To evaluate whether Holter electrocardiographic monitoring may improve the detection of ST-segment depression in patients with anginal chest pain and normal coronary arteries, we performed symptom-limited exercise testing and 24-hour Holter monitoring in a group of 38 such patients (27 women, age 54 +/- 8 years). Patients were divided into 2 groups:group X1 included 28 patients with and group X2 10 patients without significant ST-segment depression during exercise testing. There were no significant differences between the 2 groups in age, gender, characteristics of chest pain, exercise duration, heart rate (HR), and blood pressure at peak exercise, but anginal pain during exercise testing was reported by 10 patients of group X1 (36%) and 9 of group X2 (90%) (p <0.01). Episodes of ST-segment depression on Holter monitoring were found in 17 patients of group X1 (61%) and in 5 patients of group X2 (50%) (p = NS). There were no differences between the 2 groups in daily number of ST episodes (3.6 +/- 4 vs 2.8 +/- 5 episodes per patient), symptomatic episodes (8% vs 18%), and duration of the episodes. On average, HR increased significantly, in a similar way, from 15 minutes before ST-segment depression to 1-mm ST in both groups, and its value at the onset of ischemia was similar in the 2 groups (102 +/- 22 vs 109 +/- 18 beats/min, p = NS). Finally, HR at 1-mm ST during Holter monitoring was significantly lower than that observed at 1-mm ST during exercise testing (127 +/- 16 beats/min, p < or = 0.01) in group X1, and it was also lower than that observed at peak exercise (136 +/- 22 beats/min, p < or = 0.01) in group X2. In conclusion, Holter monitoring can significantly increase the detection of ST-segment depression in patients with anginal pain and normal coronary arteries, indicating a cardiac, although not necessarily ischemic, origin of the pain. Indeed, 50% of our patients with negative symptom-limited exercise testing showed spontaneous ST changes, compatible with transient myocardial ischemia, during daily activities. Differences in the response of coronary microvascular tone to exercise testing and to stimuli operating during daily life are likely to play a significant role in determining these findings.