Acute rest perfusion imaging in high risk unstable angina: association with troponin T and clinical endpoints.

Abstract

Troponin T (TnT) and rest perfusion imaging (RPI) have been reported to be important diagnostic tools for risk stratification in patients with chest pain. We investigated the association between two methods in 60 patients presenting with typical chest pain at rest within the last 6 h before admission. All patients underwent Tc-99m gated SPECT imaging and serial TnT measurements and were followed for occurrence of adverse cardiac events up to 30 days. Perfusion defect was detected in 42 patients and elevated TnT was observed in 23 patients. All of the patients with an elevated TnT have also perfusion defect in RPI. Half of the patients with normal TnT level (51%) presented a perfusion defect detected by RPI (p = NS). The patients with elevated TnT levels had more perfusion defect numbers than those with normal TnT levels (8.2 +/- 2.9 vs. 5.3 +/- 2.2; p = 0.0007). Cardiac events occurred in 38 patients (14 MI, 24 revascularisation). In predicting cardiac events, RPI and TnT had sensitivities (97 vs. 58%; p < 0.001), specificities (77 vs. 95%, p = NS), positive predictive values (PPV) (88 vs. 96%; p = NS) and negative predictive values (NPV) (94% vs. 57%; p = NS), respectively. In predicting MI, the two tests had sensitivities (93 vs. 93%; p = NS), specificities (37 vs. 78%; p < 0.001), PPVs (31 vs. 57%; p = NS) and NPVs (94 vs. 97%; p = NS), respectively. These results suggest that in patients with rest angina (1) TnT elevation is associated with the extent of myocardial perfusion defect; (2) both tests are valuable, while positive RPI is more sensitive in predicting all cardiac events irrespective of TnT; both positive TnT and positive RPI predict a high probability to have MI; (3) both negative test results predict a very low probability to have cardiac event, including MI.