Usefulness of Holter monitoring for detecting myocardial ischemia in patients with nondiagnostic exercise treadmill test.

To determine if Holter monitoring can predict cardiac risk in patients when the exercise test is nondiagnostic, a total of 90 eligible patients were monitored for 24 hours after their exercise test without alterations in baseline medications. Prospective follow-up was obtained and events were confirmed by investigators unaware of subjects' clinical data. Nineteen patients (21%) had a total of 71 episodes of ST depression, all of which were asymptomatic. During a mean follow-up of 719 days, there were 10 patients with adverse events: 3 with cardiac deaths, 3 with nonfatal myocardial infarctions, and 4 with admissions for unstable angina. Of the 10 adverse events, 9 occurred in the group of 19 with ST depression detected by Holter (relative risk 34, 95% confidence interval 10 to 114). The sensitivity of ST depression was 90%, the specificity 88%, the predictive positive value 47%, and the predictive negative value 99%. In a multivariate Cox proportional-hazards model that controlled for prior history of coronary artery disease, hypercholesterolemia, and all exercise test variables, the presence of ST depression detected by Holter was the only independent predictor of outcome. In patients with nondiagnostic exercise tests, ST depression detected by Holter monitoring identified those with an increased risk of adverse cardiac events. The absence of ST depression detected by Holter was a useful predictor of low risk.