RESUMO
OBJECTIVES/BACKGROUND: Although Duke treadmill score (DTS) is the most widely used risk stratification method in younger patients undergoing exercise treadmill test (ETT) its specific value in the elderly is not established. METHODS: 137 patients aged ≥80 years who underwent ETT myocardial perfusion imaging (MPI) were studied. DTS and MPI (including summed stress scores, SSS) related data were registered per patient and follow up was performed to document cardiac death (CD), myocardial infarction (MI) and late (>3 months) revascularization (LR). Kaplan Meir and Cox regression survival analysis were employed to determine the prognostic value of DTS in relation to MPI data for these endpoints. RESULTS: After a median follow up duration of 6.7 years 28 deaths, 7 CDs, 4 non fatal MIs and 12 LRs were observed. Incidence rates of CD/MI were significantly different only between low and high risk SSS categories (p = 0.044). Risk groups by DTS had no significant differences in survival free of CD/MI (p = 0.743) in contrast to risk groups according to SSS (p = 0.026), while both DTS and SSS based risk groups had significantly different survival free of CD/MI or LR. SSS was a significant univariate predictor of both CD/MI (HR 1.088, p = 0.019) and CD/MI or LR (HR 1.095, p < 0.001), but DTS only of the latter endpoint (HR 0.909, p = 0.003). CONCLUSIONS: In octogenarians DTS was found to be a significant predictor of the LR related endpoint but not of the hard endpoint of CD/MI, in contrast to SSS which was a powerful predictor of both soft and hard cardiac endpoints.