Catheter-based mitral valve clip repair (CBMCR) is feasible for selected patients with mitral regurgitation (MR). Two-dimensional (2D) transesophageal echocardiography (TEE) is the standard modality for evaluating MR and procedural guidance. Recently, real-time three-dimensional TEE became available. The aim of this study was to evaluate the value of combined 2D and three-dimensional TEE for CBMCR. In evaluating MR for CBMCR, the confidence of interpretation of 2D TEE was compared with that of combined imaging for the localization of major valve pathology. In patients who underwent CBMCR, the outcomes and the duration of CBMCR were compared.
In this retrospective study, MR evaluation was performed by 2D TEE alone and by combined imaging in 80 and 57 patients, respectively. CBMCR was guided by 2D TEE alone in 20 patients and by combined imaging in 39 patients.
Examination by combined imaging allowed en face visualization of mitral valve anatomy and MR jet origin. The confidence of interpretation by combined imaging was higher than for 2D TEE (1.1 ± 0.3 vs 1.8 ± 0.7, P < .001).The guidance of CBMCR by combined imaging facilitated alignment of the catheter trajectory, clip positioning, and orientation of clip arms. The procedural success and final MR grade were not different between the two study groups. However, the procedural time of CBMCR using combined imaging compared with that using 2D TEE guidance alone was shorter (241 ± 58 vs 201 ± 68 min, P = .035).
The use of combined imaging compared with 2D TEE alone appears to enhance the confidence of interpretation concerning mitral pathology and catheter-clip system location and may also reduce CBMCR time.
Suture-based surgical edge-to-edge mitral valve (MV) repair performed by approximating the middle scallops of the anterior and posterior leaflets to create a double-orifice valve was first performed by Alfieri and colleagues in 1991. A recently evolved catheter-based MV clip repair (CBMCR) uses a clip to perform a transcatheter procedure, which works as a modification of the Alfieri technique. The Endovascular Valve Edge to Edge Repair (EVEREST) I trial demonstrated the feasibility of CBMCR and the medium-term durability of the repair. In a multicenter randomized trial (EVEREST II), 279 patients with mitral regurgitation (MR) were randomized 2:1 to CBMCR or surgical repair or replacement. EVEREST II demonstrated the noninferiority of CBMCR compared with surgical MV repair or replacement. On the basis of the experience of that trial, whose enrollment was completed in 2008, in 2009 the Real world ExpAnded muLtIcenter Study of the MitralClip System (REALISM) was initiated, a prospective, multicenter, continued access registry of the EVEREST II study. Until recently, CBMCR was guided by standard, two-dimensional (2D) transesophageal echocardiography (TEE). However, 2D TEE is limited in the direct visualization of major MV pathology, the assessment of the three-dimensional (3D) location of the catheter-clip system during the procedure, and the effect of the procedure on 3D MV morphology. In April 2008, combined imaging was introduced, which included both 2D and 3D TEE, at Cedars Sinai Medical Center and at New York University Langone Medical Center, which have been participating centers in both EVEREST II and REALISM since October 2005.
A new real-time 3D transesophageal echocardiographic technology was found to be helpful in a catheter-based closure of atrial septal defects. It provides unique en face visualization of mitral anatomy and the MR orifice. Three-dimensional technology also provides unique and additive information in the transcatheter closure of mitral prosthesis paravalvular leaks.
We hypothesized that the combined use of 2D and 3D TEE would have additive value over 2D imaging in evaluating MR for CBMCR and guidance of the procedure. The objectives of this study were to compare standard 2D TEE with combined 2D and 3D TEE imaging for confidence of interpretation of major valvular pathology in evaluating MR before endovascular repair, for procedural outcomes, and for the duration of CBMCR.
In this study, we retrospectively analyzed the data of consecutive patients with moderate to severe or severe MR screened and randomized to CBMCR in EVEREST II or screened in the REALISM continued-access registry between October 2005 and September 2009. Echocardiographic evaluation by 2D TEE was performed in 80 patients, and 57 of 80 patients were also evaluated by 3D TEE. The usefulness of 2D TEE and combined imaging (2D TEE and real-time 3D TEE as well as 3D color flow Doppler [CFD]) for guidance of CBMCR was studied in 23 and 36 patients, respectively.
Echocardiographic Evaluation of MR Severity
Transthoracic echocardiography was used for the assessment of MR severity. Patients entered into the study had moderate to severe (3+) ( Table 1 ) functional or degenerative MR who were symptomatic or asymptomatic but with compromised left ventricular (LV) function (LV ejection fraction < 60% or LV end-systolic dimension > 40 mm). The criteria for moderate to severe and severe MR have been published previously, according to the recommendations of the American Society of Echocardiography. The etiology of MR could be either functional or degenerative. Patients with MR due to rheumatic fever or endocarditis were excluded, as were patients with an ejection fraction of <25% and substantial LV dilatation (end-systolic dimension > 55 mm). Phillips iE33 equipment (Philips Medical Systems, Andover, MA) was used for the standard screening 2D Doppler echocardiographic evaluation of MV morphology, MR, and LV function and size.
|Variable||2D TEE only ( n = 23)||Combined 2D and 3D TEE ( n = 57)||P|
|Age (y)||68 ± 14||71 ± 13||NS|
|Central/mixed and marginal||17/6||36/21||NS|
|MR grade||3.8 ± 0.4||3.9 ± 0.4||NS|
Our study population consisted only of patients who met eligibility criteria by transthoracic echocardiography and who then underwent additional evaluation of MV morphology and MR by TEE.
The major eligibility criterion evaluated both by 2D and 3D TEE was an MR jet originating between the middle (A2 and P2) scallops. If a second marginal jet existed, it had to be insignificant. The anatomic exclusion criteria assessed only by 2D TEE included coaptation and flail dimensions, namely. coaptation length < 2 mm and coaptation depth > 11 mm and leaflet flail gap > 10 mm and width > 15 mm. Phillips iE33 equipment was used for the 2D and 3D transesophageal echocardiographic evaluation of MR and MV morphology. Both 2D transesophageal echocardiographic biplane and multiplane imaging were performed as appropriate.
First, the MV pathology was evaluated by 2D TEE using the method described by Foster et al. The anatomy was studied in three horizontal planes with the transducer angle oriented at 0°. The midesophageal four-chamber view was used for visualization of the middle scallops of both leaflets (A2 and P2). By anteroflexion and pull-back, the lateral A1 and P1 scallops were imaged, and by retroflexion and advancement of the probe, imaging of the medial A3 and P3 scallops was attempted. MV morphology was reevaluated with the transducer angle oriented between 45° and 90° and the imaging plane oriented to the line of coaptation (commissural view) for evaluation of P1, A2, and P3 scallops. By manual clockwise rotation of the probe shaft, the anterior leaflet (A1, A2, and A3) was visualized, and by counterclockwise rotation of the shaft, the posterior leaflet (P1, P2, and P3) was evaluated ( Figure 1 A). The transgastric short-axis (multiplane angle 0°–30°) view at the MV level was used for a cross-check of the findings obtained from esophageal views.
For the evaluation of the mitral anatomy, live 3D and 3D zoom-mode imaging of the MV was performed. The 3D images were oriented to a surgical view, and each scallop of both leaflets was studied (Figures 1 B and 2 A). Then full-volume electrocardiographically triggered 3D CFD of the MR jet was acquired over seven cardiac cycles. The origin of the jet was visualized by manual cropping of the 3D CFD data. The cropping plane was moved perpendicular to the jet direction until the smallest jet cross-sectional area was obtained 5 to 10 mm immediately above the coaptation of leaflets ( Figure 2 C). In patients with atrial fibrillation, multiple 3D transesophageal echocardiographic CFD data sets were acquired for optimization of image quality, with relatively stable RR intervals (<20% RR interval variability).
CBMCR Procedural Guidance
2D and 3D TEE
For 2D transesophageal echocardiographic guidance of clip navigation within the left atrium, the alignment of the catheter perpendicular to the mitral annulus (coaxial to the left atrium long axis) and clip orientation orthogonal to the line of coaptation areis determined with repetitive cross-checking in multiple standard 2D midesophageal views, including the short-axis view at the base of the heart, the four-chamber view, the long-axis (three-chamber) view, the bicaval view, and the commissural view, as well as nonstandard views with or without color Doppler. This protocol was used to assess the 3D effect of each catheter manipulation during CBMCR ( Figure 3 A).
In the 3D TEE cohort, the x-plane modality was used to guide transseptal catheterization by the simultaneous visualization of the interatrial septum in the short-axis and bicaval views. Real-time live 3D and 3D zoom images were obtained for the special orientation of the catheter-clip system within the left atrium, for the alignment of the catheter perpendicular to the mitral annulus plane, for positioning of the clip between the middle (A2 and P2) scallops, and for the orientation of the clip arms orthogonal to the line of coaptation ( Figures 3 B and 4 ).
Echocardiographic studies including 2D and 3D transesophageal echocardiographic images were reviewed by both echocardiologists and the interventional cardiologist offline.
Intraprocedural 2D and real-time 3D transesophageal echocardiographic data were assessed online, and the 3D CFD images were processed in the catheterization laboratory immediately after image acquisition. The decision to proceed to the next procedural step was based on agreement between the interventional cardiologist and the echocardiologist after review of the 2D and 3D images.
Using this strategy, the locations of MV pathology and origin of the MR jet were classified as central, marginal, or mixed.
The confidence level of interpretation was graded 1 to 3, with highest being 1 if the involvement of any of the six scallops was possible to confirm or rule out by two or more 2D transesophageal echocardiographic views (midesophageal 0°, midesophageal 45°–90°, and transgastric 0°–30°) or by one 2D and 3D transesophageal en face view. The confidence level was intermediate or grade 2 if the status of one scallop was impossible to double confirm, and the confidence level was lowest or grade 3 if the involvement of two or more scallops could not be confirmed or ruled out.
Statistical analysis was performed using SPSS version 13.0 (SPSS, Inc., Chicago, IL). Continuous data are presented as mean ± SD. Categorical data are presented as absolute numbers. The 2D and combined 2D and 3D transesophageal echocardiographic quantitative data as well as baseline and postintervention echocardiographic findings were compared using paired Student’s t tests. P values < .05 were considered significant.
Agreement between 2D TEE and the combined modality was assessed using the overall raw agreement (percentage agreement) and the Fleiss κ statistic. The Fleiss κ coefficient is an agreement rate adjusted for chance rate of agreement, and it was calculated using the Online Kappa Calculator, developed by Randolph. For this purpose, we used Randolph’s variation of Brennan and Prediger’s free marginal κ.