The handle http://hdl.handle.net/1887/55945 holds various files of this Leiden University dissertation
Author: Compier Marieke
Title: Atrial fibrillation ablation : balancing between treatment efficacy and complications
Date: 2017-09-14
Chapter 4
Asymptomatic Cerebral Embolism following PVAC and Irrigated-tip Ablation for Atrial Fibrillation: Incidence, Potential Causes and Clinical Implications. Results from the CE-AF Trial Pilot.
Compier MG, Bruggemans EF, van Buchem MA, Middelkoop HAM, Eikenboom J, van der Hiele K, Zeppenfeld K, Schalij MJ, Trines SA
Submitted
AbstrACt
Purpose. Catheter ablation for atrial fibrillation (AF) has been indicated as a risk factor for the occurrence of asymptomatic cerebral embolism (ACE). The causes and potential clinical implications are largely unknown, however. The aim of this pilot was to study risk factors for ACE incidence and neuropsychological consequences during AF ablation with an irrigated-tip catheter (Thermocool) and a non-cooled, decapolar duty-cycled catheter (PVAC).
Methods and results. Fifteen patients with paroxysmal AF were prospectively random- ized in a 2:1 fashion to PVAC or Thermocool ablation. A diffusion-weighted cerebral MRI and extended neuropsychological testing were performed one day before and after the procedure. Blood samples were obtained for determination of endothelial damage and coagulation. Peri-procedural microembolic signals (MES) were evaluated by Transcranial Doppler ultrasonography. A 3-month follow-up MRI was performed in case of a postpro- cedural lesion on cerebral MRI.
Baseline characteristics were comparable between both groups. In the Thermocool- group, no new MRI lesions were detected. In the PVAC-group, 2 patients developed a new microembolus (<5 mm) and 1 patient a microbleed, all located in the cerebellum. A follow-up MRI showed remainder lesions of the micro-emboli.
Total MES duration was higher during ablation with PVAC compared to Thermocool (98.06±58.78 versus 37.96±12.69 sec, p=0.02). Von-Willebrand factor antigen tended to increase (1.05 to 1.3 IU/ml) and prothrombin time tended to decrease (41.2 to 37.3 sec) after PVAC , while remaining stable after Thermocool. Psychomotor test scores remained stable after PVAC ablation (32.8 to 32.7), while increasing after Thermocool possibly due to the learning effect (33.8 to 36.8, p=0.02).
Conclusions. Ablation with PVAC led to a 30% incidence of new lesions on MRI, longer total duration of MES, increased endothelial damage, a tendency towards increased coagulation and a diminished psychomotor functioning compared to Thermocool abla- tion. Altogether, the thromboembolic risk of the PVAC catheter is increased compared to the irrigated-tip catheter. The diminished psychomotor function after PVAC ablation could possibly be explained by the occurrence of the ACE in the cerebellum, which is involved in coordination of motor functioning.
4 IntroduCtIon
Catheter ablation aiming at pulmonary vein isolation (PVI) is a cornerstone procedure for the treatment of symptomatic, drug-refractory AF.1 The risk for symptomatic cerebral thromboembolic complications, including stroke and transient ischemic attack (TIA), is approximately 0.5-1%.2 The incidence of asymptomatic cerebral emboli (ACE) however, as shown with pre- and post-procedural diffusion weighted cerebral MRI, is increased more than tenfold compared to the incidence of symptomatic emboli.3-6 Among the several single-shot devices developed to increase ablation efficacy and decrease pro- cedural times, the non-cooled duty-cycled multi-electrode ablation catheter (PVAC) is widely used with acceptable outcomes.7 However, the ACE incidence on DW-MRI after PVAC ablation was found to be even four times higher compared to the incidence after cooled-tip ablation.3, 5
A prospective trial investigating potential causes of ACE after PVAC ablation indicated several procedural changes that reduced ACE incidence, including performance of the procedure under uninterrupted coagulation to prevent formation of thrombotic emboli, prevention of interaction between electrode 1 and 10 to reduce the formation of micro- bubbles and loading the catheter under water to prevent introduction of air emboli.8 Combining these changes led to a reduction of ACE to only 1.7%.8 Another study still showed an 11% ACE incidence with application of these changes, however.9 Other factors leading to ACE formation therefore seem to be involved. Indeed, both the induction of a procoagulant state and the production of gaseous microemboli measured with transcra- nial Doppler were found to be elevated during PVAC ablation and further evaluation of these factors may reveal other potential causes for thromboembolic complications.10, 11
Despite the high incidence, formation of ACE usually does not result in neurological symptoms or abnormalities on global neurological testing.6 Extended neuro(psycho) logical testing instead of global neurological investigation may reveal otherwise undis- covered subtle clinical effects of newly formed ACE after ablation.12
So far, no randomized controlled trial has been performed comparing the ACE in- cidence between different ablation catheters combined with thorough evaluation of causes and possible clinical implications of the newly formed ACE.
The aim of this pilot study was to thoroughly evaluate all aspects considering devel- opment and consequences of ACE formation during catheter-guided AF ablation. The following aspects were studied: 1. The incidence of ACE after PVAC ablation and after cooled-tip ablation was compared in patients randomized to the two treatments, 2.
The incidence of cerebral microembolic signals (MES) during ablation was compared between the two groups, 3. The induction of a procoagulant state after cooled-tip abla- tion and ablation with PVAC was evaluated and 4. The influence of newly developed ACE on the performance of patients was evaluated using extended neuropsychological tests.
Methods
Patient population
Patients scheduled for a first ablation of drug-refractory paroxysmal AF were included in this study. Patients were randomized in a 2:1 fashion to ablation with PVAC or with the Thermocool catheter. Exclusion criteria included: age below 18, previous surgical or catheter AF ablation, a contraindication for cerebral DW-MRI or the inability to perform neuropsychological testing due to mental retardation. All patients gave written informed consent before enrolling the study. Patients were pre-treated with Vitamin K antagonists for at least 2 months prior to ablation. Patients were followed-up in the outpatient clinic at 1, 3, 6, and 12 months after the procedure according to standard clinical care. The pilot is part of a larger trial currently including patients and was approved by the local ethical review board and registered at clinicaltrials.gov: NCT01361295.
Ablation procedure
During hospitalization, oral anticoagulation was continued in order to maintain an inter- national normalized ratio between 2.0 and 3.0. The ablation procedure was performed under local anaesthesia and intravenous analgesia. After venous catheterisation, 5000 IU heparin was administered. Single transseptal access using the Bard 9F Channel sheath (for the PVAC procedures, Bard Electrophysiology Division, Lowell, MA, USA) or double transseptal access using 8.5F SL0 sheaths (for the cooled-tip procedures, St. Jude Medi- cal, St. Paul, MI, USA) was obtained under guidance of intracardiac echocardiography (ICE) and subsequent pulmonary venography was performed. During the procedure, an activated clotting time (ACT) was maintained above 300 sec with additional heparin.
A reference catheter was placed in the coronary sinus (CS). Ostial ablation with the PVAC catheter was performed with a multi-channel RF generator delivering RF energy in a duty-cycled manner in a bipolar : unipolar ratio of either 4:1 or 2:1, creating both adjoining and transmural lesions (Genius 14.4, Medtronic Ablation Frontiers). Each RF application was continued for 60 s. The standard temperature-control mode was ap- plied with a maximum power of 8 W in case of 4:1 energy delivery and 10 W in case of 2:1 energy delivery. Circumferential ablation was performed with the cooled-tip catheter with a power of 30-35 W to reach a maximum temperature of 43ºC (Navistar Thermocool, Biosense Webster, Diamond Bar, CA, USA). Sheaths were removed after normalisation of ACT. No protamine was administered. Vitamin K antagonists were continued until at least three months after the ablation procedure in patients with a CHA2DS2-VASc score
<2 and indefinitely in patients with a CHA2DS2-VASc score of ≥ 2.
4 diffusion weighted cerebral MrI
MRI (1.5 Tesla, Philips, The Netherlands) was performed the day before and the eve- ning after the ablation using a dedicated head coil. The imaging protocol included a diffusion-weighted single-shot spin echo echoplanar sequence, turbo fluid attenuated inversion recovery, T2-weighted turbo spin echo sequences and susceptibility-weighted MRI (T2*) for the detection of cerebral microbleeds. Diffusion gradients were applied in three orthogonal directions during diffusion-weighted imaging. Acute embolic lesions were defined as focal diffusion abnormalities in a cortical or subcortical location, or in the vascular territory of the perforating arteries. The size and localization of focal diffu- sion abnormalities were analysed. In addition, apparent diffusion coefficient (ADC) maps were calculated based on the diffusion-weighted data in conjunction with the isotropic diffusion-weighted images. They were used for image interpretation in equivocal cases (“T2 shine through” in primary diffusion-weighted data). Patients with a focal diffusion abnormality underwent follow-up MRI using the same MRI protocol 3 months later to assess the development of structural tissue changes at the site of the index lesion to confirm the irreversibility of the previous diffusion abnormality.
Cerebral embolism after ablation was diagnosed as a new diffusion abnormality on DW-MRI of the brain. MRI scans were evaluated by 2 independent radiologists. In case of different results, consensus was sought. If consensus could not be reached, a third radiologist was consulted.
transcranial doppler
Transcranial Doppler monitoring during ablation was performed by insonation of the right middle cerebral artery (MCA) from the transtemporal window at a depth of 45 to 55 mm, using a 2 MHz pulsed wave Doppler transducer (DWL Multi Dop P, DWL Sipplingen, Germany). In case of left dominancy, the blood flow of the left MCA was examined. The Doppler probe was fixed on the skull with a headband to minimize movement artefacts.
Power and gain settings were adjusted to the lowest level to obtain blood flow spectra of low intensity. Patients were monitored continuously starting from 15 minutes before transseptal puncture until sheath removal from the left atrium. High-intensity transient signals were registered manually throughout the procedure in conjunction with the cor- responding procedural events. The raw Doppler signals were recorded on a 24-bit Wave/
MP3 recorder for off-line analysis (Eridol R-09, Roland Corporation Nakagawa, Japan).
Detection of cerebral micro-embolic signals (MES) was performed by automated detec- tion analysis. Both the amount and total duration of MES were registered. Analysis was performed for the total procedure time and for different time periods. Four periods were defined. Period 1: transseptal puncture, period 2: mapping of the left atrium (cooled-tip catheter) or contrast venography of the pulmonary veins (PVAC), period 3: ablation, period 4: waiting time after ablation before sheath removal from the left atrium.
laboratory measurements
The day before the procedure 10 ml of citrated (3.2%) blood was collected (sample 1).
Right before ablation another blood sample was collected (sample 2). A blood sample was drawn immediately after ablation (sample 3). The day after ablation, the final blood sample (sample 4) was drawn.
Laboratory measurements for coagulation parameters were performed in blood samples 1-4. VWF:Ag was measured by in-house developed ELISA using polyclonal anti- bodies. D-dimer was determined using STA Liatest D-dimer (Roche, Diagnostica Stago).
Prothrombin fragment 1+2 was determined by sandwich-type ELISA (Dade-Behring, Marburg, Germany). Fibrinogen was determined according to the Claus method. Throm- bin generation was assessed using the Calibrated Automated Thrombogram (CAT) (Thrombinoscope). APTT and PT-INR were measured by standard methods. VWF:Ag was measured as a marker for endothelial damage.
neuro(psycho)logical assessment
The neuro(psycho)logical status of the patients before and after ablation was assessed using questionnaires combined with a battery of cognitive function tests. This test bat- tery had a specific focus on memory, attention and concentration, executive function- ing and psychomotor speed. Global cognitive functioning was assessed with the Mini Mental State Examination; memory functioning was tested using the Verbal Learning and Complex Figure Tests of Rey; attention and concentration was tested using the Sustained Attention to Response Test; executive functioning was examined using the Stroop Color Word Test, Figure Fluency Test and Trail Making Test part B and psychomo- tor speed was estimated with the Letter Digit Substitution Test and Trail Making Test part A. The following questionnaires were sent to the participants: the Leiden Neuro(psycho) logical Status Inventory, the Hospital Anxiety and Depression Scale and the Neuropsy- chiatric Inventory Scale.
statistical analysis
Data were analyzed using SPSS (version 20.0, SPSS Inc., Chicago, IL, USA). Continuous data were expressed as mean ± standard deviation. All continuous parameters were checked for normality with the Shapiro-Wilk test or the Kolmogorov-Smirnov test. Base- line characteristics between groups were compared with an unpaired Student t-test or a Mann-Whitney U test for non-parametric data. A chi-square test was performed for binary data or a fisher’s exact test for binary data with low expected count.
Results from the ultrasonography measurements were compared between both groups with an independent students t-test. The results from the neuropsychological and laboratory tests between both groups were evaluated with a mixed linear model (GLM) for repeated measures (pre- and post-ablation) and independent samples (PVAC
4 vs. Thermocool). When the GLM showed a significant difference between both types of
ablation and a significant interaction, test results were evaluated within each group with a paired t-test. A p-value of <0.05 was considered statistically significant.
results
Patient and procedural characteristics
Fifteen patients were randomized to ablation with PVAC (PVAC group, n=10) or with the Thermocool catheter (Thermocool group, n=5). Baseline and procedural characteristics were comparable between both groups (table 1 ). One patient from the PVAC group had an INR of 1.1 before ablation. This patient underwent transesophageal echocardiogra- phy to rule out the presence of a thrombus. Low molecular weight heparin was supplied subcutaneously directly after sheath removal until an INR of at above 2 was obtained.
The number of patients that underwent peri-procedural cardioversion was significantly higher in the Thermocool group compared to the PVAC group. Procedural and fluoros- copy times were significantly longer for the Thermocool ablation.
table 1. baseline characteristics
baseline parameters Thermocool (n=5) PVAC (n=10) p-value
Age (years, ± SD) 58 ± 8 59 ± 8 0.81
Male gender, n (%) 5 (100) 7 (70) 0.51
Hypertension, n (%) 2 (40) 4 (40) 1.00
Coronary artery disease, n (%) 1 (20) 0 (0) 0.33
LA-diameter (mm) 41 ± 5 41 ± 6 0.89
INR 2.3 ± 0.6 2.2 ± 0.6 0.67
CHA2DS2-VASc 1.0 ± 0.8 1.0 ± 1.2 0.82
Procedural parameters
Fluoroscopy time (min) 40 ± 17 23 ± 8 0.02
Duration entire procedure (min) 257 ± 31 181 ± 44 <0.01
ACT (seconds) 383 ± 20 362 ± 25 0.11
Cardioversion % 5 (100) 4 (40) 0.04
Incidence of new cerebral lesions
In the Thermocool-group, no new MRI lesions were detected after ablation. In the PVAC- group, 2 patients developed a new microembolus (<5 mm, figure 1B) and 1 patient showed a microbleed after ablation. The microbleed was located in the right cerebellum and showed no diffusion defect. The microemboli were located in the left cerebellum.
The patient from the PVAC group with a low INR before ablation did not develop an ACE
lesion on MRI after ablation. At 3 months follow-up, MRI showed remainder lesions in the 2 patients with micro-embolism. These results were confirmed by 2 independent radiologists who were blinded to procedure type.
Periprocedural microembolic signals
The amount of MES was low during sheath introduction, increased during venography and was highest during energy delivery. An example of MES distribution in a PVAC patient is shown in figure 2.
The total duration of MES was significantly higher during ablation with PVAC (98.0±58.78 sec) as compared to Thermocool ablation (37.96±12.69 sec, p=0.02, figure 3). When comparing the total duration of MES between the patients with and without a new cerebral lesion on MRI, the total duration showed a tendency to be longer in patients with a new lesion, although this did not reach significance (66.28 versus 11.44 sec, p=0.19).
figure 1. pre- (A) and post-procedural (B) diffusion-weighted MRI with new ACE lesion in left cerebellum on post-ablation scan (ACE=asymptomatic cerebral embolism)
figure 2. example of MES incidence during the ablation procedure in 1 patient
4
laboratory measurements
The changes in laboratory measurements between both ablation groups showed no significant difference between both groups and no significant interaction. In both groups prothrombin time (PT) significantly decreased after ablation from 38.3 to 35.5 sec (p=0.03). INR significantly decreased from 2.7 to 2.5 (p=0.02). Von-Willebrand factor antigen (VWAG) significantly increased from 1.02 to 1.23 (p<0.01). APTT, fibrinogen, D- dimer and Factor 1+2 remained unchanged.
Examining the data in the PVAC and Thermocool groups separately, VWAG showed a tendency to increase (from 1.05 to 1.3 IU/ml) and PT and INR to decrease (from 41.2 to 37.3 sec and from 2.9 to 2.62 respectively) after PVAC while they remained stable after cooled-tip ablaton (VWAG from 0.99 to 1.16, PT from 35.5 to 33.7, INR from 2.5 to 2.38).
Fibrinogen had a tendency to increase after PVAC (from 2.73 to 2.91 g/l), while it had a tendency to decrease after Thermocool ablation (from 2.88 to 2.64 g/l). Prothrombin factor 1+2 had a tendency to decrease after ablation with Thermocool, (from 0.58 to 0.41) while remaining stable after PVAC ablation (from 0.47 to 0.49, p=0.72).
neuro(psycho)logical tests
The raw scores of psychomotor test remained stable after PVAC-ablation (Letter-Digit Substitution Test, from 32.8 to 32.7 correct answers in 60 seconds), while increasing after Thermocool (from 33.8 to 36.8 correct answers). These results were significantly differ- ent between both groups (p=0.02). The other tests all showed comparable results for both groups. When comparing the results before and after ablation taking both groups together showed that the scores on cognitive domains increased, while depression and anxiety remained unchanged (table 2) .
figure 3. mean total duration of MES in seconds during the entire procedure (SE=standard error of the mean, significant between-group differ- ence, p=0.016)
dIsCussIon
Main findings
This pilot study investigated the thromboembolic risk and the clinical effects of newly formed ACE after AF ablation with a duty-cycled multi-electrode catheter (PVAC) compared to ablation with a cooled-tip catheter (Thermocool). This study is the first to describe a specific correlation between ACE formation and neuropsychological dysfunction. The major findings considering the predefined aims were: 1. The amount of MES during PVAC ablation was significantly higher than during Thermocool ablation (p=0.02), 2. Ablation with PVAC resulted in a tendency towards induced endothelial damage and a procoagulant state and 3. Ablation with PVAC led to a subtle decrease in psychomotor functioning, which could be related to the newly formed ACE that were located in the cerebellum.
Incidence of asymptomatic cerebral emboli for different catheters
Schrickel et al. first reported the incidence of post-procedural asymptomatic cerebral embolism on DW-MRI after PVI with cooled RF, in concordance with ACE found after other procedures including left-sided heart catheterization.4 The incidence of clinically asymptomatic cerebral emboli after ablation was 11% (6/53 patients). Gaita et al. com- pared the incidence of ACE for 3 different ablation catheters with pre- and postproce- dural DW-MRI.3 The incidence after ablation with Thermocool was 8.3% (3/36 patients), after PVAC 38.9% (14/36 patients) and 5.6% (2/36 patients) after ablation with the cryoballoon catheter. PVI performed with PVAC increased the risk for ACE with an odds ratio of 1.48 compared to the other 2 catheters. Another study also compared the ACE incidence after PVI with the Thermocool catheter, PVAC and cryoballoon.5 A pre- and post-procedural MRI was performed and showed new lesions in 7.4% of patients after Thermocool ablation (2/27 patients), 4.3% after cryoballoon ablation (1/23 patients) and 37.5% after PVAC ablation 9/24 patients). In line with these findings, we reported a high incidence of ACE after ablation with PVAC (30%) and a low incidence after Thermocool ablation (0%).
definition and timing of ACe imaging
It is important to mention that several studies used different definitions for ACE. The studies by Gaita et al. and Herrera-Siklody et al. performed a DW-MRI 1-2 days after ablation and defined ACE as a combination of a positive lesion on diffusion weighted MRI, which is a marker for acute ischemia, and a positive FLAIR (fluid attenuated inverse recovery sequence) sequence, which marks the volume of a DW-MRI lesion.13 Schrickel et al. on the other hand, defined ACE as a lesion visible on DW-MRI only at one day after ablation. DW-MRI is used for diagnosing cerebral ischemia since it turns positive within
4 minutes after a lesion is formed.14 FLAIR sequences turn positive after approximately at
least 24 hours or even longer after lesion formation and may therefore be more useful to identify more chronic lesions.15 Identification of newly formed cerebral emboli with FLAIR sequences should be done with MRI obtained 2-7 days after ablation in order to prevent underestimating the amount of ACE,13 Identification of ACE with DW-MRI can be performed within minutes up to 1-2 days after ablation. In our study, ACE was defined as a new lesion on DW-MRI performed hours after ablation, which was therefore within the proper time frame.
factors influencing ACe occurrence
Several studies reported on factors influencing ACE incidence. Some procedural chang- es, including performance of the procedure under uninterrupted anticoagulation and maintenance of an ACT >300 seconds, were found to reduce ACE incidence and were incorporated in this study.9 Cardioversion during the ablation procedure was also found to increase the risk of ACE.16 In the current study, cardioversion was performed more often during Thermocool ablation but no ACE was reported in this group. Therefore, performance of a peri-procedural cardioversion did not explain the higher occurrence of ACE after PVAC. Submerged loading of the catheter may also reduce ACE incidence by decreasing the formation of microemboli.8 During this study, the catheter was not loaded under water. The incidence of MES was found to be highest during energy deliv- ery however and not during sheath introduction.
Further research aiming on elucidating potential causes of ACE on cerebral MRI after PVAC ablation indicated that ablation with electrode 1 (E1) and 10 (E10) in close proxim- ity can lead to bipolar radiofrequency interaction, potentially leading to a higher risk of new ACE formation.8 Wieczorek et al. investigated the relationship between procedural proximity of E1 and E10 and new ACE on DW-MRI.9 All included patients (n=37) under- went the ablation procedure under comparable circumstances, including uninterrupted oral anticoagulation, ACT > 300 sec and submerged PVAC introduction into the sheaths.
E1-E10 interaction was registered during approximately half of the procedures (49%).
The amount of newly formed ACE observed on post-procedural MRI 1 day after abla- tion was much lower in the patient group without E1-E10 interaction compared to the group with E1-E10 interaction (11% vs. 44%, p=0.029). A follow-up study by this group, including 120 patients undergoing PVAC ablation, showed that 20% of patients still had newly formed ACE when E1-E10 interaction was prevented by prohibiting simultaneous activation of electrode pair 1 (E1 and E2) or 5 (E9 and E10).17 The ACE incidence was 28.3% when a maximum of 4 electrode pairs were activated simultaneously and de- creased to 11.7% when a maximum of only 2 electrode pairs were activated at the same time. A more recent trial aimed at reducing the ACE incidence, the ERACE trial, evaluated the ACE incidence, with ACE being defined as a lesion that is shown both on DW-MRI
and on FLAIR imaging, after a number of procedural changes including performance of the procedure under uninterrupted coagulation aiming at a continuous peri-procedural ACT > 350 ms.8 The interaction between electrode 1 and 10 was prevented by switching off one of the electrode pairs. Finally, submerged loading of the catheter was performed.
Combining these changes led to a significant reduction of newly formed post-procedural ACE on DW-MRI and FLAIR 24-48 hours (mean 28.2 hours) after ablation with PVAC, with an incidence of only 1.7% (n=1 of 60 patients). As the scans were performed at a mean delay of 28 hours after the procedure and a positive FLAIR sequence was required for the diagnosis, the incidence of ACE may have been underestimated, however. Indeed, another study by Wieczorek et al. showed that with application of the same changes, a substantial amount of patients still showed newly formed ACE on post-procedural MRI performed within 24 hours after ablation, with ACE being defined as a new lesion on DW-MRI only (11%).9 Besides the difference in ACE definition, the only procedural differ- ence between the ERACE trial and the study by Wieczorek et al. was the attempted ACT, which was >350 sec and > 300 sec respectively. Wieczorek et al. evaluated whether a subclinical ACT (<300 sec) was related to formation of an increased number of ACE, but this was not the case.
factors influencing Mes formation
In a study comparing PVAC with cryoballoon ablation, the amount of cerebral MES was found to be higher after PVAC ablation when compared to cryoballoon ablation (p<0.01), even when the ACT was above 350 seconds during the entire procedure.11, 18 The composition of these emboli were found to be mainly gaseous. A possible explana- tion for the increased amount of MES was also found to be E1-E10 interaction.19 Indeed, when E1-E10 interaction was prohibited the amount of MES significantly decreased.20 Altogether, E1-E10 interaction seems to be a risk factor for the formation of MES and ACE and prevention of interaction reduces the incidence of both. In this study, no E1-E10 deactivation occurred. The formation of MES and consequently possible ACE formation may thus have been influenced by electrode 1 and 10 interaction. The formation of MES during irrigated radiofrequency ablation was also found to be influenced by tis- sue temperature.21 Since the electrodes of PVAC are non-cooled, higher temperatures are reached compared to the cooled Thermocool catheter. This may also be a (partial) explanation for the higher MES incidence after PVAC ablation found in this study.
Not much is known about the relation between the incidence of MES and the occur- rence of ACE after AF ablation. In contrast, in stroke patients a positive association be- tween the formation of MES and new lesions on cerebral DW-MRI has been described.22 In this study, patients with acute stroke (n=37) underwent TCD to detect newly formed MES. The number of small ischemic lesions on DW-MRI was significantly higher in patients who were MES positive during TCD compared to patients that did not show
4 any MES during TCD (p=0.02), Since our study showed a higher incidence of MES dur-
ing PVAC ablation compared to Thermocool ablation, this may have led to the higher incidence of ischemic lesions on DW-MRI in patients that underwent ablation with PVAC.
Induction of a procoagulant state
A recent study comparing markers of a procoagulant state and endothelial damage after PVAC and cryoballoon showed that endothelial damage was more prominent after PVAC ablation while activation of coagulation increased after both ablation procedures.10 Vitamin K antagonists were withdrawn 2-3 days before ablation in this study and hepa- rin was used to bridge the procedure. During the ablation procedure, ACT levels were kept between 250 and 350 seconds. Although ablation was performed under vitamin K antagonists and with a continuous ACT level > 300 seconds in our study, markers of a procoagulant state still showed a tendency to increase after ablation in the PVAC group, while they remained stable after Thermocool ablation. Also, endothelial damage showed a tendency to increase after ablation with PVAC. Thrombus formation induced by radiofrequency energy has previously been described, probably caused by the dis- ruption of the endothelium. This may have contributed to thromboembolic events after PVAC ablation.
Clinical implications of ACe resulting from Af ablation
Schwarz et al. investigated the effect of AF catheter ablation on neuropsychological performance at 3 months after ablation and compared the performance to the results of AF patients without ablation.23 New ischemic lesions on post-procedural MRI after PVI with the Thermocool catheter were documented in 3/21 patients. Patients that underwent ablation showed neuropsychological decline at 3-months follow-up when compared to AF-patients without ablation, while baseline performance was comparable between both groups. For the patients with a new lesion, one patient was excluded due to a severe brain infarction. Of the other 2 patients, one showed in a mild decrease in neuropsychological performance, while the other patient showed an increase in perfor- mance after ablation. Medi et al. evaluated the effect of AF-ablation on neurocognitive functioning after ablation.12 At 90-day follow-up, 13% of patients showed subtle post- operative neurocognitive dysfunction compared to AF-patients without ablation. The incidence of new lesions on MRI was not investigated during this study, so no relation could be detected with potentially newly formed ACE. Herm et al. showed that the oc- currence of ischemic lesions after cryoballoon ablation or ablation with a MESH catheter did not influence neuropsychological performance at 6 months follow-up.24
Kochhäuser et al. evaluated the influence of ablation technique and MES formation on neurocognitive functioning by extensive neuropsychological testing comparable to the test battery used in this study.25 Eighteen NPO test parameters were evaluated before
and after ablation with either PVAC (n=40) or irrigated RF (n=36). PVAC ablation caused an increased amount of peri-procedural MES, which was in line with the results of our study. An increased number of MES was significantly associated with impaired results on 4 of the 18 NPO tests. No DW-MRI was performed however, so whether the increased amount of MES resulted in the formation of ACE, which in turn may have led to reduced scores on neuropsychological functioning is unknown.
In line with the studies from Medi et al. and Schwarz et al., this study found a subtle difference in post-procedural psychomotor functioning between both groups indicat- ing better results after Thermocool ablation. The location of the new MRI lesions, the cerebellum, is indeed involved in coordination of motor functioning and might be a possible explanation for the difference between both groups. Since the same tests were executed pre- and post-ablation, patients were expected to perform better during the second test battery due to a learning curve. No increase was found for psychomotor functioning after PVAC ablation however, despite the learning curve. This may have been caused by the formation of ACE located in the cerebellum.
limitations
This study was single-center and only discusses the results of the pilot of the study, including a total of 15 patients. Taking predefined risk factors for ACE formation into account, no submerged loading of catheters occurred, and E1-E10 interaction was not prohibited by deselecting electrode pair 1 or 5. This could have resulted in a higher ACE incidence. Also, the attempted ACT was > 300 ms instead of 350 ms, which also could have been a risk factor for ACE formation. Despite of this, the median ACT-levels eventu- ally were > 350 ms however.
Although the number of included patients was small, significant differences between the effect of PVAC ablation compared to Thermocool ablation could already be found.
Ablation with PVAC resulted in shorter procedure and fluoroscopy times, which is an ad- vantage over other procedures. The thromboembolic risk was also increased however.
The full prospectively randomized controlled trial should and will be performed with the currently available PVAC Gold catheter to confirm the results described in this study.
ConClusIons
Ablation with PVAC led to a 30% incidence of new lesions on MRI while no new lesions were found after ablation with the Thermocool catheter. When evaluating potential causes of these ACE, longer duration of MES, increased endothelial damage and a tendency towards increased coagulation were found after PVAC ablation compared to Thermocool ablation. A small decrease in psychomotor functioning was observed after
4 PVAC ablation compared to Thermocool ablation. Altogether, the thromboembolic risk
of the PVAC catheter seems to be increased compared to the irrigated-tip catheter. Al- though several procedural adjustments have been suggested for PVAC ablation, leading to a decrease in ACE incidence, this complication still occurs and needs to be considered when developing new ablation catheters and before performing an ablation procedure, especially in patients with an increased risk for thromboembolic complications.
referenCe lIst
1. Camm AJ, Kirchhof P, Lip GY, Schotten U, Savelieva I, Ernst S, Van Gelder IC, Al-Attar N, Hindricks G, Prendergast B, Heidbuchel H, Alfieri O, Angelini A, Atar D, Colonna P, De CR, De SJ, Goette A, Gorenek B, Heldal M, Hohloser SH, Kolh P, Le Heuzey JY, Ponikowski P, Rutten FH. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Europace 2010 Octo- ber;12(10):1360-1420.
2. Cappato R, Calkins H, Chen SA, Davies W, Iesaka Y, Kalman J, Kim YH, Klein G, Natale A, Packer D, Skanes A, Ambrogi F, Biganzoli E. Updated worldwide survey on the methods, efficacy, and safety of catheter ablation for human atrial fibril- lation. Circ Arrhythm Electrophysiol 2010 February;3(1):32-38.
3. Gaita F, Leclercq JF, Schumacher B, Scaglio- ne M, Toso E, Halimi F, Schade A, Froehner S, Ziegler V, Sergi D, Cesarani F, Blandino A.
Incidence of silent cerebral thromboem- bolic lesions after atrial fibrillation ablation may change according to technology used:
comparison of irrigated radiofrequency, multipolar nonirrigated catheter and cryo- balloon. J Cardiovasc Electrophysiol 2011 September;22(9):961-968.
4. Schrickel JW, Lickfett L, Lewalter T, Mitt- man-Braun E, Selbach S, Strach K, Nahle CP, Schwab JO, Linhart M, Andrie R, Nickenig G, Sommer T. Incidence and predictors of silent cerebral embolism during pulmo- nary vein catheter ablation for atrial fibril- lation. Europace 2010 January;12(1):52-57.
5. Herrera SC, Deneke T, Hocini M, Lehrmann H, Shin DI, Miyazaki S, Henschke S, Fluegel P, Schiebeling-Romer J, Bansmann PM, Bourdias T, Dousset V, Haissaguerre M, Arentz T. Incidence of asymptomatic in- tracranial embolic events after pulmonary vein isolation: comparison of different
atrial fibrillation ablation technologies in a multicenter study. J Am Coll Cardiol 2011 August 9;58(7):681-688.
6. Neumann T, Kuniss M, Conradi G, Janin S, Berkowitsch A, Wojcik M, Rixe J, Erkapic D, Zaltsberg S, Rolf A, Bachmann G, Dill T, Hamm CW, Pitschner HF. MEDAFI-Trial (Mi- cro-embolization during ablation of atrial fibrillation): comparison of pulmonary vein isolation using cryoballoon technique vs. radiofrequency energy. Europace 2011 January;13(1):37-44.
7. Malmborg H, Lonnerholm S, Blomstrom P, Blomstrom-Lundqvist C. Ablation of atrial fibrillation with cryoballoon or duty-cycled radiofrequency pulmonary vein ablation catheter: a randomized controlled study comparing the clinical outcome and safety; the AF-COR study. Europace 2013 November;15(11):1567-1573.
8. Verma A, Debruyne P, Nardi S, Deneke T, DeGreef Y, Spitzer S, Balzer JO, Boersma L.
Evaluation and reduction of asymptomatic cerebral embolism in ablation of atrial fi- brillation, but high prevalence of chronic silent infarction: results of the evaluation of reduction of asymptomatic cerebral embolism trial. Circ Arrhythm Electrophysiol 2013 October;6(5):835-842.
9. Wieczorek M, Lukat M, Hoeltgen R, Condie C, Hilje T, Missler U, Hirsch J, Scharf C.
Investigation into causes of abnormal cerebral MRI findings following PVAC duty-cycled, phased RF ablation of atrial fibrillation. J Cardiovasc Electrophysiol 2013 February;24(2):121-128.
10. Malmborg H, Christersson C, Lonnerholm S, Blomstrom-Lundqvist C. Comparison of effects on coagulation and inflammatory markers using a duty-cycled bipolar and unipolar radiofrequency pulmonary vein ablation catheter vs. a cryoballoon cath- eter for pulmonary vein isolation. Europace 2013 June;15(6):798-804.
4
11. Sauren LD, Van Belle Y, De Roy L, Pison L, La Meir M, Van Der Veen FH, Crijns HJ, Jordae- ns L, Mess WH, Maessen JG. Transcranial measurement of cerebral microembolic signals during endocardial pulmonary vein isolation: comparison of three different ablation techniques. J Cardiovasc Electro- physiol 2009 October;20(10):1102-1107.
12. Medi C, Evered L, Silbert B, Teh A, Halloran K, Morton J, Kistler P, Kalman J. Subtle post- procedural cognitive dysfunction after atrial fibrillation ablation. J Am Coll Cardiol 2013 August 6;62(6):531-539.
13. Deneke T, Jais P, Scaglione M, Schmitt R, DI BL, Christopoulos G, Schade A, Mugge A, Bansmann M, Nentwich K, Muller P, Krug J, Roos M, Halbfass P, Natale A, Gaita F, Haines D. Silent cerebral events/lesions related to atrial fibrillation ablation: a clini- cal review. J Cardiovasc Electrophysiol 2015 April;26(4):455-463.
14. Di Biase L, Gaita F, Toso E, Santangeli P, Mohanty P, Rutledge N, Yan X, Mohanty S, Trivedi C, Bai R, Price J, Horton R, Galling- house GJ, Beheiry S, Zagrodzky J, Canby R, Leclercq JF, Halimi F, Scaglione M, Cesarani F, Faletti R, Sanchez J, Burkhardt JD, Natale A. Does periprocedural anticoagulation management of atrial fibrillation affect the prevalence of silent thromboembolic lesion detected by diffusion cerebral mag- netic resonance imaging in patients un- dergoing radiofrequency atrial fibrillation ablation with open irrigated catheters?
Results from a prospective multicenter study. Heart Rhythm 2014 May;11(5):791- 798.
15. Grosse-Dresselhaus F, Galinovic I, Villringer K, Audebert HJ, Fiebach JB. Difficulty of MRI based identification of lesion age by acute infra-tentorial ischemic stroke. PLoS One 2014;9(3):e92868.
16. Gaita F, Caponi D, Pianelli M, Scaglione M, Toso E, Cesarani F, Boffano C, Gandini G, Valentini MC, De PR, Halimi F, Leclercq JF.
Radiofrequency catheter ablation of atrial
fibrillation: a cause of silent thromboem- bolism? Magnetic resonance imaging assessment of cerebral thromboembo- lism in patients undergoing ablation of atrial fibrillation. Circulation 2010 October 26;122(17):1667-1673.
17. Wieczorek M, Hoeltgen R, Brueck M.
Does the number of simultaneously activated electrodes during phased RF multielectrode ablation of atrial fibrillation influence the incidence of silent cerebral microembolism? Heart Rhythm 2013 July;10(7):953-959.
18. Nagy-Balo E, Tint D, Clemens M, Beke I, Kovacs KR, Csiba L, Edes I, Csanadi Z.
Transcranial measurement of cerebral mi- croembolic signals during pulmonary vein isolation: a comparison of two ablation techniques. Circ Arrhythm Electrophysiol 2013 June;6(3):473-480.
19. Haines DE. ERACEing the risk of cerebral embolism from atrial fibrillation ablation.
Circ Arrhythm Electrophysiol 2013 Octo- ber;6(5):827-829.
20. Zellerhoff S, Ritter MA, Kochhauser S, Dittrich R, Kobe J, Milberg P, Korsukewitz C, Dechering DG, Pott C, Wasmer K, Leitz P, Guner F, Eckardt L, Monnig G. Modified phased radiofrequency ablation of atrial fibrillation reduces the number of cerebral microembolic signals. Europace 2013 Sep- tember 26.
21. Wood MA, Shaffer KM, Ellenbogen AL, Ownby ED. Microbubbles during radio- frequency catheter ablation: composi- tion and formation. Heart Rhythm 2005 April;2(4):397-403.
22. Nakajima M, Kimura K, Shimode A, Miyas- hita F, Uchino M, Naritomi H, Minematsu K. Microembolic signals within 24 hours of stroke onset and diffusion-weighted MRI abnormalities. Cerebrovasc Dis 2007;23(4):282-288.
23. Schwarz N, Kuniss M, Nedelmann M, Kaps M, Bachmann G, Neumann T, Pitschner HF, Gerriets T. Neuropsychological decline after
catheter ablation of atrial fibrillation. Heart Rhythm 2010 December;7(12):1761-1767.
24. Herm J, Fiebach JB, Koch L, Kopp UA, Kunze C, Wollboldt C, Brunecker P, Schultheiss HP, Schirdewan A, Endres M, Haeusler KG. Neu- ropsychological effects of MRI-detected brain lesions after left atrial catheter ablation for atrial fibrillation: long-term results of the MACPAF study. Circ Arrhythm Electrophysiol 2013 October;6(5):843-850.
25. Kochhauser S, Lohmann HH, Ritter MA, Leitz P, Guner F, Zellerhoff S, Korsukewitz C, Dechering DG, Banken J, Peters NM, Eckardt L, Monnig G. Neuropsychological impact of cerebral microemboli in ablation of atrial fibrillation. Clin Res Cardiol 2015 March;104(3):234-240.
