Dobutamine echocardiography (DE) has been shown to be safe, feasible, and accurate for identification of coronary artery disease (CAD) in mixed populations. The purpose of this study was to examine gender differences in physiologic response and accuracy of DE. We studied 2,886 consecutive DEs, performed in 2,748 patients, 1,209 of whom (44%) were women. A standard incremental protocol (5 to 40 microg/kg/min in 3-minute stages) was followed by atropine and/or an additional stage with 50 microg/kg/min, if the heart rate response was inadequate. Hemodynamic and echocardiographic findings were recorded at each stage. Three hundred sixty-nine patients without previous cardiac intervention (including 135 women) also underwent cardiac catheterization within 1 year of DE. Significant coronary stenoses (defined angiographically as >50% diameter) were present in 67% of women and 65% of men, of whom 55% and 65%, respectively, had multivessel disease. Women had a higher baseline heart rate (76 +/- 13 vs 73 +/- 14 beats/min, p <0.0001), and showed a more rapid increase in heart rate at low dose, with a higher age-predicted maximum heart rate at peak. This led to test termination at target heart rate but a submaximum dose in 22% of women versus 15% of men (p <0.0001) and less frequent administration of atropine (29% vs 34%, p <0.01). Dose-limiting side effects (8% vs 7%) and submaximum heart rate responses (14% vs 17%) were comparable in men and women. Even after the exclusion of negative DE at submaximal heart rate responses, the overall sensitivity was significantly lower in women than men (78% vs 88%, p <0.05), both for single (72% vs 78%, p <0.05) and for multivessel disease (82% vs 93%, p <0.05). The low specificity in both genders (55% vs 46%) probably reflected post-test referral bias. Thus, physiologic responses to dobutamine stress are comparable in men and women, except for a more rapid heart rate response in women, but the accuracy of DE for diagnosis of CAD in women is less than in men.