Abstract
Aims
The optimal anticoagulation regimen in patients with end-stage kidney disease (ESKD) undergoing atrial fibrillation (AF) catheter ablation is unknown. We sought to describe the real-world practice of peri-procedural anticoagulation management in patients with ESKD undergoing AF ablation.
Methods and results
Patients with ESKD on haemodialysis undergoing catheter ablation for AF in 12 referral centres in Japan were included. The international normalized ratio (INR) before and 1 and 3 months after ablation was collected. Peri-procedural major haemorrhagic events as defined by the International Society on Thrombosis and Haemostasis, as well as thromboembolic events, were adjudicated. A total of 347 procedures in 307 patients (67 ±9 years, 40% female) were included. Overall, INR values were grossly subtherapeutic [1.58 (interquartile range: 1.20–2.00) before ablation, 1.54 (1.22–2.02) at 1 month, and 1.22 (1.01–1.71) at 3 months]. Thirty-five patients (10%) suffered major complications, the majority of which was major bleeding (19 patients; 5.4%), including 11 cardiac tamponade (3.2%). There were two peri-procedural deaths (0.6%), both related to bleeding events. A pre-procedural INR value of 2.0 or higher was the only independent predictor of major bleeding [odds ratio, 3.3 (1.2–8.7), P = 0.018]. No cerebral or systemic thromboembolism occurred.
Conclusion
Despite most patients with ESKD undergoing AF ablation showing undertreatment with warfarin, major bleeding events are common while thromboembolic events are rare.
Keywords: Atrial fibrillation, End-stage kidney disease, Haemodialysis, Ablation, Anticoagulation
Graphical Abstract
Graphical Abstract.
What’s new?
In a real-world setting, a vast majority of patients with end-stage kidney disease (ESKD) undergoing atrial fibrillation (AF) ablation were undertreated with warfarin, with more than 90% of patients having subtherapeutic international normalized ratio in the peri-procedural period.
Despite undertreatment with warfarin, the rate of haemorrhagic complications was higher than most AF ablation historical cohorts treated with warfarin, and no cerebral or systemic thromboembolism occurred.
These findings suggest that international guideline recommendations for peri-procedural anticoagulation may not be applicable to ESKD patients undergoing AF ablation and may even call into question the role of anticoagulation in this particular population.
Introduction
Atrial fibrillation (AF) and chronic kidney disease (CKD) frequently coexist, and many studies have shown causal links between the two conditions.1–3 The prevalence of AF increases in more advanced forms of CKD, and 10–20% of patients with end-stage kidney disease (ESKD) are reported to have AF.4 As many antiarrhythmic drugs undergo renal metabolism or excretion, catheter ablation plays a growing role in this patient population.5–7 The importance of peri-procedural anticoagulation management in AF ablation has been highlighted in the seminal Coumadin in Preventing Thromboembolism in Atrial Fibrillation (COMPARE) trial.8 It is now widely accepted that patients undergoing AF ablation should be put under therapeutic anticoagulation throughout the peri-procedural period, either with continuous warfarin therapy with a target international normalized ratio (INR) of 2.0–3.0 or uninterrupted direct oral anticoagulants (DOACs).9,10
An important knowledge gap exists regarding anticoagulation therapy in patients with ESKD. Although warfarin has traditionally been the preferred method of long-term anticoagulation in patients with ESKD and AF, one recently published meta-analysis found that warfarin was not associated with decreased risk of stroke, while increasing the risk of intracranial haemorrhage in this population.11 Since ESKD patients were excluded from most of the key clinical trials assessing anticoagulation strategy in AF ablation,9,10,12 the optimal anticoagulation method in ESKD patients undergoing AF ablation remains uncertain. It has been proposed that patients with ESKD on haemodialysis therapy are at risk of life-threatening haemorrhage owing to impaired platelet function as well as an altered coagulation cascade, while systemic anticoagulation during haemodialysis may exert a protective effect against atrial thrombus formation.13 Literature focusing on AF ablation in ESKD patients is scarce, with only a few single-centre studies of 20–30 patients being available to date.5,14,15 Hence, it is unclear if peri-procedural warfarin therapy with a target INR of 2.0–3.0, as endorsed by the most recent European Society of Cardiology (ESC) guidelines,16 is equally effective and safe in ESKD patients undergoing AF ablation. To address this issue, we conducted a multicentre study aiming to describe the real-world practice on peri-procedural anticoagulation strategy and its association with potential adverse outcomes following ablation.
Methods
Study participants and data collection
All patients with ESKD on haemodialysis who underwent catheter ablation for paroxysmal or persistent AF in 12 referral centres between December 2006 and April 2022 were identified and included in the study. Patients who had cavotricuspid isthmus ablation as a stand-alone procedure were excluded. Anonymized patient demographic data were collected and sent to a central laboratory (Tokyo Medical and Dental University Hospital). The study protocol was approved by the institutional review board of the Tokyo Medical and Dental University Hospital. The need for patient consent was waived due to the anonymized and retrospective nature of the study (opt-out method). The study complied with the Declaration of Helsinki.
Ablation and peri-procedural anticoagulation
Generally, ablation was performed under continued warfarin therapy, although target INR value was left to each operator’s discretion. Pulmonary vein isolation (PVI) was performed in patients with paroxysmal or persistent AF. In radiofrequency ablation, types of three-dimensional mapping systems and ablation catheters were left to the discretion of the operator. After the introduction of cryoballoon ablation, many centres adopted this approach in paroxysmal AF patients. Patients who also had documented typical atrial flutter underwent ablation of the cavotricuspid isthmus. A small subset of patients with atypical atrial flutter, or persistent atrial fibrillation, had left atrial linear ablations. In left atrial ablation procedures, the common practice was to maintain activated coagulation time (ACT) between 300 and 400 s. During each dialysis session before and after the procedure, unfractionated heparin was used for anticoagulation. Commonly a bolus of 1000 units was given at the start of dialysis, followed by continuous infusion of 500 units/h, although the exact dose followed each institution’s protocol.
Study outcomes
We collected data on INR before and 1 and 3 months after the ablation procedure. The INR value and proportions of the patients whose INR were within therapeutic range (2.0–3.0 as endorsed by international guidelines) at three time points were outcomes of interest. Time in therapeutic range (TTR) was calculated by taking the number of INR measurements within the range of 2.0–3.0 divided by the total number of INR measurements as previously described.17 Intraprocedural ACTs were also collected when available. Other study outcomes included major complications, defined as those that were life-threatening, resulted in permanent harm, required intervention or significantly prolonged hospitalization,18 and occur up to 3 months after the procedure. Bleeding events were adjudicated as defined by the International Society on Thrombosis and Haemostasis (ISTH)19 and Bleeding Academic Research Consortium (BARC).20
Statistical analysis
Continuous variables are expressed as mean ± standard deviation or median and interquartile ranges depending on the distribution. Categorical variables are expressed by percentages. For continuous variables, comparisons were made by student’s t-test for variables with normal distribution and Wilcoxon rank-sum test for variables with skewed distribution. Categorical variables were compared using chi-squared test or Fisher’s exact test when the expected number in any cell was below five. Logistic regression was performed to probe the independent predictors of major bleeding. All variables that showed a significant association in the univariate analysis were included in the multivariate analysis. Spearman rank correlation test was used to assess for correlations. All tests were two-sided and P values < 0.05 were considered statistically significant. All analyses were performed using JMP (version 12; SAS Institute Inc., Cary, NC, USA).
Results
Baseline patient characteristics
A total of 347 procedures in 307 patients were analysed. Thirty patients underwent two procedures, and five patients had three procedures. Table 1 shows the baseline patient characteristics. The mean age was 67 ± 9 years. The aetiology of ESKD was most commonly chronic glomerulonephritis (32%), followed by diabetic nephropathy (27%). The median time on dialysis before ablation was 78 (24–180) months. Seventy per cent of the patients were hypertensive, and 10% had history of ischemic stroke. The median CHADS2 score was 1 (IQR:1–2), CHA2DS2-VASc score was 3 (2–4), and the HAS-BLED score was 3 (2–4). Seventy-three per cent of the patients had paroxysmal AF, 21% had persistent AF, 5% had persistent atrial flutter or atrial tachycardia, and 1% had paroxysmal atrial flutter or atrial tachycardia. PVI using radiofrequency catheters and cryoballoon were performed in 68% and 31% of the cases, respectively. Fifty-one per cent of the patients underwent cavotricuspid isthmus ablation. Pre-procedural transesophageal echocardiography was available in 175 patients (50%), and median left atrial appendage flow velocity was 44 (29–61) cm/s. In most cases (87%), ablation was undertaken the day after the last dialysis session.
Table 1.
Baseline patient characteristics
| n = 347 procedures in 307 patients | |
|---|---|
| Age (years) | 67 ± 9 |
| Female, n (%) | 138 (40%) |
| Time on dialysis before ablation (months) | 78 (IQR: 24–180) |
| Height (cm) | 162 (155–168) |
| Weight (kg) | 56 (49–66) |
| BMI (kg/m2) | 22 (19–25) |
| CKD aetiology | |
| Benign nephrosclerosis, n (%) | 60 (17%) |
| Chronic glomerulonephritis, n (%) | 110 (32%) |
| Diabetic nephropathy, n (%) | 92 (27%) |
| Polycystic kidney disease, n (%) | 18 (5%) |
| Other, n (%) | 67 (19%) |
| Congestive heart failure, n (%) | 91 (26%) |
| Hypertension, n (%) | 242 (70%) |
| Diabetes, n (%) | 94 (27%) |
| Stroke, n (%) | 33 (10%) |
| Antiplatelet agents or NSAIDs, n (%) | 125 (36%) |
| Proton-pump inhibitors, n (%) | 255 (74%) |
| Left atrial appendage emptying flow velocity (cm/s) | 44 (IQR: 29–60) |
| D-dimer (μg/mL) | 0.8 (IQR: 0.5–1.6) |
| CHADS2 score | 1 (IQR: 1–2) |
| CHA2DS2-VASc score | 3 (IQR:2–4) |
| HAS-BLED score | 3 (IQR:2–4) |
| LVEF (%) | 62 (50–69) |
| Left atrial diameter (mm) | 43 (39–48) |
| Arrhythmia treated | |
| Paroxysmal AF, n (%) | 252 (73%) |
| Persistent AF, n (%) | 73 (21%) |
| Paroxysmal atrial flutter or atrial tachycardia, n (%) | 4 (1%) |
| Persistent atrial flutter or atrial tachycardia, n (%) | 17 (5%) |
| Ablation procedure | |
| Radiofrequency PVI, n (%) | 236 (68%) |
| Cryoballoon PVI, n (%) | 109 (31%) |
| Cavotricuspid isthmus, n (%) | 178 (51%) |
| Other, n (%) | 141 (41%) |
Values are reported as the mean ± standard deviation, median with interquartile ranges (IQR), or number of patients (%) unless otherwise noted.
AF, atrial fibrillation; BMI, body mass index; CKD, chronic kidney disease; LVEF, left ventricular ejection fraction; NSAIDs, non-steroidal anti-Inflammatory drugs; PVI, pulmonary vein isolation.
Peri-procedural anticoagulation management
Figure 1 depicts the INR before and 1 and 3 months after the procedure. Generally, INR values were grossly subtherapeutic throughout the peri-procedural period. Before ablation, the median INR was 1.58 (1.20–2.00). At 1 month after the procedure, it was 1.54 (1.22–2.02) and 1.22 (1.01–1.71) 3 months after the procedure. The proportion of patients with INR between 2.00 and 3.00, between 1.50 and 2.00 and <1.50 were 23, 30, and 43% at the time of ablation, 21, 27, and 48% 1 month after the procedure, and 11, 24, and 61% 3 months after the procedure, respectively. In only 22 patients (6.3%) was the INR above 2.0 both before and 1 month after the procedure, yielding a a TTR of 20%. This undertreatment was largely consistent across most of the participating centres, as shown in Figure 2.
Figure 1.
International normalized ratio (INR) values before and 1 and 3 months after ablation. The thin gray band indicates the INR values recommended by the guidelines, corresponding to an INR of 2.0–3.0. The bottom and top of each box represent 25th and 75th quartiles, and whiskers indicate 25th quartile—1.5*IQR and 75th quartile + 1.5*IQR.
Figure 2.
International normalized ratio (INR) values before and 1 and 3 months after ablation in each participating sites. The number on the abscissa corresponds to individual institutions. Blue, red, and green boxes and whiskers represent values before (baseline) and at 1 and 3 months, respectively. The thin gray band indicates the INR values recommended by the guidelines, corresponding to an INR of 2.0–3.0. The bottom and top of each box represent 25th and 75th quartiles, and whiskers indicate 25th quartile—1.5*IQR and 75th quartile + 1.5*IQR.
The median amount of unfractionated heparin used during the procedure was 10 000 (8000–13 000) units. There was a significant negative correlation between the heparin dose used and pre-procedural INR value (R2 = 0.09, P < 0.0001). The averaged ACT during the procedure was 318 (291–355) s.
Following ablation, warfarin was discontinued in the majority of cases within 3 months after the procedure. The median duration of warfarin therapy after the procedure was 2 (1–3) months. In 74 patients (21%), warfarin was continued beyond 3 months after ablation.
Complications
Overall, major complications occurred in 35 patients (10%) including two peri-procedural deaths (0.6%). Major bleeding events as defined by ISTH accounted for more than half of the total complications (19 patients, 5.4%). Detailed descriptions of the major bleeding events and INR values before ablation are available in the Supplementary Table. The most frequent bleeding event was cardiac tamponade (11 patients, 3.2%). Most of the cases were managed by pericardiocentesis, whereas one patient required sternotomy and surgical repair. Intracranial haemorrhage in the peri-procedural period occurred in two patients. Retroperitoneal haemorrhage occurred in one patient. The two peri-procedural deaths were both related to the bleeding events; one was due to intracranial haemorrhage shortly after ablation, the other succumbed to multiorgan failure, following rapid deterioration of her condition after bleeding from the subclavian venous puncture site.
Bleeding events classified as BARC 2–5 occurred in 40 patients (12%). Apart from those described in the Supplementary Table, this included 13 cases of puncture site haemorrhages (3.7%), three cases of femoral artery pseudoaneurysm requiring surgical repair (0.8%), subclavian arterial pseudoaneurysm resulting in endovascular treatment by interventional radiology (n = 1), spontaneous haemorrhage from an artery of the abdominal wall requiring endovascular treatment by interventional radiology (n = 1), and haemorrhage in the renal cyst and peri-nephric haematoma requiring re-hospitalization (n = 1).
Non-haemorrhagic major complications included worsening of heart failure symptoms (n = 1), air embolism resulting in transient ST elevation (n = 1), bradycardia requiring temporary pacing (n = 2), puncture site infection (n = 1), and resuscitated cardiopulmonary arrest following protamine infusion (n = 1). Importantly, there was no cerebral or systemic thromboembolism in our cohort. Incidence of each specific complication is summarized in Table 2.
Table 2.
Incidence of each specific type of complications
| n = 347 procedures in 307 patients | |
|---|---|
| Death, n (%) | 2 (0.6%) |
| Stroke, TIA, or systemic embolism, n (%) | 0 (0%) |
| Haemorrhagic complications | |
| Cardiac tamponade, n (%) | 11 (3.2%) |
| Intracranial haemorrhage, n (%) | 2 (0.6%) |
| Retroperitoneal haemorrhage, n (%) | 1 (0.3%) |
| Puncture site haemorrhage, n (%) | 13 (3.7%) |
| Pseudoaneurysm, n (%) | 4 (1.2%) |
| Other major haemorrhage, n (%) | 5 (1.4%) |
| Non-haemorrhagic complications | |
| Exacerbation of heart failure, n (%) | 1 (0.3%) |
| Temporary pacing for bradycardia, n (%) | 2 (0.6%) |
| Puncture site infection, n (%) | 1 (0.3%) |
| Anaphylactic shock after protamine infusion, n (%) | 1 (0.3%) |
Values are reported as number of patients (%).
TIA, transient ischemic attack.
Predictors of major bleeding events
The baseline and procedural characteristics of the patients with and without ISTH major bleeding events are given in Table 3. When compared to patients without major bleeding events, those with major bleeding events were more often female (62% vs. 37%, P = 0.03) and weighed significantly less [51 (41–57) vs. 57 (49–67) kg, P = 0.03]. The proportion of patients who underwent PVI with radiofrequency catheters, PVI with cryoballoon, cavotricuspid isthmus ablation, or any other ablation was not statistically different between the two groups, although there was a trend towards less bleeding with cryoballoon compared to radiofrequency (16% vs. 32%, P = 0.10). The proportion of patients whose INR was 2.0 or higher at the time of ablation was significantly greater in patients with major bleeding events than those without (47% vs. 25%, P = 0.04). Major bleeding occurred in 9 out of 90 patients (10%) when pre-procedural INR was 2.0 or higher, and in 10 out of 257 (3.9%) when INR was below 2.0. Of those whose INR was 2.0 or higher suffering major bleeding events, one had a supratherapeutic INR of 3.1 and all other patients had therapeutic INR between 2.0 and 3.0 (see also Supplementary Table). In the multivariate analysis, a pre-procedural INR of 2.0 or higher was the only independent predictor of major bleeding [odds ratio, 3.3 (1.2–8.7), P = 0.018]. The mean ACT was similar between patients with and without major haemorrhage [320 (286–380) vs. 318 (291–355) s, P = 0.72]. There was no significant variation in the rate of major bleeding across different centres (P = 0.34).
Table 3.
Baseline and procedural characteristics in patients with and without major bleeding events
| With major bleeding (n = 19) | Without major bleeding (n = 328) | P value | |
|---|---|---|---|
| Age (years) | 66 ± 10 | 67 ± 9 | 0.63 |
| Sex (female), n (%) | 12 (63%) | 129 (38%) | 0.03 |
| Height (cm) | 155 (151–164) | 162 (155–168) | 0.07 |
| Weight (kg) | 51 (41–57) | 57 (49–67) | 0.03 |
| BMI (kg/m2) | 21 (18–23) | 22 (19–25) | 0.17 |
| CKD aetiology | 0.08 | ||
| Benign nephrosclerosis, n (%) | 0 (0%) | 60 (18%) | |
| Chronic glomerulonephritis, n (%) | 7 (37%) | 103 (31%) | |
| Diabetic nephropathy, n (%) | 5 (26%) | 87 (27%) | |
| Polycystic kidney disease, n (%) | 0 (0%) | 18 (5%) | |
| Other, n (%) | 7 (37%) | 60 (18%) | |
| Time on dialysis before ablation (months) | 72 (24–180) | 96 (38–239) | 0.27 |
| Ablation protocol | |||
| Radiofrequency PVI, n (%) | 14 (74%) | 222 (68%) | 0.58 |
| Cryoballoon PVI, n (%) | 3 (16%) | 106 (32%) | 0.10 |
| Cavotricuspid isthmus, n (%) | 9 (47%) | 169 (52%) | 0.72 |
| Other ablation, n (%) | 8 (42%) | 133 (41%) | 0.89 |
| Congestive heart failure, n (%) | 4 (21%) | 87 (27%) | 0.59 |
| Hypertension, n (%) | 10 (53%) | 232 (71%) | 0.11 |
| Diabetes, n (%) | 5 (26%) | 89 (27%) | 0.93 |
| History of ischemic stroke, n (%) | 4 (21%) | 29 (9%) | 0.12 |
| History of significant bleeding, n (%) | 2 (11%) | 31 (9%) | 0.88 |
| Antiplatelet agent or NSAID use, n (%) | 8 (42%) | 117 (36%) | 0.57 |
| INR before ablation | 2.01 (1.48–2.34) | 1.57 (1.18–2.00) | 0.05 |
| INR 2.0 or higher before ablation | 9 (47%) | 81 (25%) | 0.04 |
| Mean ACT (s) | 320 (286–380) | 318 (291–355) | 0.71 |
| LVEF (%) | 62 (58–72) | 62 (50–69) | 0.40 |
| Left atrial diameter (mm) | 41 (36–46) | 43 (39–48) | 0.21 |
| CHADS2 score | 1 (1–3) | 1 (1–2) | 0.78 |
| CHA2DS2-VASc score | 2 (1–5) | 3 (2–4) | 0.70 |
| HAS-BLED score | 3 (2–4) | 3 (2–4) | 0.99 |
Values are reported as the mean ± standard deviation, median with interquartile ranges (IQR) or number of patients (%) unless otherwise noted.
ACT, activated coagulation time; BMI, body mass index; CKD, chronic kidney disease; LVEF, left ventricular ejection fraction; NSAIDs, non-steroidal anti-inflammatory drugs; PVI, pulmonary vein isolation.
Rhythm outcomes
Overall, during a median follow-up period of 211 (61–606) days, 40% and 36% of patients with paroxysmal and persistent AF, respectively, had documented arrhythmia recurrence beyond the 90 day blanking period after a single procedure.
Discussion
To the best of our knowledge, this is the only study focusing on anticoagulation management in ESKD patients undergoing AF ablation. The principal findings of our study were as follows:
In a real-world setting, patients with ESKD undergoing AF ablation are grossly undertreated from the viewpoint of anticoagulation guideline recommendations, with 94% of patients having subtherapeutic INR at some point during the peri-procedural period.
Nonetheless, a high rate of haemorrhagic complications was observed in this population.
A pre-procedural INR of 2.0 or higher was the only independent predictor of major bleeding.
No patient had peri-procedural cerebral or systemic thromboembolism.
Vitamin K antagonist (VKA) therapy in AF patients on haemodialysis has been controversial, as no randomized trial has shown its benefit over no anticoagulation. Although apixaban is increasingly used in some countries, the recently published randomized Renal Hemodialysis Patients Allocated Apixaban Versus Warfarin in Atrial Fibrillation (RENAL-AF) study failed to show a superiority of apixaban over warfarin, with numerically higher rate of bleeding in the apixaban group.21 Because it is well established that patients on haemodialysis are at increased risk of life-threatening haemorrhage while on anticoagulation,22 some authors have suggested a role for left atrial appendage closure in this population.23 In the absence of strong evidence, guidelines give inconsistent recommendations on this issue. While ESC guidelines do not provide a firm recommendation,16 Kidney Disease: Improving Global Outcomes guidelines recommend against anticoagulating patients with AF on haemodialysis.24 These conflicting guideline recommendations may partly account for the result of the present study, where the vast majority had subtherapeutic INR at the time of ablation.
It has not been entirely clear if patients with ESKD were at increased risk of complications from catheter ablation due to the lack of studies including large numbers of patients. Previous studies comparing the outcomes of catheter ablation for AF in patients with and without ESKD reported no statistically significant difference in the rate of complications, although these studies included relatively small number of patients.5,6 The rate of major bleeding events and cardiac tamponade in our cohort was 5.4% and 3.2%, which is higher than what has been reported in most historical cohorts from the warfarin arms of the previously published prospective randomized clinical trials. The major bleeding rate and cardiac tamponade incidence in the on-warfarin arm of the COMPARE trial was 0.88% and 0.5%, respectively.8 An active-controlled multicentre study with blind adjudication designed to evaluate the safety of uninterrupted rivaroxaban and uninterrupted vitamin K antagonists in subjects undergoing catheter ablation for non-valvular AF (VENTURE-AF) reported a major bleeding rate of 0.8% and no cardiac tamponade.12 In the anticoagulation using the direct factor Xa inhibitor apixaban during AF catheter ablation, in comparison to VKA therapy (AXAFA–AFNET 5), the major bleeding rate was 4.4%, and cardiac tamponade occurred in 1.6%.9 As previously suggested, platelet dysfunction and abnormal coagulation cascade found in ESKD patients may be responsible for the high bleeding risk in our study. Additionally, drug interactions by antiarrhythmic agents such as amiodarone, which is often used in patients with renal impairment, cannot be excluded.
Of note, the use of ultrasound-guided puncture was at each operator’s discretion in this study. In general, ultrasound-guided punctures are infrequently performed in Japan because, due to low prevalence of obesity, femoral punctures are straightforward in most cases. Although increased or routine use of ultrasound-guided puncture may have the potential to reduce puncture site bleeding, we believe the high rate of bleeding in ESKD patients is due to an inherent difference between ESKD and non-ESKD patients. We think this because vascular complications are infrequent in a nationwide registry of non-ESKD patients, where BARC ≥2 bleeding complications occurred in 1.12%.25 This figure is substantially smaller than that of the present cohort, where we observed BARC ≥2 bleeding in 12%. The order of magnitude greater rate of puncture site bleeding in our cohort compared to non-ESKD counterparts cannot be due to differences in the use of ultrasound-guided puncture.
To our surprise, despite a vast majority of our cohort having subtherapeutic INR throughout the peri-procedural period, we did not observe any thromboembolic events. In the COMPARE trial, discontinuation of warfarin 2–3 days before the procedure was strongly associated with peri-procedural stroke even if bridged with low-molecular weight heparin. Additionally, in the same study, all patients who had peri-procedural stroke in the on-warfarin arm had subtherapeutic INR.8 Another study found that subtherapeutic INR at the time of ablation was a predictor of silent thromboembolic lesions detected by post-procedural magnetic resonance imaging.26 Several important considerations are worth mentioning on this issue. First, although most of the patients in our cohort had at least one conventional stroke risk, the original CHADS2 and CHA2DS2-VASc scores were mainly derived from non-ESKD cohorts, and currently little is known about the predictors of stroke in patients with ESKD and AF. It was previously suggested that CHA2DS2-VASc score was a poor predictor of ischaemic stroke in ESKD patients with AF.27 Second, the endpoint of this study was adverse events only in the peri-procedural period, which is fundamentally different from the naturally occurring embolic events those clinical scores are designed for. Third, patients with ESKD undergo intermittent systemic anticoagulation on a regular basis at the time of haemodialysis. This may confer some level of protection against the formation of left atrial thrombus and partially explain the low rate of thromboembolic events in our cohort.
Clinical implications
Based on the observed high bleeding risk in patients with ESKD undergoing AF ablation, we believe that these patients should be anticoagulated with extreme caution and further studies are needed to determine the optimal anticoagulation regimen, as well as the safest target INR range. It is worth emphasizing that risks and benefits should be carefully discussed with the patients when considering AF ablation.
Strengths and limitations
The strength of the present study is the large number of patients included compared to previous studies, which allowed us to capture relatively infrequent events such as haemopericardium or intracranial haemorrhage. One major limitation is the lack of a control group of non-ESKD patients, making it difficult to assess if the observations made were truly specific to ESKD patients. During the study period, DOACs largely replaced warfarin in non-ESKD patients, making it almost impossible to generate a control group treated with warfarin. Another limitation was the heterogeneity of treatment strategies across different centres. However, our findings were largely consistent across centres. Additionally, because all of the patients in the study were Japanese, our results may not be generalizable to patients of other races. Finally, because apixaban is not currently approved for use in patients on haemodialysis in Japan, its potential usefulness in ESKD patients undergoing AF ablation remains to be determined.
Conclusions
We described the real-world practice of peri-procedural anticoagulation in patients with ESKD undergoing AF ablation in 12 referral centres in Japan. INR values were grossly subtherapeutic in a vast majority of patients. Nonetheless, haemorrhagic complications were common, while no thromboembolic complications were seen during the peri-procedural period. These findings suggest that international guideline recommendations for peri-procedural anticoagulation may not be applicable to ESKD patients undergoing AF ablation and may even call into question the role of anticoagulation in this particular population.
Supplementary material
Supplementary material is available at Europace online.
Supplementary Material
Contributor Information
Tasuku Yamamoto, Department of Cardiovascular Medicine, Tokyo Medical and Dental University, Yushima 1-5-45, Bunkyo-ku, Tokyo 113-8510, Japan.
Shinsuke Miyazaki, Department of Cardiovascular Medicine, Tokyo Medical and Dental University, Yushima 1-5-45, Bunkyo-ku, Tokyo 113-8510, Japan.
Yasuaki Tanaka, Department of Cardiology, Yokosuka Kyosai Hospital, Yonegahama-dori 1-16, Yokosuka-shi, Kanagawa 238-8558, Japan.
Toshikazu Kono, Department of Cardiology, Japanese Red Cross Saitama Hospital, Shintoshin 1-5, Chuo-ku, Saiatama-shi, Saitama 330-8553, Japan.
Tadanori Nakata, Cardiovascular Division, Tsuchiura Kyodo Hospital, Otsuno 4-1-1, Tsuchiura-shi, Ibaraki 300-0028, Japan.
Akira Mizukami, Department of Cardiology, Kameda Medical Center, Higashicho 929, Kamogawa-shi, Chiba 296-8602, Japan.
Daisetsu Aoyama, Department of Cardiology, University of Fukui Hospital, Shimoaizuki 2303, Matsuoka, Eiheiji-cho, Yoshida-gun, Fukui 910-1193, Japan.
Hirofumi Arai, Department of Cardiology, Japanese Red Cross Yokohama City Bay Hospital, Shinyamashita 3-12-1, Naka-ku, Yokohama-shi, Kanagawa 231-8682, Japan.
Yuta Taomoto, Department of Cardiology, Musashino Red Cross Hospital, Sakaiminami-cho 1-26-1, Musashino-shi, Tokyo 180-8610, Japan.
Tomoki Horie, Department of Cardiology, Musashino Red Cross Hospital, Sakaiminami-cho 1-26-1, Musashino-shi, Tokyo 180-8610, Japan.
Rintaro Hojo, Department of Cardiology, Tokyo Metropolitan Hiroo Hospital, Ebisu 2-34-10, Shibuya-ku, Tokyo 150-0013, Japan.
Shiho Kawamoto, Heart Center, Tokyo Yamate Medical Center, Hyakunin-cho 3-22-1, Shinjuku-ku, Tokyo 169-0073, Japan.
Kento Yabe, Department of Cardiology, Ome Municipal General Hospital, Higashiome 4-16-5, Ome-shi 198-0042, Japan.
Kikou Akiyoshi, Department of Cardiology, Hiratsuka Kyosai Hospital, Higashiome 4-16-5, Ome-shi 198-0042, Japan.
Nobutaka Kato, Department of Cardiology, Hiratsuka Kyosai Hospital, Higashiome 4-16-5, Ome-shi 198-0042, Japan.
Yuichi Ono, Department of Cardiology, Ome Municipal General Hospital, Higashiome 4-16-5, Ome-shi 198-0042, Japan.
Atsushi Suzuki, Heart Center, Tokyo Yamate Medical Center, Hyakunin-cho 3-22-1, Shinjuku-ku, Tokyo 169-0073, Japan.
Seiji Fukamizu, Department of Cardiology, Tokyo Metropolitan Hiroo Hospital, Ebisu 2-34-10, Shibuya-ku, Tokyo 150-0013, Japan.
Yasutoshi Nagata, Department of Cardiology, Musashino Red Cross Hospital, Sakaiminami-cho 1-26-1, Musashino-shi, Tokyo 180-8610, Japan.
Yasuteru Yamauchi, Department of Cardiology, Japanese Red Cross Yokohama City Bay Hospital, Shinyamashita 3-12-1, Naka-ku, Yokohama-shi, Kanagawa 231-8682, Japan.
Hiroshi Tada, Department of Cardiology, University of Fukui Hospital, Shimoaizuki 2303, Matsuoka, Eiheiji-cho, Yoshida-gun, Fukui 910-1193, Japan.
Hitoshi Hachiya, Cardiovascular Division, Tsuchiura Kyodo Hospital, Otsuno 4-1-1, Tsuchiura-shi, Ibaraki 300-0028, Japan.
Osamu Inaba, Department of Cardiology, Japanese Red Cross Saitama Hospital, Shintoshin 1-5, Chuo-ku, Saiatama-shi, Saitama 330-8553, Japan.
Atsushi Takahashi, Department of Cardiology, Yokosuka Kyosai Hospital, Yonegahama-dori 1-16, Yokosuka-shi, Kanagawa 238-8558, Japan.
Masahiko Goya, Department of Cardiovascular Medicine, Tokyo Medical and Dental University, Yushima 1-5-45, Bunkyo-ku, Tokyo 113-8510, Japan.
Tetsuo Sasano, Department of Cardiovascular Medicine, Tokyo Medical and Dental University, Yushima 1-5-45, Bunkyo-ku, Tokyo 113-8510, Japan.
Funding
No funding was received for this research.
Data availability
The data underlying the findings of this article are available upon reasonable request to the corresponding author.
References
- 1. Bansal N, Fan D, Hsu CY, Ordonez JD, Marcus GM, Go AS. Incident atrial fibrillation and risk of end-stage renal disease in adults with chronic kidney disease. Circulation 2013;127:569–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Alonso A, Lopez FL, Matsushita K, Loehr LR, Agarwal SK, Chen LYet al. Chronic kidney disease is associated with the incidence of atrial fibrillation. Circulation 2011;123:2946–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Nabauer M, Oeff M, Gerth A, Wegscheider K, Buchholz A, Haeusler KGet al. Prognostic markers of all-cause mortality in patients with atrial fibrillation: data from the prospective long-term registry of the German Atrial Fibrillation NETwork (AFNET). Europace 2021;23:1903–12. [DOI] [PubMed] [Google Scholar]
- 4. Zebe H. Atrial fibrillation in dialysis patients. Nephrol Dial Transplant 2000;15:765–8. [DOI] [PubMed] [Google Scholar]
- 5. Sairaku A, Yoshida Y, Kamiya H, Tatematsu Y, Nanasato M, Hirayama Het al. Outcomes of ablation of paroxysmal atrial fibrillation in patients on chronic hemodialysis. J Cardiovasc Electrophysiol 2012;23:1289–94. [DOI] [PubMed] [Google Scholar]
- 6. Takigawa M, Kuwahara T, Takahashi A, Kobori A, Takahashi Y, Okubo Ket al. The impact of haemodialysis on the outcomes of catheter ablation in patients with paroxysmal atrial fibrillation. Europace 2014;16:327–34. [DOI] [PubMed] [Google Scholar]
- 7. Potpara TS, Ferro C, Lip GYH, Dan GA, Lenarczyk R, Mallamaci Fet al. Management of atrial fibrillation in patients with chronic kidney disease in clinical practice: a joint European heart rhythm association (EHRA) and European renal association/European dialysis and transplantation association (ERA/EDTA) physician-based survey. Europace 2020;22:496–505. [DOI] [PubMed] [Google Scholar]
- 8. Di Biase L, Burkhardt JD, Santangeli P, Mohanty P, Sanchez JE, Horton Ret al. Periprocedural stroke and bleeding complications in patients undergoing catheter ablation of atrial fibrillation with different anticoagulation management. Circulation 2014;129:2638–44. [DOI] [PubMed] [Google Scholar]
- 9. Kirchhof P, Haeusler KG, Blank B, de Bono J, Callans D, Elvan Aet al. Apixaban in patients at risk of stroke undergoing atrial fibrillation ablation. Eur Heart J 2018;39:2942–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Calkins H, Willems S, Gerstenfeld EP, Verma A, Schilling R, Hohnloser SHet al. Uninterrupted dabigatran versus warfarin for ablation in atrial fibrillation. N Engl J Med 2017;376:1627–36. [DOI] [PubMed] [Google Scholar]
- 11. Randhawa MS, Vishwanath R, Rai MP, Wang L, Randhawa AK, Abela Get al. Association between use of warfarin for atrial fibrillation and outcomes among patients with end-stage renal disease: a systematic review and meta-analysis. JAMA Netw Open 2020;3:e202175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Cappato R, Marchlinski FE, Hohnloser SH, Naccarelli GV, Xiang J, Wilber DJet al. Uninterrupted rivaroxaban vs. uninterrupted vitamin K antagonists for catheter ablation in non-valvular atrial fibrillation. Eur Heart J 2015;36:1805–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Shah M, Tsadok MA, Jackevicius CA, Essebag V, Eisenberg MJ, Rahme Eet al. Warfarin use and the risk for stroke and bleeding in patients with atrial fibrillation undergoing dialysis. Circulation 2014;129:1196–203. [DOI] [PubMed] [Google Scholar]
- 14. Hayashi M, Kaneko S, Shimano M, Ohashi T, Kubota R, Takeshita Ket al. Efficacy and safety of radiofrequency catheter ablation for atrial fibrillation in chronic hemodialysis patients. Nephrol Dial Transplant 2014;29:160–7. [DOI] [PubMed] [Google Scholar]
- 15. Hayashi T, Murakami M, Yokota S, Yamada T, Mashimo Y, Miyashita Het al. Comparison between cryoballoon ablation and radiofrequency catheter ablation for atrial fibrillation in patients on hemodialysis. Indian Pacing Electrophysiol J 2021;21:67–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Hindricks G, Potpara T, Dagres N, Bax JJ, Boriani G, Dan GAet al. 2020 ESC guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European association for cardio-thoracic surgery (EACTS). Eur Heart J 2021;42:373–498. [DOI] [PubMed] [Google Scholar]
- 17. Schmitt L, Speckman J, Ansell J. Quality assessment of anticoagulation dose management: comparative evaluation of measures of time-in-therapeutic range. J Thromb Thrombolysis 2003;15:213–6. [DOI] [PubMed] [Google Scholar]
- 18. Hoyt H, Bhonsale A, Chilukuri K, Alhumaid F, Needleman M, Edwards Det al. Complications arising from catheter ablation of atrial fibrillation: temporal trends and predictors. Heart Rhythm 2011;8:1869–74. [DOI] [PubMed] [Google Scholar]
- 19. Schulman S, Kearon C. Subcommittee on control of anticoagulation of the scientific and standardization committee of the international society on thrombosis and haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost 2005;3:692–4. [DOI] [PubMed] [Google Scholar]
- 20. Mehran R, Rao SV, Bhatt DL, Gibson CM, Caixeta A, Eikelboom Jet al. Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the bleeding academic research consortium. Circulation 2011;123:2736–47. [DOI] [PubMed] [Google Scholar]
- 21. Pokorney SD, Chertow GM, Al-Khalidi HR, Gallup D, Dignaco P, Mussina Ket al. Apixaban for patients with atrial fibrillation on hemodialysis: a multicenter randomized controlled trial. Circulation 2022;146:1735–45. [DOI] [PubMed] [Google Scholar]
- 22. Siontis KC, Zhang X, Eckard A, Bhave N, Schaubel DE, He Ket al. Outcomes associated with apixaban use in end-stage kidney disease patients with atrial fibrillation in the United States. Circulation 2018;138:1519–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Turakhia MP, Blankestijn PJ, Carrero JJ, Clase CM, Deo R, Herzog CAet al. Chronic kidney disease and arrhythmias: conclusions from a kidney disease: improving global outcomes (KDIGO) controversies conference. Eur Heart J 2018;39:2314–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Herzog CA, Asinger RW, Berger AK, Charytan DM, Díez J, Hart RGet al. Cardiovascular disease in chronic kidney disease. A clinical update from kidney disease: improving global outcomes (KDIGO). Kidney Int 2011;80:572–86. [DOI] [PubMed] [Google Scholar]
- 25. Kusano K, Yamane T, Inoue K, Takegami M, Nakao YM, Nakai Met al. The Japanese catheter ablation registry (J-AB): annual report in 2019. J Arrhythm 2021;37:1443–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Di Biase L, Gaita F, Toso E, Santangeli P, Mohanty P, Rutledge Net al. Does periprocedural anticoagulation management of atrial fibrillation affect the prevalence of silent thromboembolic lesion detected by diffusion cerebral magnetic resonance imaging in patients undergoing radiofrequency atrial fibrillation ablation with open irrigated catheters? Results from a prospectivemulticenter study. Heart Rhythm 2014;11:791–8. [DOI] [PubMed] [Google Scholar]
- 27. Shih CJ, Ou SM, Chao PW, Kuo SC, Lee YJ, Yang CYet al. Risks of death and stroke in patients undergoing hemodialysis with new-onset atrial fibrillation: a competing-risk analysis of a nationwide cohort. Circulation 2016;133:265–72. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying the findings of this article are available upon reasonable request to the corresponding author.