We appreciate Dr. Sanchez's comments and fully agree that for the general internist practicing in 1994, little can be done to identify risk among persons with no evidence of underlying disease, other than doing standard preventive techniques aimed at decreasing the development of coronary artery disease. After the disease has developed, specific markers are available for at least some of the risk factors associated with sudden death. Multiple strategies will probably be necessary, depending on the size and risk of the targeted population base. The hope for the future is the evolution of simple screening techniques, which could allow the identification of high-risk clusters hidden within larger populations. For example, a currently useful clinical technique for identifying a high-risk subgroup in a general population is the response of the Q-T interval on the standard 12-lead electrocardiogram to class I-A antiarrhythmic drugs such as quinidine. Patients with idiosyncratic exaggerated Q-T prolongation appear to be at an increased risk for potentially fatal torsade de pointes [1]. This example is of limited importance because it occurs infrequently.

Recent observations of specific T-wave changes in response to ischemia and reperfusion [2] may soon provide a marker identifying larger patient clusters at risk for fatal arrhythmias during ischemic events. Experimental studies suggest that subgroups may have specific ion channel patterns controlling the response to ischemia and reperfusion. These patterns may predispose to such T-wave changes and to fatal arrhythmias [3].

In regard to general screening, the use of ultrafast computed tomography is problematic. The technique appears to provide a noninvasive means of identifying patients with emerging structural disease of the coronary arteries [4], but it is not yet useful for screening for risk for sudden cardiac death in the general population. In our analysis of the role of dynamic risk factors in sudden cardiac death [5], tests that identify the consequences of structural disease (that is, ischemia) appear to be more useful than those that simply identify an anatomic abnormality.

In response to the question about the CAST study, no data from CAST suggest that a serum magnesium abnormality contributed to outcome, and we believe that, based on existing knowledge, the routine use of magnesium supplementation to prevent sudden cardiac death (or malignant ventricular arrhythmias) is not warranted for the general population.