Background In patients taking digoxin, the exercise electrocardiogram has a lower specificity for detecting coronary artery disease. However, the effect of digoxin on adenosine-induced ST-segment depression is unknown. The purpose of this study was to evaluate the specificity of the electrocardiogram during adenosine myocardial perfusion imaging in patients taking digoxin. Methods Between May 1991 and September 1997, patients (n = 99) taking digoxin who underwent adenosine stress imaging with thallium-201 or technetium-99m sestamibi and coronary angiography within 3 months were retrospectively identified. Exclusion criteria included prior myocardial infarction, coronary artery angioplasty or bypass surgery, left bundle branch block, paced ventricular rhythm, or significant valvular disease. Twelve-lead electrocardiograms were visually interpreted at baseline, during adenosine infusion, and during the recovery period. The stress electrocardiogram was considered positive if there was ≥1 mm additional horizontal or downsloping ST-segment depression or elevation 0.08 seconds after the J-point compared with the baseline tracing. Results ST-segment depression and/or elevation occurred in 24 of 99 patients. There were only 2 false-positive stress electrocardiograms, yielding a specificity of 87% and positive predictive value of 92%. All 8 patients with ≥2 mm STsegment depression had multivessel disease by coronary angiography. Conclusions ST-segment depression or elevation during adenosine myocardial perfusion imaging in patients taking digoxin is highly specific for coronary artery disease. Marked (≥2 mm) ST-segment depression and/or ST-segment elevation is associated with a high likelihood of multivessel disease. (Am Heart J 2000;140:937-40.)