Etiological diagnosis of left ventricular dysfunction: computed tomography compared with coronary angiography and cardiac magnetic resonance.

INTRODUCTION AND OBJECTIVES To evaluate the capability of multidetector computed tomography to diagnose the coronary etiology of left ventricular dysfunction compared with using invasive coronary angiography and magnetic resonance. METHODS Forty consecutive patients with left ventricular dysfunction of uncertain etiology underwent invasive coronary angiography and contrast magnetic resonance. All patients were evaluated with multidetector computed tomography including coronary calcium presence and score, noninvasive coronary angiography, and myocardial tissue assessment. RESULTS The sensitivity and specificity of the presence of coronary calcium to identify left ventricular dysfunction was 100% and 31%, respectively. If an Agatston calcium score of >100 is taken, specificity increases to 58% with sensitivity still 100%. Sensitivity and specificity for coronary angiography by multidetector computed tomography was 100% and 96%, respectively; for identifying necrosis in contrast acquisition it was 57% and 100%, respectively; and in late acquisition, 84% and 96%, respectively. To identify coronary ventricular dysfunction with necrosis, the sensitivity and specificity was 92% and 100%, respectively. CONCLUSIONS Of all the diagnostic tools available in multidetector computed tomography, coronary angiography is the most accurate in determining the coronary origin of left ventricular dysfunction. A combination of coronary angiography and myocardial tissue study after contrast allows a single test to obtain similar information compared with the combination of invasive coronary angiography and contrast magnetic resonance.