PURPOSE: Evaluation of the incidence and clinical implications of perfusion defects that appear worse in the rest than the stress myocardial perfusion imaging with technetium-99m isonitrile (MIBI), similar to the reverse redistribution described with thallium-201. METHODS: The studies of 730 consecutive patients with known or suspected coronary artery disease who underwent either diagnostic or prognostic (after myocardial infarction) technetium-99mMIBI planar scans were reviewed. Usual planar images were subjective and semi-quantitative analyzed. Patients were divided in two groups according to the indication of the test. The incidence of the reverse reperfusion pattern, the correlative findings with the angiographic coronary anatomy, when available, and its clinical implications were assessed. RESULTS: Three in the 540 (0.55%) patients of the diagnostic group and 6 out of the 190 (3.15%) patients of the myocardial infarction group have shown the "reverse reperfusion" pattern, with an overall incidence of 1.23%. Perfusion defects were anterior in 2 and inferior in 1 patient of the diagnostic group, compared to 4 anterior and 2 inferior within the prognostic population patients. One patient of the diagnostic group and 4 of the prognostic group showed ST-T changes on the exercise ECG. All of them achieved at least 85% of the maximum predicted heart rate during the stress test. The only patient with the "reverse reperfusion" pattern in the diagnostic group had no coronary disease at angiography, while the 6 patients in the post myocardial infarction group have shown either an occluded (2 cases) or recanalized infarct related artery (2 cases) and remote coronary disease (2 cases). Three of them underwent successful coronary angioplasty based on clinical and laboratory evidences of ischemia. CONCLUSION: The "reverse reperfusion" with technetium 99-m MIBI is an uncommon finding and may be associated with jeopardized myocardium after infarction, although its predictive value for recanalized infarct related artery seems to be very low. It seems to provide no added value towards coronary artery disease diagnosis. Additional experience is required in selected subgroups of patients in order to clarify the clinical value of the "reverse reperfusion" pattern with the technetium-99m MIBI.