To establish the real nature of 201Tl defects in the assessment of myocardial viability (e.g., fixed versus reversible), 201Tl reinjection was evaluated in a multicenter trial involving 402 consecutive patients with ischemic heart disease and exercise 201Tl defects. METHODS Twelve hospitals, using the same type of gamma camera and computer software, adopted one of the two most widely used reinjection protocols. In 230 patients (Group A), reinjection was performed immediately after stress-redistribution planar imaging; in 172 patients (Group B), reinjection was performed on a separate day and followed by rest-redistribution imaging. The images were interpreted by three blinded observers in a core laboratory on a five-point qualitative scale; the reproducibility in visual scoring was excellent. RESULTS Groups A and B had a similar prevalence of myocardial segments with abnormal uptake at stress (39%, 40%), as well as with reversible (16%, 17%), partially reversible (21%, 19%) and irreversible (63%, 64%) defects at redistribution. After reinjection, 201Tl uptake improved in 27% and 36% of both partially reversible and irreversible defects in Groups A and B. No differences were found when comparing early and delayed reinjection imaging in Group B. CONCLUSIONS This study confirms the validity of 201Tl reinjection in a large, unselected population, but the discordance with stress/redistribution is less than has been previously reported for both 201Tl reinjection protocols, the prevalence of improved segments after reinjection was higher with the separate day approach.