This study aimed to investigate the anatomical characteristics complicating cavotricuspid isthmus (CTI) ablation and the effectiveness of various procedural strategies.
This study aimed to investigate the anatomical characteristics complicating cavotricuspid isthmus (CTI) ablation and the effectiveness of various procedural strategies.
This study included 446 consecutive patients (362 males; mean age 60.5 ± 10.4 years) in whom CTI ablation was performed. A total of 80 consecutive patients were evaluated in a preliminary study. The anatomy of the CTI was evaluated by multidetector row-computed tomography (MDCT) prior to the procedure. A multivariate logistic regression analysis revealed that the angle and mean wall thickness of the CTI, a concave CTI morphology, and a prominent Eustachian ridge, were associated with a difficult CTI ablation (P < 0.01). In the main study, 366 consecutive patients were divided into 2 groups: a modulation group (catheter inversion technique for a concave aspect, prominent Eustachian ridge, and steep angle of the CTI or increased output for a thicker CTI) and nonmodulation group (conventional strategy). The duration and total amount of radiofrequency energy delivered were significantly shorter and smaller in the modulation group than those in the nonmodulation group (162.2 ± 153.5 vs 222.7 ± 191.9 seconds, P < 0.01, and 16,962.4 ± 11,545.6 vs 24,908.5 ± 22,804.2 J, P < 0.01, respectively). The recurrence rate of type 1 atrial flutter after the CTI ablation in the nonmodulation group was significantly higher than that in the modulation group (6.3 vs 1.7%, P = 0.02).
Changing the procedural strategies by adaptating them to the anatomical characteristics improved the outcomes of the CTI ablation.
The cavotricuspid isthmus (CTI) is defined as the area between the tricuspid valve (TV) and inferior vena cava (IVC), and is contiguous in anatomy to the triangle of Koch. The CTI is a critical component of the reentry circuit for CTI-dependent atrial flutter (AFL).[1-5] Radiofrequency (RF) catheter ablation targeting the CTI is the optimal treatment for CTI-dependent AFL.[6, 7] Despite high overall success rates, the ablation procedure is occasionally difficult due to variations in the anatomical characteristics of the CTI.[8, 9] Multidetector row-computed tomography (MDCT) has become a widely used substitute in cardiac angiography studies.
In a preliminary study, the CTI anatomy was evaluated using 64-row MDCT, and the anatomical characteristics of the CTI that complicate the ablation procedure were identified. Subsequently, adaptations to the CTI ablation strategies according to the anatomical information provided by the preprocedural MDCT were prospectively examined in a main study. The goal was to improve and minimize the difficulty of the ablation procedure.
The institutional review board approved the study, and written informed consent was obtained from all patients. Patients with a serum creatinine level of 1.2 mg/dL or more were excluded from the study. Patients in which adequate anatomical information necessary to perform the measurements was not obtained during the MDCT scan were excluded. Fortunately, the scans in all subjects provided the necessary anatomical information and it was not necessary to exclude any subjects due to an uninterpretable scan quality.
In total, 80 patients (63 males [79.7%], aged 59.8 ± 10.4 years) in whom CTI ablation was successfully performed using an 8 mm tip ablation catheter were included from the preliminary study. MDCT was performed within 24 hours before the ablation procedure in all patients. Typical AFL was clinically documented in 23 patients (AFL only: n = 7; both atrial fibrillation (AF) and AFL: n = 16). The remaining 57 patients with persistent or paroxysmal AF underwent a combined AF and CTI ablation with no evidence of AFL. Of the 73 patients with AF in whom circumferential pulmonary vein isolation (CPVI) was performed, paroxysmal AF was evident in 63 patients and persistent AF in 10.
Three hundred and sixteen consecutive patients (males: n = 255, 80.7%, aged 60.5 ± 10.2 years) scheduled for a CTI ablation were prospectively enrolled in the main study. Typical AFL was clinically documented in 22 patients. In the remaining 294 patients (including 56 patients with AFL) with persistent or paroxysmal AF, a combined AF and CTI ablation was performed with no evidence of AFL.
MDCT data sets were acquired using a 64-slice CT scanner (Lightspeed VCT; GE Healthcare, Waukesha, WI, USA) with retrospective ECG-gated scans using a dual-shot-type injector (Nemoto-Kyorindo, Tokyo). To satisfy these conflicting issues, we used a multiphasic contrast material injection protocol. In the routine protocol, the first phase of the multistepwise protocol, 144 mgI/kg were administered during 7 seconds. In the second phase, the initial contrast concentration was 370 mgI/mL; it was gradually decreased by dilution with saline during 15 seconds. The total iodine dose in the second phase was 155 mg/kg. The scan protocol methods used in this study have been described previously. Image analysis software (Virtual Place Advance; AZE, Tokyo, Japan) allowed 3-dimensional (3D) viewing of multiplanar reconstruction images reformatted as cross-sectional images. The optimal image was selected during the end-diastole phase of the right atrium, which was defined as the image immediately before the opening of the tricuspid valve. The images were also reconstructed using electrocardiography (ECG) edited at the level of the anomalies of the ECG signal, which were caused by premature beats, AF, and mis-triggering. Multiplanar reconstructions of the axial images were obtained by operating a cursor manually on selected cross-sectional source images orthogonal to the source images during the latest right atrial diastolic phase of the cardiac cycle [30–40% of the interbeat (R–R) interval].
The quantative measurements and morphologies of the CTI were analyzed in all patients. The values for the length (Fig. 1A) and angle (Fig. 1B) were obtained in long-axis 2-chamber right anterior oblique views parallel to the septum. The length of the CTI was defined as the segment beginning from the base of the end of the Eustachian valve to the base of the tricuspid valve. The angle was constructed by the intersection of the line passing through the center of the IVC and the line extending to the orifice of the IVC from the tricuspid annulus. The presence of a prominent Eustachian ridge was confirmed in the same view. The myocardial thickness of the CTI was measured at the center of the 3 parallel isthmic levels (paraseptal, central, and inferolateral) in the short-axis left anterior oblique views (Fig. 1C,D). The mean myocardial thickness of the CTI was calculated from the values of those 3 sites. The angle of the CTI was classified into 2 types (steep: <90° and gentle: ≥90°). The CTI morphology was classified into 3 types; straight (≤3 mm), concave (>3 to ≤5 mm), and pouchlike (>5 mm), according to the depth of the CTI. Also, the analysis software of this study could measure only to the first decimal place. Two blinded cardiologists experienced in reading MDCT images independently analyzed all the data sets. Quantitative measurements were performed in triplicate.
Continuous monitoring was performed using endocardial electrocardiography (bandpass filtered 30–500 Hz) and surface electrocardiography. Data were stored on a digital amplifier and recorder system (Bard Electrophysiology, Lowell, MA, USA). A 6F multipolar (15-pole) catheter (Irvine Biomedical, Irvine, CA, USA) was utilized with the distal poles (poles 1−10) placed within the coronary sinus (CS) and the proximal electrodes (poles 11−15) located from the superior vena cava to the superior right atrium via the right internal jugular vein. A decapolar electrode catheter was positioned on the lateral wall of the right atrium (Irvine Biomedical) inserted from the right femoral vein. A decapolar electrode catheter (Biosense Webster, Diamond Bar, CA, USA) was introduced via the femoral vein into the region of the bundle of His.
After intravenous administration of heparin (bolus injection of 3000 IU followed by an hourly injection of 1000 IU), a 7F quadripolar deflectable catheter with an 8 mm tip electrode (Navister; Biosense Webster) was positioned on the isthmus using a preshaped long sheath with a 180° curve (RAMP curve, St. Jude Medical, Daig Division, Inc., Minnetonka, MN, USA). In the beginning, the ablation catheter was positioned at the ventricular aspect of the CTI at the 6 o'clock position in the LAO 45° projection. Ablation was carried out with the temperature set at 55 °C and energy output at 40−50 W using a dragging catheter technique under anatomic and electrographic guidance.[4, 12, 13] RF was applied with voltage outputs modulated by impedance monitoring to avoid an increase due to excessive heating. Ablation of the CTI was performed during either sinus rhythm (pacing from the CS) or ongoing typical AFL with the procedural endpoint of complete bidirectional isthmus block. Reversal of the right atrial depolarization sequence was established by a complete CTI map using a decapolar mapping catheter straddling the line of block and by recording widely separated (>120 milliseconds) local double potentials along the ablation line during atrial pacing. Differential pacing was performed to distinguish complete from incomplete bidirectional isthmus block. When complete conduction block of the CTI was not achieved after first-line ablation, the conduction gaps were mapped and ablated searching for single or narrow-split potentials with the ablation catheter moving along the original ablation line. If conduction block was not still achieved, the catheter was repositioned on the ventricular side of the CTI at the 7 or 5 o'clock position. If bidirectional block status was maintained over the 30-minute period following the achievement of the bidirectional block, the procedure was considered complete.
In this study, the procedure time of the CTI ablation was defined as the duration from the beginning of the first RF application to confirmation of the bidirectional block of the CTI, excluding the observation time. The CTI was divided into a proximal part (from the coronary sinus ostium to the Eustachian valve) and a distal part (from the tricuspid annulus to the coronary sinus ostium) according to a catheter placed within the CS. The region in which a complete bidirectional conduction block was successfully achieved was recorded. For patients in whom a combined AF and CTI ablation was performed, CPVI was performed at the beginning of the procedure. The patients were mildly sedated without intubation during the procedure using short acting intravenous drugs.
The patients were prospectively randomized and assigned to either a modulation or nonmodulation group. In the modulation group (n = 183), the ablation strategies were modulated according to a flow chart that was based on the results of the preliminary study (Fig. 2). In brief, if the thickness of the CTI was ≥2.7 mm, the CTI ablation was initiated with a high output (60 W) and high-temperature (60 ºC) setting from the beginning. If the morphology of the CTI was concave, angle of the CTI was ≤86.4°, or a prominent Eustachian ridge was found, the catheter was inverted until the outer curve was in contact with the isthmus at the proximal part of the CTI (Fig. 3). Thus, the catheter came in contact with the outer surface of the CTI in these cases. In the absence of any of these anatomical features, the CTI ablation was performed by the conventional method. In the nonmodulation group (n = 183), the ablation lines were created conventionally, similar to the method in the preliminary study. Verification of the bidirectional block in the CTI was also performed as in the preliminary study.
The patients were examined in our outpatient clinic 2 weeks after performance of the ablation procedure. Follow-up continued every 1−2 months thereafter. Patients were examined using 12-lead ECG and Holter monitoring. Interviews were conducted regarding clinical symptoms and residual palpitations.
The data were tested for normal (Gaussian) distribution using the Kolmogorov−Smirnov test. Normally distributed continuous variables are presented as means ± SD. In the case of a non-Gaussian distribution, median values and quartiles are given. Comparisons between groups were performed using Student's t-test. In cases of continuous data with a non-Gaussian distribution, the Mann−Whitney U test was used. Categorical variables were compared using the Fisher's exact test. A univariate analysis of the patient characteristics was conducted for a comparison between the difficult and straightforward cases, and a logistic regression analysis was performed to detect any independent significant predictors by adjusting for multiple variables. The multiple pairwise comparisons among the patient groups were tested using a Tukey–Kramer test for continuous variables. Statistical significance was defined as a P < 0.05 using a 2-sided comparison. Statistical analyses were performed using JMP 9 software (SAS Software, Cary, NC, USA).
The baseline characteristics of the 80 subjects and ablation results of the preliminary study are summarized in Table 1. For the myocardial thickness of the CTI, there was no significant difference between the 3 parallel isthmic levels (paraseptal: 2.61 ± 0.46 vs central: 2.41 ± 0.51 vs inferolateral: 2.38 ± 0.43) by Tukey–Kramer multiple comparison tests. Bidirectional conduction block in the CTI was achieved in all patients. The subjects in the preliminary study were divided into 2 groups according to the median procedure time: the difficult cases (n = 40; procedure time ≥16 minutes) and straightforward cases (n = 40; procedure time <16 minutes).
|Parameters||Results (n = 80)|
|Gender male, n (%)||63 (79.7)|
|Age, year||59.8 ± 10.4|
|BMI, kg/m2||23.9 ± 2.9|
|Left ventricular ejection fraction, %||64.0 ± 9.9|
|Hypertrophic cardiomyopathy, n (%)||2 (2.5)|
|Dilated cardiomyopathy, n (%)||1 (1.2)|
|Coronary artery disease, n (%)||4 (5.0)|
|Valvular heart disease, n (%)||4 (5.0)|
|Anatomical data collected from MDCT|
|Length of the CTI, mm||29.9 ± 7.6|
|Angle of the CTI, degree||97.6 ± 26.7|
|Angle type of the CTI|
|Steep type, n (%)||16 (20.0)|
|Gentle type, n (%)||64 (80.0)|
|Mean thickness of the CTI, mm||2.5 ± 0.6|
|Prominent Eustachian ridge, n (%)||11 (13.8)|
|Morphology of the CTI|
|Straight aspect, n (%)||46 (57.5)|
|Concave aspect, n (%)||20 (25.0)|
|Pouch-like aspect, n (%)||14 (17.5)|
|Ablation during AFL, n (%)||3 (3.8)|
|Procedure time, minutes||21.8 ± 20.9|
|Total RF application duration, seconds||346.4 ± 321.4|
|Total RF energy, J||26,482.5 ± 21,403.4|
|Fluoroscpic time, minutes||13.1 ± 11.8|
A multivariate logistic regression analysis revealed that the mean thickness of the CTI, steep angle of the CTI, concave aspect of the CTI morphology, and prominent Eustachian ridge were characteristics significantly associated with the difficulty of the CTI ablation (Table 2).
|Variables||Odds Ratio (95% Confidence Interval)||P Value||Odds Ratio (95% Confidence Interval)||P Value|
|Age, per year||1.00 (0.97–1.03)||0.96||0.99 (0.94–1.04)||0.75|
|Male sex||1.69 (0.71–4.14)||0.24||1.30 (0.35–4.73)||0.69|
|BMI, per kg/m2||1.01 (0.90–1.16)||0.75||1.04 (0.87–1.27)||0.62|
|Length of the CTI, per mm||1.04 (0.99–1.09)||0.15||1.05 (0.97–1.11)||0.31|
|Mean thickness of the CTI, per mm||6.35 (2.67–16.68)||<0.01||7.10 (2.39–24.45)||<0.01|
|Steep angle of the CTI||8.5 (2.64–38.11)||<0.01||11.47 (2.65–66.21)||<0.01|
|Straight aspect||0.42 (0.20–0.86)||0.02||2.52 (0.24–33.56)||0.45|
|Concave aspect||5.79 (2.49–14.57)||<0.01||24.56 (2.21–419.32)||<0.01|
|Pouch-like aspect||0.57 (0.21–1.46)||0.24||2.59 (0.20–36.34)||0.47|
|Prominent Eustachian ridge||18.31 (3.45–338.75)||<0.01||37.42 (5.45–790.63)||<0.01|
The area under the curve was 0.73 (P < 0.01) for the angle of the CTI and 0.95 (P < 0.01) for the thickness of the CTI. The optimal cutoff values according to this analysis were 86.4° for the angle of the CTI (sensitivity: 55%, specificity: 90%) and 2.67 mm for the thickness of the CTI (sensitivity: 92.5%, specificity: 87.5%). These values were useful in predicting the procedural difficulty of the CTI ablation.
The procedure time of the CTI ablation was significantly longer in the patients whose critical point for a successful ablation was the proximal part than that in the patients whose critical point of a successful ablation was the distal part (P < 0.05). Especially in terms of the variations in the anatomical characteristics of the CTI including a concave aspect (38.7 ± 35.0 vs 17.5 ± 15.3 minutes, P = 0.02), prominent Eustachian ridge (27.4 ± 19.7 vs 9.1 ± 7.7 minutes, P = 0.04), and steep angle of the CTI (57.5 ± 32.1 vs 30.7 ± 14.2 minutes, P = 0.02), achievement of bidirectional isthmus block in the proximal part was significantly more difficult than that of the distal part.
In this prospective, randomized study, no significant difference was evident in the baseline and anatomical characteristics between the nonmodulation and modulation groups (Table 3). In the results of the CTI ablation in both groups (Table 3), the total CTI ablation procedure time was significantly shorter in the modulation group than that in the nonmodulation group. Further, the total RF duration was significantly shorter and total amount of RF energy required to establish bidirectional conduction block at the CTI was significantly smaller in the modulation group than that in the nonmodulation group. In addition, the fluoroscopic time was significantly shorter in the modulation group than that in the nonmodulation group. The results of the comparison of the procedure time between the modulation and nonmodulation groups according to the anatomical characteristics complicating the CTI ablation are shown in Fig. 4. Tukey–Kramer tests revealed that the procedure time in the patients with a concave aspect, thickness of the CTI ≥2.7 mm, angle of the CTI ≤86.4° or prominent Eustachian ridge was significantly longer than that in those without in the nonmodulation group (30.4 ± 23.6, 31.6 ± 23.6, 41.2 ± 26.5, 52.6 ± 37.3 vs 9.1 ± 6.8 minutes, P < 0.05). However, in the modulation group, there was no significant difference in the procedure time between the patients with anatomical difficulty and those without (13.6 ± 8.5, 12.8 ± 9.2, 13.2 ± 9.4, 12.4 ± 8.6 vs 9.3 ± 8.2 minutes).
|Variables||Modulation Group (n = 183)||Nonmodulation Group (n = 183)||P Value|
|Gender male, n (%)||152 (83.1)||147 (80.3)||0.50|
|Age, years*||60.9 ± 10.4||60.3 ± 10.3||0.61|
|BMI, kg/m2*||24.2 ± 3.0||24.2 ± 3.2||0.81|
|Left ventricular ejection fraction, %*||61.9 ± 6.1||61.5 ± 5.8||0.51|
|Hypertrophic cardiomyopathy, n (%)||2 (1.1)||1 (0.6)||0.56|
|Dilated cardiomyopathy, n (%)||1 (0.6)||1 (0.6)||1.00|
|Coronary artery disease, n (%)||3 (1.6)||4 (2.2)||0.70|
|Valvular heart disease, n (%)||8 (4.4)||7 (3.8)||0.79|
|Anatomical data collected from MDCT|
|Length of the CTI, mm*||31.9 ± 8.0||32.3 ± 7.3||0.93|
|Angle of the CTI, degree*||104.1 ± 22.5||101.3 ± 21.4||0.22|
|Angle type of the CTI|
|Steep type, n (%)||34 (18.6)||32 (17.5)||0.79|
|Gentle type, n (%)||149 (81.4)||151 (18.5)||0.79|
|Mean thickness of the CTI, mm*||2.6 ± 0.3||2.6 ± 0.5||0.27|
|Morphology of the CTI|
|Straight aspect, n (%)||105 (57.4)||111 (60.7)||0.52|
|Concave aspect, n (%)||48 (26.2)||50 (27.3)||0.81|
|Pouch-like aspect, n (%)||30 (16.4)||22 (12.0)||0.23|
|Prominent Eustachian ridge, n (%)||15 (8.2)||20 (10.9)||0.37|
|Thickness of the CTI ≥2.7 mm, n (%)||58 (31.7)||62 (33.9)||0.66|
|Angle of the CTI ≤86.4, n (%)||30 (16.4)||28 (15.3)||0.77|
|Ablation during AFL, n (%)||8 (4.4)||10 (5.5)||0.63|
|Procedure time, minutes*||10.0 ± 8.7||18.1 ± 17.8||<0.01|
|Total RF application duration, seconds*||166.2 ± 153.5||222.7 ± 191.9||<0.01|
|Total RF energy, J*||16,962.4 ± 11,545.6||24,908.5 ± 22,804.2||<0.01|
|Fluoroscpic time, minutes*||6.8 ± 5.7||14.2 ± 12.0||<0.01|
|Success rate n (%)†||178 (98.9)||176 (96.2)||0.09|
|Recurrence (%)||3 (1.7)||11 (6.3)||0.02|
During the 28.8 ± 12.5 months follow-up periods, the recurrence rate of type 1 AFL after the CTI ablation in the nonmodulation group was significantly higher than that in the modulation group (6.3 vs 1.7%, P = 0.02).
In this study, we first identified the anatomical characteristics of the CTI that caused difficulty in the ablation procedure by analyzing the preprocedural MDCT images. Those were the angle and mean wall thickness of the CTI, a concave CTI morphology, and a prominent Eustachian ridge. We continuously demonstrated that changing the ablation strategy based on the anatomical information of the CTI by the analysis MDCT data improved the CTI ablation procedure, especially in the difficult cases.
Knecht et al. reported the thickness of the CTI as an anatomic parameter strongly associated with procedural difficulty in the achievement of transmural lesions  similar to the results of our study. Heidbüchel et al. noted that the presence of a Eustachian ridge (valve) and a concave aspect was associated with a significantly higher number of RF applications to achieve conduction block of the CTI. Catheter inversion may improve the catheter contact in these cases. Sporton et al. and Wieczorek et al. reported that untreated single electrograms could be identified predominantly in the proximal portions of the CTI following the inversion of the ablation catheter.[17, 18] Scavée et al. demonstrated higher efficacy of an externally irrigated tip ablation catheter for rapid achievement of CTI block compared with an 8 mm tip catheter. In contrast, the 8 mm tip catheter was demonstrated to be more effective for ablation in the case of a straight isthmus morphology where a constant high-blood flow is obtained compared with irrigated catheters. Further, the stratified CTI morphology (“straight,” “concave,” and “pouch-like”) included mostly the straight type in this study. Therefore, we used an 8 mm tip catheter throughout the study. The usefulness of a voltage-guided technique to target the high-voltage electrograms along the CTI to ablate the functionally important anatomical muscle bundles was reported recently.[21, 22] The results of this study may be useful for overcoming difficult cases with this approach. When the catheter contact is poor, the catheter inversion technique will be necessary to map the target electrograms. Various imaging modalities (e.g., right atrial angiography, 3D mapping system, intracardiac echocardiography) have been utilized to characterize the CTI morphology and improve the AFL ablation procedure.[20, 23, 24] MDCT has the advantage of providing views of the CTI anatomy from every angle, and allows accurate measurements of the parameters of the CTI compared with the other imaging modalities.
This was a single-center study. Patients with extreme right atrial enlargement and hemodynamic factors such as severe tricuspid regurgitation or pulmonary hypertension were not included among the study subjects. Those factors may cause difficulty during ablation procedures. No mechanical valve replacements, pacemakers, or implantable cardioverter defibrillators were implanted in any of the patients included in our study because of the artifact interference during the analysis of the MDCT images. Only an 8 mm tip ablation catheter and shaped sheath were used for the ablation and we did not perform a comparison with other catheters, including contact sensing irrigated catheters, when investigating the effectiveness of the various procedural strategies. Use of different catheters and sheaths may therefore result in different outcomes for CTI ablation.
A high output, high-temperature setting, and strong contact using the catheter inversion technique (modulation strategies) was not attempted in every case in this study. Modulation strategies may make accomplishment of CTI block easy in any case, but increase the risk of cardiac injury especially in a thin right atrium near the IVC region.
An MDCT scan with a contrast injection provides an accurate method for preoperative evaluation of the highly variable anatomy of the CTI. Variable catheter ablation strategies based on the MDCT findings significantly improved the adaptation of the ablation catheter to the anatomical characteristics associated with the procedural difficulties with the CTI ablation compared with the conventional strategy. However, we have to consider that the additional exposure from the MDCT may be detrimental to the patients. Our point was not that all patients had better undergo preprocedural MDCT before the CTI ablation. The current results suggest that if we encounter challenging cases of a CTI ablation and cannot complete the CTI block by the usual conventional strategy, we can give thought to these anatomical difficulties and promptly shift to change the strategy. If no widespread double potentials or a decrease in the local electrogram is encountered in spite of sufficient electrode-tissue contact, we have to consider that the myocardial wall of the CTI wall is thick and should adjust the RF power output accordingly. If the RF power suddenly drops at a site, the ablation catheter may be positioned in the low blood flow area such as a pouch. In that case, we had better consider changing the location of the ablation line. In cases in whom complete bidirectional conduction block of the CTI cannot be achieved despite universal widespread double potentials along the CTI line, the electrode-tissue contact may be insufficient in the proximal part of the CTI. In those cases, the catheter inversion technique is useful and recommended.
The authors thank Masao Kiguchi, R.T., Department of Radiology, for his excellent technical assistance.