Accuracy of Imaging Modalities in Detection of Baffle Leaks in Patients Following Atrial Switch Operation

Abstract

Aims: Patients with dextro-transposition of the great arteries (d-TGA) status post atrial switch operation are vulnerable to complications such as baffle leaks. The best noninvasive imaging modality to detect baffle leaks is unknown. The purpose of this study was to determine the sensitivity and specificity of different noninvasive imaging modalities in the detection of baffle leaks in this population. Methods and Results: A single center retrospective chart review of atrial switch patients was performed. Sensitivity, specificity, negative predictive value, and positive predictive value for detecting leaks were calculated for transthoracic echocardiogram (TTE) with and without agitated saline, transesophageal echocardiogram (TEE) with and without agitated saline, and cardiac magnetic resonance imaging (cMRI). Studies were included if performed within 1 year of catheterization. Angiography via catheterization was used as the gold standard for the detection of baffle leaks. Fifty-eight atrial switch patients (54 Mustards: four Sennings) from a single pediatric center, undergoing 76 catheterizations, were analyzed. Thirty-nine catheterizations documented a baffle leak. Overall combination of sensitivity and specificity was better in agitated saline studies (TTE: sensitivity 71.4%, specificity 100%; TEE: sensitivity 100%, specificity 92.3%) versus nonagitated studies (TTE: sensitivity 50.0%, specificity 100%; TEE: sensitivity 83.3%, specificity 77.8%; or cMRI: sensitivity 66.7%, specificity 100%). Conclusion: TTE or TEE with agitated saline is superior to cardiac magnetic resonance imaging or nonagitated saline TTE and TEE in detecting baffle leaks in atrial switch patients. Agitated saline studies should be performed in this population when looking for baffle leaks.