Caudal vena cava isolation using ablation index‐guided radiofrequency catheter ablation (CARTO™ 3) to treat sustained atrial tachycardia in horses

Abstract

Background

Myocardial sleeves of the caudal vena cava are the predilection site for atrial tachycardia (AT) in horses. Caudal vena cava isolation guided by the ablation index, a lesion quality marker incorporating power, duration and contact force, might improve outcome.

Objectives

Describe the feasibility and outcome of caudal vena cava isolation using ablation index‐guided radiofrequency catheter ablation (RFCA) to treat AT in horses.

Animals

Ten horses with sustained AT.

Methods

Records from 10 horses with sustained AT treated by three‐dimensional electro‐anatomical mapping and ablation index‐guided RFCA (CARTO™ 3) were reviewed.

Results

Three‐dimensional electro‐anatomical mapping of the right atrium identified a macro‐reentry circuit in the caudomedial right atrium (n = 10). Point‐by‐point RFCA was performed to isolate the myocardial sleeves of the caudal vena cava in power‐controlled mode with a mean of 17 ± 7 applications. The ablation index target was 400‐450. A median ablation index of 436 (range, 311‐763) was reached using a median maximum power of 35 (range, 24‐45) W for a median duration of 20 (range, 8‐45) seconds, with a median contact force of 10 (range, 3‐48) g. Sinus rhythm was restored in all 10 horses. To date, 9‐37 months post‐ablation, none of the horses have had recurrence.

Conclusions and Clinical Importance

Caudal vena cava isolation using ablation index‐guided RFCA was feasible and effective to permanently treat sustained AT in horses. Ablation index guidance ensured efficient lesion creation, and isolation of the caudal vena cava eliminated the arrhythmogenic substrate, thereby minimizing the risk of recurrence.

Abbreviations

3D EAM

three‐dimensional electro‐anatomical mapping

ACT

activated clotting time

AF

atrial fibrillation

AI

ablation index

AT

atrial tachycardia

CF

contact force

cTnI

cardiac troponin I

RF

radiofrequency

RFCA

radiofrequency catheter ablation

VCG

vectorcardiogram

1. INTRODUCTION

Atrial tachycardia (AT) is a supraventricular tachyarrhythmia characterized by a rapid atrial rhythm with identifiable P′ waves on the surface ECG. The underlying mechanism can be focal AT, caused by a rapidly firing focus, or macro‐reentry AT, also called atrial flutter. ¹ , ² The criteria for differentiating focal and macro‐reentry AT using a surface ECG are not clearly defined in horses. Therefore, three‐dimensional electro‐anatomical mapping (3D EAM) is required to identify the underlying mechanism. ¹ , ² , ³ , ⁴ Recently, 3D EAM in horses with sustained AT identified macro‐reentry AT in the caudal right atrium as the underlying mechanism in the majority of horses. ¹

In the past, AT was treated similarly as atrial fibrillation (AF) using quinidine sulfate or transvenous electrical cardioversion. ⁴ Treatment of AT however can be more difficult in horses, with a higher recurrence rate compared with AF. Because ablation treats the underlying cause of the arrhythmia, it could markedly decrease recurrence risk. ¹ , ⁵ , ⁶ Therefore, both in horses and in humans, radiofrequency catheter ablation (RFCA) is the recommended treatment for AT. This procedure is based on the application of radiofrequency (RF) energy to the myocardium. By a combination of resistive and conductive heating, the targeted tissue will experience irreversible coagulation necrosis, which will evolve into a nonconducting scar. ⁷ The efficacy of RFCA depends on the formation of durable lesions that are transmural and contiguous. ⁸ The contact force (CF) between catheter tip and tissue is a key determinant for lesion formation and procedural efficacy. Successful treatment of AT by RFCA in 8 horses has been described, but no CF monitoring system was used, and recurrence still was seen in 3 horses. ¹ Also, in that study, RF energy was applied at the zone of slow conduction, and ablation was terminated as soon as sinus rhythm was restored. The area of slow conduction probably was not always fully interrupted, leaving arrhythmogenic substrate and resulting in recurrence in some horses. Contact force‐sensing technology was not available during the previously reported procedures, so it was unknown if the ablation catheter had sufficient contact with the atrial wall during energy delivery. Low CF leads to ineffective tissue heating and fails to induce durable lesions, which increases the likelihood of recurrence. ⁸ , ⁹ , ¹⁰ , ¹¹ Besides CF, power and duration of RF application are crucial for lesion formation. Therefore, the ablation index (AI), a lesion quality marker incorporating power, CF and duration in a weighted formula, has been introduced. ¹² The use of AI‐guided ablation enhances outcomes after RFCA of AF, ¹³ , ¹⁴ atrial flutter ¹⁵ and focal AT ¹⁶ in human patients. The feasibility of AI‐guided ablation to isolate the caudal vena cava and pulmonary veins has been shown experimentally in horses. ¹⁷ We describe the feasibility and outcome of isolation of the caudal vena cava myocardial sleeves using 3D EAM and AI‐guided ablation to treat sustained AT in 10 horses.

2. MATERIAL AND METHODS

Medical records of 10 horses with sustained AT that were examined using 3D EAM and treated by RFCA using a CF‐sensing system (CARTO™ 3) by the Equine Cardioteam Ghent of Ghent University were reviewed.

In each horse, a general clinical examination and standard echocardiographic examination was performed. Atrial tachycardia was diagnosed using a base‐apex ECG (Figure 1). ¹⁸ Subsequently, a 12‐lead ECG was recorded (Contec, Qinhuangdao, China) in the standing horse and the derived vectorcardiogram (VCG) was calculated (Matlab R2018b, MathWorks, Eindhoven, The Netherlands) to identify the approximate origin of the arrhythmia. ¹⁹ , ²⁰

FIGURE 1.

Base‐apex ECG of a horse with sustained atrial tachycardia, characterized by rapid, monomorphic P′ waves, normal QRS morphology but irregular RR intervals. Notice that in lead I, P′ waves are not that clear to distinguish and can therefore be misinterpreted as atrial fibrillation.

2.1. Mapping and ablation procedure

In the standing horse, a fixed curve sheath (HeartSpan, Merit Medical, Etterbeek, Belgium) with protruding J‐tipped guidewire was inserted through the left jugular vein and placed in the coronary sinus under transthoracic echocardiographic guidance. ²¹ This sheath served for insertion of a decapolar catheter into the coronary sinus after induction of anesthesia.

Three‐dimensional electro‐anatomical mapping and RFCA were performed under general anesthesia. The horses were positioned in left lateral recumbency, with the magnetic field generator placed underneath the horse, as previously described. ¹ During positioning on the table, 2 in‐parallel grounding patches for the ablation generator were placed on the left dorsal thorax. Six reference patches of the mapping system were placed on the thorax, 3 on the left side and 3 on the right side, at the corners of the generated magnetic field. A decapolar deflectable catheter (Webster CS Bi‐Directional Catheter, Biosense Webster, Diegem, Belgium) was placed through the fixed curve sheath into the coronary sinus. The electrograms from this catheter served as a timing reference for atrial activation during the mapping procedure. An 8.5F bi‐directional deflectable guiding sheath (Vizigo, Biosense Webster, Diegem, Belgium or Senovo Bi‐Flex, Biotronik, Vilvoorde, Belgium) was placed in the right jugular vein through which the mapping catheter (Pentaray Nav eco Catheter, Biosense Webster, Diegem, Belgium) was inserted. A 3D EAM of the right atrium was made to determine the location and mechanism of AT. Both the sheath and the mapping catheter were continuously irrigated with heparinized (1 IU/mL) saline using pressurized bags. Once the origin and mechanism of AT were identified, the mapping catheter was removed and an 8F irrigated, deflectable CF‐sensing ablation catheter (Thermocool SmartTouch, Biosense Webster, Diegem, Belgium) was inserted through the guiding sheath into the right atrium. Point‐by‐point ablation was performed at the areas of slow conduction, dorsal and ventral to the line of conduction block, resulting in complete isolation of the caudal vena cava myocardial sleeves. Radiofrequency energy was delivered in power‐controlled mode with power set at 35‐40 W. The CF target was 10‐15 g. Power was decreased when catheter tip temperature reached 43°C. Energy was delivered until an AI of 400‐450 was reached. During energy delivery, the ablation catheter was irrigated with saline at 30 mL/min. Successful caudal vena cava isolation was confirmed by remapping the caudal right atrium with the mapping catheter and demonstrating conduction block. In the absence of complete conduction block, additional ablation lesions were delivered, targeting the sites of electrical breakthrough.

Unfractionated heparin (Heparine LEO, LEO pharma, Lier, Belgium) and sodium penicillin (20 000 IU/kg, Penicillin 5.000.000IE, KELA pharma, Hoogstraten, Belgium) were administered IV at the start of the procedure. In the first 4 horses, 60 IU/kg of heparin were administered, followed by subsequent administrations of 30 IU/kg. In the last 6 horses, 80 IU/kg were administered, followed by 40 IU/kg repeat administrations. Activated clotting time (ACT) was measured throughout the procedure using a point‐of‐care device (i‐STAT, Abbott, Zaventem, Belgium) in 8 of 10 horses. The timing of repeat heparin administration was adjusted to achieve and maintain a target ACT of 300‐350 seconds. In the 2 horses in which ACT was not measured, heparin administration was repeated every 2 hours (60 IU/kg followed by 30 IU/kg). Serum cardiac troponin I (cTnI) concentration was measured before, immediately after and 4 hours after ablation (Alinity i‐STAT Sensitive Troponin‐I assay, Abbott, Zaventem, Belgium).

Once a complete line of block was confirmed, the catheters were withdrawn and the horses were recovered using rope‐assisted recovery. The horses were clinically evaluated daily.

2.2. Data analysis

Normality of descriptive variables was evaluated using the Shapiro‐Wilk test. Normally distributed data are reported as mean ± SD and non‐normally distributed data as median and range.

3. RESULTS

Ten horses (8 Warmblood, 1 Quarter Horse, 1 Andalusian; 4 mares, 5 geldings, and 1 stallion) were referred to the clinic for diagnosis and treatment of AT using 3D EAM and RFCA between June 2021 and October 2023. General information of the horses, electrocardiographic and echocardiographic findings are presented in Table S1. The horses had a mean ± SD age of 8 ± 4 years, weight of 528 ± 64 kg and height of 165 ± 7 cm. None of the horses had been previously treated for AT. Horses 1, 5, 8, 9, and 10 were asymptomatic and an irregular heart rhythm was detected during a routine or prepurchase examination. Horse 2 showed poor performance and tachypnea during recreational exercise. Horse 3 had collapsed during a show jumping competition. Horse 4 had epistaxis and showed decreased performance as a show jumper. Horse 6 was a 2‐year‐old stallion and an arrhythmia was detected during examination for weight loss. In Horse 7, AT was detected during examination for ataxia and weakness caused by presumed West Nile virus infection.

Auscultation identified an arrhythmia in all horses, but often included episodes with a regular rhythm, especially when heart rate was slightly increased. A left‐sided holosystolic murmur was noted in Horse 5 (4/6), Horse 6 (4/6), and Horse 9 (3/6). Horse 8 had a mid‐to‐end systolic 4/6 bilateral murmur. An early‐mid systolic murmur was detected in Horse 2 (left 3/6 and right 1/6), Horse 3 (left 2/6), and Horse 10 (left 1/6).

In 9 of 10 horses, 12‐lead ECG and vectorcardiogram (VCG) identified a left and cranioventral direction of the initial atrial wavefront, suggesting a caudo‐dorsal right atrial origin of AT (Figure 2). The spatial directions of the mean electrical axis from the 10 horses were consistent, except for the second half of the P′ wave in Horse 9 (Figure 3). In Horse 9, the VCG first showed a rapid right‐to‐left initial atrial activation, followed by a right and cranioventral direction of the atrial depolarization. Therefore, it was uncertain if the AT origin was located in the right or left atrium. Isoelectric intervals between the P′ waves were present in all leads in Horses 1 and 2. Horse 3 had isoelectric lines in 8 leads (II, III, aVL, aVF, V1, V2, V3, V5). Horses 6, 8, and 10 had a continuously undulating pattern present in all leads, whereas Horses 4 and 9 had this undulating pattern in 11 leads and Horses 5 and 7 in 10 leads. Differentiation between isoelectric lines and a continuously undulating pattern was sometimes difficult. The mean atrial rate was 187 ± 26 depolarizations per minute in the standing horses and 184 ± 22 depolarizations per minute during anesthesia, averaged over 10 P′P′ intervals on the base‐apex surface ECG.

FIGURE 2.

Twelve‐lead ECG (A) and corresponding vectorcardiogram (B) of sustained atrial tachycardia from Horse 5. (A) Most leads have a continuous undulating pattern, except for lead II and V2 which show an isoelectric line between P′ waves. (B) The vectorcardiogram axes are calculated from the 12‐lead ECG with X representing the right‐left, Y the cranial‐caudal, and Z the ventral‐dorsal axes. The first half of the P′ wave is represented by the red lines, while the second half of the P′ wave is represented by the green lines. The vectorcardiogram shows a cranial, ventral, and left direction of the initial atrial wavefront, suggesting a caudodorsal right atrial origin of atrial tachycardia.

FIGURE 3.

Spatial directions of the mean electrical axis (MEA) of the first (left panel) and second (right panel) half of the P′ wave visualized with a Lambert azimuthal equal‐area plot. Dots inside the green circle represent a dorsal MEA, while dots outside represent a ventral MEA. Left is 0° and caudal is 90°. P′ waves of different horses or indicated by different colors. All horses show a left and cranioventral direction of the initial atrial wavefront, except for 1 horse (dark blue dot, Horse 9) in which the second halve of the P′ wave shows a right and cranioventral direction.

3.1. Three‐dimensional electro‐anatomical mapping

Mapping and ablation results, including biophysical ablation parameters, are presented in Table S2. A detailed 3D activation map of the AT circuit including a mean of 6055 ± 2156 local electrograms was completed in a mean mapping time of 48 ± 26 minutes. In all horses, the activation map identified a macro‐reentry circuit in the caudal right atrium, between the fossa ovalis and the caudal vena cava. Despite the uncertain origin based on the VCG in Horse 9, the entire reentry circuit also was located in the caudomedial right atrium. Therefore, myocardial sleeves probably connected the reentry circuit transmurally with the caudal left atrium, which resulted in prompt left atrial activation. The reentry circuit rotated in clockwise (n = 8) or counterclockwise (n = 2) direction around a line of conduction block. The line of block was identified by double potentials. Areas of slow conduction, dorsal and ventral to the line of block, were represented by fractionated signals and were the target of ablation (Figure 4).

FIGURE 4.

Panel (A) shows a left lateral view on a coherent map of the right atrium. Coherent mapping is a module of the CARTO™ 3 system, integrating local activation time with conduction velocity vectors, thereby helping to identify areas of conduction slowing or block. Bold vectors indicate slow conduction and areas without conduction velocity vectors (zones of conduction block) appear as brown on the map. The colors follow the rainbow spectrum from red (earliest activation) to purple (latest activation). A clockwise reentry circuit in the caudomedial aspect of the right atrium rotates around a line of conduction block (brown line). A dorsal and ventral isthmus can be identified and these are associated with zones of slow conduction, represented by the bold vectors. Panels (B), (C), and (D) show recordings during atrial tachycardia at the corresponding locations indicated on panel (A). Each panel shows the surface ECG lead I, II, III on top (yellow), 2 coronary sinus electrograms (dark blue traces), and 4 electrograms from the mapping catheter (4 traces at the bottom). (B) The mapping catheter is located in the caudal right atrium, proximal to the line of conduction block. Electrograms from the mapping catheters show sharp, high‐amplitude deflections, indicating healthy myocardium. (C) The mapping catheter is located at the area dorsal to the line of conduction block. Electrograms from the mapping catheter show fractionated signals, which represent slow conduction. (D) The mapping catheter is located at the line of conduction block. Intracardiac recordings from the mapping catheter, represent double potentials (highlighted by the yellow frame), which corresponds to the line of conduction block.

In some horses, the mapping system had difficulty tracking the different splines of the mapping catheter in the ventral caudal right atrium because of the large impedance gradients. The issue was resolved by exchanging the mapping catheter for the ablation catheter, which allowed detailed mapping of that area.

3.2. Radiofrequency catheter ablation

The areas of slow conduction, dorsal and ventral of the line of conduction block, were targeted for RFCA. By transecting both areas, the caudal vena cava myocardial sleeves were completely isolated. Caudal vena cava isolation was completed in a mean time of 41 ± 15 minutes and required a mean of 17 ± 7 RF applications. The median maximum power was 35 (range, 24‐54) W for a median duration of 20 (range, 8‐45) seconds at a median CF of 10 (range, 3‐48) g. The median impedance drop was 7 (range, 1‐32) Ω. A median AI of 436 (range, 311‐763) was reached. Successful electrical isolation was achieved in all horses and was confirmed by demonstration of entrance block. Isolation implies the absence of electrograms distal to the ablation line where electrograms were recorded before ablation, and was identified either by remapping the caudal right atrium (n = 8) or by recording with the ablation catheter (n = 2; Figure 5). First pass isolation was achieved in 5 horses (Horses 2, 3, 7, 8, 10). In the other 5 horses (Horses 1, 4, 5, 6, and 9), the remap showed that conduction was still present distal to the ablation line and persistent conduction through a gap was sought. Gaps arose because of insufficient spacing between lesions. These gaps were targeted by additional ablation so that complete isolation of the caudal vena cava was achieved. The same AI target was used for the additional lesions. Sinus rhythm was restored after a median of 3 (range, 1‐10) RF applications in all horses, except for Horse 2. In Horse 2, AF occurred before the start of ablation because of manipulation of the ablation catheter in the caudal right atrium. Ablation was continued in AF and, after successful isolation, amiodarone (5 mg/kg IV over 30 minutes; Cordarone 150 mg/3 mL, Sanofi‐Aventis, Machelen, Belgium) was administered to restore sinus rhythm.

FIGURE 5.

Left lateral view of the right atrium. The left panel shows an activation map of the right atrium before (A) and after (B) ablation. The right panel shows the surface ECG lead I, II, III (yellow), electrograms from the coronary sinus catheter (dark blue), followed by the local electrograms recorded with the mapping catheter at the myocardial sleeves of the caudal vena cava. (A) Activation map before ablation revealed a clockwise reentry circuit that rotates around a line of conduction block (white line). The mapping catheter records a local electrogram at the myocardial sleeves of the caudal vena cava, highlighted by the yellow box. The local electrogram corresponds to the location on the map highlighted by the yellow dot. (B) Activation map after ablation. Point‐by‐point ablation (red and pink dots) was performed to isolate the caudal vena cava. After ablation of the areas of slow conduction, remapping showed incomplete conduction block. Therefore, additional lesions were placed to connect the dorsal and ventral area of slow conduction and to obtain permanent conduction block. Successful isolation was demonstrated by remapping the caudal right atrium, focusing on the area distal of the ablation line. The lack of color distal to the ablation line means that no electrograms could be recorded, which proves entrance block. The electrograms, highlighted by the yellow box, correspond to the location on the map highlighted by the yellow circle.

Baseline cTnI, measured at the start of the procedure, was below the limit of quantification in all horses. Immediately after ablation, cTnI increased in 1 horse to 0.018 ng/mL and remained below the limit of quantification in the others. Four hours after ablation, cTnI concentrations were below the limit of quantification in 1 horse and increased in the other 9 horses, reaching concentrations between 0.013 and 0.05 ng/mL. All results were still within the reference range (<0.06 ng/mL).

3.3. Complications

Horse 4 experienced a myopathy of the left triceps after recovery and often was in right lateral recumbency in the next 24 hours, which resulted in right facial nerve paralysis. The horse was treated with dexamethasone IV (initial dose 0.06 mg/kg, followed by 0.03 mg/kg; Rapidexon, Dechra, Lille, Belgium) and flunixin IV (1.1 mg/kg; Emdofluxin, Emdoka, Hoogstraten, Belgium) for 4 days, followed by flunixin PO (1.1 mg/kg; Finadyne, Intervet International, Brussels, Belgium) for the next 5 days. Vitamins B1 and E (1 mg/kg and 7.5 IU/kg, respectively) were administered PO. The triceps muscle was treated locally with prednisolone ointment (Ekyflogyl, Audevard, Clichy, France) 4 times a day and with daily physiotherapy. The clinical signs of the triceps myopathy resolved over the following days, and the right facial nerve paralysis resolved over the following weeks.

Horses 7 and 9 developed a fever the day after the procedure. Ultrasonography of the thorax identified B‐lines and consolidations which resolved with 35 and 21 days of antibiotic treatment, respectively. These horses were discharged 31 and 7 days after the procedure, respectively.

3.4. Follow‐up

If no complications occurred, the horses could leave the clinic 5 days post‐ablation. Box rest and hand‐walking were advised the first week. After that, the horse could go into a paddock or small pasture. After 1 month, sinus rhythm was confirmed by the referring veterinarians in all horses. Subsequently, exercise could be resumed gradually over a period of 1 month.

The heart rhythm is checked every month by the owner using the KardiaMobile 6L (AliveCor, California, USA). To date, 9‐37 months post‐ablation, none of the horses experienced recurrence. All horses are currently back to their full training regimen.

4. DISCUSSION

These cases describe the first use of AI‐guided RF ablation to treat AT in horses. In all horses, the caudal vena cava myocardial sleeves could be successfully isolated, which resulted in restoration of sinus rhythm. In addition, none of the horses experienced recurrence after a follow‐up period of 9‐37 months. Successful RFCA has been described previously for the treatment of AT in 8 horses, but 3 of 8 experienced recurrence. ¹ Treatment in those horses was probably suboptimal because of the lack of CF‐sensing technology and termination of ablation upon restoration of sinus rhythm, before full isolation of the caudal vena cava was achieved.

In humans and dogs, the ablation target for macro‐reentry AT is the narrowest isthmus of the reentry circuit. ²² , ²³ , ²⁴ Contiguous ablation lesions should be placed to transect the critical isthmus and connect unexcitable barriers. ²² In the previous case series of AT ablation in horses, ablation was performed at the area of slow conduction, but once sinus rhythm was restored, the ablation procedure was terminated. ¹ Most likely, this slow conducting area was not entirely interrupted by a point‐by‐point ablation line, which is a possible explanation for the few recurrences observed. In our cases, the strategy was different than ablation of macro‐reentry AT in humans and dogs where only the narrowest isthmus is transected. Instead, we continued point‐by‐point ablation until both the dorsal and ventral areas were transected, even if sinus rhythm had been established. Closing 1 zone of slow conduction already interrupts the reentry circuit, but by closing both zones, we included a precaution in case 1 of the 2 would reconnect. By closing both areas, we obtained full isolation of the caudal vena cava myocardial sleeves, which eliminated the arrhythmogenic substrate and has prevented recurrence so far. This strategy is comparable to pulmonary vein isolation as a treatment for AF in human medicine, which prevents rapidly firing foci in the pulmonary veins from initiating AF. ²⁵ , ²⁶ In our case series, the caudal right atrium was remapped after ablation to demonstrate entrance block. Doing so means that atrial impulses can no longer be conducted into the myocardial sleeves of the caudal vena cava and proves successful isolation of these sleeves. This is also the endpoint of pulmonary vein isolation for AF treatment in human medicine. ²⁷ , ²⁸

In human medicine, arrhythmia recurrence after RFCA is mostly caused by non‐transmural lesions or gaps in the ablation line. ²⁹ Variables such as catheter‐tissue contact, power and duration are crucial for lesion formation and transmurality. With adequate CF, the catheter is embedded in the myocardium and allows more energy to be delivered into the tissue, instead of being lost in the blood pool. ¹⁰ Several studies observed increasing lesion size with increasing CF, ³⁰ , ³¹ , ³² , ³³ which resulted in the development of CF‐sensing catheters to provide real‐time assessment of CF during each energy delivery. ⁸ , ³⁴ In our previous case study, CF‐sensing technology was not available, which could have resulted in insufficient tissue contact and non‐transmural lesions. ¹ Because of the large size and motion of the horse's heart, it is probably even more difficult to obtain stable and sufficient catheter‐tissue contact in horses compared with humans. Besides CF, lesion size is also dependent on power and duration. ¹⁴ The AI is a novel lesion quality marker that incorporates CF, power and duration into a weighted formula with improved assessment of lesion size and lower incidence of AF recurrence. ¹² , ¹³ , ¹⁴ , ³⁵ A minimal CF of 10 g was targeted in our cases, based on recommendations in human medicine. ³⁶ Doing so was feasible in most RF applications, with a median CF of 10 (3‐48) g. Occasionally, a lower CF was obtained which was compensated by a longer energy application, so that the AI target was reached at the end of the application. A target power of 35‐40 W was used until an AI of 400‐450 was achieved. Depending on the location and atrial wall thickness, a target of 550 for the anterior/roof segments and 400 for posterior/inferior segments is used for pulmonary vein isolation in human patients. ³⁷ The posterior wall in humans is a thin structure with a reported thickness of 0.88‐1.74 mm, requiring less energy. ³⁸ Because the myocardial sleeves in the caudal vena cava of horses are thin‐walled structures, with a mean thickness of ±1.6 mm, ³⁹ an AI target of 400‐450 was used to avoid unnecessary RF energy application and improve safety. Novel ablation techniques could potentially further improve safety and decrease procedure time. High‐power, short‐duration (45‐70 W, 5‐10 seconds) is an emerging strategy in human medicine and limits collateral tissue damage by increasing the resistive heating phase and decreasing passive conductive heating. ⁴⁰ Doing so leads to lesions with wider diameter and less depth, which is beneficial for creating contiguous lesions in thin‐walled structures. It also limits catheter instability and decreases procedure times. ⁴¹ Pulsed field ablation is a novel energy source and employs a non‐thermal mechanism in which ultra‐short high voltage pulses are delivered to induce pores in the targeted cells. It has high selectivity for myocardial tissue and therefore limits collateral damage. ⁴² , ⁴³ Both techniques should be further investigated in equine cardiology.

Impedance drop is a specific marker of local tissue heating and can be used as a real‐time marker of lesion creation. ⁷ , ⁴⁴ In human medicine, a 5 to 10 Ω impedance decrease is clinically used to ensure effective ablation lesions and lesions with a small impedance decrease are associated with conduction recovery. ⁴⁵ , ⁴⁶ In our patients, a median impedance decrease of 7 Ω was reached. In 2 horses, however, the median impedance decrease was only 4 Ω but both reached the AI target, which is known to be strongly correlated with impedance decrease. ⁴⁷ One possible explanation for the limited impedance decrease is that CF variability, which is not included in the AI, was higher in those 2 cases. ⁴⁸ On the other hand, the sensitivity of impedance decrease as a marker of lesion formation is limited in which energy applications with a lower impedance decrease still can result in transmural lesions, especially in thin tissue such as the caudal vena cava. ⁴⁴

Atrial tachycardia was associated with decreased performance, epistaxis or collapse during exercise in 3 horses. All 3 were used for recreational exercise and show jumping, suggesting that clinical signs are not necessarily restricted to horses in physically demanding sport disciplines. In 7 horses, AT was not associated with obvious clinical signs. In 1 of these 7, AT was detected during examination for weakness and ataxia, which were attributed to presumed West Nile virus infection. However, staggering and weakness previously have been reported in a horse with AT. ¹ Atrial tachycardia is characterized by identifiable P′ waves on the surface ECG. Depending on the lead, electrode positioning and settings, P′ waves are sometimes challenging to detect and AT can be misinterpreted as AF (Figure 1). Multiple lead recording with strategically‐placed electrodes is helpful to identify a lead with clearly identifiable P′ waves. ¹⁹ It has been shown that a 12‐lead‐derived VCG can aid in identifying the presumed anatomical region of origin of atrial ectopy in horses. ¹⁹ Predicting the origin of an arrhythmia, especially distinguishing between right‐ and left‐sided origin, is useful to plan the procedure. A left‐sided arrhythmia requires a transseptal puncture, requiring special equipment and involving longer anesthesia time. If the region of origin is known beforehand by the VCG, mapping can be focused on this region, which decreases mapping time and speeds up the procedure. In 9 of 10 horses, the anatomical site of origin predicted by the 12‐lead ECG matched with 3D EAM. However, in 1 horse the AT origin was not clear on the 12‐lead ECG and VCG. In this horse, we first mapped the right atrium and caudal vena cava, because they are the predilection sites for AT in horses. Indeed, 3D EAM identified the origin of AT in the typical caudomedial right atrium. The myocardial sleeves probably also conducted to the adjacent caudal left atrial myocardium, explaining the caudal left atrial origin on the VCG. ⁴⁹ In humans and dogs, 12‐lead ECG also can be used to differentiate focal from macro‐reentry AT. Isoelectric lines between P′ waves in all leads are associated with focal AT, whereas continuously undulating P′ waves are associated with macro‐reentry AT, although overlap exists: in rare occasions the isoelectric lines can represent areas of slow conduction in macro‐reentry cases. ²² , ⁵⁰ Focal AT also can present with continuous undulation in the case of a very rapid AT or a broad P′ wave caused by intra‐atrial conduction disturbances. ⁵¹ In horses, no differentiating criteria exist. ⁴ Three of our horses showed an isoelectric line in all leads (Horses 1, 3) or the majority of the leads (Horse 2). These 3 horses had a slow atrial rate, whereas the horses with a continuous undulating pattern had a higher atrial rate. In horses, the macro‐reentry circuit is localized in the myocardial sleeves of the caudal vena cava, a thin structure that might not contain enough myocardial mass to induce a continuous undulation on the surface ECG.

In all 10 horses, 3D EAM identified a very similar macro‐reentry circuit located caudal to the fossa ovalis, at the transition from caudomedial right atrium to caudal vena cava, which could be regarded as the typical atrial flutter in horses. This location is consistent with a previous report, where in 8 of 9 horses, AT originated from the same region. ¹ This region seems to act as a hotspot for AT in horses, which can be explained by the presence of irregularly delineated myocardial sleeves that have been identified macroscopically and histologically. ³⁹ In addition, a previous study found that healthy horses can have a fibrous island in this area, which might result in a line of block as observed on the 3D EAM, around which the reentry rotates. ³⁹ The combination of disorganized fibers in myocardial sleeves and the fibrous islands creates an ideal substrate for reentry in the caudal right atrium. This situation is in contrast with both humans and dogs, where the most common type of macro‐reentry AT, also called typical atrial flutter, consists of a circuit rotating around the tricuspid annulus with an area of slow conduction at the cavotricuspid isthmus. ²³ , ⁵⁰ In humans and dogs, the flutter circuit covers almost the entire right or left atrium. In horses, however, the mechanism was a local reentry circuit that only covered a small area in the caudal right atrium, involving the myocardial sleeves. The reentry circuit path length in the horses was approximately the same compared to the circuit path length of atrial flutter in human patients. The inferior vena cava is a rare arrhythmogenic site that can initiate AF in humans. ⁵² In horses, it has been demonstrated that AT can deteriorate into AF. ¹ This observation means that AT can be the initiating trigger for AF, and isolation of the caudal vena cava might be a useful treatment to decrease AF recurrence in some horses. ¹⁷

Time needed to complete the 3D EAM varied among horses and ranged between 22 and 104 minutes. Long mapping durations were caused by tracking issues of the different splines of the mapping catheter in the ventral caudal vena cava. The Pentaray mapping catheter has 20 electrodes, 1 mm in size and separated by 2 mm inter‐electrode spacing, arranged in 5 radiating splines, which allows sampling multiple points simultaneously and provides mapping with high resolution. A sensor in the catheter shaft enables magnetic tracking of the catheter tip and records data about the impedance field. The electrodes on the splines are tracked based on impedance technology only. ⁵³ The tracking issues of the different splines possibly are caused by high variations in impedance field in the ventral caudal vena cava. In this specific region, the ablation catheter functioned well, because the ablation catheter is tracked by a combination of magnetic and impedance technology and therefore is less reliant on impedance.

Complications in our study were related to general anesthesia and included triceps myopathy in 1 horse and temporary fever in 2 horses. Myopathy or neuropathy is a commonly reported post‐anesthetic complication in horses. ⁵⁴ Triceps myopathy probably was related to the long duration of general anesthesia (180 minutes) and lateral recumbency. ⁵⁵ In the 2 horses with fever, B‐lines were observed during lung ultrasonography. A recent study showed that B‐lines increased significantly after general anesthesia for elective surgery in horses. ⁵⁶ The most common complications of RFCA in human medicine include vascular complications, hematoma, perforation and cardiac tamponade, thromboembolism or collateral injury such as phrenic nerve palsy or atrio‐esophageal fistula. ⁵⁷ , ⁵⁸ None of these complications were observed in our horses. A recent study described caudal vena cava and pulmonary vein isolation in horses with a power of 45 W and target AI of 450‐500. Ablation lesions were evaluated at necropsy and no adverse effects such as crater formation or tissue disruption were noticed. ¹⁷ Cardiac troponin I is an accurate marker for assessment of myocardial injury after RFCA in humans. ⁵⁹ In 9 of 10 horses, cTnI concentrations exhibited a small increase 4 hours post‐ablation, still within the reference range, and in 1 horse cTnI remained below the limit of quantification. These findings indicate that RFCA seems to be safe and only results in minor local injury, which is the intention of the procedure.

In human patients, thromboembolism is an important complication of RFCA because of precipitation of thrombi on catheters or within sheaths. ⁵⁸ , ⁶⁰ Therefore, anticoagulation was implemented during the procedure by administration of unfractionated heparin at the start and during the procedure, and the sheath and mapping catheter were irrigated with heparinized saline. ¹⁷ , ⁶¹ Activated clotting time was measured every hour in 8 of 10 horses. Based on guidelines used in human medicine, heparin was administered to achieve and maintain an ACT target of 300‐350 seconds. ²⁹ None of the horses experienced adverse effects of hemorrhage or thromboembolism. However, numbers are small and right‐sided thrombi can cause pulmonary emboli that might be clinically silent. More research is needed to establish an optimal anticoagulation protocol for cardiac procedures in horses.

Caudal vena cava isolation using ablation index‐guided RFCA restored sinus rhythm in all horses and did not result in recurrence. However, sample size was small and larger studies with longer follow‐up time are necessary to establish the benefit of this technique. Ours was a non‐randomized study, without comparison with CF only‐guided ablation. Increased operator experience and evolving ablation techniques also may have improved the outcomes. The AI is a proprietary formula, and therefore the results cannot be generalized to other ablation systems.

5. CONCLUSION

The mechanism of AT was a macro‐reentry circuit around a line of conduction block localized in the myocardial sleeves of the caudal vena cava in all horses. Ablation index‐guided RFCA using the CARTO™ 3 system was technically feasible, restored sinus rhythm and was effective to treat the cause of sustained AT in all horses. Besides the use of CF‐sensing catheters and AI guidance to ensure efficient lesion creation, the ablation protocol also was adapted to achieve full isolation of the myocardial sleeves in the caudal vena cava. This new treatment strategy was developed to minimize recurrence risk and resulted in absence of recurrence for at least 9‐37 months post‐ablation.

CONFLICT OF INTEREST DECLARATION

Authors declare no conflicts of interest.

OFF‐LABEL ANTIMICROBIAL DECLARATION

Authors declare no off‐label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION

Authors declare no IACUC or other approval was needed.

HUMAN ETHICS APPROVAL DECLARATION

Authors declare human ethics approval was not needed for this study.

Supporting information

Table S1: General information, electrocardiographic and echocardiographic findings of 10 horses with sustained atrial tachycardia. Dpm, depolarizations per minute.

Table S2: Three‐dimensional electro‐anatomical mapping and ablation data, including procedure duration, type of atrial tachycardia (AT), biophysical ablation parameters, time of follow‐up and recurrence of 10 horses with sustained AT. Ablation data are presented as median [interquartile range].

Buschmann E, Van Steenkiste G, Vernemmen I, et al. Caudal vena cava isolation using ablation index‐guided radiofrequency catheter ablation (CARTO™ 3) to treat sustained atrial tachycardia in horses. J Vet Intern Med. 2025;39(1):e17251. doi: 10.1111/jvim.17251