1 Improvement in renal and endothelial function after catheter ablation in patients with persistent atrial fibrillation Keisuke Okawa (MD)

Improvement in renal and endothelial function after catheter ablation in patients with persistent atrial fibrillation

Keisuke Okawa (MD)^a,*, Toru Miyoshi (MD, PhD, FJCC)^b, Masahiro Sogo (MD)^a, Shohei Hara (MD)^a, Yuya Sudo (MD)^a, Satoko Ugawa (MD)^a, Masahiko Takahashi (MD)^a, Masayuki Doi (MD, PhD)^a, Hiroshi Morita (MD, PhD)^c, Hiroshi Ito (MD, PhD, FJCC)^b

a Department of Cardiovascular Medicine, Kagawa Prefectural Central Hospital, Takamatsu, Kagawa, Japan

b Department of Cardiovascular Medicine, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan

c Department of Cardiovascular Therapeutics, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan

*Corresponding author:

Keisuke Okawa, MD

Department of Cardiovascular Medicine Kagawa Prefectural Central Hospital

1-2-1 Asahi-machi, Takamatsu City, Kagawa 760-8557, Japan Tel: +818019030665; Fax: +81878342148

E-mail: [email protected]

Keywords:

Paroxysmal atrial fibrillation; Persistent atrial fibrillation; Catheter ablation; Renal function;

Endothelial function

ABSTRACT

Background: Cardiovascular events in patients with atrial fibrillation (AF) can be lowered by catheter ablation. We hypothesized the underlying mechanism was improvement in renal and endothelial function corresponding to AF burden, and investigated whether restoration of sinus rhythm (SR) after ablation affected these functions according to AF type.

Methods and Results: We prospectively measured estimated glomerular filtration rate (eGFR), urinary albumin-to-creatinine ratio (UACR), and reactive hyperemia index (RHI) in 358

consecutive patients with AF before and 6 and 12 months after the ablation. For each AF type [paroxysmal AF (PAF), n=229, and persistent AF (PeAF), n=129], we evaluated changes in these markers and influence of chronic kidney disease (CKD). The eGFR and natural logarithm-

transformed (ln) UACR improved at 6 months in the PeAF group (68.7±18.7 to 71.8±18.9 mL/min/1.73 m², p=0.003 and 3.1±1.6 to 2.8±1.5, p<0.001, respectively) and remained

unchanged in the PAF group. Among the PeAF patients, recurrent AF was identified in 41, but only transiently in 38 patients. PeAF at baseline independently predicted increased eGFR [odds ratio (OR)=2.16, 95% confidence interval (CI) 1.18-4.01, p=0.012] and decreased UACR (OR=2.39, 95% CI 1.17-4.94, p=0.017). In the PeAF patients with CKD, ln-RHI significantly increased at 6 months after the ablation, and the change (Δ) in ln-RHI was significantly correlated with the ΔeGFR (r=0.35, p=0.03).

Conclusions: SR restoration after ablation was associated with an improved eGFR and UACR in PeAF patients, but not PAF patients. In PeAF patients with CKD, an improved endothelial function after ablation was associated with an improved renal function.

INTRODUCTION

The prognosis of patients with atrial fibrillation (AF) principally depends on the incidence of cardiovascular events because thromboembolic events can be effectively prevented with anticoagulation therapy [1]. The risk of cardiovascular events in these patients is high due to multiple comorbidities, including hypertension, diabetes mellitus, obesity, and chronic kidney disease (CKD). Among these risk factors, CKD is a strong predictor of cardiovascular events and increases the AF risk. [2] Recent studies have demonstrated that catheter ablation is associated with decreased cardiovascular events, especially heart failure, and mortality in patients with AF [3-5]. Catheter ablation is a well-established rhythm control therapy even in patients with structural heart disease, such as surgically corrected valvular AF [6]. Moreover, restoring sinus rhythm (SR) with ablation improves renal function, as assessed by the estimated glomerular filtration rate (eGFR) [7], and vascular endothelial function in patients with AF [8]. These findings suggest that improvements in renal and vascular endothelial function after ablation would lower the risk of cardiovascular events in patients with AF. However, it is not clear if the beneficial effects of ablation would be consistent for the different types of AF. A meta-analysis reported a higher risk of a poor prognosis for patients with persistent AF (PeAF) than for those with paroxysmal AF (PAF) [9]. In a previous study, we demonstrated that PeAF, compared to PAF, was associated with a worse endothelial function, which was quantified by the reactive hyperemia index (RHI) [10]. If AF persisting or a high AF burden regulate the endothelial

dysfunction and cardiovascular events, catheter ablation may provide a greater benefit to patients with PeAF than to those with PAF. In this study, we hypothesized that the effects of ablation on renal function also depend on the AF burden. Therefore, we investigated the differences in the temporal changes, before and after ablation, in the eGFR and urine albumin-to-creatinine ratio

(UACR), of which both are independent renal markers of cardiovascular events, between the patients with PAF and those with PeAF. Additionally, for each AF type, we evaluated the associations between the renal and vascular endothelial function (RHI) with respect to the presence of CKD, and the significance of transient AF recurrences after ablation on the 3

markers, for a clearer understanding of the impact on the SR restoration or AF burden reduction.

METHODS

Study population

We prospectively enrolled 422 consecutive patients with AF who underwent catheter ablation between May 2013 and March 2016 at Kagawa Prefectural Central Hospital. We included patients who underwent a first session of ablation for AF and could be followed-up thereafter for a period of 12 months. We excluded patients who had undergone previous ablation and those on concurrent dialysis. Of the initial 422 patients enrolled, 47 were excluded, 10 were lost to follow-up, the RHI was not measured in 4, and 3 experienced AF recurrences and non- maintenance of sinus rhythm (SR). Ultimately, the data from 358 patients, in whom SR was maintained for 12 months, were analyzed (Fig. 1).

For the main analysis, we divided the patients into 2 groups, based on the AF type before the ablation (PAF, n=229, and PeAF, n=129) and compared the parameters between the groups.

The diagnosis of PAF or PeAF was based on the American Heart Association/American College of Cardiology/Heart Rhythm Society guidelines definitions [11]. We then subdivided the patients into 4 groups based on the presence or absence of CKD at baseline to evaluate the association between the renal and endothelial function in patients with AF, and based on AF recurrences, to

clarify the impact of SR restoration, especially on AF burden reduction in PeAF patients. AF recurrence was considered transient if SR was maintained.

We obtained written informed consent from all participants. This study conformed to the 1975 Declaration of Helsinki and was approved by the Clinical Ethics Committee of Kagawa Prefectural Central Hospital.

Study protocol

We measured the renal parameters (eGFR and UACR) and endothelial function (RHI) on the day of admission, before the ablation. Transthoracic echocardiography was performed within 2 weeks before the ablation. We measured the left atrial (LA) diameter in the parasternal long- axis view at the end of ventricular systole. We measured the left ventricular ejection fraction (LVEF) in the apical two- and four-chamber views, using the disc summation method. In patients with PeAF, the LVEF was expressed as the average over 5 cardiac cycles.

We performed the catheter ablation according to the established procedures. We recreated the 3-dimensional geometry of the LA and pulmonary veins using a 3-dimensional mapping system (CARTO; Biosense-Webster, Diamond Bar, CA, USA) and mapped all pulmonary veins with a decapolar circular catheter. We performed electrical pulmonary vein isolation (PVI) using an open-irrigation 3.5-mm-tip deflectable catheter (Thermocool; Biosense- Webster). We delivered radiofrequency energy at a maximum power output of 35 W and

maximum temperature of 45°C. We determined that endpoint of the procedure was no PV reconnection 20 minutes after the initial PVI under an intravenous isoproterenol administration.

We also confirmed dormant conduction using adenosine triphosphate after the initial PVI, as needed. We performed a second ablation if AF recurred and could not be controlled with

antiarrhythmic drugs. During the second ablation, we checked for reconnections of the blocks done in the first ablation, and performed re-isolation if reconnections were identified, adding a posterior wall isolation, low-voltage zone, or trigger site ablation, as needed.

Patients were prospectively followed for 12 months via monthly outpatient clinic visits, with an electrocardiogram obtained at each visit. We measured the eGFR, UACR, and RHI at 6 and 12 months, and repeated transthoracic echocardiography at 6 months. Long-term (2 weeks) electrocardiogram monitoring was performed using an auto-triggered loop recorder (SpiderFlash- T, Sorin, Saluggia, Italy) at 6 months after the ablation (Fig. 1).

Measurement of the renal and endothelial markers

The eGFR was determined using the serum cystatin-C levels, measured using the latex immunonephelometry method (BM6070; Nihon Denshi, Tokyo, Japan), based on the following formulas adopted by the Japanese Society of Nephrology:

eGFRcys (mL/min/1.73 m²) = (104×cystatin-C^−1.019×0.996^age) – 8, for men or (104×cystatin-C^−1.019×0.996^age×0.929) – 8, for women [12].

UACR (mg/g Cr) was determined as:

urinary albumin (mg)×urinary creatinine⁻¹ (g)

The urinary albumin and creatinine levels were measured using immunonephelometry and enzyme methods (Nihon Denshi). CKD was defined by an eGFR of ≤60 mL/min/1.73 m². A UACR ≥10 mg/g Cr was defined as positive albuminuria.

We assessed the endothelial function by measuring the RHI with RH-peripheral arterial tonometry (RH-PAT) using an EndoPAT2000 (Itamar Medical, Caesarea, Israel), which measures the digital hyperemic response. RH-PAT is a well-established assessment approved by the US

Food and Drug Administration. The RHI is an independent predictor of adverse cardiac events, beyond the traditional Framingham risk score [13]. The principle underlying RH-PAT has been previously described [14] and its accuracy during AF has been validated in our previous study [10].

Diagnosis of AF recurrences

AF recurrences were defined as AF recorded for >30 s or AF symptoms persisting after the blanking period. For AF recurrences, we performed pharmacological or electrical

cardioversion and prescribed antiarrhythmic drugs or repeated the ablation to maintain SR.

Statistical analyses

Values are expressed as the mean ± standard deviation (SD) or median [interquartile range, (IQR)], based on their distributions. Categorical variables are shown as percentages, and between-group comparisons were conducted using chi-squared and Fisher’s exact tests. The natural logarithm-transformed (ln) UACR and ln-RHI were used to provide a normal distribution for these variables. The values of the eGFR, ln-UACR, and ln-RHI were compared among 3 measurements, at baseline and 6 and 12 months after the ablation, using a repeated measures analysis of variance (ANOVA) followed by a Tukey’s test. A multiple logistic regression analysis was used to determine the independent predictors of the changes in the eGFR (≥1 mL/min/1.73 m²) and UACR (≤10 mg/gCr). A single logistic analysis was performed to analyze the changes in the eGFR and UACR, as well as for the following categorical variables: an age >65 years, female sex, smoking, heart failure, vascular disease, eGFR ≤60 mL/min/1.73 m², hypertension, diabetes mellitus, dyslipidemia, obesity (body mass index ≥25 kg/m²), use of statins or renin-angiotensin aldosterone system (RAAS) inhibitors, LVEF <55%, the PeAF at baseline, and AF recurrence.

Thereafter, to determine the factors associated with the improvements in the renal function, a

multivariate logistic regression analysis, including variables with a p-value <0.05 in single logistic analyses, was performed. Associations between the change in the (Δ) ln-RHI and either the ΔeGFR or Δln-UACR were evaluated using a Pearson’s correlation analysis. A p <0.05 was considered statistically significant for all analyses. Statistical analyses were performed using JMP version 12 software (SAS Institute Inc., Cary, NC, USA).

RESULTS

The clinical characteristics and laboratory data were compared between the PAF and PeAF groups (Table 1). The proportion of male patients and frequencies of smoking, diabetes mellitus, and heart failure were higher in the PeAF than PAF group. The high-sensitivity C- reactive protein, brain natriuretic peptide, LA diameter, and UACR were also significantly higher in the PeAF than PAF group. In contrast, the eGFR and LVEF were significantly lower in the PeAF than PAF group.

Changes in the renal markers after the ablation

In the PeAF group, the eGFR significantly increased and the ln-UACR significantly decreased 6 months after the ablation. In the PAF group, however, these two parameters did not significantly change by 6 months after the ablation (Fig. 2, Online Table 1). PeAF at baseline independently predicted an increase in the eGFR [odds ratio (OR)=2.13, 95% confidence interval (CI) 1.35-3.4, p=0.001; Table 2] and a decrease in the UACR (OR=1.94, 95% CI 1.05-3.58, p=0.033; Table 3) after the ablation.

At 12 months after the ablation, all patients could be followed up in this study and SR was maintained. In both groups, the eGFR and ln-UACR remained comparable to the values obtained at 6 months (Online Table 1).

Changes in the LVEF after the ablation

The LVEF in the PeAF group was lower in the PAF group at baseline (59±8.4% versus 64.4±7.4%, p <0.001) and significantly increased 6 months after the ablation (63.1±6.5% versus baseline, p <0.001; Online Fig. 1). In the PeAF group, the ΔLVEF was significantly correlated with the ΔeGFR (r=0.28, p=0.002). In the PAF group, the LVEF was unchanged 6 months after the ablation, and the ΔLVEF was not correlated with the ΔeGFR (Online Fig. 1).

Impact of CKD on the changes in the renal and endothelial function after the ablation In the PeAF group, the AF duration until the ablation did not significantly differ between the patients with and without CKD (13±17 versus 13±14 months, respectively, p=0.6). In the patients without CKD, the eGFR and UACR improved by 6 months after the ablation in the PeAF group, but were unchanged after the ablation in the PAF group. In the patients with CKD, the eGFR improved from baseline to 6 months after the ablation in both the PAF and PeAF groups, with the UACR improving only in the PeAF group (Table 4).

At baseline, the ln-RHI was lowest in the PeAF with CKD group, with a significant increase by 6 months after the ablation in this group. After the ablation, the ln-RHI remained unchanged among the PAF with and without CKD groups, and the PeAF without CKD group (Table 4).

There was a significant correlation between the Δln-RHI and ΔeGFR in the PeAF with CKD group (r=0.35, p=0.03). For the PeAF without CKD and PAF groups, the Δln-RHI did not correlate with the ΔeGFR.

Only in the PeAF with albuminuria group, did the RHI, which had the lowest value at baseline, significantly increase after the ablation. The ln-RHI among the PAF with and without albuminuria groups, and the PeAF without albuminuria group remained unchanged after the ablation (Table 5).

AF recurrence and renal and endothelial function

AF recurrence up to 6 months after the ablation was observed in 41 (18%) and 26 (20%) patients in the PAF and PeAF group, respectively. Among the 67 patients with an AF

recurrence, SR was restored with cardioversion or naturally in 64, and was maintained with medications or re-ablation. All second ablation procedures were performed 6 to 12 months after the first procedure in 16 patients in the PAF group (7%) and 16 in the PeAF group (12%).

Therefore, the AF recurrence in these 64 patients was regarded as transient. In the remaining 3 patients, SR could not be maintained, despite treatment, and they were excluded from further analysis. We compared the temporal changes in the eGFR, ln-UACR, and ln-RHI between patients with and without a transient AF recurrence (Table 6). The eGFR, ln-UACR, and ln-RHI significantly improved after the ablation in the PeAF without a recurrence group. The eGFR and ln-RHI tended to improve after the ablation in the PeAF with a transient recurrence group. In this group, the ln-UACR decreased insignificantly after the ablation, compared to that at baseline.

These three parameters remained unchanged at 6 months after the ablation in the PAF group, regardless of a transient AF recurrence.

DISCUSSION

The main findings of this study were as follows. First, the eGFR and UACR improved 6 months after the ablation in the patients with PeAF, but not in those with PAF, and PeAF at baseline independently predicted an eGFR increase and UACR decrease 6 months after the ablation. Among the patients with PeAF, the endothelial function (RHI) was worse in the

individuals with CKD, and the RHI increased only in the patients with PeAF and CKD 6 months after the ablation, and the ΔRHI correlated with the ΔeGFR. Transient AF recurrence did not prevent the recovery of the renal and vascular endothelial function after the ablation in the PeAF patients.

Improvement in the renal function after ablation in patients with PeAF

Previous studies have indicated that catheter ablation of AF might improve the eGFR [7, 15]. However, it is unknown whether the effect of catheter ablation on the renal function varies

according to the AF type. Our results showed that the renal function, assessed by the eGFR and UACR, improved after ablation in patients with PeAF but not in those with PAF, indicating that the AF burden may exacerbate renal dysfunction. Therefore, catheter ablation may be used to improve the renal function and prognosis in patients with PeAF. Improvements in the renal function may partially explain the reduction in cardiovascular events after catheter ablation in patients with AF.

Several mechanisms may explain the improvements in the renal function after ablation among patients with PeAF. First, a SR-related hemodynamic improvement, including the renal circulation, could increase the eGFR. In this study, the LVEF significantly improved after

ablation only in the PeAF group, with a significant correlation between the ΔeGFR and ΔLVEF.

An improved endothelial function after ablation may partially explain the increased eGFR.

Several studies have suggested that the eGFR and UACR are associated with the vascular endothelial function [16, 17]. In our previous study, we demonstrated that SR restoration after catheter ablation was associated with an increase in the RHI in PeAF patients, but not in those with PAF [10]. In this study, we confirmed this finding and further identified a significant correlation between the ΔeGFR and Δln-RHI in patients with PeAF and CKD.

Impact of catheter ablation on albuminuria

Albuminuria and proteinuria are independent risk factors for cardiovascular events. A recent study reported that proteinuria increased the risk of the incidence of AF, even in patients with a normal eGFR [18]. Several studies demonstrated that increased albuminuria was

associated with a higher incidence of AF [2] and all-cause mortality, irrespective of the eGFR level [19]. Therefore, a decreased eGFR and albuminuria are thought to occur via different mechanisms. Our results indicated that SR restoration by ablation was also associated with a UACR decrease in patients with PeAF. Glomerular hypertension and endothelial dysfunction are required for albuminuria to occur [20]. These two phenomena are closely associated with the RAAS, and its activation is associated with AF development. Goette et al. demonstrated that the level of angiotensin-converting enzyme expression in the atrium is greater in patients with AF than in those in SR [21], suggesting that activation of the RAAS is greater with PeAF than with PAF because of the larger AF load. Thus, the glomerular pressure might increase during AF and decrease after SR restoration. We also found that the RHI significantly increased after ablation in patients with PeAF and albuminuria. Therefore, reducing the glomerular pressure, through the

suppression of the RAAS, and an improved endothelial function after ablation may lead to a reduction in the albuminuria among patients with PeAF.

Albuminuria may be a therapeutic target and a surrogate marker for predicting cardiovascular events. A previous study has reported that medications that reduce albuminuria also reduce renal and cardiovascular adverse events [22]. In our study, SR restoration by ablation significantly reduced the UACR in patients with PeAF, indicating that AF aggravates

albuminuria. However, albuminuria in patients with PeAF may be reversed. Although further studies are necessary, SR control after ablation could become a reassuring non-pharmacological option for reducing albuminuria and cardiovascular events in patients with PeAF.

Influence of transient AF recurrences on the outcomes

Navaravong et al. reported that patients with AF and stage 2 and 3 CKD had an increase in the eGFR after catheter ablation, regardless of a recurrence [15]. They suggested that the reduction in the AF burden possibly explains the increase in the eGFR, and our results are consistent with theirs. We found that the eGFR and RHI increased after ablation in patients with PeAF, irrespective of a transient recurrence. Our results indicated that a transient AF recurrence had few adverse effects on the improvement in the renal and vascular endothelial function. Our study also suggested that the AF burden was an important factor related to the eGFR and the change from PeAF to PAF could have beneficial renal and endothelial effects.

Study limitations

There were several limitations to this prospective study. First, data on patients without AF, patients with AF who did not undergo catheter ablation, patients with maintenance of SR only with anti-

arrhythmic drugs, and patients with persistent AF recurrence were unavailable or insufficient and this precluded comparisons. Second, the multivariate analysis was limited because some covariates and confounding variables might not have been considered or included. Third, a change in the comorbid conditions, including hypertension and diabetes mellitus, or a change in the prescriptions may have modified our results. However, the effect of those confounding factors on the renal or endothelial function would be small. This is because the outpatient care, such as dietary counselling and changes in the medications, was performed according to the clinical guidelines without distinction depending on the AF type. Finally, although we demonstrated an improvement in the renal and endothelial function in patients with PeAF, it was unknown whether catheter ablation was associated with a reduction in cardiovascular events because of the small number of patients and mid-term follow-up period, in addition to the possibility of a regression to the mean due to the small variation in the measured parameters. However, we considered it was important that the renal function improvement was maintained for at least 12 months, because AF increases the risk of a renal function decline. SR maintenance might suppress any progression of renal dysfunction [23].

Further study is needed to determine if the reduction in the cardiovascular events in the patients that received ablation is related to the improvements in these functions.

CONCLUSIONS

Maintaining SR with ablation improves the eGFR and UACR in patients with PeAF.

Since cardiovascular events are the primary cause of death associated with AF, catheter ablation may be a therapeutic option for reducing cardiovascular accidents and improving the prognosis, especially in patients with PeAF.

Funding: No financial support

Conflicts of interest: All authors have no conflicts of interest

REFERENCES

[1] Gomez-Outes A, Lagunar-Ruiz J, Terleira-Fernandez AI, Calvo-Rojas G, Suarez-Gea ML, Vargas-Castrillon E. Causes of death in anticoagulated patients with atrial fibrillation. J Am Coll Cardiol 2016;68:2508-21.

[2] Alonso A, Lopez FL, Matsushita K, Loehr LR, Agarwal SK, Chen LY, et al. Chronic kidney disease is associated with the incidence of atrial fibrillation: the Atherosclerosis Risk in Communities (ARIC) study. Circulation 2011;123:2946-53.

[3] Lin YJ, Chao TF, Tsao HM, Chang SL, Lo LW, Chiang CE, et al. Successful catheter ablation reduces the risk of cardiovascular events in atrial fibrillation patients with CHA2DS2-VASc risk score of 1 and higher. Europace 2013;15:676-84.

[4] Friberg L, Tabrizi F, Englund A. Catheter ablation for atrial fibrillation is associated with lower incidence of stroke and death: data from Swedish health registries. Eur Heart J 2016;37:2478-87.

[5] Ichijo S, Miyazaki S, Kusa S, Nakamura H, Hachiya H, Kajiyama T, et al. Impact of catheter ablation of atrial fibrillation on long-term clinical outcomes in patients with heart failure. J Cardiol 2018;72:240-6.

[6] Kim JO, Shim J, Lee SH, Yu HT, Kim TH, Uhm JS, et al. Clinical characteristics and rhythm outcome of catheter ablation of hemodynamically corrected valvular atrial fibrillation. J Cardiol 2019;73:488-96.

[7] Takahashi Y, Takahashi A, Kuwahara T, Okubo K, Fujino T, Takagi K, et al. Renal function after catheter ablation of atrial fibrillation. Circulation 2011;124:2380-7.

[8] Yoshino S, Yoshikawa A, Hamasaki S, Ishida S, Oketani N, Saihara K, et al. Atrial fibrillation- induced endothelial dysfunction improves after restoration of sinus rhythm. Int J Cardiol 2013;168:1280-5.

[9] Ganesan AN, Chew DP, Hartshorne T, Selvanayagam JB, Aylward PE, Sanders P, et al. The impact of atrial fibrillation type on the risk of thromboembolism, mortality, and bleeding: a systematic review and meta-analysis. Eur Heart J 2016;37:1591-602.

[10] Okawa K, Miyoshi T, Tsukuda S, Hara S, Matsuo N, Nishibe N, et al. Differences in endothelial dysfunction induced by paroxysmal and persistent atrial fibrillation: Insights from restoration of sinus rhythm by catheter ablation. Int J Cardiol 2017;244:180-5.

[11] January CT, Wann LS, Alpert JS, Calkins H, Cigarroa JE, Cleveland JC, Jr., et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2014;64:e1-76.

[12] Japan nephrology society. [Special issue: Clinical practice guidebook for diagnosis and treatment of chronic kidney disease 2012]. Nihon Jinzo Gakkai shi 2012;54:1034-191.

[13] Rubinshtein R, Kuvin JT, Soffler M, Lennon RJ, Lavi S, Nelson RE, et al. Assessment of endothelial function by non-invasive peripheral arterial tonometry predicts late cardiovascular adverse events. Eur Heart J 2010;31:1142-8.

[14] Kuvin JT, Patel AR, Sliney KA, Pandian NG, Sheffy J, Schnall RP, et al. Assessment of peripheral vascular endothelial function with finger arterial pulse wave amplitude. Am Heart J 2003;146:168-74.

[15] Navaravong L, Barakat M, Burgon N, Mahnkopf C, Koopmann M, Ranjan R, et al.

Improvement in estimated glomerular filtration rate in patients with chronic kidney disease undergoing catheter ablation for atrial fibrillation. J Cardiovasc Electrophysiol 2015;26:21-7.

[16] Recio-Mayoral A, Banerjee D, Streather C, Kaski JC. Endothelial dysfunction, inflammation and atherosclerosis in chronic kidney disease--a cross-sectional study of predialysis, dialysis and kidney-transplantation patients. Atherosclerosis 2011;216:446-51.

[17] Stehouwer CD, Henry RM, Dekker JM, Nijpels G, Heine RJ, Bouter LM. Microalbuminuria is associated with impaired brachial artery, flow-mediated vasodilation in elderly individuals without and with diabetes: further evidence for a link between microalbuminuria and endothelial dysfunction--the Hoorn Study. Kidney Int Suppl 2004;92:S42-4.

[18] Molnar AO, Eddeen AB, Ducharme R, Garg AX, Harel Z, McCallum MK, et al. Association of proteinuria and incident atrial fibrillation in patients with intact and reduced kidney function. J Am Heart Assoc 2017;6:e005685.

[19] Matsushita K, van der Velde M, Astor BC, Woodward M, Levey AS, de Jong PE, et al.

Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet 2010;375:2073-81.

[20] Nakagawa T, Sato W, Glushakova O, Heinig M, Clarke T, Campbell-Thompson M, et al.

Diabetic endothelial nitric oxide synthase knockout mice develop advanced diabetic nephropathy.

J Am Soc Nephrol 2007;18:539-50.

[21] Goette A, Staack T, Rocken C, Arndt M, Geller JC, Huth C, et al. Increased expression of extracellular signal-regulated kinase and angiotensin-converting enzyme in human atria during atrial fibrillation. J Am Coll Cardiol 2000;35:1669-77.

[22] de Zeeuw D, Remuzzi G, Parving HH, Keane WF, Zhang Z, Shahinfar S, et al. Albuminuria, a therapeutic target for cardiovascular protection in type 2 diabetic patients with nephropathy.

Circulation 2004;110:921-7.

[23] Beyer-Westendorf J, Kreutz R, Posch F, Ay C. The CHA2DS2-VASc score strongly correlates with glomerular filtration rate and predicts renal function decline over time in elderly patients with atrial fibrillation and chronic kidney disease. Int J Cardiol 2018;253:71-7.

FIGURE LEGENDS

Figure 1 Flowchart of the study participants. A total of 422 patients with AF were admitted to Kagawa Prefectural Central Hospital for catheter ablation between May 2013 and March 2016.

Ultimately, the eGFR, UACR, and RHI were compared, before and after ablation, in 358 patients with both types of AF.

AF, atrial fibrillation; ECG, electrocardiogram; eGFR, estimated glomerular filtration rate; PAF, paroxysmal atrial fibrillation; PeAF, persistent atrial fibrillation; RHI, reactive hyperemia index;

UACR, urinary to-albumin-creatinine ratio.

Figure 2 Changes in the eGFR, UACR, and RHI after the ablation in patients with PAF and PeAF. In patients with PeAF, the eGFR and ln-UACR significantly improved 6 months after the ablation as compared to the baseline measurements. In the patients with PAF, these three parameters remained unchanged. Values were compared by a repeated measures ANOVA.

*p<0.05 vs. before the ablation.

eGFR, estimated glomerular filtration rate; ln, natural logarithm-transformed; PAF, paroxysmal atrial fibrillation; PeAF, persistent atrial fibrillation; RHI, reactive hyperemia index; UACR, urinary albumin-to-creatinine ratio.

Online Figure 1 Changes in the LVEF after the ablation, and the relationship between the ΔLVEF and ΔeGFR in each patient with AF. The LVEF was significantly lower at baseline in the patients with PeAF than in those with PAF, and significantly increased from baseline after the ablation only in the patients with PeAF. The change in the LVEF was significantly correlated with the change in the eGFR in patients with PeAF, but not in those with PAF.

AF, atrial fibrillation; eGFR, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; PAF, paroxysmal atrial fibrillation; PeAF, persistent atrial fibrillation.

Patients scheduled for catheter ablation n=422

Total recruited n=375

eGFR, UACR, and RHI measurements

Catheter ablation

eGFR, UACR, and RHI measurements 6 and 12 months after catheter ablation ECG check performed at each clinical visit

Excluded n=47 [Previous catheter ablation (n=42) Concurrent dialysis (n=5)]

Excluded n=13 [Lost to follow-up (n=10) Non-maintenance of sinus rhythm (n=3)]

Long-term ECG monitoring

Excluded n=4 [Unable to obtain RHI measurements (n=4)]

Data available for analysis n=358

Figure 1: Flow chart of the study participants

0 1 2 3 4 5

PAF PeAF

30 40 50 60 70 80 90 100

PAF PeAF

0.0 0.2 0.4 0.6 0.8

PAF PeAF

eGFR Ln-UACR Ln-RHI

ml/min/1.73m²

Figure 2: Changes in the eGFR, UACR, and RHI after the ablation in patients with PAF and PeAF

Before the ablation

6 months after the ablation 12 months after the ablation

* * * *

Table 1. Baseline characteristics

PAF (n=229) PeAF (n=129) p

Age (years) 68 (63-74) 69 (63-73) 0.69

Male sex 141 (62) 94 (73) 0.037

BMI (kg/m²) 23.6 (21.4-25.8) 24.2 (22.4-26.0) 0.10

Smoking habits 10 (4) 17 (13) 0.003

Hypertension 138 (60) 77 (60) 1

Diabetes mellitus 36 (16) 32 (25) 0.049

Dyslipidemia 159 (69) 82 (64) 0.29

Heart failure 9 (4) 26 (20) <0.001

Vascular disease 32 (14) 18 (14) 1

Hs-CRP (mg/dL) 0.06 (0.03-0.11) 0.09 (0.05-0.16) 0.002

BUN (mg/dL) 16.0 ± 5.2 16.7 ± 4.8 0.22

Creatinine (mg/dL) 0.81 ± 0.24 0.87 ± 0.25 0.021

Cystatin-C (mg/L) 1.06 ± 0.68 1.08 ± 0.31 0.69

eGFR (mL/min/1.73 m²) 72.1 ± 17.7 68.7 ± 18.7 0.084

UACR (mg/g) 9.8 (5.8-22.8) 14.2 (7.3-38.5) 0.003

BNP (pg/mL) 49.0 (23.3-97.3) 143.4 (78.0-213.5) <0.001

LVEF 64.4 ± 7.4 59 ± 8.4 <0.001

LA dimension (mm) 35.3 ± 6.4 39.9 ± 5.3 <0.001

Medication

Statins 61 (26) 29 (22) 0.45

ACEIs or ARBs 93 (41) 55 (43) 0.74

Values are presented as the mean ± SD or median (interquartile range) in accordance with the distribution, and number of patients (%).

Abbreviations: ACEI= angiotensin-converting enzyme inhibitor; ARB= angiotensin II receptor blocker; BMI= body mass index; BNP= brain natriuretic peptide; BUN= blood urea nitrogen; eGFR= estimated glomerular filtration rate; Hs-CRP= high-sensitivity C- reactive protein; LA= left atrium, LVEF= left ventricular ejection fraction; PAF=

paroxysmal atrial fibrillation; PeAF= persistent atrial fibrillation; SD= standard deviation; UACR= urine albumin-to-creatinine ratio

Table 2. Factors related to an increase in the eGFR at 6 months after the ablation

Univariate Multivariate

Factor OR (95% CI) p OR (95% CI) p

PeAF at baseline 2.28 (1.46-3.57) < 0.001 2.13 (1.35-3.4) 0.001 Age>65 years old 0.91 (0.59-1.42) 0.679

Female sex 0.89 (0.58-1.38) 0.616

Smoking 2.54 (1.18-5.43) 0.012 2.22 (1.04-5.08) 0.04 Heart failure 0.69 (0.42-1.13) 0.142

Vascular disease 0.82 (0.45-1.48) 0.5 eGFR≤60

ml/min^/1.73m²

2.17 (1.31-3.57) 0.002 1.84 (1.09-3.12) 0.021

Hypertension 1.48 (0.97-2.27) 0.07

Diabetes mellitus 2.21 (1.26-3.86) 0.004 1.80 (1.00-3.29) 0.048 Dyslipidemia 1.06 (0.68-1.65) 0.802

Obesity

(BMI≥25kg/m²)

1.58 (0.97-2.56) 0.064

Use of statin 1.2 (0.74-1.93) 0.466

Use of ACEI/ARB 1.25 (0.82-1.9) 0.307 LVEF<55% 1.62 (0.91-2.88) 0.099 AF recurrence 0.86 (0.51-1.47) 0.588

Abbreviations: ACEI= angiotensin-converting enzyme inhibitor; AF: atrial fibrillation; ARB= angiotensin II receptor blocker; BMI= body mass index; CI:

confidence interval; eGFR= estimated glomerular filtration rate; LVEF= left ventricular ejection fraction; OR: odds ratio; PeAF= persistent atrial fibrillation

Table 3. Factors related to a decrease in the UACR at 6 months after the ablation

Univariate Multivariate

Factor OR (95% CI) p OR (95% CI) p

PeAF at baseline 1.91 (1.15-3.18) 0.013 1.94 (1.05-3.58) 0.033 Age>65 years old 1.28 (0.74-2.2) 0.374

Female sex 1.31 (0.78-2.2) 0.304 Smoking 0.68 (0.27-1.71) 0.399 Heart failure 1.47 (0.83-2.55) 0.187

Vascular disease 2.8 (1.49-5.28) 0.002 1.62 (0.75-3.41) 0.215 eGFR≤60

ml/min^/1.73m²

2.86 (1.67-4.88) <0.001 1.68 (0.89-3.16) 0.109

Hypertension 2.98 (1.66-5.36) <0.001 2.51 (1.19-5.49) 0.015 Diabetes mellitus 2.54 (1.43-4.51) 0.002 2.07 (1.06-4.00) 0.033 Dyslipidemia 2.54 (1.43-4.51) 0.002 1.24 (0.65-2.41) 0.523 Obesity

(BMI≥25kg/m²)

1.75 (0.99-3.09) 0.054

Use of statin 1.52 (0.88-2.64) 0.138

Use of ACEI/ARB 1.91 (1.15-3.16) 0.012 1.01 (0.52-1.98) 0.97 LVEF<55% 2.33 (1.27-4.29) 0.008 1.93 (0.93-3.98) 0.076 AF recurrence 1.58 (0.82-3.07) 0.184

Abbreviations: ACEI= angiotensin-converting enzyme inhibitor; AF: atrial fibrillation; ARB= angiotensin II receptor blocker; BMI= body mass index; CI:

confidence interval; eGFR= estimated glomerular filtration rate; LVEF= left ventricular ejection fraction; OR: odds ratio; PeAF= persistent atrial fibrillation

Table 4. Changes in the eGFR, UACR, and RHI for each AF type according to the

presence or absence of CKD

BL 6 months after ablation

p

vs. BL

eGFR mL/min^/1.73 m²

PAF without CKD (n=179) 78.9±12.7 78.5±13.5 0.23

with CKD (n=50) 47.7±9.5 50.0±11.7 0.002

PeAF without CKD (n=90) 78.1±13 80.6±14.1 0.009

with CKD (n=39) 46.8±9.1 51.4±11.7 <0.001

ln-UACR

PAF without CKD (n=179) 2.2±1.2 2.2±1.3 0.26

with CKD (n=50) 3.6±1.7 3.6±1.6 0.42

PeAF without CKD (n=90) 2.6±1.3 2.4±1.3 0.016

with CKD (n=39) 4.0±2 3.6±1.8 0.013

ln-RHI

PAF without CKD (n=179) 0.58±0.25 0.56±0.28 0.21

with CKD (n=50) 0.56±0.31 0.58±0.27 0.23

PeAF without CKD (n=90) 0.53±0.28 0.55±0.27 0.3

with CKD (n=39) 0.34±0.27 0.47±0.25 0.004

Values are presented as the mean±SD.

Abbreviations: BL= baseline; eGFR= estimated glomerular filtration rate; CKD= chronic kidney disease; Ln= natural logarithm-transformed; PAF= paroxysmal atrial fibrillation;

PeAF= persistent atrial fibrillation; RHI= reactive hyperemia index; UACR= urine albumin-to-creatinine ratio

Table 5. Changes in the eGFR, UACR, and RHI for each AF type according to the

presence or absence of albuminuria

BL 6 months after ablation

p

vs. BL

eGFR mL/min^/1.73 m²

PAF without albuminuria (n=116) 77.2±15.2 77.2±14.9 0.46

with albuminuria (n=113) 67.1±18.7 67.2±18.9 0.37

PeAF without albuminuria (n=42) 77.1±16 80.3±15.1 0.01

with albuminuria (n=87) 64.6±18.7 67.7±19.3 <0.001

ln-UACR

PAF without albuminuria (n=116) 1.5±0.7 1.8±1 <0.001

with albuminuria (n=113) 3.5±1.2 3.3±1.6 <0.001

PeAF without albuminuria (n=42) 1.6±0.6 1.7±0.8 0.17

with albuminuria (n=87) 3.7±1.5 3.3±1.6 <0.001

ln-RHI

PAF without albuminuria (n=116) 0.57±0.26 0.56±0.28 0.34

with albuminuria (n=113) 0.58±0.27 0.57±0.27 0.83

PeAF without albuminuria (n=42) 0.58±0.27 0.58±0.25 0.99

with albuminuria (n=87) 0.42±0.28 0.50±0.27 0.012

Values are presented as the mean±SD.

Abbreviations: BL= baseline; eGFR= estimated glomerular filtration rate; Ln= natural logarithm-transformed; PAF= paroxysmal atrial fibrillation; PeAF= persistent atrial fibrillation; RHI= reactive hyperemia index; UACR= urine albumin-to-creatinine ratio

Table 6: Changes in the eGFR, UACR, and RHI for each AF type according to the presence or absence of a transient AF recurrence

BL 6 months after ablation

p

vs. BL eGFR ml/min^/1.73 m²

PAF without tAF rec (n=188) 73.1±17.3 73.6±17.2 0.36 with tAF rec (n=41) 67.4±18.8 66.1±18.6 0.25 PeAF without tAF rec (n=103) 67.4±19 70.7±19.4 0.001 with tAF rec (n=26) 73.7±16.8 76.2±16.5 0.051 Ln UACR

PAF without tAF rec (n=188) 2.4±1.4 2.5±1.5 0.11

with tAF rec (n=41) 2.8±1.6 2.6±1.8 0.19

PeAF without tAF rec (n=103) 3.1±1.7 2.9±1.5 0.002

with tAF rec (n=26) 2.8±1.4 2.5±1.6 0.46

Ln RHI

PAF without tAF rec (n=188) 0.57±0.26 0.56±0.28 0.48 with tAF rec (n=41) 0.60±0.28 0.58±0.23 0.93 PeAF without tAF rec (n=103) 0.47±0.29 0.52±0.27 0.047 with tAF rec (n=26) 0.48±0.28 0.56±0.24 0.058

Values are presented as the mean±SD.

Abbreviations: AF= atrial fibrillation; BL= baseline; eGFR= estimated glomerular filtration rate; Ln= natural logarithm-transformed; PAF= paroxysmal atrial fibrillation;

PeAF= persistent atrial fibrillation; RHI= reactive hyperaemia index; tAF rec= transient AF recurrence; UACR= urine albumin-to-creatinine ratio;

Supplementary Table. Changes in the eGFR, UACR, and RHI after the ablation.

Baseline 6 months after the ablation

12 months after the ablation

p

eGFR mL/min^/1.73 m²

Total (n=358) 70.9 ± 18.1 72.1 ± 18.1* 71.7 ± 18.5 0.015

PAF (n=229) 72.1 ± 17.7 72.3 ± 17.6 72.3 ± 18.0 0.66

PeAF (n=129) 68.7 ± 18.7 71.8 ± 18.9* 70.9 ± 19.3* <0.001

ln-UACR

Total (n=358) 2.7 ± 1.5 2.6 ± 1.5 2.7 ± 1.5 0.24

PAF (n=229) 2.7 ± 1.3 2.8 ± 1.3 2.7 ± 1.4 0.17

PeAF (n=129) 3.1 ± 1.6 2.8 ± 1.5* 2.9 ± 1.7* 0.029

ln-RHI

Total (n=358) 0.54 ± 0.28 0.55 ± 0.27 0.54 ± 0.26 0.73

PAF (n=229) 0.57 ± 0.26 0.56 ± 0.28 0.56 ± 0.29 0.84

PeAF (n=129) 0.47 ± 0.29 0.53 ± 0.26 0.51 ± 0.22 0.10

*p <0.05 vs. baseline.

Values are presented as the mean±SD.

Abbreviations: eGFR= estimated glomerular filtration rate; ln= natural logarithm- transformed; PAF= paroxysmal atrial fibrillation; PeAF= persistent atrial fibrillation, RHI= reactive hyperemia index; UACR= urine albumin-to-creatinine ratio

30

20

10

0

-10

-20

-15 -10 -5 0 5 10 15 20 25 30 0

10 20 30 40 50 60 70 80

PAF PeAF

Before ablation 6 months after

p=0.15 p <0.001 LVEF (%)

(n=229) (n=129)

Supplementary Figure: Changes in the LVEF after the ablation, and the relationship between the ΔLVEF and ΔeGFR in each patient with AF

ΔeGFR (ml/min/1.73m

)

0

n=129 r=0.28 p=0.002 PeAF

0

30

20

10

0

-10

-20

-15 -10 -5 0 5 10 15 20 25 30 PAF

n=229

r=0.06

p=0.37