cholesterol screening in groups at lower risk should obscure those very important facts. Although the authors rely heavily on the ability of clinical trials to be used to estimate with precision the benefits of an intervention in the general population, they are also somewhat fluid in their use of such criteria. For example, they imply that the original NCEP guidelines "may have led to the increased use of gemfibrozil," a conclusion for which there is not a shred of evidence of cause and effect. In another recent report, one of the authors is more enthusiastic about a dietary intervention as a means of avoiding cholesterol-lowering drug therapy, even given the limitations of diet trials.4 In that report, the suggestion is made that the need for cholesterol-lowering drugs could be further reduced in women with the use of estrogen therapy, even though no clinical trial clearly demonstrating the efficacy of estrogen therapy in reducing vascular disease incidents has been completed or published. There apparently is an art to the flexible application of evidence to guideline development that escapes all except a few of us.
In both the NCEP and ACP guidelines, the concept of using drugs
sparingly and only for those at the highest risk is clearly
articulated. The NCEP guidelines, however, reject the notion that
knowledge of cholesterol levels will inevitably lead to drug
intervention. Even if there were any evidence for that, the proper
approach would be physician education rather than the avoidance of
cholesterol screening. There is, however, no evidence that physicians
are overprescribing cholesterol-lowering medications in low-risk
populations. One of us (T.A.P.) carried out a survey of the medical
records of 3120 randomly selected adults from 16 ambulatory care
practices. We could find no evidence of the overprescription of
cholesterol-lowering drugs. Rather, almost half of these patients had
not been screened with a total cholesterol level, and of those eligible
for cholesterol-lowering drug therapy, 
37 patients were prescribed it
(N Wasey, M Myerson, C Lewis, M Nichols, P Jenkins, TA Pearson. Quality
assurance in the management of lipid disorders: are we over- or
under-treating patients in primary care? To be presented at the 4th
International Conference on Preventive Cardiology, Montreal, Canada,
July 1997). The problem is not the obligatory use of
cholesterol-lowering drugs, as proposed by Garber and Browner, but
rather underscreening of clinical populations and undertreatment of
subgroups in whom drug efficacy and cost-effectiveness have been
demonstrated.
No group has advocated the wholesale use of drugs in low-risk populations. Because it is clear, however, that elevated cholesterol levels even in young adults predict coronary disease in later life,5 the knowledge of cholesterol levels contributes to the overall assessment of risk in patients in whom hygienic interventions can be of value in the prevention of atherosclerosis. Cholesterol measurements are not different from measurements of blood pressure, weight, waist-to-hip ratios, blood sugar, cigarette consumption, and other inexpensive predictors of coronary risk. Would the authors recommend that blood pressure be left unassessed in young adults because they rarely have strokes before middle age? Given the fact that atherosclerosis has been demonstrated in autopsy studies of young people long before it produces symptomatic clinical vascular disease, should we not provide increased emphasis on hygienic interventions for those with elevated cholesterol levels? The authors complain that they have been unjustly accused of ignoring the fact that cholesterol levels in young adults predict midlife coronary disease, even though they cited the reference in their background paper. Unfortunately, citation of a reference is not the same as recognition of its implications, which the authors have steadfastly ignored.
Garber and Browner1 continue to underestimate the role of cholesterol screening as part of a community-based program to increase cholesterol awareness and to motivate individuals at all levels of risk toward a healthier diet. First, the authors reiterate that dietary changes will yield population cholesterol level reductions of only 2% and that for any reduction in coronary events to occur, cholesterol-lowering drugs must therefore be used. In a recent analysis of the cost-effectiveness of population-wide approaches to reduce serum cholesterol levels in the US population, Tosteson et al6 used data from community-based interventions that lowered population-wide cholesterol levels by 1% to 4% and estimated, in part on the basis of the same Framingham Heart Study database referenced by Garber and Browner, the impact of population-wide cholesterol lowering through diet in 35- to 85-year-olds. They estimated that a nationwide program with costs of $4.95 per year per patient (which might include the cost of cholesterol screening once every 5 years) and a serum cholesterol reduction of an average of 2% would save 624 000 life-years and $2.1 billion in healthcare costs. Even costs of $16.55 per year per patient would be relatively cost-effective compared with many medical interventions in common use.6
Cholesterol screening clearly is one strategy to bring about such population-wide reductions in cholesterol levels. It should be emphasized that the models of Tosteson et al6 showed even lower costs per year of life saved or actual costs savings if cholesterol level reductions of 2% to 4% were achieved. A recent study in northern Sweden involved cholesterol screening by primary care physicians, followed by dietary advice for those with elevated levels, as a major part of their community-based risk reduction strategy (L. Weinehall et al, unpublished observations). After 8 years, population-wide cholesterol levels dropped by >10% in comparison with a reference population that did not receive cholesterol screening. With the use of the models of Tosteson et al,6 the cost savings of a program of this magnitude would be enormous in the US population.
The authors also reiterate the notion that in older individuals, cholesterol levels may be safely ignored, despite the fact that such individuals carry the greatest burden of silent atherosclerosis. They appear to be unaware of evidence that cholesterol fractions continue to be used to predict coronary risk even in the oldest elderly.7 8 It is disingenuous to state that cholesterol may not be a risk factor in the elderly, when it is clear that total–to–HDL cholesterol ratios or HDL cholesterol levels alone are predictive factors.
The authors also cite the reduced relative risks but ignore the increased attributable risks of an elevated cholesterol level in the elderly. Because age is the strongest risk factor for cardiovascular disease, even the groups of older persons with normal cholesterol levels have sizeable risks of disease. The relative risk, therefore, appears to be less in the elderly. Even more important, however, is the absolute or attributable risk (ie, the number of excess cases of coronary disease attributable to an elevated serum cholesterol level).
When considering attributable risk, elevated serum cholesterol is still an important contributor to coronary disease in the elderly.8 As an example, a large number of cardiac deaths prevented in the 4S Study9 and the West of Scotland Study10 occurred in the older age groups, despite relative risk reductions due to cholesterol lowering, which were less than those in younger age strata.
Because cholesterol fractions continue to be used to predict risk in the elderly, they should be measured in a population whose risk of cardiovascular disease is high. To exclude individuals from the potential benefits of intervention simply because of chronological age makes no sense. Furthermore, it is important in this population to recognize that prolongation of life is not the only benefit to be gained from cholesterol or other medical interventions. The avoidance of coronary or other vascular events is very likely to improve the quality of life during the years remaining for an older patient. For example, most recent trials of HMG-CoA reductase inhibitors document significant reductions in stroke.9 10 The prevention of this major contributor to disability in the elderly must be factored into any discussion of screening for and treatment of hypercholesterolemia in the elderly.
Although the authors of the ACP guidelines appear loathe to engage in rational extrapolation, they are silent about recommendations regarding racial and ethnic minorities. Virtually none of the clinical trials of cholesterol lowering have included in any significant numbers African Americans, Hispanics, Asians, or Native Americans. Is it their recommendation that these groups be excluded from cholesterol screening and, when appropriate, from cholesterol interventions? Should we not screen cholesterol levels in such individuals even if they have coronary disease, since no clinical trials have demonstrated a clinical lowering benefit in these groups? What guidelines do they propose for extrapolation of data from one group to other groups?
In summary, the ACP guidelines are based on the selective application of selected evidence. Moreover, they are based on a false premise that physicians are unable to stop themselves from indiscriminate prescribing of cholesterol-lowering drugs when elevated cholesterol levels are found in low-risk groups. This is the fundamental premise of their guidelines. Cost-effectiveness is not an issue in cholesterol screening; it is an issue in the indiscriminate use of cholesterol-lowering drugs. The NCEP guidelines provide very explicit and conservative recommendations for the use of cholesterol-lowering drug therapy. As such, they are not in need of revision. To withhold useful prognostic information about a patient's ultimate risk of developing vascular disease cripples the ability of the patient's physician to prescribe and promote preventive interventions that might obviate the use of drug therapy, as Browner pointed out in another article.4 At the very least, such information helps to identify individuals who must be singled out for more intensive educational efforts and monitoring before they develop the atherosclerotic lesions that eventually lead to disability and death. That is the basis of preventive medical practice, a concept that is supported by the ACP but not in their cholesterol-screening guidelines.
|
References |
|---|
|
TopIntroductionReferences |
|---|
1. Garber AM, Browner WS. Cholesterol screening guidelines: consensus, evidence, and common sense. Circulation. 1997;95:1642-1645.
2.
Garber AM, Browner WS. American College of
Physicians guidelines for using serum cholesterol, high-density
lipoprotein cholesterol, and triglyceride levels as screening tests for
preventing coronary heart disease in adults. Ann Intern
Med.. 1996;124:515-517.
3.
Summary of the second report of the National
Cholesterol Education Program (NCEP) Expert Panel on Detection,
Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult
Treatment Panel II). JAMA.. 1993;269:3015-3023.
4.
Avins AL, Browner WS. Lowering risk without
lowering cholesterol: implications for National Cholesterol
Policy. Ann Intern Med.. 1996;125:502-506.
5.
Klag MJ, Ford DE, Mead LA, He J, Whelton PK, Liang KY,
Levine DM. Serum cholesterol in young men and subsequent cardiovascular
disease. N Engl J Med. 1993;328:313-318.
6.
Tosteson ANA, Weinstein MC, Hunink MGM, Mittleman MA,
Williams LW, Goldman PA, Goldman L. Cost-effectiveness of
population-wide educational approaches to reduce serum cholesterol
levels. Circulation.. 1997;95:24-30.
7.
Zimetbaum P, Frishman WH, Ooi WL, Derman MP, Aronson
M, Gidez LI, Eder HA. Plasma lipids and lipoproteins and the incidence
of cardiovascular disease in the very elderly: the Bronx Aging Study.
Arterioscler Thromb.. 1992;12:416.
8. Manolio T, Pearson TA, Wenger NK, Barrett-Conner E, Payne GH, Harlan WR. Cholesterol and heart disease in older persons and women: review of an NHLBI workshop. Ann Epidemiol.. 1992;2:161-176. [Medline] [Order article via Infotrieve]
9. Scandinavian Simvastatin Survival Study Group. Randomized trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet.. 1994;344:1383-1389. [Medline] [Order article via Infotrieve]
10.
Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR,
MacFarlane PW, McKillop JH, Packard CJ, for the West of Scotland
Coronary Prevention Study Group. Prevention of coronary heart disease
with provastatin in men with hypercholesterolemia. N Engl
J Med.. 1995;333:1301-1307.
This article has been cited by other articles:
 |
|
 |
|
  J. C. LaRosa Prevention and Treatment of Coronary Heart Disease: Who Benefits? Circulation, October 2, 2001; 104(14): 1688 - 1692. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. L. Navas-Nacher, L. Colangelo, C. Beam, and P. Greenland Risk Factors for Coronary Heart Disease in Men 18 to 39 Years of Age Ann Intern Med, March 20, 2001; 134(6): 433 - 439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. P. Lowe, P. Greenland, K. J. Ruth, A. R. Dyer, R. Stamler, and J. Stamler Impact of Major Cardiovascular Disease Risk Factors, Particularly in Combination, on 22-Year Mortality in Women and Men Arch Intern Med, October 12, 1998; 158(18): 2007 - 2014. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. N. Lemaitre, C. D. Furberg, A. B. Newman, S. B. Hulley, D. J. Gordon, J. S. Gottdiener, R. H. McDonald Jr, and B. M. Psaty Time Trends in the Use of Cholesterol-Lowering Agents in Older Adults: The Cardiovascular Health Study Arch Intern Med, September 14, 1998; 158(16): 1761 - 1768. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||