Computed tomography evaluation of right heart dysfunction in patients with acute pulmonary embolism

To evaluate the role of qualitative assessment of right heart dysfunction on multidetector computed tomography (CT) in patients with acute pulmonary embolism.

Seventy-four consecutive adults with pulmonary embolism diagnosed on multidetector nongated CT were identified between July 2002 and March 2004. There were 47 women and 27 men, with a mean age of 62 years. Each CT scan was jointly reviewed by 2 of 3 reviewers in consensus. The CT scans were qualitatively assessed for dilatation of the right ventricle and the position of the interventricular septum. Scans were considered positive for right heart dysfunction if, on visual integration of multiple axial images, the right ventricle was dilated or the interventricular septum was straightened or bowed into the left ventricle. The extent of pulmonary vascular obstruction was graded using the CT clot burden scoring system. Reports of echocardiograms (n = 30) were reviewed when available. The sensitivity and specificity of CT and echocardiography in demonstrating right heart dysfunction were calculated and compared using pulmonary vascular obstruction of > or =30% as the reference standard.

Sixty-six percent (49 of 74 patients) with pulmonary embolism had right heart dysfunction on CT, with right ventricular dilatation in 38 patients and septal straightening or bowing in 44 patients. Forty-nine percent (36 of 74 patients) had pulmonary vasculature obstruction of > or =30%. There was a significant difference between the mean clot burden of patients with (12.8) and without (7.5) right heart dysfunction on CT (P = 0.0021). The sensitivity and specificity of CT in demonstrating right heart dysfunction were 81% (29 of 36 patients) and 47% (18 of 38 patients), respectively. Forty-one percent (30 of 74 patients) had technically adequate echocardiograms within 48 hours of CT. Fifty-seven percent (17 of 30) of the echocardiograms were positive for right heart dysfunction. There was no significant difference between the mean clot burden of patients with (12.7) and without (10.3) right heart dysfunction on echocardiography. Echocardiography had a sensitivity of 56% (10 of 17 patients) and a specificity of 42% (5 of 13 patients) in demonstrating right heart dysfunction.

Qualitative assessment of the cardiac chambers is a quick and practical means of evaluating for right heart dysfunction on CT. Computed tomography findings of right heart dysfunction in patients with acute pulmonary embolism compare favorably with echocardiography and correlate with a higher mean pulmonary arterial clot burden. Because most patients do not undergo echocardiography, chest CT often provides the only opportunity to evaluate for right heart dysfunction in patients with acute pulmonary embolism.