Slit-based irrigation catheters can reduce procedure-related ischemic stroke in atrial fibrillation patients undergoing radiofrequency catheter ablation

Open irrigation ablation catheters are now the standard in radiofrequency catheter ablation (RFCA) of atrial fibrillation (AF). Among various irrigation catheters, laser-cut slit-based irrigation system (Cool Flex and FlexAbility) has a unique design to cool the catheter tip more efficiently. We aimed to assess the safety of slit-based irrigation catheters regarding prevention of procedure-related ischemic complication in AF patients undergoing RFCA. The analysis was performed with Korea University Medicine Anam Hospital RFCA registry. Procedure-related ischemic complication was defined as ischemic stroke or transient ischemic attack (TIA) occurring within 30 days after RFCA. Patients were divided into 3 groups: non-irrigation, hole-based irrigation, and slit-based irrigation catheter groups. A total of 3,120 AF patients underwent first RFCA. Non-irrigation, non-slit-based irrigation, and slit-based irrigation catheters were used in 290, 1,539, and 1,291 patients, respectively. As compared with non-irrigation and non-slit-based irrigation catheter groups, slit-based irrigation catheter group had significantly older age, higher prevalence of non-paroxysmal AF, large left atrial size, and decreased left atrial appendage flow velocity. The CHA2DS2-VASc score was not different among the 3 groups. Procedure-related ischemic complication occurred in 17 patients (0.54%) with 16 ischemic strokes and 1 TIA event: 5/290 (1.72%), 11/1,539 (0.71%), and 1/1,291 (0.08%) events in non-irrigation, non-slit-based irrigation, and slit-based irrigation catheter groups, respectively (p = 0.001). Slit-based irrigation catheter was superior in direct comparison with non-slit-based irrigation catheter (0.71% vs. 0.08%; p = 0.009). Slit-based irrigation catheters were highly effective in preventing procedure-related ischemic complications.


Introduction
Radiofrequency catheter ablation (RFCA) for symptomatic atrial fibrillation (AF) is an established treatment which is associated with significantly improved quality of life [1]. Recent clinical trials also suggest that RFCA can improve not only quality of life but also hard clinical outcomes including all cause death in patients with both AF and heart failure [2,3]. Claim data based retrospective data also suggest that RFCA is associated with reduced risk of ischemic stroke [4,5]. Despite various benefits of RFCA in AF patients, RFCA is associated with several catastrophic complications such as procedure-related stroke or atrio-esophageal fistula [6,7]. Time interval between occurrence of procedure-related stroke and correct diagnosis due to patient sedation delays appropriate therapy and significant proportion of patients with procedure-related stroke have permanent neurologic consequences. Optimal treatment modalities for procedure-related stroke are also not established since majority of these strokes are due to char formation during radiofrequency energy delivery [8,9]. Char formation during radiofrequency energy delivery is thought to be related with tissue overheating and acute cellular damage [10]. Irrigation catheters were developed in order to reduce tissue overheating and char formation [11][12][13]. Among irrigation catheters, open-irrigation catheters showed lower interface temperature, thrombus, and steam pop compared with closed-loop irrigation catheters indicating greater interface cooling capability [14]. Open-irrigation catheters are further classified according to their shape: 6-holes, 12-holes, 56-holes, and slit-based irrigation system. Slitbased irrigation catheters have the largest irrigation space compared to other open-irrigation catheters. However, whether slit-based irrigation catheters are superior to other open-irrigation catheters in terms of stroke prevention remains largely unknown. We aimed to compare the risk of procedure-related ischemic complication among non-irrigation, non-slit-based irrigation, and slit-based irrigation catheters.

Patients
Consecutive AF patients undergoing RFCA at Korea University Medicine Anam Hospital between June 1998 and April 2019 were analyzed retrospectively. A total of 3,120 patients underwent their first RFCA for AF during the study period. There was no specific exclusion criteria and all patients who underwent their first-time RFCA in our institution were included. This study was approved by the Institutional Review Board of Korea University Medicine Anam Hospital (approval number: 2020AN0165). Written informed consent was waived due to the retrospective nature of the current study. The study protocol adheres to the ethical guidelines of the 2008 Declaration of Helsinki.

Ablation procedure and anticoagulation
The precise protocol for RFCA in our institution is published elsewhere [6]. Pre-procedural anticoagulation was performed with either warfarin or non-vitamin K oral anticoagulant (NOAC). Some patient was not prescribed with anticoagulants. However, we performed transesophageal echocardiography (TEE) in 92.9% of patients to rule out any thrombus or sludge in left atrium (LA) of left atrial appendage (LAA). Emptying, filling, and average flow velocity of the LAA were measured during TEE evaluation. Protocols of our pre-procedural imaging evaluation are published elsewhere [15]. Either computed tomography (CT) or magnetic resonance imaging (MRI) study was performed to assess the anatomy of the pulmonary veins and to create a three-dimensional reconstruction map using either EnSite NavX or CARTO systems. During RFCA, intravenous heparin was administered to maintain activated coagulation time between 300-350 seconds.
After index RFCA, anticoagulation with either warfarin or NOAC was performed for at least two months after the procedure. After two months, anticoagulation therapy was given to patients with CHA 2 DS 2 -VASc score � 2. Anticoagulation was discontinued, at the operator's discretion, if no AF was documented on regular Holter monitoring (every 3 months for the first year and every 6 months thereafter).

Definitions
Procedure-related ischemic complication was defined as any ischemic stroke or transient ischemic attack (TIA) which occurred within 30 days post-RFCA. Immediate procedurerelated ischemic complication was defined as ischemic stroke or TIA within 3 days post-RFCA. Ischemic stroke was defined as any neurologic symptom that persisted for more than 24 hours which could not be explained by other medical conditions. Transient ischemic attack was defined as any neurologic symptom that was not attributable to other medical cause, but resolved completely within 24 hours. If acute infarction was observed on brain CT or MRI, the event was classified as ischemic stroke despite complete restoration of neurologic symptoms. Neurologists confirmed the final diagnosis of ischemic stroke and TIA. Neurologic consequence of procedure-related stroke was classified into three stages: (i) None: no neurologic symptoms; (ii) Minimal: permanent neurologic symptoms which does not limit normal daily life; (ii) Significant: permanent neurologic symptoms which have significant limitation on normal daily life or occupation.

Irrigation system
Irrigation system of ablation catheters were classified into three groups: non-irrigation, nonslit-based irrigation, and slit-based irrigation catheters. Non-slit-based irrigation catheters included closed-loop irrigation catheters and hole-based open-irrigation catheters. Slit-based irrigation catheters were consisted of Cool Flex and FlexAbility catheter (Abbott, Chicago, IL, USA). Flow rate of the irrigation fluid was based on the manufacturer's recommendations.

Statistical analysis
Continuous variables are expressed as mean ± standard deviation. Categorical variables are presented as percentile value. Unpaired t-test was used to compare continuous variables. Categorical variables were compared using either the chi-square test or Fisher's exact test, as appropriate. Multivariate logistic regression analysis was performed to evaluate the impact of different type of irrigation system on procedure-related ischemic complications. Covariates were included in the multivariate model if significant difference was observed in the baseline characteristics or were a known risk factor for ischemic stroke or TIA. Statistical significance was based on p-value less than 0.05. SPSS version 24.0 (SPSS Inc., Armonk, NY, USA) was used for all statistical analyses.

Patients
During June 1998 to April 2019, a total of 3,120 patients underwent their first RFCA for AF with 290 non-irrigation, 1,539 non-slit-based irrigation, and 1,291 slit-based irrigation ablation catheters. Baseline characteristics of the entire cohort are summarized in Table 1. Mean age was 55.74 ± 10.96 years and 78.9% were male. Previous history of thromboembolic events was observed in 8.4% of the patients and mean CHA 2 DS 2 -VASc score was 1.27 ± 1.26.

Procedure-related ischemic complication
Procedure-related ischemic complication occurred in 17 patients: six patients had no permanent neurologic consequences and five patients had permanent neurologic symptoms which significantly limited their normal daily life. Mean CHA 2 DS 2 -VASc score was 1.35 ± 1.12 and 14 patients were male. Eleven (64.7%) patients had non-paroxysmal AF and substrate modification in addition to pulmonary vein isolation was performed in 64.7% of patients. Characteristics of these 17 patients with procedure-related stroke are summarized in Table 2. Baseline characteristics of patients with and without procedure-related stroke are compared in Table 1. Patients with procedure-related ischemic complication had larger LA diameter (44.38 ± 5.44 vs. 41.18 ± 6.04 mm; p = 0.035), higher proportion of non-paroxysmal AF (64.7% vs. 40.8%; p = 0.046).
Distribution of patients who were treated with antiplatelets, warfarin, NOAC, or no antithrombotic agents before and after RFCA is depicted in Fig 1A and 1B. Type of antithrombotic medication did not affect the incidence of procedure-related ischemic complication (p = 0.261 for pre-RFCA and 0.810 for post-RFCA; Fig 2A and 2B).

Discussion
This study revealed that slit-based irrigation catheters can be effective in preventing procedure-related ischemic complication. Non-irrigation catheters showed highest risk for procedure-related ischemic complication and peri-procedural anticoagulation did not influence the risk for procedure-related ischemic complication. This data is the largest study to date analyzing safety issue of different irrigation systems and the first report demonstrating that irrigation systems can affect actual clinical stroke and TIA.

Ischemic complication during RFCA
Together with atrio-esophageal fistula, peri-procedural ischemic complication is most dreaded complication in AF patients undergoing RFCA [16,17]. Since patients are in deep sedation or

PLOS ONE
Importance of the cooling system under general anesthesia, they cannot complain any neurologic symptoms when their cerebral arteries are occluded by thrombus or char. It takes several hours to complete the procedure and to recover from sedation or general anesthesia. Therefore, immediate revascularization was often not feasible. Furthermore, char formation during radiofrequency energy delivery makes harder thrombus material which makes revascularization therapy even more difficult. Although massive thrombus and char embolization during RFCA is a rare event, it is usually unrecoverable once occurred. In our registry, five patient had significant neurologic sequelae defined as permanent neurologic symptoms limiting their normal daily life or occupation. Making ischemic stroke during treatment process of AF is an irony and 0.16% (5 among 3,120 patients) risk of having significant neurologic sequelae cannot be ignored.

Irrigation system
Local temperature elevation at the distal tip of ablation catheter during RFCA can result in temperature-dependent coagulum and char formation [10]. Histological review of these deposits revealed that it was consisted of denaturized and aggregated proteins rather than classical thrombus [10]. Therefore, reducing the temperature of distal electrode tip can potentially reduce char and thrombus formation.
In our registry, patients who were ablated with non-irrigation catheters showed highest risk of having procedure-related ischemic complication (1.72%) followed by non-slit-based irrigation (0.71%) and slit-based irrigation catheters (0.08%). Irrigation catheters were developed for two major reasons: formation of deeper lesions and reduction of ischemic complications [11]. Previous studies demonstrated that irrigation catheters were capable of making deeper lesions as compared with non-irrigation catheters [18,19]. Although irrigation catheters have shown that coagulum formation can be reduced as compared to non-irrigation catheters, whether irrigation catheters can reduce clinical ischemic complications remains largely unknown [11]. Previous study reported that irrigated tip catheter did not reduce the incidence of symptomatic peri-procedural ischemic stroke [20]. However, the study enrolled patients from February 2001 to January 2008 where no slit-based irrigation catheters were available and our study demonstrated a lower incidence of procedure-related ischemic complication in slitbased irrigation catheters. Our study also showed lower incidence of procedure-related ischemic complication in non-slit-based irrigation catheters compared non-irrigation catheters (0.71% vs. 1.72%). The discrepancy between the two studies can be due to small sample size or different type of non-irrigation catheter (predominantly 8mm tip in the study by Scherr et al. and 4mm tip in ours).

Anticoagulation
Previous studies reported that continuation of anticoagulation in peri-operative period can reduce ischemic complication after RFCA [21,22]. However, pre-or post-RFCA anticoagulation did not affect the incidence of peri-procedural ischemic complication in our study. Since embolic material formed by catheter ablation is mainly consisted of denaturized and aggregated proteins rather than classical thrombus, peri-procedural ischemic complication can still occur despite optimal anticoagulation therapy [10]. Indeed, heat induced protein denaturation and aggregation occurred independently from heparin concentration [10]. Nevertheless, optimal anticoagulation therapy should be done in peri-procedural period since denaturized material formed by radiofrequency energy can activate coagulation cascade. Both irrigation and anticoagulation are essential.

Limitations
Our results are based on retrospective analysis and therefore, are not free from intrinsic limitation of such analysis. Differences in baseline characteristics such as prevalence of vascular disease, AF type, LA diameter, and LAA flow velocity were observed between slit-based vs. non-slit based catheter groups although it was adjusted in the multivariate model. Unmeasured confounders might exist in our cohort. Systematic variations in the ascertainment of stroke or TIA over time might exist since our cohort extends more than 20 years. However, all stroke or TIA events were diagnosed by neurologists and this was centrally reviewed during data collection process. Although total patient number was large, the number of periprocedural ischemic complication was small limiting statistical power. Slit-based irrigation catheters do not have contact-force sensing capability. Therefore, ablation parameters such as ablation index or lesion size index which has shown to improve ablation outcome cannot be utilized when using slit-based irrigation catheters [23,24]. This can be a critical limitation of slit-based irrigation catheters and contact-force sensing ability should be added in the near future.

Conclusion
Slit-based irrigation catheters may reduce the risk of peri-procedural ischemic complication compared with non-irrigation or non-slit-based irrigation catheters. Our results should be tested in future clinical trials.
Supporting information S1 Fig. Incidence of peri-procedural ischemic complication according to catheter type. (PDF) S1