Symptoms and health‐related quality of life 5 years after catheter ablation of atrial fibrillation

Abstract Objectives To investigate the effect of catheter ablation (CA) on symptoms and health‐related quality of life (HRQoL) after 5 years, and analyze predictors of recurrence of symptoms. Background The primary indication for CA of atrial fibrillation (AF) is to reduce symptoms and improve HRQoL where long‐term follow‐up are sparse. Methods In this observational, long‐term, single‐center study, patients were recruited from Linköping University Hospital, Sweden. They were aged ≥18 years and had been referred for CA from November 2011 until June 2019. Arrhythmia‐specific symptoms and HRQoL were assessed by patient‐reported outcome measures (PROMs) with the Arrhythmia‐Specific questionnaire in Tachycardia and Arrhythmia (ASTA). Results In the study were 1521 patients, 69% men, mean age 62 years. At baseline, 87% of the patients and at the 5‐year follow‐up 80% of those eligible filled out the ASTA questionnaire. At follow‐up, 50% reported freedom from symptoms, 18% had >50% symptom reduction, 14% had a minor reduction, while 18% reported no effect or a worsening of symptoms. Factors predicting symptoms were female gender (hazard ratio [HR]: 1.8; 1.2–2.8), body mass index ≥ 35 (HR: 3.9; 1.6–9.8), and ischemic heart disease (IHD) (HR: 2.6; 1.2–5.9). After 5 years, breathlessness during activity, weakness/fatigue, and tiredness were still the most common symptoms; regarding HRQoL they were impaired physical ability and deteriorated life situation. Conclusions and Clinical Implications This clinical cohort of patients with AF evaluated through PROMs showed that CA had long‐lasting effects on symptoms and HRQoL and that the use of PROMs in clinical routines was feasible. Factors predicting symptoms after CA were female gender, IHD, and obesity, an important reminder to encourage lifestyle management.


| INTRODUCTION
Treatment outcomes assessed by patient-reported outcome measures (PROMs) need to be addressed in patients with atrial fibrillation (AF) and should be included in clinical routines in connection with catheter ablation (CA). 1-3 AF-related symptoms can be graded either with the physician-assessed European Heart Rhythm Association score or with PROMs. 4 According to the European Society of Cardiology (ESC) guidelines, treatment with CA is indicated in symptomatic AF patients with negatively affected daily life. Importantly, guidelines from 2020 highlight the use of PROMs and recommend disease-specific evaluations in clinical routines. 5,6 One such tool for assessment is the Arrhythmia-Specific questionnaire in Tachycardia and Arrhythmia (ASTA), developed and validated in a Swedish patient population with AF, supraventricular and ventricular arrhythmias, and which assesses symptom burden and health-related quality of life (HRQoL). 7,8 Even though the primary indication for CA in patients with AF is symptom relief, most studies report results based upon rhythm rather than symptoms and effects on HRQoL, and the follow-up is seldom long-term. The aim of this paper was to investigate the effect of CA on symptoms and HRQoL and to analyze possible predictors of remaining symptoms 5 years after CA.

| Study design
This was an observational, single-center, long-term cohort study in a Swedish patient population treated with CA of AF at the University Hospital, Linköping, Sweden, between November 2011 and June 2019.

| Patients and follow-up
Consecutive patients referred for CA of AF were asked to complete the ASTA questionnaire before and 5 years after CA, as part of the clinical routine. Patients received written instructions approximately 2 weeks before the CA on how to access and fill out the ASTA questionnaire. At the time of follow-up, the patients received written instructions on how to access the questionnaire. Those not responding within one month received a reminder. Patients included were those aged ≥18 years with sufficient knowledge of the Swedish language, capable of independently filling out the questionnaire, and willing to participate.

| PROMs with the arrhythmia-specific ASTA questionnaire
The first part of the ASTA questionnaire evaluates the latest episode of arrhythmia and current medication. The second part assesses symptom burden with a nine-item symptom scale (ASTA symptom scale) with the response alternatives: "No (0), Yes, to a certain extent (1), Yes, quite a lot (2) or Yes, a lot (3)." There are also questions regarding the experiences of arrhythmia (palpitations) and on whether the respondent has ever fainted or come close to fainting. Part three assesses HRQoL with a 13-item scale, with the same four-point response scale (ASTA HRQoL scale). The ASTA HRQoL scale is divided into a seven-item physical subscale and a six-item mental subscale. Scale scores are calculated where values range from 0 to 100. Higher scores reflect a higher symptom burden and a worse effect on HRQoL due to arrhythmia. When patients answer "No, I have no arrhythmia" at follow-up the recommended scoring for the ASTA scales is zero 7-9 (Supporting Information 1).

| Ethical considerations
The Regional Ethical Review Board at the Faculty of Health Sciences, Linköping, Sweden approved the extraction and publication of the clinical data (DNR 2016-349-31). The study complies with the Declaration of Helsinki. 10 This evaluation is a part of the clinical routine, and oral informed consent was considered adequate.

| Statistical analysis
For baseline data, continuous variables were expressed as means and standard deviation (±SD). Categorical data were reported as counts with percentages within brackets. The Kolmogorov-Smirnov test was used to test the normality of the samples. A χ 2 test was used for All reported p values were two-sided and a p-value of less than .05 was considered statistically significant. The analyses were performed using the SPSS 24.0 (SPSS).

| Patients
At baseline, 1323 patients out of 1521 filled out the ASTA questionnaire; 69% were men, with a mean age of 62 years (±10.18).
One thousand patients were not eligible for the 5-year follow-up, as the time that had elapsed from the index procedure to the end of recruitment was shorter than 5 years. Finally, 521 patients were eligible for the 5-year follow-up. All patients received the ASTA questionnaires and 418 responded (80%), of whom 185 completed all parts of the ASTA questionnaire, that is, those still experiencing symptoms in relation to arrhythmia.
There were no differences in background characteristics between those who responded and the nonresponders. The mean number of CA procedures per patient during the follow-up was 1.6. The most commonly used antiarrhythmic medication was a beta-blocker, followed by amiodarone, and Class IC, mostly flecainide. The most common concomitant disease was hypertension (Table 1).

| Treatment effects of CA over time
At the 5-year follow-up, 52% reported freedom from AF-related symptoms. A major reduction (>50% symptom reduction compared to baseline) was seen in 18%, and 14% had a minor reduction (10%-50%). Only 9% reported no treatment effect on symptoms and another 9% experienced a worsening of symptoms after the CA  men and women at the 5-year follow-up: −5.6 (95% CI: −7.6 to −3.6; p < .0001; Table 3).
Older patients (age ≥ 75 years) and younger patients (age < 75) reported the same significant improvement in the symptom scale score at the 5-year follow-up (p = .295), that is, there were no differences between the age groups.

| Arrhythmia-related symptoms at baseline and at the 5-year follow-up
The most commonly reported symptoms at baseline were weakness/ fatigue, breathlessness during activity and tiredness and at the 5-year follow-up, the same symptoms were experienced but to a   significantly lesser degree. Worry/anxiety in connection with arrhythmia was reported by 886 patients (67%) at baseline and was still present in 138 (59%) after 5 years (see Table S1).

| Effect of catheter ablation-ASTA HRQoL scale scores 5 years after the initial procedure
The ASTA HRQoL scale score was significantly lower 5 years after CA   Table S2).

| Predictors of arrhythmia-related symptoms 5 years after the initial procedure
Factors predicting the presence of symptoms 5 years after the initial CA were female gender, with an 80% increased risk of reporting  Table 4).

| DISCUSSION
Our main findings were that half of the patients undergoing CA for AF were free from arrhythmia-related symptoms at the time of follow-up, and in those still reporting symptoms, more than 80% experienced a symptom reduction after 5 years. The three significant predictors for remaining symptomatic at the time of follow-up were female gender, obesity, and IHD.
To the best of our knowledge, this is the largest observational

| Effect of catheter ablation on symptoms-comparisons over time
Altogether, more than 80% of the study participants experienced an improved arrhythmia-related situation, consistent with the overall aim of ablation therapy in patients with AF. 5  This may suggest that the symptoms could be due to extra beats rather than AF, which is consistent with an overall reduction in symptom burden in those still experiencing symptoms. It is well known that there is a limited relationship between symptoms and actual heart rhythm, and this has to be considered when interpreting this finding. 12,13 Three symptoms were dominant: breathlessness during activity, weakness/fatigue, and tiredness but to a lesser extent after CA.
The ASTA symptom scale score decreased over time, and differences were seen between genders both before and after CA. As shown earlier, women are usually more symptomatic, have a higher heart rate during AF, are less likely to undergo CA, have the worse functional capacity, and experience more anxiety than men, all of which are seen for all types of AF. 3,[14][15][16][17] Seventy-five years of age was chosen as a cutoff for defining elderly patients, as this age relates to the risk of thromboembolic complications in the CHA 2 DS 2 VASC scoring. 18 Elderly patients had the same long-term treatment benefits as younger patients; thus our Note: Multiple logarithmic regression analysis was performed to determine possible predictors of the presence of symptoms 5 years after catheter ablation of atrial fibrillation. The model was fitted by an enter method where all variables were entered into the original model and then variables with probability values of .05 were removed. The predictors that turned out to be significant are presented in this table.
| 47 data strengthen the argument that patients with AF should not be excluded from CA solely because of age. This finding is in line with a previous study with a shorter follow-up, demonstrating a clear benefit of CA treatment for the elderly. 19

| Effect of catheter ablation on HRQoL-comparisons over time
HRQoL in patients with AF is influenced by typical AF-related symptoms but also several other factors, such as worry, potential side-effects of medication, risk of recurrence, and the need for repeat procedures after CA. Several patient-related factors also influence HRQoL, such as age and gender as well as comorbidities common in patients with AF. Information about what arrhythmia-specific concerns patients experience in a long-term perspective after CA is sparse and the definition of successful treatment may vary depending on patient or healthcare provider perspectives.
On a group level, HRQoL improved significantly over time, without any differences found between genders or age groups.
Two studies included similar cohorts to the present study comparing CA to medical therapy-CABANA and CAPTAF. 20,21 In CABANA, 2204 patients were randomized, and the study focuses on data from the 12-month follow-up, with assessments performed with two arrhythmiaspecific PROMs (AFEQT and MAFSI). The positive improvements seen at 12 months were sustained over the 5-year follow-up period. The study had a high response rate, equal to the present study. In CAPTAF, 155 patients were randomized and evaluated with a generic questionnaire (SF-36) with a focus on general health over a 12-month period.
Both studies favored treatment with CA.
At baseline the four most commonly reported items with a negative impact on HRQoL were: "The ability to perform physical activities," "deteriorated life situation," "avoided planning for things they would like to do," and "Feel unable to work, study or carry out daily activities" and these were still the most commonly reported items with a negative impact on HRQoL. These items in the HRQoL scale were also the most frequently reported in a previously published large study on patients with AF before CA. 3 In one study making use of a disease-specific questionnaire and an implantable loop recorder the reduction or absence of documented AF was closely related to the reduction or elimination of AF-related symptoms. 22 It is reasonable to suspect that the remaining arrhythmia was the most probable reason for the observed negative impact on HRQoL still present after 5 years.

| Predictors of arrhythmia at 5 years after the initial procedure
There were three independent predictors of negative long-term treatment outcome: female gender, obesity (BMI ≥ 35), and IHD. Our data is in line with previous studies that demonstrate that female gender and obesity predict arrhythmia-related symptoms in patients with AF treated with CA. [23][24][25][26] As discussed previously, women usually have a higher symptom burden than men and often experience a more negatively influenced HRQoL. 3,17,27 However, we did not find any gender differences regarding HRQoL at follow-up. One explanation may be an adaptation to the symptoms, that is, subjects may have found coping strategies and thus experienced less negative influence on HRQoL. Another possibility is that the heart rate during episodes slows down over time. 15 The mean BMI in our cohort was 27.8-equal between the sexesconfirming that many patients with AF referred for CA are overweight, some being obese, which is in line with previously reported cohorts. 23,28,29 Obesity is known to be a significant predictor for reduced success rate after CA and has a negative impact on HRQoL, physical health status, and symptom burden during AF episodes. [23][24][25][26]30,31 The importance of self-management strategies with lifestyle interventions is pointed out in the guidelines and is important to consider before referral for CA. 5,29 There is solid evidence regarding the positive effects of lifestyle management, which can lead to a reduced symptom burden, absence of AF, and positive effects on comorbidities. 26,28 The third predictive factor seen was IHD, giving a threefold increased risk of having arrhythmia-related symptoms five years after the ablation procedure. It is known that IHD in combination with AF worsens the prognosis and has been described as "double trouble." [23][24][25]32 5.4 | Take-home messages/clinical significance in the present study

| Methodological considerations, strengths and limitations
This is the largest long-term follow-up observational study in patients with AF undergoing CA, representing the patients' real-life situation assessed with PROMs in a clinical setting. The current study, in contrast to many other studies, includes a nonselected patient group representing a common population with AF planned for CA, with only one exclusion criterion, which was difficulties with the Swedish language.
Large-scale clinical follow-up is a challenge since rhythm, symptoms, and HRQoL ideally should all be evaluated long term. The follow-up was carried out 5 years after the procedure, irrespective of whether the patient was waiting for a new evaluation or even waiting for a repeat procedure. At the 5-year follow-up, 80% of eligible patients responded to the ASTA questionnaire, which constitutes a very high response rate. We strived to conduct the assessments via the ASTA web version, thus avoiding missing data and making possible direct data import into the database, something that also minimizes input errors.
The aim of the study was to explore long-term follow-up data, seen from the patients' perspective. Data assessed was a part of clinical routine, based on PROMs without any information on actual rhythm. An electrocardiogram provides information about the rhythm "here and now" but does not capture the overall picture.

| CONCLUSIONS
This large clinical cohort of patients with AF, presenting real-life data in a 5-year long-term perspective with a high response rate, showed that CA has a long-lasting positive effect on both symptoms and HRQoL and to some extent changes the perception of AF.
Half of the patients did not experience any symptoms and more than 80% of those still having symptoms experienced an improvement.
Factors predicting remaining symptoms were female gender, the presence of IHD, and obesity, the latter serving as an important reminder to encourage lifestyle management. The use of diseasespecific PROMs for long-term follow-up in a routine clinical setting was feasible and added important clinical information about the care of patients with AF.