A total of 55 patients (39 males, mean age 57±8 years) with drug-refractory symptomatic paroxysmal AF were enrolled from 2015 to 2016. A total of 15 consecutive patients (MediGuide group) who received radiofrequency catheter ablation for the first time under the guidance of a MediGuide™ mapping system were studied. A control group of 40 age- and sex-matched patients (Conventional group) who received radiofrequency catheter ablation for the first time under the guidance of an EnSite Velocity™ mapping system were also investigated. Catheter ablation was performed in MediGuide group and Conventional group during the same period. All patients received circumferential PVs isolation and CTI ablation. Ethical approval was granted by the Institutional Review Board (IRB) of the Veterans General Hospital, Taipei, Taiwan (IRB No. 2017-03-001AC). All subjects gave written informed consent.

The LA and PVs were evaluated with an electrocardiographically gated, 64-slice multidetector computed tomography (CT) scanner (Aquilion 64 CFX; Toshiba Medical System, Tokyo, Japan) before catheter ablation. All patients underwent contrast-enhanced CT scanning during sinus rhythm. The mean duration from CT scan to ablation procedure was 1.8±1.5 days. There was no significant difference between the MediGuide group (2.1±1.2 days) and the Conventional group (1.8±1.6 days). The detailed protocol and analysis of cardiac CT imaging have been described before.7) In brief, the patients were instructed to hold their breath to acquire the images, which covered an area from the superior margin of the pulmonary hilum to the cardiac apex (collimation 64×0.5 mm, gantry rotation time 350 ms, table speed 6.3 mm/rotation, tube voltage 120 kV, and effective tube current 545 mA). The acquisition time was 8 to 12 seconds depending on the heart rate. The LA and PVs were segmented and reconstructed in 3D from CT slices using EnSite Verismo™ software (St. Jude Medical). We excluded the patients with PV anatomical variants (ex: common PV ostium or existence of middle PV) from this study. The CT reconstruction was not fused in Velocity™ system during ablation procedure in all patients.

The MediGuide technology has been previously described.³⁾⁴⁾ Briefly, this system is comprised of a transmitter unit generating a low-powered electromagnetic field and a miniaturized single-coil sensor (<1 mm³) contained in the distal end of a specialized catheter such as diagnostic catheter and ablation catheter. The transmitter unit is installed in the fluoroscopy detector of a conventional X-ray imaging system (Siemens Artis, Erlangen, Germany) and can provide real-time 3D electromagnetic catheter localization. At the beginning of the ablation procedure, 2 short fluoroscopy loops (3 seconds each) are acquired. The position and orientation of catheter tips are non-fluoroscopically visualized on the previously acquired fluoroscopic images. A reference sensor fixed on the patient's chest and a real-time electrocardiogram (ECG) signal allows compensation for cardiac and respiratory motions or patient movement. The mapping data of this system are linked to EnSite Velocity™, which can offer electromagnetic scaled-data to impedance scaled-data.

Before the electrophysiological study and radiofrequency catheter ablation, antiarrhythmic agents were ceased for a minimum of 5 half-lives (except for amiodarone), and transesophageal echocardiography was performed to exclude any LA thrombi and evaluate cardiac anatomy. A standardized electrophysiology study was performed for all patients in the fasting state without sedation and local anesthesia was given during the procedure for pain relief. The surface electrocardiogram and bipolar intracardiac electrogram were stored on a digital recording system (LabSystem PRO™; Bard Electrophysiology, Lowell, MA, USA). In MediGuide group, the catheters utilized for mapping were the following: 1) A sensor-equipped decapolar electrophysiology catheter (MediGuide-Enabled Livewire™; St. Jude Medical), 2) a multipolar circular mapping catheter (Inquiry™ Afocus™ II; St. Jude Medical), and 3) a 4-mm open-irrigated tip catheter (CoolPath™ DUO MediGuide-enabled; St. Jude Medical). The sensor-equipped decapolar catheter was advanced into the heart from the right jugular vein, tracked, and visualized on the pre-recorded cine loop (right anterior oblique 30 degree and left anterior oblique 60 degree). The catheters were non-fluoroscopically inserted into coronary sinus on the dynamic models. In the Conventional group, the catheters utilized for mapping were the following: 1) A conventional decapolar electrophysiology catheter (RESPONSE™; St. Jude Medical) inserted from the right jugular vein to the coronary sinus, 2) a multipolar circular mapping catheter (Inquiry™ Afocus™ II), and 3) a 4-mm open-irrigated tip catheter (FlexAbility™; St. Jude Medical).

Before a transseptal approach for LA access and isolation of PVs, right atrial angiography and esophagography were simultaneouly performed to delineate the location of both atria and esophagus. After that, a transseptal puncture by modified Brockenbrough technique (BRK™ Transseptal needle; St. Jude Medical) was performed using fluoroscopic landmarks, and an 8.5F long vascular sheath (St. Jude Medical) was introduced into the LA. The coronary sinus catheter was used as a positional reference for the 3D mapping system. In the MediGuide group, the sensor-equipped ablation catheter was carefully used to define PV branches and PV ostia and create the outline of LA body on the pre-recorded cine loop. These mapping data were integrated with the EnSite Velocity™ system. Additional mapping points were collected using the multipolar circular mapping catheter on the Ensite Velocity™ screen and the original geometry of LA was completed (Figure 1). In the Conventional group, the Ensite Velocity™ with the multipolar circular mapping catheter was used to create the outline of the LA body and the PVs under fluoroscopic guidance. Additional mapping points were collected using the ablation catheter and the original geometry of LA was completed (Figure 2). After the creation of the original LA geometry, a field scaling algorithm was applied to the original geometry. This field scaling adjusts for the non-linearity of the geometry and compensates for the distortion of the geometry.8) In the present study, 3D geometry of LA was created based on the same impedance field scaling algorithm in both groups.

In the MediGuide group, the electromagnetic field-scaled and impedance field-scaled geometries (M-EM-GEO and M-IM-GEO) were created as shown in Figure 1. In the Conventional group, the impedance field-scaled geometry (C-IM-GEO) was created as shown in Figure 2. To evaluate the accuracy of cardiac anatomy between magnetic-based geometry and impedance-based geometry in MediGuide technology, M-EM-GEO, and M-IM-GEO were compared. In addition, to validate the superiority of non-fluoroscopic magnetic-based mapping compared with the conventional impedance-based mapping, M-EM-GEO and C-IM-GEO were also compared.

The measurement of PV angles has been previously described.9) In brief, the longitudinal axis of the LA was defined as the line between the center points of the ipsilateral PVs. Right superior pulmonary vein (RSPV), right inferior pulmonary vein (RIPV), left superior pulmonary vein (LSPV), and left inferior pulmonary vein (LIPV) angles were measured between the main trunk of each PV and the longitudinal axis of the LA (Figure 1, ①–④). Additionally, we measured the posterior LA surface area, which was defined as the area encircled with the posterior PV-LA junction, roof line and floor line (Figure 1, ⑤). In the present study, anterior LA points were not used for the evaluation of accuracy because of the inherent mobility of this structure. We calculated the actual change {(EM-GEO or IM-GEO)−CT-based geometry} and the relative change (actual change)/(CT-based geometry) of 4 PV angles and the posterior LA surface area in each field scaled geometry, compared to those of CT-based geometry. Cardiac CT has been considered as a gold standard for assessment of the anatomy of the LA.

In the present study, to assess spatial accuracy of the electromagnetic and impedance field scaling, the anterior-posterior diameter of the LA was also measured as the distance between the central part of the posterior wall to the central part of anterior part in MediGuide group (Figure 1, ⑥).

After the creation of the 3D geometry of the LA, continuous circumferential lesions were created to encircle the right and left PV ostia. When the patients experienced intolerable pain during the ablation procedure, they received local anesthesia and/or mild sedation with midazolam and fentanyl for pain relief. In the MediGuide group, the sensor-equipped ablation was visualized on the pre-recorded cine loop. The mapping and ablation data were integrated with the EnSite Velocity™ System and non-fluoroscopic PV isolation was performed. In the Conventional group, PV isolation was performed using the EnSite Velocity™ system under fluoroscopic guidance. All procedures were performed by well-experienced operators (cumulative AF case number >400). The method of PV isolation has been previously described in our previous publications.10)11) Radiofrequency energy was continuously applied using an irrigated tip ablation catheter while repositioning the catheter tip every 40 seconds. The standard ablation setting consisted of a power of 25–30 W, and irrigation flow rate of 30 mL/min. When a radiofrequency energy delivery to the posterior wall near the esophagus was necessary, power delivery was reduced to 20–25 W. After completion of the circumferential lesion set, the ipsilateral superior and inferior PVs were mapped carefully by the multipolar circular mapping catheter. Additional ablation applications were applied along the circumferential lines close to the earliest ipsilateral PV potentials. Furthermore, ablation of the residual PV potentials was performed using the electrogram-guided approach to obtain entrance block. Successful circumferential PV isolation was demonstrated by the absence of any electrical activity inside the PV or dissociation of the PV potentials (entrance and/or exit blocks). In all procedures, fluoroscopy and procedure times (from transseptal puncture to the completion of ablation) were measured.

The standard ablation setting in CTI ablation consisted of a power of 40 W, and irrigation flow rate of 30 mL/min. In the MediGuide group, CTI ablation was performed under the pre-recorded cine loop guidance. In the Conventional group, CTI ablation was performed under fluoroscopy guidance. Bi-directional isthmus block was defined as the endpoint of the ablation procedure. In all procedures, fluoroscopic and procedure times were measured.

Patients were followed up in the cardiology outpatient clinic with 12-lead ECGs and 24-hour Holter monitoring (2 weeks after the catheter ablation, then every 1–3 months). Antiarrhythmic drugs were prescribed for 8 weeks to prevent the early recurrence of paroxysmal AF. For patients who could not come for outpatient follow-up in our institution, they were followed up by the referring physicians and also contacted over telephone for recurrent symptoms and recurrent arrhythmias. These patients were also advised to complete follow-up screening and the medical reports were obtained from these referring physicians. Arrhythmia recurrence after ablation followed Heart Rhythm Society/European Heart Rhythm Association/European Cardiac Arrhythmia Society guidelines.12) Recurrence was defined as any office ECG demonstrating AF or atrial tachycardia, or any episode of AF or atrial tachycardia >30 seconds on 24-hour Holter monitoring after a 90-day blanking period.

Continuous data were reported as mean values ± standard deviation. Categorical data were reported as absolute values and percentages. Numerical variables were compared by the Mann-Whitney U test. Categorical data were compared by the χ² test or Fisher's exact test. The p<0.05 was considered to indicate statistical significance. Single-measure intraclass correlation coefficients (ICCs) and associated 95% confidence intervals (CIs) were estimated using a 2-factor random effect model to estimate the test-retest reliability of electromagnetic- and impedance-scaled geometry and CT-based geometry. ICC in the range 0 to 0.20 is slight reliability, 0.21 to 0.40 is fair reliability, 0.41 to 0.80 is moderate reliability, and 0.81 to 1.00 is good reliability.13) Event-free rate for arrhythmia recurrence was evaluated with the Kaplan-Meier method, and analyzed by a log-rank test. The analysis was performed by a senior biostatistician using SPSS statistical software (version 22.0; SPSS Institute, Chicago, IL, USA).

The baseline clinical characteristics of the study subjects are shown in Table 1. There were no statistically significant differences between the 2 groups in age, gender, body mass index, or in the presence of hypertension, diabetes mellitus, or coronary artery disease. The LA dimension and left ventricular ejection fraction based on the echocardiographic findings, and the baseline medical therapy were not different between the 2 groups. In addition, the 4 PV angles and the posterior LA surface area measured from CT image were similar between the 2 groups.

The study data of the PV angles and posterior LA surface area are shown in Table 2 and Table 3. The PV angles and posterior LA surface area were measured in MediGuide group (M-EM-GEO and M-IM-GEO) and Conventional group (C-IM-GEO). The actual change and the relative change of 4 PV angles and the posterior LA surface area in each field-scaled geometry were calculated, compared to those of CT-based geometry. The actual change and the relative change of all PV angles and posterior LA surface area were significantly smaller in M-EM-GEO than those data in M-IM-GEO and C-IM-GEO. However, the actual change and the relative change of all PV angles and posterior LA surface area were similar between the two groups. Table 4 shows ICC and associated 95% CI between EM- or IM-GEO and CT-based geometry. ICCs of all PV angles and posterior LA surface area in M-EM-GEO demonstrated “good reliability”. In contrast to M-EM-GEO, ICCs of all PV angles and posterior LA surface area in M-IM-GEO and C-IM-GEO demonstrated “moderate reliability”.

In the MediGuide group, the actual change and the relative change of the anterior-posterior diameter of the LA were significantly smaller in M-EM-GEO than those data in M-IM-GEO (actual change, 3.3±2.1 vs. 7.8±4.7 mm; relative change, 9.3±5.5 vs. 22.3±11.3%; p<0.05 and p<0.01), respectively. ICC and associate 95% CI between M-EM-GEO and CT-based geometry were 0.934 and 0.775–0.982, respectively. On the other hand, the ICC and associated 95% CI between M-IM-GEO and CT-based geometry were 0.755 and 0.318–0.482, respectively.

Successful circumferential PVs isolation and CTI ablation were achieved in all patients (100%). Although the procedure time of PVs isolation and CTI ablation were similar between the 2 groups, the fluoroscopy time was significantly shorter in MediGuide group compared to Conventional group (Figure 3A and B).

Arrhythmia recurrence was observed in 10 patients (18%) during a follow-up of 12.2±4.5 months. There was no difference in arrhythmia recurrence between MediGuide group (13%) and Conventional group (20%). In the Kaplan-Meier analysis, the event-free for arrhythmia recurrence did not differ between the two groups as shown in Figure 4. During the ablation procedure and the follow-up period, the procedure related complications (e.g., death, pericardial effusions, strokes, PVs stenosis, esophageal injury, phrenic nerve palsy, and vascular access complications) were not observed in each group.