Computed tomography validated right ventricular mid‐septal lead implantation using right ventricular angiography

Abstract Background Right ventricular (RV) mid‐septal pacing has been proposed as an alternative to RV apical pacing. Fluoroscopic and electrocardiogram criteria are unreliable for predicting the RV mid‐septal lead position. This study aimed to define the optimal RV mid‐septal pacing site using RV angiography. Methods We randomized patients undergoing pacemaker implantation (PPM) to the RV angiography‐guided group (Group A) or conventional fluoroscopy‐guided group (Group F). In Group A, we performed an angiogram in right anterior oblique (RAO 30°), left anterior oblique (LAO 40°), and left lateral (LL) views. We made a 5‐segment grid in RAO 30° and LL views and a 3‐segment grid in LAO 40° on the angiographic silhouette to define the lead position. Computed tomography (CT) was used to validate the lead tip position in both groups. Results We enrolled 53 patients (Group A: 26, Group F: 27) with a mean age of 55.9 ± 12.2 years. CT images validated the lead position in the mid‐septum (Group A, 23 [88.5%]; Group F, 11 [40.7%], P = .0003) and anteroseptal (Group A, 3 [11.5%]; Group F, 5 [18.5%], P = .24). In Group F, the lead was in the anterior wall in 9 patients (33.3%) and the right ventricular outflow tract in 2 (7.4%) patients and none in these two positions in Group A. The lead tip in segment one on the angiographic 5‐segment grid in RAO 30° and LL views indicated a mid‐septal lead position on CT. Conclusions RV angiography is safe and may be used to confirm the mid‐septal lead position during PPM.

of a steep learning curve, availability of the hardware from a single vendor, the necessity of an electrophysiology recording system, and knowledge of the conduction system's anatomy limiting it to specialized centers. A meta-analysis of several studies comparing RV apical to nonapical sites showed mixed medium and long-term effects on left ventricular function. 5 The most significant limitation of the studies on nonapical RV pacing is the nonuniformity of the area of ventricular septal stimulation, along with a lack of accuracy of the septal lead placement. Studies on mid-septal pacing have used fluoroscopic criteria for the lead tip position in the left anterior oblique (LAO) 40°6 and the right anterior oblique (RAO) 30° views. 7 A negative QRS complex or a q wave in lead I on the surface electrocardiogram (ECG) is supposed to indicate RV septal pacing. [8][9][10] Three-dimensional (3D) echocardiography and computed tomography (CT) scans have shown the fluoroscopic and ECG criteria are inaccurate for predicting the mid-septal lead position. [11][12][13] Placement of RV lead in the mid-septum using fluoroscopy alone is challenging because of the interventricular septum's complex anatomy and the inability to define the mid-septal area. Angiography has been suggested as an additional imaging technique during permanent pacemaker implantation (PPM) to determine the septal anatomy in real-time and improve mid-septal lead implantation accuracy. 12,14,15 The study aimed to define the optimal site for RV mid-septal lead implantation using contrast RV angiography. We compared the contrast angiography-guided mid-septal lead implantation with the previously described fluoroscopic-guided lead implantation technique 6,7 using computed tomography.

| ME THODS
The present study is a single-center, randomized study conducted in a tertiary care teaching hospital. We enrolled adult patients (age 21-80 years) with symptomatic atrioventricular (AV) block, in sinus rhythm, with normal left ventricular function (Echo LVEF >55%), and no contraindication for contrast agents. Patients were excluded based on the following criteria: baseline LV ejection fraction (LVEF) ≤ 55%, presence of atrial arrhythmias, renal dysfunction (Serum creatinine >1.3 mg/L), aged <21 or >80 years, current pregnancy, history of contrast allergy, or unwillingness to participate. Patients with atrial fibrillation not requiring a dual-chamber pacemaker and sick sinus syndrome with normal AV conduction were also excluded from the study. Patients who gave written informed consent were prospectively enrolled in the study. All patients underwent detailed clinical, electrocardiographic, biochemical, and two-dimensional echocardiographic evaluations before the procedure. A research nurse randomized the patients to angiography-guided (Group A) or conventional fluoroscopic-guided (Group F) pacemaker implantation. The institutional clinical research and ethics committee approved the study.
Pacemaker implantation was performed after 6 hours of fasting by the patient. Intravenous normal saline was started at 60 mL/h from 6 hours before the procedure. We performed blood urea and serum creatinine estimations before and 24 hours after the patients' procedure. During the implant procedure, two separate venous accesses were obtained by extrathoracic axillary vein puncture. All patients received a dual-chamber pacemaker using a standard 58 cm, 7-French, bipolar, steroid-eluting, active-fixation lead (Medtronic CapSureFix Novus 5076, Medtronic Inc., Minneapolis, MN, USA) for the right ventricle and a 52 cm bipolar, active fixation lead (Medtronic CapSureFix Novus 5076, Medtronic Inc) for the right atrium.
In Group A, two sheaths were introduced over the guidewire, with one sheath being a 7F peel-away introducer and the other a 7F angiographic sheath with a sidearm (AVANTIR+, Cordis Corporation, Santa Clara, CA, USA). The active fixation pacing lead was introduced through the peel-away introducer and advanced into the right atrium using a straight stylet. A 7F angled pigtail catheter was advanced through the angiographic sheath and into the right ventricle over a 0.035" Teflon guidewire and positioned in the mid-RV. The lead was then advanced into the pulmonary artery using a manually shaped two-dimensional (2D) stylet. The 2D stylet was then exchanged for a manually fashioned dual-curve 3D stylet as described previously. 16 Using the posterior-anterior (PA) or RAO 30° fluoroscopic views, the lead with a 3D stylet was withdrawn using a gentle counterclockwise torque until the lead tip fell below the RVOT with an abrupt leftward movement. The lead was then gently advanced, maintaining the torque to direct the tip to the mid-septum. Before In Group F, the pacing lead was advanced to the pulmonary artery using a 2D stylet in the PA projection. The 2D stylet was exchanged for a manually fashioned 3D stylet, and the lead retracted into the mid-septum as described above using the defined fluoroscopic views. The screw was deployed once the fluoroscopic views indicated the mid-septal lead position in the LAO 40° and RAO 30° views. As described previously in the literature, the lead tip was considered in the mid-septum if, in the LAO 40° view, the lead tip faced the spine with an angulation between the horizontal plane and the axis of the distal part of the lead between 0° and 60°. In the RAO 30° view, the lead tip was in the middle quadrant. 3,7,17 A 12-lead ECG was recorded at the end of the procedure using multichannel electrophysiology (EP) analysis system (EP tracer, Schwarzer Cardiotek GmbH, Heilbronn, Germany). The surface ECG recording was filtered at 0.05-150 Hz, 10 mm/1 mV amplitude, and measurements were recorded at a sweep speed of 100 mm/s. The ECG parameters were analyzed on the 12-lead ECG by an independent operator and included QRS duration, QRS axis, presence or absence of QRS notching in limb leads, presence or absence of a q wave, or negative QRS complex in the lead I, and a QRS transition zone in the precordial leads.
A cardiac CT scan was performed after the implantation during the outpatient follow-up. A dual-source, 64-slice Siemens Definition Flash CT scanner (Siemens, Erlangen, Germany) was used to acquire images with a tube voltage between 100 and 120 kV and a tube current of 200-300 mA, depending on patient size. While performing the CT scan, the pacemaker was programmed to VVI 60 bpm to reduce the motion artifact. Scans were performed after 75 mL of Omnipaque 350 was injected at 6 mL/s, followed by 50 mL of saline. During image acquisition, prospective ECG gating was done using a phase window (70%-80%) and an image matrix of 512 × 512 pixels, and initial reconstruction was performed at the 75% phase (0.75-mm slice thickness and 0.5 mm intervals) using B26 (soft) and B46 (hard) kernels. Offline analysis was performed with data sets that were transferred to an external workstation. For accurate localization of the RV lead axial slices, oblique reconstructions and maximum-intensity projection images were used. Two experienced radiologists blinded to the other's results performed the analysis, and any disagreement between the two was resolved by consensus.
RV lead positioning was defined in short-and long-axis planes as designated by Pang et al. 12 The RV lead tip positions were divided into mid-septum, anteroseptal junction, anterior, and free wall. The fluoroscopic and angiographic lead tip positions were compared and validated using CT scan images.

| RE SULTS
A total of 57 patients were initially enrolled in the study. Four patients did not consent for a follow-up CT scan and were not con- ECG characteristics of the two groups are shown in Table 3.
Group A patients had significantly narrower QRS complexes, lesser q waves in the lead I, a more leftward axis, and earlier QRS transitions in the chest leads. Patients in Group F had significantly more q waves in the lead I, a more rightward axis, and a later QRS transition on chest leads. There was no significant difference in the notching of Q waves in the inferior leads.

| D ISCUSS I ON
This study makes a novel contribution to the literature by defining specific areas of the RV septum, and especially the mid-septal area, using real-time angiography, corroborated by CT scans, as opposed to conventional fluoroscopy. The study's principal findings are as follows: first, the use of right ventricular angiography helps delineate the RV anatomy during PPM and helps to target the midseptal area accurately, as validated by a CT scan. Second, the use of and chest x-rays were in only modest agreement with echocardiographic data. 19 It is now evident that CT scanning is more reliable and accurate to identify the RV septal lead position than echocardiography or magnetic resonance imaging. 13 The original ECG criteria for RV septal pacing using fluoroscopic landmarks included: (i) paced QRS duration <140 ms, (ii) precordial transition earlier than lead V4, (iii) absence of QRS notching in inferior leads, and (iv) a q wave, QS, or an isoelectric QRS in lead I. 8,10 Burri et al., using 3D electroanatomical mapping to validate the ECG criteria for RV septal pacing, found that no single criteria, including a negative QRS in lead I, could accurately distinguish RV mid-septal pacing from anterior wall pacing. 20 In the present study, a narrower QRS complex, lesser q waves in the lead I, a more leftward axis, and earlier QRS transitioning in chest leads were suggestive of midseptal pacing. Pang et al., who validated the mid-septal lead position by CT scan, found that the q wave in the lead I was more common with anteroseptal lead position than the septal lead position. 12 In the study by Burri et al., negative QRS complex or the presence of a q wave in lead I was more common when pacing from the anterior wall compared with from the mid-septum. This is because of a more leftward position of the lead tip while pacing from the anteroseptal or anterior wall compared with the mid-septum that has a rightward bulge. 20 The narrower QRS complex while pacing the mid-septal  region is because of the earlier engagement of the His-Purkinje system. 21 According to the fluoroscopic criteria for septal pacing, the RV lead tip should face the spine, with the angle between the horizontal plane and the lead between 0° and 60° in the LAO 40° view. In the RAO 30° view, the lead tip should be in the middle or inferior quadrants. 6,7,10 Even though the information gained from CT is retrospective, it is more accurate than echocardiography or magnetic resonance imaging for defining the lead tip position and is considered a "gold standard." 13,22 In the study by Osmancik et al., mid-septal lead positioning was achieved in only 41% of patients in whom the LAO 40° criteria for mid-septum placement were met. 17 Pang et al. 12 found that only 21% of the leads were in the true mid-septum when implanted using conventional fluoroscopic criteria validated by CT.

| Limitations
There are several limitations to our study. First, this was a singlecenter study performed on a selected group of patients with normal LV function and AV block. Patients with severe LV dysfunction, valvular heart diseases, or congenital heart diseases were excluded because the dilatation and distortion of the chambers could confound the implant attempts. Another limitation is the small cohort size, which could have led to broader confidence intervals in the statistical analysis. Additionally, in patients in the angiographic group, the lead was repositioned to a better position after reviewing the angiography. This could not be avoided since the purpose of the study was to use angiography to position the lead in the mid-septal position accurately. We did not compare the ECG between the anterior and septal RVOT pacing, as it was not the study's purpose and may be a topic of future research. Hemodynamic assessment, pressurevolume loops, and cardiac output were not assessed in the study, either. Since this was a study to determine RV angiography's utility to define the mid-septal area and position the lead, we did not assess left ventricular function on follow-up.

| CON CLUS ION
RV angiography in the RAO 30°, LL, and LAO 40° fluoroscopic views define the complex 3D RV anatomy in real-time during PPM.
According to the present study, the appropriate area for mid-septal lead implantation is the proximal upper segment (segment 1) in the RV angiography-based 5-segment grid on the cardiac silhouette in RAO 30°and LL fluoroscopic views. Angiography is a safe and effective method that can be used to confirm the mid-septal lead position.
In difficult anatomical situations or if there is doubt regarding midseptal lead positioning during PPM, RV angiography can be used to confirm mid-septal lead position.

ACK N OWLED G EM ENTS
The authors wish to thank Editage for their assistance in editing and proofreading this manuscript.

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to declare.