Author + information
- Received January 25, 2008
- Revision received February 21, 2008
- Accepted February 27, 2008
- Published online July 29, 2008.
- Jay N. Cohn, MD⁎ ( and )
- Daniel A. Duprez, MD, PhD
- ↵⁎Reprint requests and correspondence:
Dr. Jay N. Cohn, Cardiovascular Division, Mayo Mail Code 508, University of Minnesota Medical School, 420 Delaware Street SE, Minneapolis, Minnesota 55455.
Efforts to prevent atherosclerotic morbid events have focused primarily on risk factor prevention and intervention. These approaches, based on the statistical association of risk factors with events, have dominated clinical practice in the last generation. Because the cardiovascular abnormalities eventuating in morbid events are detectable in the arteries and heart before the development of symptomatic disease, recent efforts have focused on identifying the presence of these abnormalities as a more sensitive and specific guide to the need for therapy. Advances in noninvasive techniques for studying the vasculature and the left ventricle now provide the opportunity to use early disease rather than risk factors as the tool for clinical decision making. A disease scoring system has been developed using 10 tests of vascular and cardiac function and structure. More extensive data to confirm the sensitivity and specificity of this scoring system and to demonstrate its utility in tracking the response to therapy are needed to justify widespread application in clinical practice.
Efforts to prevent atherosclerotic disease morbidity during the past generation have emphasized risk factor prevention and treatment. Indeed, the recent commentary in the Journal by Greenland and Lloyd-Jones (1) advocates greater national attention to risk factor prevention to counteract what could otherwise be a future increase in cardiovascular morbid events in the U.S.
The problem with the message to prevent and treat risk factors is that it may be good public health policy but does not translate into a useful strategy for individual patient care. It is certainly clear that levels of blood pressure (BP) and cholesterol are directly associated with the risk of atherosclerotic morbid events, and obesity increases the risk of diabetes. Furthermore, certain drug treatments that lower BP and lower cholesterol have a favorable effect on the incidence of these morbid events. Most data, however, indicate that diet has limited long-term effects on low-density lipoprotein cholesterol or BP (2). Indeed, the observational link between BP or cholesterol levels and morbid events identifies hazard ratios that may be impressive from a population perspective but misleading for a physician dealing with an individual patient (3). Does a 20% or 30% higher risk in one patient justify an intervention that is not to be used in the patient without such increased risk? That seems to be a remarkably crude and imprecise approach to decision making. Risk factors are not the disease.
The investigators cite the Third National Health and Nutrition Examination Survey (NHANES-3) data that more than 99% of men between the ages of 35 and 74 years have at least one risk factor (4). If true, what value are risk factors in defining risk? At best they would have a sensitivity in the 90%+ range but a specificity near 0. So what is the physician charged with doing to advise or treat his or her patients? If one waits to treat until BP or cholesterol reach the threshold for guideline-mandated interventions, then many morbid events will occur before treatment is instituted and many patients not destined to suffer morbid events may be treated unnecessarily. The TROPHY (Trial of Preventing Hypertension) trial suggests that development of arterial hypertension could be prevented if drug treatment were started earlier (5).
Relying on ubiquitous risk factors in defining treatment strategies to prevent morbid events has been abandoned in the cancer field. Recent evidence of a steep decline in cancer deaths in the U.S. has been attributed largely to identifying early disease in asymptomatic individuals by routine screening with techniques such as mammography, colonoscopy, Pap smears, and prostate antibody screening (6). No such screening for early cardiovascular disease (CVD) has been advocated or reimbursed by insurance companies in the U.S.
The traditional approach to preventing CVD has been the modest primary prevention risk reduction strategies proposed by Greenland and Lloyd-Jones (1) or the pharmacologic secondary prevention widely recognized to prevent recurrent events in patients who have already suffered heart attacks or strokes (7). Where is a strategy akin to the success in cancer prevention: early recognition and aggressive intervention?
Atherosclerotic disease antedates morbid events by years. It can now be recognized by easily performed noninvasive testing. We have developed a global model for assessing vascular and cardiac health and identifying early disease. It consists of 7 vascular and 3 cardiac functional and structural tests performed in 1 h by one technician in a single room (8). The 7 vascular tests include large and small artery elasticity (compliance), sitting BP, BP response during a moderate treadmill exercise test, optic fundus photography, measurement of carotid intimal-media thickness and microalbuminuria. The 3 cardiac tests include an electrocardiogram, left ventricular (LV) ultrasonography for LV volume and mass, and determination of blood N-terminal pro-B-type natriuretic peptide level. These individual cardiovascular tests were selected on the basis that they are predictive of cardiovascular morbidity and mortality (9). Table 1 summarizes the disease score, identified as the Rasmussen score, named for the benefactor of our center, and the scoring of each test. Each of the 10 tests is scored as follows: 0 for normal, 1 for borderline abnormal, and 2 for abnormal. The total score for any individual test may therefore range from 0 to 20. We have now screened 1,500 patients. One-third of this population has a Rasmussen disease score between 0 and 2 (low disease risk), one-third between 3 and 5, and one-third 6 and above, which is considered high risk. Our preliminary data indicate that this disease score is far superior to risk factor assessment and Framingham 10-year risk scores in identifying those destined for morbid events. Why has the profession not embraced this approach in identifying individuals in need of pharmacopreventive therapy? Indeed, all of the recent prospective trial data show the dramatic benefit on outcome of drugs to lower cholesterol and BP but less documentation of benefits from risk factor reduction through lifestyle changes alone.
The time has come to enter a new era in which we focus on early disease rather than risk factors. We now know how to identify and track the arterial and cardiac disease that will eventually make people sick. Recently we showed in the DETECTIV (Detection and Treatment of Early Cardiovascular Disease Trial: Intervention with Valsartan) study that the angiotensin II receptor blocker valsartan could reduce the Rasmussen disease score in asymptomatic high-risk patients with pre-hypertension or BP controlled to <140/90 mm Hg (10). We should seize the opportunity to identify asymptomatic individuals with the disease and treat it. That is something physicians can and should do. It would likely be more effective and understandable than the vague public health approaches we have advocated for the last generation. Clinical trials to document the cost effectiveness of this approach are urgently needed.
Dr. Cohn is a director of and holds equity in Cohn Prevention Centers, LLC.
- Abbreviations and Acronyms
- blood pressure
- cardiovascular disease
- left ventricular
- Received January 25, 2008.
- Revision received February 21, 2008.
- Accepted February 27, 2008.
- American College of Cardiology Foundation
- Greenland P.,
- Lloyd-Jones D.
- Law M.R.,
- Wald N.J.
- Smith S.C. Jr..,
- Allen J.,
- Blair S.N.,
- et al.,
- American Heart Association; American College of Cardiology; National Heart, Lung and Blood Institute
- Duprez D.A.,
- Florea N.D.,
- Jones K.,
- Cohn J.N.