In this issue of JCR, Zoghbi et al 1 present a model of risk stratification that compares the American Association of Cardiovascular and Pulmonary Rehabilitation Association (AACVPR) risk stratification process (ARSE as labeled by the authors) with a modified version of the Charlson comorbidity index (CMI). The authors specifically address the impact of noncardiac comorbidities on the ability of these tools to predict symptoms (eg, chest pain, dyspnea, fatigue) and events causing a change in a patient's condition or therapy (eg, uncontrolled hypertension, hypo- or hyperglycemia, brady/tachyarrhythmias, ST segment changes). Hard events such as myocardial infarction, restenosis, or death were not investigated.
The findings demonstrate an expected relation across both the ARSE and the CMI. The proportion of patients with at least one event increased from the lowest to the highest risk categories with both tools. Both tools proved to be independent predictors of events in an age-, gender-, and race-adjusted logistic regression model. However, interestingly, the R2 in a model containing the ARSE and the CMI was 0.065, indicating that these tools explained only 6.5% of the variability in events among groups. This finding emphasizes the importance of unidentified characteristics that increase the risk for recurrent events among patients with documented coronary heart disease (CHD).
Overall, the authors have conducted an interesting investigation of an important, and perhaps underappreciated, purpose of cardiovascular rehabilitation (CVR) medical surveillance designed to identify patients more likely to experience untoward events during exercise therapy, particularly during the early recovery period after a clinical cardiac event. This article illustrates, however, how imprecise the process of risk stratification can be. It also emphasizes the fact, as pointed out in a 1991 JCR editorial, that no patient with CHD is at low risk. 2 All patients with CHD are four to eight times more likely to experience a recurrent event than their age-matched peers without CHD are to experience a first event. 3
Studies on the safety of CVR have demonstrated that individuals who experience a cardiac arrest during exercise therapy are not necessarily the "highest" risk patients. 4,5 In the first 21 years of the Boone Hospital Center (Columbia, Missouri) CVR program, there were seven cardiac arrests, representing a rate of approximately 1 per 125,000 patient hours of exercise therapy (unpublished data). Five of these events occurred during phase 3 and two occurred during phase 2. Only two of these patients were in the "highest" ARSE risk category.
An explanation for this frequently observed phenomenon may be the fact that many patients who experience a coronary occlusion have relatively minimal obstructive disease (<50% lesions) and likely would be classified in "lowest" risk category by standard criteria. 6 A discussion of these findings is beyond the scope of this editorial, but they suggest that emerging imaging techniques such as intravascular ultrasound or magnetic resonance, which can identify "vulnerable" plaques, could greatly enhance the ability of risk stratification procedures to identify patients at greater risk of an untoward event, particularly recurrent angina, myocardial infarction, or cardiac arrest, while participating in CVR exercise therapy. 7
When considering risk stratification, it is important to note that one purpose of ARSE is to identify patients at risk for mortality during the first year after a cardiac event. The lowest-risk patients have a 1-year mortality risk of 2% to 4%, a moderate risk of 10% to 15%, and a highest risk of 25% or greater. Thus, how well this procedure may predict which patients will experience a non-life threatening change in symptoms or condition during CVR exercise therapy is unknown. The findings from the current study suggest that neither the ARSE nor the CMI is very effective for this purpose.
The stratification process for disease progression is based on established clinical guidelines, such as the National Cholesterol Education Program, and scientific evidence, which demonstrate that stabilization or regression of coronary artery disease (CAD) and reduction in clinical events occur when optimal levels of traditional coronary risk factors are achieved. 8-10 It is designed to identify patients whose disease is "more likely" to progress, and subsequently to facilitate the implementation of targeted interventions to minimize a patient's risk for worsening CAD. Neither process is designed to identify the patients who are going to experience non-life threatening events while participating in CVR exercise therapy.
Recent studies confirm the utility of the traditional risk factors included in the AACVPR risk stratification process for disease progression. 11,12 Greenland et al 11 reported that among a total cohort of more than 400,000 men and women (mostly men), 87% to 100% of individuals who experienced fatal CHD were exposed to at least one major CAD risk factor (hypertension, high cholesterol, diabetes, or smoking). In the group 40 to 59 years of age, 87% to 94% of the individuals with fatal CHD had at least one major risk factor. For nonfatal myocardial infarction (MI), prior exposure to one major risk factor was documented in 92% of the men ages 40 to 59 years and 87% of the women in this age group. A second report by Khot et al 12 found that among patients with CHD, at least one of the four major risk factors was present in 84.6% of 34,589 women and 80.6% of 87,869 men. Among men younger than 55 years and women younger than 65 years, only 10% to 15% of patients with CHD lacked any of these conventional risk factors. If the other two of American Heart Association's six major risk factors for CAD (physical inactivity and obesity) were included in an analysis of this nature, certainly nearly all patients with CAD/CHD would be exposed to at least one major risk factor, and most likely two or more.
Other studies have confirmed that aggressive management of traditional risk factors retards or reverses CAD progression and improves myocardial perfusion. 13-15 A recent study by Sdringola et al 16 showed that patients with CAD treated maximally (with exercise, diet, and medical therapy to achieve target levels such as a low-density lipoprotein level lower than 85 mg/dL) experienced significantly fewer recurrent clinical events over a 5-year follow-up period than patients with CAD who received moderate treatment with medical therapy and traditional dietary therapy or those who received poor treatment, as evidenced by essentially uncontrolled levels of traditional risk factors. Only 6.6% of the maximally treated group experienced a clinical event, as compared with 20.8% of the moderate treatment group and 30.6% of the poor treatment group. Clearly, the AACVPR risk stratification process for disease progression is a prudent process that facilitates the identification of patients at higher risk for CAD progression and is critical to their long-term management and risk reduction.
This discussion is not intended to detract from the importance of assessing comorbidities. In fact, this is recommended in several chapters and sections of the AACVPR guidelines (ie, Guideline 12, page 44, Table 5.1, page 54). 17 It is known, for example, that individuals with various forms of arthritis are at increased risk for CAD/CHD. The article by Zoghbi et al 1 does not address the nature of the observed relations between several noncardiac comorbidities and the risk of CHD events. My personal bias is that many of these patients are inactive and overweight, with high levels of traditional CAD risk factors, all accounting for much of their increased risk.
Overall, Zoghbi et al 1 have made a positive contribution to the body of knowledge on risk stratification and medical surveillance of post-CAD patients. It is imperative that CVR practitioners of CVR stratify patients for risk of 1 year mortality, recurrent nonfatal events, and disease progression. Comorbidities also must be identified and addressed to ensure an optimally safe and effective secondary prevention program for all patients.
Other factors shown to increase the risk for clinical events during exercise such as high exercise capacity, exceeded target heart rate levels, type A personality, and others also should be identified, monitored, and addressed. Close supervision of all patients with CAD on a daily basis as described in the AACVPR guidelines (pages 50 and 51) is essential to the optimization of safety and long-term outcomes. 17
References