Diagnostic Utility of Cardiac Troponin-I Levels in Patients with Suspected Pulmonary Embolism

Although positive troponin-I (TnI) assays have been reported in patients with pulmonary embolism (PE), TnI levels in patients with suspected PE have not been evaluated systematically. The purpose of this study was to evaluate the diagnostic utility of TnI measurements in patients with suspected PE. Consecutive patients with suspected PE were identified in whom nuclear ventilation/perfusion (V/Q) scans were performed and TnI levels were measured. TnI levels in patients with and without positive V/Q scans were compared by use of t tests. After categorizing TnI levels as positive (TnI-pos, > or = 0.40 ng/mL) or negative, chi-square tests were used to relate these values to V/Q scan results. Separate comparisons were made for subjects with high-probability V/Q scans (V/Q-high, > or = 90% likelihood of PE) and intermediate- or high-probability V/Q scans (V/Q-pos, > or = 50% likelihood of PE). The mean TnI level in the 10 subjects with V/Q-high scans was 0.39 +/-0.79 ng/mL. The mean TnI level in the 81 subjects without V/Q-high scans was 0.36 +/-0.66 ng/mL (p=0.89). TnI levels did not differ between the 22 V/Q-pos subjects and the 69 subjects with negative V/Q scans (p = 0.86). A positive TnI in the setting of V/Q-pos had a sensitivity of 32%, specificity of 71%, positive predictive value of 26%, and a negative predictive value = 77% (chi-square = 0.06, p = 0.80). Elevated TnI levels are not associated with positive V/Q scans. The TnI assay is not a useful test in patients suspected of having PE, unless used to exclude myocardial ischemia or infarction.