Background: Dipyridamole (DIP) is the most commonly employed pharmacological stressor during myocardial perfusion tomography (SPECT) in patients unable to reach an adequate work load. Aim: To assess the predictive capacity of DIP SPECT on survival. Material and methods: We included 985 adults aged 66 ±11 years (45% women) with rest and DIP-SPECT. The main indications for the procedure were coronary artery disease (CAD) screening in 66% and known CAD in 33%. Participants were followed for a median of 65 months (interquartile range 54 to 86 month) During the follow up, 261 deaths were recorded and 98% had a specified cause in their death certificate. Results: Myocardial SPECT was abnormal in 44% of participants. Transient ischemic defects were observed in 34%, definitive defects concordant with infarction in 27% and post-stress systolic dysfunction in 23%. Twenty five percent of deaths were attributable to cardiac or ischemic cause and 22% to cancer. In a bivariate analysis, the hazard ratio (HR) of death of any cause was lower in females and higher in the presence of CAD. The multivariate analysis showed that being older than 46 years increased the HR of death of any cause. In a bivariate analysis, the HR for cardiac death was higher when the myocardial SPECT showed ischemia, necrosis or left ventricular dilation. In the multivariate analysis, post-stress left ventricular systolic function was associated with a lower risk of cardiac death. Conclusions: An abnormal myocardial SPECT, perfusion abnormalities, left ventricular systolic function or dilation are independent predictors of cardiac death in these participants.Background: Dipyridamole (DIP) is the most commonly employed pharmacological stressor during myocardial perfusion tomography (SPECT) in patients unable to reach an adequate work load. Aim: To assess the predictive capacity of DIP SPECT on survival. Material and methods: We included 985 adults aged 66 ±11 years (45% women) with rest and DIP-SPECT. The main indications for the procedure were coronary artery disease (CAD) screening in 66% and known CAD in 33%. Participants were followed for a median of 65 months (interquartile range 54 to 86 month) During the follow up, 261 deaths were recorded and 98% had a specified cause in their death certificate. Results: Myocardial SPECT was abnormal in 44% of participants. Transient ischemic defects were observed in 34%, definitive defects concordant with infarction in 27% and post-stress systolic dysfunction in 23%. Twenty five percent of deaths were attributable to cardiac or ischemic cause and 22% to cancer. In a bivariate analysis, the hazard ratio (HR) of death of any cause was lower in females and higher in the presence of CAD. The multivariate analysis showed that being older than 46 years increased the HR of death of any cause. In a bivariate analysis, the HR for cardiac death was higher when the myocardial SPECT showed ischemia, necrosis or left ventricular dilation. In the multivariate analysis, post-stress left ventricular systolic function was associated with a lower risk of cardiac death. Conclusions: An abnormal myocardial SPECT, perfusion abnormalities, left ventricular systolic function or dilation are independent predictors of cardiac death in these participants.