Clinical risk factor analysis and electrocardiographic evaluation modalities

CAD, the leading cause of death in Western industrialized nations, is the foremost differential diagnosis for adults presenting with chest pain syndromes. When angina pectoris is classic in its presentation, this symptom alone provides strong evidence for the presence of significant coronary atherosclerosis. Diamond and Forrester²⁶ calculated that 89% of patients presenting with typical angina, defined as anterior chest, neck, shoulder, jaw, or arm discomfort precipitated by physical exertion or psychic stress and relieved by nitroglycerin within minutes, have angiographic evidence of CAD. The prevalence of CAD in patients with nonanginal and atypical chest pain is 16% and 50%, respectively. Unfortunately, the visceral appreciation and description of chest pain sydromes by the patient may vary greatly in their accuracy, and may often present a difficult diagnostic problem for even the most skilled clinician. For this reason, more objective markers of CAD have been developed to corroborate the clinical history. The exercise test, initially described by Bruce and McDonough in 1969, has been used for more than 1.5 million Americans annually since 1977. The initial enthusiasm for exercise ECG stress testing has been tempered by a significant incidence of false-positive and false-sensitive tests. In the preoperative evaluation of asymptomatic high-risk patients, or patients with chest pain syndromes, the end-point of clinical evaluation and noninvasive testing is a decision regarding the need for selective coronary arteriography. Indications for coronary arteriography vary among physicians and institutions, and may be based on both local availability of the procedure and the degree of acceptance of noninvasive test data. The selection of patients for coronary arteriography must be individualized, and should evade application of any simple algorithm. Neither Bayesian nor discriminant function analysis of patient characteristics will directly denote which patients require coronary arteriography.²⁷ These rigorous analytic tests provide estimates of CAD probability, but clinical judgement must be allowed to fulfill its role as the final arbiter,^28,29 and must weigh alternative diagnoses and the implications of these diagnoses in terms of future therapy and prognosis. Thus, the most cost-effective approach in the asymptomatic patient with low disease prevalence is to use the most accurate available noninvasive diagnostic modality. In this population, selected patients may be adequately evaluated by a standard exercise ECG stress test. Exercise ventricular function evaluation using echocardiography or radionuclide ventriculography has low specificity, but has a sensitivity comparable to that of exercise myocardial perfusion imaging. Ambulatory ECG monitoring is generally not warranted in the asymptomatic patient. Bayesian probability analysis, computer algorithms, and clinical diagnosis have comparable accuracy for disease detection. ^28,29 As such, practicing cardiologists do not usually require tables or computers to assist them in deciding which patients have CAD, or which patients require coronary arteriography. Their clinical impressions must be intelligently incorporated with diagnostic test results (such as the exercise ECG and other noninvasive tests) to arrive at the most cost-effective approach to the preoperative diagnosis of CAD, and to determine whether there is a need for conary arteriography. An estimation of the disease prevalence, based on symptom complex, age, sex, etc, is essential to the use of any diagnostic test. It is important that the ability to discriminate between the presence of extensive versus mild disease exist for a diagnostic test. The confounding effect(s) of coexisting diseases and patient selection bias must also be taken into consideration when evaluating the diagnostic accuracy of any test. The 1989 report of the Society of Cardiac Angiography listed rates of major complications and death resulting from diagnostic angiography used for CAD detection as 10 and 1 (per 10,000 patients), respectively (Table 9). Comparable rates of complications and death resulting from exercise testing are 4 to 9 and 0.5 to 1 (per 10,000 patients). Asymptomatic patients with a high pretest probability (more than 90%) of significant CAD should proceed directly to coronary arteriography, while noninvasive evaluation should be applied to patients in a low CAD prevalence (less than 10%) subgroup. Unfortunately, the 35% to 65% prevalence of significant CAD in patients with PVD creates a "gray zone" that makes either diagnostic approach defensible. In this setting, the availability of maximal pharmacologic stress myocardial perfusion imaging is an attractive alternative, particurlarly when one considers that cardiac catheterization complications increase in patients with extensive peripheral and cerebrovascular disease.