Author + information
- Received October 10, 2010
- Revision received March 7, 2011
- Accepted March 17, 2011
- Published online July 1, 2011.
- François Regoli, MD, PhD⁎,
- Francesco F. Faletra, MD⁎,
- Gaetano Nucifora, MD⁎,
- Elena Pasotti, MD⁎,
- Tiziano Moccetti, MD⁎,
- Catherine Klersy, MD† and
- Angelo Auricchio, MD, PhD⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Angelo Auricchio, Division of Cardiology, Fondazione Cardiocentro Ticino, Via Tesserete 48, CH-6900 Lugano, Switzerland
Objectives The aim of this study was to evaluate the feasibility and acute efficacy of real-time 3-dimensional transesophageal echocardiography (RT3DTEE)–guided ablation of the cavotricuspid isthmus (CVTI).
Background The use of RT3DTEE to guide a transcatheter radiofrequency ablation procedure has never been systematically investigated.
Methods Seventy consecutive patients with CVTI-dependent atrial flutter underwent CVTI ablation. Procedural monitoring using RT3DTEE was assigned to patients who requested general anesthesia for the procedure (n = 21 [30%]). In the other 49 patients (the control group), the procedures were monitored using the standard fluoroscopic approach. Procedural time was considered as skin-to-skin electrophysiological procedure duration, not including anesthesia preparation; adequate radiofrequency ablation applications (with fixed temperature and power settings) were considered as lesions lasting ≥ 60 s.
Results RT3DTEE allowed visualization of the CVTI and identified related structures in most patients (20 of 21); anatomic features such as long CVTI (n = 11), prominent Eustachian ridge (n = 9), prominent Eustachian valve (n = 6), septal recess (n = 8), and pectinate muscles (n = 10) were frequent. Also, RT3DTEE allowed continuous visualization of ablation catheter movement and contact. Compared with the control group, RT3DTEE was equally effective in achieving CVTI bidirectional block (100% in both groups), and no complications occurred. RT3DTEE shortened procedural time (median 73.0 min, interquartile range [IQR] 60.0 to 90.0 min, vs. median 115.0 min, IQR 85.0 to 133.0 min, p < 0.001), reduced radiation exposure (median fluoroscopy time 4.2 min, IQR 3.1 to 8.4 min, vs. median 19.3 min, IQR 12.9 to 36.4 min, p < 0.001; median fluoroscopy dose 575.4 cGy · cm2, IQR 428.5 to 1,299.4 cGy · cm2, vs. median 3,520.7 cGy · cm2, IQR 1,700.0 to 6,709.0 cGy · cm2, p < 0.001), and reduced the number of radiofrequency applications to achieve bidirectional block (median 7, IQR 6 to 10, vs. median 12, IQR 10 to 22, p = 0.007). A strong learning curve was detected by comparing procedural data between the first and last patients treated using RT3DTEE.
Conclusions RT3DTEE-guided ablation of CVTI was feasible, allowing real-time detailed morphological CVTI characterization as well as continuous visualization of the ablation catheter during radiofrequency ablation. This approach entailed marked reductions in procedural time, radiation exposure, and the number of radiofrequency applications.
- atrial flutter ablation
- cavotricuspid anatomy
- electrophysiologic imaging
- 3-dimensional echocardiography
Radiofrequency ablation (RFA) of the cavotricuspid valve isthmus (CVTI) is a widely applied, highly successful nonpharmacological therapy for atrial flutter (1,2). It is well recognized, however, that varying anatomical and morphological aspects of the CVTI may at times render conduction block across the CVTI through catheter ablation difficult, thus resulting in prolonged ablation and fluoroscopy times or even procedural failure. Continued efforts have therefore been oriented toward developing further innovative ablative (3,4) and mapping (5,6) strategies as well as imaging modalities (7) to ensure incremental procedural success while reducing procedural durationand risk; also, procedure-related biological risk by prolonged exposure to ionization radiation represents a growing concern (8).
Real-time 3-dimensional transesophageal echocardiography (RT3DTEE) is a recently developed technique that provides 3-dimensional images of unprecedented quality. Because the right atrium is close to the transducer, this allows visualization of morphology and anatomy with great detail, as outlined in a recent contribution (9). In the specific context of CVTI ablation, RT3DTEE may help accurately characterize anatomical barriers limiting efficacious RFA delivery. Although RT3DTEE monitoring of pulmonary vein isolation procedure has been reported anecdotally (10), whether RT3DTEE may precisely delineate the CVTI and be suitable for guiding ablation in this region has not been demonstrated so far.
Therefore, the objectives of the present study were 3-fold: 1) to assess the feasibility of RT3DTEE to image the CVTI and evaluate the ability to navigate with the mapping catheter within the CVTI region; 2) to verify the acute efficacy of RT3DTEE-guided CVTI ablation by comparison with the conventional fluoroscopy-guided approach; and 3) to evaluate potential reductions of radiation exposure using RT3DTEE.
This was a single-center, prospective, observational study including consecutive patients presenting electrocardiographically documented typical atrial flutter who underwent RFA of the CVTI at Cardiocentro Ticino (Lugano, Switzerland). After obtaining patients' written and oral consent to the procedure, and at patients' request, RFA was carried out either with or without deep sedation. When under general anesthesia, patients were monitored using RT3DTEE. Patients in whom CVTI RFA had previously failed or who presented associated atrial tachyarrhythmias with common atrial flutter (atrial fibrillation, atrial arrhythmias, re-entry tachycardia, and so on), were excluded from the study. Between January 2009 and November 2010, 70 consecutive patients underwent isolated, first-time CVTI-dependent atrial flutter ablation. Twenty-one patients (30%) requested general anesthesia (the RT3DTEE group). Patients in whom CVTI ablation was carried out according to the fluoroscopy-based approach represented the control group.
CVTI ablation procedure
In all patients, antiarrhythmic drugs as well as oral anticoagulation therapy were discontinued before the ablation procedure. The ablation procedure was conducted by positioning 2 standard decapolar deflectable catheters via the femoral vein along the right atrial free wall and another in the coronary sinus. A conventional 8-mm mapping ablation catheter was then used. Intracavitary digitalized electrograms and 12-lead surface electrocardiograms were recorded and displayed on a multichannel recorder and stored on magneto-optical discs (Prucka CardioLab; GE Marquette Medical Systems, Milwaukee, Wisconsin). RFA was performed with either EPT-1000 XP (EP Technologies, Mountain View, California) or Cordis-Stockert (Cordis-Webster, Miami Lakes, Florida) radiofrequency (RF) generators, allowing the delivery of continuous, unmodulated current, with power output set at 80 W and a pre-set temperature of 65°C. The end point of the ablation procedure was the achievement of complete bidirectional CVTI block (11,12), persisting for a waiting period of 30 min.
The following procedural data were collected: total procedural time, total RF application time and number of RF applications (applications ≥ 60 s in duration ), fluoroscopy time, and cumulative exposure dose. Procedural time in the RT3DTEE group did not include anesthesia preparation time but was limited to the duration of the electrophysiological procedure.
Real-time 3-dimensional transesophageal echocardiographic acquisition and analysis
RT3DTEE was performed using a commercially available fully sampled matrix-array transesophageal echocardiographic transducer and ultrasound system (X7-2t Live 3D transducer, iE33; Philips Medical Systems, Andover, Massachusetts). First, a real-time 3-dimensional image from the 2-dimensional 4-chamber plane was obtained, using a pyramidal dataset large enough to include the entire right atrium. Subsequently, by using the auto-crop function, the anterior half of the atrium was removed, and an arbitrary crop was applied to remove remaining nonrelevant anatomic structures. The image was then rotated to provide a perspective similar to the fluoroscopic left anterior oblique view (Fig. 1). From this position, a slight clockwise rotation and a deeper cut of the frontal plane were applied to provide a perspective similar to the fluoroscopic right anterior oblique view. Both views allowed evaluation of the length of the CVTI and the details of its anatomical features (Figs. 2 and 3).⇓⇓ Real-time images were viewed by both the echocardiographer and the electrophysiologist during the RFA procedure.
Quantitative analysis of CVTI length was performed offline using dedicated software (QLAB Cardiac 3DQ; Philips Medical Systems). The multiplanar reconstruction mode of the software was used to identify the exact extensions of the right atrial myocardium for accurate CVTI length determination; the length of CVTI was then measured at end-systole as the linear distance between the tricuspid annulus and the end of the right atrial myocardial sleeve.
All remaining qualitative analyses were performed online during the acquisition of real-time 3-dimensional images. In particular, the Eustachian ridge was evaluated as either prominent or normal, while the Eustachian valve was judged as prominent, moderate, or unremarkable. The presence of a septal recess was recorded, and it was assessed as large (when the recess extended to reach the mid-CVTI), regular, or absent. Finally, encroachment of pectinate muscles into the CVTI was determined and reported as abundant, of moderate extent, or absent (Figs. 2 and 3).
Reproducibility of echocardiographic analysis
The datasets of 10 patients were randomly selected and analyzed again 3 months later by the original observer and by a second observer who was blinded to the results of the previous analysis. Intraobserver and interobserver agreement was assessed for the quantitative measurements of CVTI length and for the qualitative evaluations of the Eustachian ridge, Eustachian valve, septal recess or pouch, and pectinate muscles.
Continuous variables are expressed as mean ± SD or as median (interquartile range [IQR]) if appropriate. Normally distributed data were compared using independent Student t tests. Otherwise, comparisons between groups were performed using Mann-Whitney U tests. Categorical values are stated as absolute and relative frequencies and were compared using Fisher exact tests. All tests were 2 tailed.
Intraobserver and interobserver agreement in the measurements of CVTI length was assessed using Bland-Altman analysis and expressed as the mean difference between the 2 measurements ± 2 SDs. For qualitative evaluation of anatomical features such as the Eustachian valve, kappa statistics were used, and the level of agreement was interpreted as follows: 0 to 0.2 = poor to slight, 0.21 to 0.4 = fair, 0.41 to 0.6 = moderate, 0.61 to 0.8 = substantial, and 0.81 to 1.0 = nearly perfect. A linear regression model for either fluoroscopy time or dose was fitted, including ablation time (log-transformed), group, and their interaction. Stata version 11 (StataCorp LP, College Station, Texas) was used for computation. A p value <0.05 was considered statistically significant.
Comparison of the baseline characteristics between the 2 groups is presented in Table 1. Patients in the control group were significantly older than those in the RT3DTEE group and presented a trend toward slightly lower left ventricular ejection fractions. The prevalence of atrial fibrillation, however, was higher in the RT3DTEE group.
Reproducibility of echocardiographic analysis
Intraobserver and interobserver agreement for the measurement of CVTI length was good; according to Bland-Altman analysis, the mean differences ± 2 SDs for CVTI length were 1.6 ± 4.0 mm and 1.8 ± 7.0 mm, respectively.
Intraobserver and interobserver agreement for the qualitative evaluation of the Eustachian ridge (kappa = 1.0 for both), Eustachian valve (kappa = 1.0 for both), septal recess or pouch (kappa = 0.80 for both), and pectinate muscles (kappa = 1.0 and 0.80, respectively) was nearly perfect.
Feasibility of RT3DTEE to visualize the CVTI
In all but 1 patient (5%), intraprocedural visualization of the CVTI was possible. The reason for failure was a prominent aortic bulb with a marked protrusion into a small-sized right atrium. In the remaining 20 patients, CVTI imaging was always possible, and visualization was considered of good quality in most cases (75%). The median length of the CVTI was 40 mm; in 11 patients (55%), CVTI length was higher than the median value, and it was therefore labeled “long CVTI.” With the exception of 2 patients (10%) in whom the CVTIs were found to be morphologically normal, the other 18 patients (90%) presented at least 1 prominent anatomical feature (Table 2).
Feasibility of RT3DTEE to visualize the ablation catheter
RT3DTEE allowed visualization of the mapping catheter when moved from the septal to the lateral position and vice versa (Fig. 4A). Typical drag movement (anterior to posterior movement along the CVTI) could be equally appreciated; however, limited visibility of the catheter tip was encountered when the tip reached the proximal portion of the CVTI, before entering the inferior vena cava. Visualization of catheter tip contact was possible with RT3DTEE in many patients (50%) but was limited by 2 key factors, namely, different acoustic resonance yielded by different catheter manufacturers (Fig. 4A) and acoustic shadowing of the catheter due to the artifact beam (Fig. 4B). These difficulties could be corrected by gentle rotation of the ultrasound probe and adjustment of acoustic gain and cropping or by selecting 1 particular catheter manufacturer (Online Video 1).
Efficacy and safety of the RT3DTEE-guided approach for ablation
Achievement of bidirectional CVTI block was possible in all patients regardless of the monitoring approach (fluoroscopy or RT3DTEE). Patients monitored by RT3DTEE had significantly shorter procedural times and were exposed to significantly less radiation. Also, a smaller median number of RF applications was needed to achieve bidirectional block compared to the control group (Table 3). The association between radiation exposure and ablation time (Fig. 5) was not different when comparing the control and RT3DTEE groups (nonsignificant interaction). On bivariate analysis, both ablation time and group were significantly (both p values <0.001) and independently associated with fluoroscopy time and dose, with markedly lower values for the RT3DTEE group.
As experience with RT3DTEE increased, procedural and fluoroscopy times further reduced in the last 10 cases compared with the first 10 cases, thus indicating a strong learning curve effect (Table 2, Fig. 6). Of note, significant reduction of RFA time was observed in the last RT3DTEE subgroup (median 7.0 min, IQR 6.4 to 8.4 min) compared with the control group (median 11.7 min, IQR 8.0 to 20.0 min) (p = 0.036). In 8 of the 9 patients (89%) with complex CVTI anatomy (multiple prominent features), CVTI conduction block was obtained by ablating along the inferolateral isthmus line.
No major or minor complications ensued from the ablation procedure in either group.
To the best of our knowledge, ours is the first study to systematically evaluate the use of RT3DTEE to guide a transcatheter ablation procedure, and it may therefore be considered as a proof of concept for the use of RT3DTEE in guiding RFA of atrial arrhythmias. RT3DTEE reliably visualized the CVTI in nearly all patients, providing enough temporal and spatial resolution for tracking ablation catheter movements, thus enabling successful treatment of CVTI-dependent atrial flutter ablation in all patients. In comparison with the conventional fluoroscopy-guided approach, RT3DTEE was equally effective in achieving conduction block across the CVTI, with a smaller number of RF applications, and entailed a significant reduction of radiation exposure and procedural time.
Anatomical characterization of the CVTI in patients with typical common atrial flutter
Anatomic data on the CVTI from autopsy (13) and from cardiac computed tomographic (14) evaluations in general patient cohorts without specific atrial arrhythmic conditions have described a high anatomical heterogeneity of this structure. More specifically, data on the characterization of CVTI anatomy in patients with common atrial flutter are derived mostly from angiography of the right atrium (15,16). The major advantage of RT3DTEE is the ability to precisely evaluate anatomical and morphological characteristics in vivo and in real time. In the patient cohort of this study with atrial flutter, RT3DTEE confirmed that the CVTI is a complex and heterogenous structure. The incidence of prominent Eustachian ridge and valve (60%) was greater than in the general population (20% ) but in line with other studies performed in patients with common atrial flutter (15). Moreover, in this study, prominent Eustachian ridge and valve, a medial recess, and the presence of pectinate muscles sometimes coexisted (20%). This finding emphasizes that structural and morphological heterogeneity may represent an important arrhythmogenic substrate for triggering and maintaining macroreentry.
Improving navigation along the CVTI while reducing procedural time
Constant efforts have been oriented toward rendering procedural time for CVTI ablation more predictable (3–7,15,17–22) (Table 4). A number of studies have explored the importance of adequate structural characterization through either angiography (15,16), magnetic resonance imaging (7,23), or the use of nonfluoroscopic mapping techniques (5,6,21). Assessment of remote magnetic navigation has also been investigated, yielding only modest results, because this technique, while reducing radiation exposure, increases RF application time (3,4).
RT3DTEE allowed the simultaneous and immediate characterization of the CVTI, and in most cases with complex “septal” anatomy (8 of 20 patients), an inferolateral line was performed to obtain CVTI block (Fig. 7,Online Video 2) (14,16,24,25). A prominent Eustachian valve was almost always associated with a prominent Eustachian ridge. Other important obstacles were considered to be the length of the isthmus (7,15,22) as well as the CVTI–to–inferior vena cava angle (7). With regard to the latter feature, our preliminary experience with nonquantitative RT3DTEE did not assess the CVTI–to–inferior vena cava angle in real time.
As observed, the learning curve associated with the novel approach played a role. This was shown by reduced dispersion of procedural times, radiation exposure, and ablation times, defined as a narrowing of the IQR around the median values in the latter cases (Fig. 6). This suggests that RT3DTEE may render procedural time highly predictable and RF delivery more precise, because ablation of the CVTI is performed by circumventing anatomic obstacles.
Reduction of biological risk from radiation exposure
The excitement generated by cardiovascular imaging has recently been tempered by a string of high-impact publications (26–30) that have raised concerns about increased medical procedure radiation exposure and the potential risk for cancer. Most atrial arrhythmias are not life threatening, and the procedural purpose is mostly improvement in quality of life. In this view, the potential curative intent should be weighed against procedural success rate, risk for complications, and the long-term risk for a biological effect. Thus, following the principle of “as low as reasonably achievable,” which has guided the U.S. Food and Drug Administration's approval process and developments in the field of cardiac imaging, our data show that RT3DTEE may well serve this goal (31). RT3DTEE resulted in at least a 3-fold reduction in radiation exposure compared with the conventional fluoroscopy-assisted approach.
The reduction in fluoroscopy time observed during RT3DTEE was greater than that obtained with either the remote magnetic (3,4) or the electroanatomically guided (21) approach. In the latter case, mean fluoroscopy time was approximately 8 min, and as with RT3DTEE, the use of fluoroscopy was limited to the diagnostic part of the procedure, whereas ablation was guided mainly by navigation without fluoroscopy.
Issues and challenges in the use of RT3DTEE for monitoring catheter ablation procedures
The findings of this study substantiate the expanding role of RT3DTEE from a diagnostic modality to a tool for monitoring invasive procedures. Compared with other interventional cardiologic procedures (percutaneous valve repair or replacement, closure of atrial and ventricular defects, left atrial appendage closure), standard electrophysiological and ablation procedures possess many particularities. The relatively rapid catheter movement, the relatively narrow visualization field at an acceptable high frame rate (25 Hz), and metallic artifacts due to the large catheter tip may all represent technical challenges for routine adoption of the technique. In addition, the finding that different ablation catheter manufacturers present different acoustic impedance suggests that some catheters are more ultrasound compatible than others; this issue merits further investigation. Despite these challenges and the burden involved in using RT3DTEE to monitor an electrophysiological procedure, there are certain situations, such as pregnancy or in younger patients, in which a strong incentive to reduce harmful radiation exposure while minimizing procedural risks exists. To this end, the finding that RT3DTEE may reliably and effectively monitor an ablation procedure with only a short associated learning curve may be interesting and helpful. Finally, RT3DTEE may provide a unique way to further define key anatomical structures responsible for a macroreentry in the CVTI region.
This study was a pilot, single-center, nonrandomized study with a relatively small patient population. The RT3DTEE approach was represented by the need for general anesthesia, which lengthened overall procedural duration and could potentially increase risk; such a limitation may be overcome in the future when small-sized transesophageal echocardiographic probes become available (32). Also, general anesthesia and the use of mild sedation and pain control medication in the conventional approach to limit patient pain and movement may each have effects on procedural outcome data. For this reason, in the computation of the number of RF lesions and time of ablation, only RF applications lasting ≥60 s were considered.
RT3DTEE assessment of CVTI anatomy was mostly qualitative. Clearly, a validation process with other imaging modalities is needed. Despite some modest baseline differences between the 2 study groups, it is unlikely that these may have influenced acute procedural outcomes. More studies are needed to establish correlates between CVTI anatomical and morphological features, clinical variables, and their influence on CTVI ablation.
This study has demonstrated that RT3DTEE-guided ablation of the CVTI is feasible, yielding comparable acute efficacy to the standard fluoroscopy-guided approach. This imaging modality conferred high-resolution images, allowing navigation without fluoroscopy and precise RF delivery in target anatomical areas of the CVTI, thus allowing considerable reductions of procedural time, radiation exposure, and the number of RFA applications. Further investigations are needed to explore this new promising anatomically driven, fluoroscopy-free navigation approach for guiding the ablative treatment of cardiac arrhythmias.
For supplementary videos and their legends, please see the online version of this article.
Dr. Faletra has received speaking fees from Philips Medical Systems. Dr. Auricchio is a consultant for Sorin Group, Medtronic, Biotronik, EBR System, Merck, Biosense Webster, BDS Cordis, Philips, Impulse Dynamics, and has received speaker fees from Sorin Group, Medtronic, Biotronik, St. Jude Medical, and Abbott. All other authors have reported that they have no relationships to disclose.
- Abbreviations and acronyms
- cavotricuspid isthmus
- interquartile range
- radiofrequency ablation
- real-time 3-dimensional transesophageal echocardiography
- Received October 10, 2010.
- Revision received March 7, 2011.
- Accepted March 17, 2011.
- American College of Cardiology Foundation
- Blomström-Lundqvist C.,
- Scheinman M.M.,
- Aliot E.M.,
- et al.
- Pérez F.J.,
- Schubert C.M.,
- Parvez B.,
- Pathak V.,
- Ellenbogen K.A.,
- Wood M.A.
- Vollmann D.,
- Lüthje L.,
- Seegers J.,
- Hasenfuss G.,
- Zabel M.
- Kottkamp H.,
- Hügl B.,
- Krauss B.,
- et al.
- Kirchhof P.,
- Ozgün M.,
- Zellerhoff S.,
- et al.
- Efstathopoulos E.P.,
- Katritsis D.G.,
- Kottou S.,
- et al.
- Faletra F.,
- Ho S.W.,
- Auricchio A.
- Tada H.,
- Oral H.,
- Sticherling C.,
- et al.
- Cauchemez B.,
- Haissaguerre M.,
- Fischer B.,
- Thomas O.,
- Clementy J.,
- Coumel P.
- Heidbüchel H.,
- Willems R.,
- van Rensburg H.,
- Adams J.,
- Ector H.,
- Van de Werf F.
- Da Costa A.,
- Faure E.,
- Thévenin J.,
- et al.
- Hindricks G.,
- Willems S.,
- Kautzner J.,
- et al.,
- EuroFlutter Investigators
- Da Costa A.,
- Romeyer-Bouchard C.,
- Dauphinot V.,
- et al.
- Hoffmann B.A.,
- Koops A.,
- Rostock T.,
- et al.
- Scaglione M.,
- Caponi D.,
- Di Donna P.,
- et al.
- Faletra F.F.,
- D'Angeli I.,
- Klersy C.,
- et al.
- Laskey W.K.,
- Feinendegen L.E.,
- Neumann R.D.,
- Dilsizian V.
- Gerber T.C.,
- Gibbons R.J.
- Halliburton S.S.,
- Schoenhagen P.
- Shaw L.J.,
- Achenbach S.,
- Chandrashekhar Y.,
- et al.
- U.S. Food and Drug Administration, Center for Devices and Radiological Health
- Stec S.,
- Zaborska B.,
- Sikora-Frac M.,
- Krynski T.,
- Kulakowski P.