Detection of pulmonary vein stenosis by transesophageal echocardiography: Comparison with multidetector computed tomography

Objective: The objective of this study is to compare the use of transesophageal echocardiography (TEE) vs multidetector computed tomography (MDCT) for detecting pulmonary vein stenosis. Background: Pulmonary vein isolation is increasingly used to treat atrial fibrillation. Pulmonary vein stenosis remains a potential complication of pulmonary vein isolation and ideal methods for detection of stenosis are still to be determined. Methods: Thirty-six subjects who underwent pulmonary vein isolation returned for follow-up MDCT and TEE. Percent diameter loss was reported for each pulmonary vein stenosis by MDCT. A 50% narrowing was considered as an indication of a stenosis. Pulsed-wave Doppler using TEE was used to measure peak velocities of all pulmonary veins. Results: Multidetector computed tomography and TEE were performed in all subjects (58 ± 10 years) at 4 ± 2 months after pulmonary vein isolation. Atrial fibrillation was present in 14% at time of follow-up. Multidetector computed tomography was able to evaluate all 4 (100%) pulmonary veins in 36 subjects, whereas full interrogation by TEE was possible in 138 (96%) of 144 veins. Pulmonary vein stenosis >50% by MDCT was present in 7 pulmonary veins. Analysis of the receiver operating curve for TEE showed that it had optimum detection of pulmonary vein stenosis at peak velocities ∼100 cm/s with 86% sensitivity and 95% specificity. Area under the curve for TEE was 0.93. Clinically significant stenosis was observed in 2 subjects and was detected by both TEE and MDCT. Conclusions: Transesophageal echocardiography was able to detect most pulmonary veins with good sensitivity and specificity in comparison to MDCT. Pulmonary veins may be visualized more frequently by MDCT; however, TEE provides additional data about the functional significance of a pulmonary vein stenosis.