Assessment of Mitral Paravalvular Leakage After Mitral Valve Replacement Using Cardiac Computed Tomography: Comparison With Surgical Findings.

Abstract

BACKGROUND: The diagnostic performance of cardiac computed tomography (CT) for detection of paravalvular leakage (PVL) after mitral valve replacement has not been investigated in a large population. We aimed to investigate the diagnostic accuracy of CT for diagnosis of mitral PVL using surgical findings as the standard reference and to compare the diagnostic performance of CT with those of transthoracic echocardiography (TTE) and transesophageal echocardiography (TEE). METHODS AND RESULTS: A total of 204 patients with previous mitral valve replacement who underwent cardiac CT were retrospectively included. The presence of mitral PVL was analyzed on CT, TTE, and TEE. In 78 patients who underwent redo-surgery, diagnostic performance for the detection of PVL for CT, TTE, and TEE were compared with surgical findings as the standard reference. The location of mitral PVL on CT and TEE was compared with surgical findings. Mitral PVL was present in 18.1% (37/204) on CT, in 16.2% (32/198) on TTE, and in 42.6% (29/68) on TEE. On the surgical field, PVL was identified in 41.0% (32/78). The sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy for detection of PVL were 96.9%, 97.8%, 96.9%, 97.8%, and 97.4% for CT; 81.3%, 95.6%, 92.9%, 87.8%, and 89.6% for TTE; and 96.2%, 95.8%, 96.2%, 95.8%, and 96.0% for TEE. CT and TEE identified the correct location of PVL in 75.9% (22/29) and 85.6% (19/23). CONCLUSIONS: Cardiac CT may have better diagnostic accuracy compared with TTE for the detection of mitral PVL and may be comparable to TEE for the detection and localization of PVL.