Panic disorder (PD) and panic-like anxiety have been associated with an increased risk of cardiovascular death. No study has specifically examined the association between panic attacks and ischemia in patients who have coronary artery disease (CAD). We hypothesized that panic attacks would induce myocardial perfusion defects in patients who have CAD and PD. Sixty-five patients who had CAD and positive results with nuclear exercise stress testing (35 with PD and 30 without PD served as controls) underwent a well-established panic challenge test (1 vital capacity inhalation of a gas mixture containing 35% carbon dioxide and 65% oxygen) and were injected with technetium-99m sestamibi at inhalation. Single-photon emission computed tomography was used to assess per-panic challenge perfusion defects, and heart rate, blood pressure, and 12-lead electrocardiogram were continuously measured during the procedure. Patients were not withdrawn from their cardiac medications. Patients who had PD were significantly younger than the controls; otherwise groups did not differ with respect to gender, cardiac medications, nuclear exercise test results, and baseline heart rate and blood pressure. Seventy-four percent of patients (26 of 35) who had PD had a panic attack at inhalation versus 6.7% of controls (2 of 30, p <0.001). As hypothesized, patients who had PD and demonstrated a panic attack were more likely to develop a reversible myocardial perfusion defect than were controls who did not have an attack (80.9% vs 46.4% p = 0.009). Thus, despite being on their cardiac medications, panic attacks preferentially induced significant perfusion defects in patients who had CAD and PD. In conclusion, panic attacks in patients who have CAD appear to be bad for the heart.