Dong Hyun Yang MD, Presenter: Nothing to Disclose

Joon-Won Kang MD, Abstract Co-Author: Research Grant, Guerbet SA

Jae Won Lee MD, Abstract Co-Author: Nothing to Disclose

Jae-Kwan Song MD, PhD, Abstract Co-Author: Nothing to Disclose

Tae-Hwan Lim MD, PhD, Abstract Co-Author: Nothing to Disclose

To assess the value of cardiac multislice computed tomography (CT) for localization of mitral valve prolapse (MVP) in comparison with transesophageal echocardiography (TEE) and intraoperative finding

During 10 months, 60 patients with MVP underwent mitral valve repair. Of them, 45 patients (mean age, 54 years; 28 males; 13 with atrial fibrillation) underwent CT with retrospective electrocardiography gating for preoperative evaluation of coronary artery and cardiac valves using second generation dual-source CT. CT images were analyzed by two independent readers using multiplanar reformation technique and volume rendering from the atrial point of view (so called ‘surgical view’). MVP was defined as a greater than 2-mm displacement of leaflets below the annulus plane, and prolapsed scallop was recorded (anterior leaflet: from lateral to medial A1, A2, A3; posterior leaflet: P1, P2, P3). For evaluation of diagnostic performance of CT for localization of MVP, intraoperative findings (n=45) and preoperative TEE (n=38) were used as references, respectively. Interobserver agreement of CT analysis was tested with the kappa statistic, and disagreements were solved by consensus. 

The diagnostic performance of CT for the localization of prolapsed scallop in patients with MVP was as follows: sensitivity 86% / 77% (intraoperative findings / TEE as references), specificity 95% / 95%, positive predictive value 86% / 87%, negative predictive value 95% / 90% on a scallop-by-scallop comparison. On a per-patients analysis, location of prolapsed scallop was falsely assigned in only one patient (CT: A3, surgical finding: P3). Of 10 scallops showing false negative results in comparison with intraoperative finding, 9 scallops were posterior mitral leaflet. Interobserver agreement was excellent (kappa value = 0.82). Mean effective dose of CT was 15.3 millisievert.

Cardiac CT showed good results in identifying the location of mitral valve prolapse and could be applied in preoperative planning and exclusion of coronary artery disease before surgery.

CT may be used as a complementary or alternative imaging modality in localization of MVP and could be applied in preoperative planning and exclusion of coronary artery disease before surgery.

Yang, D, Kang, J, Lee, J, Song, J, Lim, T, Cardiac Computed Tomography in Localization of Mitral Valve Prolapse: Scallop-by-Scallop Comparison with Transesophageal Echocardiography and Intraoperative Findings.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12026915.html