4.2 ICE images for evaluation of LAAC
TEE has long been the standard for guiding LAAC. It can exclude LAA thrombus during the operation, guiding transseptal puncture, facilitating placement of pigtail, sheath and device, and assessing the device position and residual flow. Precise assessment of implanted device position and residual flow was commonly based on optimal long-axis views of the device displayed by TEE at different imaging planes. However, limited by the varied anatomical relationship between the esophagus and the LAA, the long-axis images of the LAA and implanted device in some patients cannot be displayed satisfactorily, mostly at 135°, which impacts the accuracy in assessing LAA sizing, device position and residual leakage. The use of ICE has been reported to be able to obtain sufficient images as TEE to guide LAAC with similar effectiveness and safety5,6,8,9. In the early studies, ICE probe was placed in the right atrium, coronary sinus, right ventricular outflow tract, or pulmonary artery to guide LAAC2,4,8. However, ICE examination conducted in the right atrium failed to show consistent optimal images from LA/LAA because of the inherent far-field limitation. The LAA images obtained with the ICE probe positioned in LA were dramatically improved3,5,6,9-11. However, the systematic approach for manipulating ICE probe to achieve consistent, comparable multi-angled images as TEE exam was not well described. Frangieh et al9and Korsholm et al5 did not discuss the detailed approach to navigate ICE probe in LAAC. Kim et al10 placed the ICE probe at a single position in LA to guide LAAC which might affect assessment accuracy. Masson et al and Alkhouli et al3,12 reported using 2 ICE views guide LAAC while Hemam et al6 went further to show that ICE could be placed in 4 different positions including the top of LA , the left upper and lower pulmonary veins, and through the mitral valve to better evaluate the LAA and the device. But these approaches, without using the LA geometry map described by us, were limited by the varied anatomical relationships among LAA, pulmonary veins and mitral valves and the technical difficulty in acquiring reproducible, comparable and omni-plane views similar to those obtained by TEE counterparts.
TEE allows to assess the LAA and the device from a full 0-180° sweep of imaging probe. To compare the performance of ICE versus TEE, we proposed the step-by-step ”FLAVOR” approach in this study which consistently obtained high quality, reproducible images at 4 proposed imaging angles. In our study, all the proposed views could be obtained with good quality in ICE group. The acquisition of images was independent of esophageal-LAA anatomical relation which clearly limits the TEE performance, with more than one-third cases in TEE group were unable to display the long-axis image at one of the four angles, mostly at 135°. We suggest that with proper training on the “FLAVOR” approach, intra-LA ICE imaging could be an attractive alternative for guiding LAAC. While the cost of ICE catheter remains a barrier limiting its wide application, it would be appealing for patients undergoing combined AF ablation and LAAC because the catheter would be needed for transseptal punctures, LA geometry mapping and guiding ablation anyway.
4.3 The effectiveness of LAAC with ICE imaging
Previous studies5,6,10 compared the ICE-guided LAAC under local anesthesia with the TEE-guided LAAC under general anesthesia, and found that both of them had a very high procedural success rate. The application of ICE did not reduce the success rate, nor did it increase device recapture times or resizing rate. Our results are consistent with these findings, showing LAAC success rate of 100% for both approaches. Intra-operative residual leak <3mm was observed in 2 cases in the ICE group and 3 cases in the TEE group after device release. The device recapture and resizing rates were similar in the 2 groups. Whether the use of ICE has a favorable impact on total procedural time, radiation time, radiation dose and contrast agent dose remained controversial5,6,9,10. This is largely attributed to different workflows during the procedure. In contrast to previous studies, we performed all the procedures under local anesthesia. We conducted a brief TEE examination only at the end of the procedure to verify the device position and the degree of residual leak. In ICE group, we used the LA geometry map to facilitate ICE catheter manipulation to obtain the optimal imaging plane thereby sparing the operator the need for intense fluoroscopy control. This approach leads to reduced fluoroscopy time and radiation exposure without a increase in procedure time in ICE cohort.