Prediction of mortality in patients with coronary artery disease undergoing vasodilator stress testing:: A comparison between 99mTc-tetrofosmin and 99mTc-sestamibi

Purpose To compare the prognostic ability of the imaging agents Tc-99m-sestamibi versus Tc-99m-tetrofosmin to predict mortality outcomes in patients with documented coronary artery disease and undergoing vasodilator stress testing. Materials and methods The study included 2147 consecutive patients who underwent rest and stress single photon emission computed tomographic (SPECT) examination with either 99mTc-sestamibi (n= 1128) or Tc-99m-tetrofosmin (n=1019). Information relating to all-cause death and cardiovascular death was collected over a 4-year study period. Unadjusted Kaplan-Meier estimates were compared for the two imaging agents. Cox proportional hazard models were examined to determine the incremental contribution of SPECT sum stress score (SSS) and the imaging agent after adjusting for clinical and demographic characteristics. Additionally, the interaction between SSS and agent was examined to determine if the effect of SSS on prognosis was different for the two agents. Results Vasodilator agents were used for stress testing in all patients who received Tc-99m-tetrofosmin and Tc-99m-sestarnibi. Despite differences in patient risk factors Kaplan-Meier estimates were similar for the two groups of patients. Resulting P-values for differences between models for the end points of (1) death from any cause and (2) cardiovascular death showed that SSS combined with clinical index was significantly better than a model that adjusted for only baseline characteristics (P < 0.0001 for both endpoints). The addition of imaging agent (Tc-99m-tetrofosmin or 99mTc-sestamibi) to the model containing both SSS and the clinical characteristics did not show further significant improvement (P=0.62, P=0.96 for death from any cause and cardiovascular death, respectively). Conclusion The type of clinically available Tc-99m-labelled myocardial perfusion agents did not affect interpretation of results for prognostic assessment.