Background: Heart transplant recipients require serial assessment of coronary arteries due to a risk of cardiac allograft vasculopathy or atherosclerosis. Currently available non-invasive imaging methods are of a limited value for the detection of coronary stenoses, and thus invasive coronary angiography (ICA) is recommended.
Aim: We evaluated diagnostic accuracy and clinical usefulness of dual-source computed tomography (DSCT) as a potential alternative to ICA for the detection of coronary stenoses.
Methods: DSCT was performed in 20 consecutive heart transplant recipients (15 males, mean age 47.5 years) who were scheduled for ICA. Exclusion criteria included renal dysfunction with creatinine clearance <45 mL/min and lack of patient consent. All examinations were performed using a first generation dual-source scanner and a retrospectively ECG-gated protocol. Data sets were routinely reconstructed in best-systolic and best-diastolic phases. We evaluated presence of a >50% stenosis in a vessel with a diameter of >1.0 mm, image quality of each segment, and radiation dose delivered to the patient. Sensitivity, specificity, positive and negative predictive values, and diagnostic accuracy were calculated in per segment, per vessel, and per patient analyses, with ICA considered the reference method.
Results: All DSCT and ICA examinations were diagnostic and performed without any complications. Mean heart rate was 85 bpm (range 63-114), and was stable in 85% of patients. Significant stenoses (>50%) were diagnosed by DSCT in 4 of the 287 segments, and these findings were confirmed by ICA in 2 segments. Sensitivity, specificity, and diagnostic accuracy were: (1) in the per segment analysis, 100%, 99%, and 99%, respectively, for the left coronary artery; and 100% each for the right coronary artery; (2) 100%, 97%, and 97%, respectively, in the per vessel analysis; and (3) 100%, 94%, 95%, respectively, in the per patient analysis. In diastolic reconstructions, right coronary segments were significantly more commonly nondiagnostic than left coronary segments (25% vs. 11.5%, p = 0.003). In contrast, right coronary segments showed better quality than left coronary segments in systolic reconstructions (63.5% vs. 42.2%, p <0.001). Mean effective radiation dose was 12.7 (range 5.4-18.7) mSv.
Conclusions: DSCT is a clinically useful alternative to invasive coronary angiography for excluding significant coronary stenoses in heart transplant recipients. The negative predictive value of this modality is very high. Sensitivity, specificity and diagnostic accuracy is acceptably high. Imaging of coronary arteries in patients with high heart rates in technically feasible, but require modifications of routine exam protocol. Using of modern prospectively ECG-triggered protocols is not reccommended.