Left ventricular (LV) cavity obliteration during dobutamine echocardiography (DE) indicates a vigorous inotropic response to stress. Such a response may suggest the absence of coronary artery disease (CAD), but a small LV cavity may also preclude recognition of wall motion abnormalities. We sought to determine the frequency, correlates, accuracy, and prognostic value of the LV cavity obliteration response in 336 consecutive patients who underwent coronary angiography within 1 year of DE. Cavity obliteration was defined by contact of the opposite walls in the apical views during DE, and ischemia by detection of a new or worsening wall motion abnormality. Sensitivity was based on comparison with coronary anatomy in 220 patients without prior revascularization. The prognostic implications of cavity obliteration were examined by follow-up of 324 patients (96%) over 23 +/- 9 months for death, myocardial infarction, and late revascularization. Cavity obliteration was present in 86 of the 336 DE studies (26%). Baseline and stress hemodynamics were not predictive of cavity obliteration, which was associated with LV hypertrophy and female gender (p <0.0001), and inversely related to LV systolic dysfunction and use of angiotensin-converting enzyme inhibitors or diuretics (p <0.02). The sensitivity of DE was less in patients with cavity obliteration than the remainder, especially in single vessel (46% vs 92%, p <0.001) but also in multivessel CAD (73% vs 95%, p = 0.01). Irrespective of DE and angiographic results, cavity obliteration was a negative predictor for cardiac events (RR 0.42, 95% confidence interval [CI] 0.21 to 0.87, p = 0.02) and death (RR 0.14, 95% CI 0.02 to 1.09, p = 0.06). Even after exclusion of patients with LV dysfunction, cavity obliteration was an independent predictor of freedom from events (RR 0.41, 95% CI 0.19 to 0.88, p = 0.02). Thus, LV cavity obliteration is a frequent response to DE, which compromises the sensitivity of DE but is correlated paradoxically with a favorable clinical outcome.