Relationship between quality of life and burden of recurrent atrial fibrillation following ablation: CAPCOST multicentre cohort study Free

Abstract

Aims

Atrial fibrillation (AF) significantly impairs patients’ quality of life (QOL). We performed this study to investigate the effect of AF-ablation success and atrial fibrillation burden (AFB) on QOL measures.

Methods and results

Overall, 230 patients with paroxysmal AF refractory to antiarrhythmic drugs were enrolled and underwent ablation in a multicentre, prospective cohort. Electrocardiogram, 48-h Holter, Canadian Cardiovascular Society Severity of Atrial Fibrillation (CCS-SAF), short form-12 (SF-12), and Atrial Fibrillation Effect on Quality of life (AFEQT) scales were used to assess patients. Atrial fibrillation burden was defined as total duration of AF during the month prior to each visit (h/month). The change in AFB was calculated as the difference between the month prior to the 12-month post-ablation and the baseline pre-ablation. The Minimal Clinically Important Difference (MCID) was considered as a 19-point change for AFEQT and 3–5-point change for SF-12 scores. There was significant rise in the AFEQT and SF12 and decrease in CCS-SAF score post-AF ablation; however, the magnitude of these changes was greater in patients without AF recurrence (P < 0.05). The QOL score that best differentiated patients with and without recurrence was AFEQT, while, CCS-SAF was the most specific score. Patients with AFB decrease >19 h/month had significantly greater change in QOL scores. Atrial fibrillation burden < 24 h/month at 12-months post-ablation was associated with significant changes in QOL and CCS-SAF when adjusting for baseline scores and other covariates. These changes were consistent with the MCID of these measures.

Conclusion

Patients experience significant improvements in QOL post-ablation, which correlate with a decrease in AFB despite ongoing brief recurrences of AF.

Clinical Trial Registration

NCT01562912. https://www.clinicaltrials.gov/ct2/show/NCT01562912? term=capcost&rank=1

Introduction

Atrial fibrillation (AF) is the most common sustained arrhythmia, which significantly impairs patients’ quality of life (QOL) and increases morbidity and mortality.^1–3 Pulmonary vein antrum isolation (PVAI) is the cornerstone treatment of patients with AF refractory to antiarrhythmic drugs. The main goal of treatment with PVAI is alleviating symptoms and improving QOL.^1,4 Although AF ablation has demonstrated efficacy over drug therapy, some studies have shown that QOL improves regardless of procedural success, which could imply a response bias in patients undergoing the procedure.^5,6 However, procedural success is often based on an endpoint of absence of any AF recurrence more than 30 s. This is a very high standard for procedural success that may be useful for clinical trials, but it is not necessarily an ideal measure of the clinical success of ablation. Many patients may experience significant reduction in atrial fibrillation burden (AFB) with accompanying improvement in their QOL despite having ongoing brief AF recurrences.^1,4–7 A sub-study of STAR AF⁴ showed patients could continue experiencing more than 2 h of AF per month and still have an improvement in QOL. The correlation of ablation success, residual AFB, and QOL improvement remains under-studied.^1,4–6 The purpose of this study is to investigate the effect of PVAI success on AFB and QOL measures.

Methods

Study design

This is a secondary analysis of the CAPCOST multicentre, open label, prospective cohort study performed from 2012 to 2017 at six tertiary centres (NCT01562912). Patients over 18 years of age with symptomatic paroxysmal AF refractory to at least one antiarrhythmic medication undergoing first time catheter ablation for AF were recruited. Patients with persistent AF, contraindication to systemic anticoagulation, left atrial size greater than 55 mm, and those who were or may potentially have been pregnant were excluded. Overall, 230 patients (mean age of 57.30 ± 10.80 years and 70.40% male) who met inclusion criteria were enrolled to undergo PVAI via multipolar duty cycled phased radiofrequency (RF) ablation (PVAC, Medtronic Inc.) or traditional point-by-point open-irrigated RF ablation. For the purpose of this study, all of the patients are pooled regardless of the specific technology used for ablation (see consort flow diagram, Supplementary Material online, S1).

Study endpoint

The endpoint of this study was to investigate the effects of PVAI success and AFB reduction on QOL measures.

Baseline and follow-up assessment

All patients underwent baseline pre-ablation assessment including record of comorbidities, medication history, morphological data, calculation of AFB, and assessment of symptom severity and QOL measures using Canadian Cardiovascular Society Severity of Atrial Fibrillation (CCS-SAF) scale, Atrial Fibrillation Effect on Quality of life (AFEQT), and short form 12 (SF12) health survey questionnaires.

Follow-up assessment occurred at 3, 6, 9, and 12 months post-ablation using a minimum of a 48-h Holter and a 12-lead electrocardiogram (ECG). If symptoms occurred between visits, extended loop recording, transtelephonic monitoring, or interrogation of implanted devices were also used where applicable.

Atrial fibrillation burden calculation

Quantification of AFB was performed using a simplified patient-reported scale (Supplementary material online, S2) including frequency of episodes, mean duration of episodes, and mean severity in the preceding month at baseline and each follow-up visit. Total duration of AFB estimation for the month prior to each visit was used for analysis and reported as h/month. The change in AFB was calculated as the difference in the duration of AFB between the month prior to the 12-month visit and the month prior to the baseline pre-ablation visit.

Recurrence

Arrhythmia recurrence was defined by symptoms and data from Holter, ECG, or loop recorder. Recurrence, for the primary efficacy endpoint of the study, was defined as manifestation of any atrial arrhythmia >30 s, excluding episodes observed during the initial 3 month blanking period.

Quality of life instruments

The CCS-SAF, SF12 health survey, and AFEQT scales were used to assess patients at baseline and at 3, 6, 9, and 12-months post-ablation.

The CCS-SAF score ranges from 0 (asymptomatic) to 4 (severe symptoms impacting QOL and daily activities) and is determined by recording possible AF related symptoms, symptom–rhythm correlation, and the impact of these symptoms on QOL and daily activities.⁸

The SF12 is a subset of the short form 36 (SF36) health survey, which is a validated general QOL measure for several conditions including AF. It consists of 12 subsets of 36 items in SF36 questionnaire, assessing 8 health concepts including general health perception, physical functioning, social functioning, role limitation due to physical problems, bodily pain, mental health, role limitation due to emotional problems, and vitality. Two summary measures are aggregated as mental (MCS) and physical component summaries (PCS) that are normalized to overall population mean of 50 ± 10.^7,9 The Minimal Clinically Important Difference (MCID) of MCS and PCS scores have been reported as 3- to 5-point change in other studies.^10,11

The AFEQT is an AF-specific QOL measure consisting of 20 items that assess symptoms, functional impairment, treatment concerns, and satisfaction.² The MCID of AFEQT score has been reported as 19-point change. We therefore used this cut-off to categorize ΔAFEQT score (12 months-baseline).¹²

Statistical analysis

Descriptive statistics were reported as mean ± standard deviation for continuous variables and frequency (percentage) for categorical variables. Comparison of QOL score changes between patients who maintained sinus rhythm and those who recurred, was performed using independent sample T-tests in normally distributed data and Mann–Whitney U test in non-normally distributed data. Comparison of QOL score changes among categories of AFB was performed using one-way Analysis of variance (ANOVA) in normally distributed data [post hoc test: Fisher least significant difference (LSD)], and Kruskal–Wallis test in non-normally distributed data (post hoc test: Bonferroni). Univariate and multivariate linear regression analyses were performed to assess predictors of change in QOL scores. Receiver operator curve (ROC) analysis was performed to measure the cut-off point, sensitivity, and specificity of QOL scores at 12 months and QOL changes (Δ_{12months-baseline}) to differentiate patients with or without AF recurrence. All statistical analyses were performed using SPSS software (SPSS, Inc., version 23.0). A P-value less than 0.05 was considered statistically significant.

Results

Patient demographic and baseline clinical characteristics are summarized in Table 1. After enrolment, 223 patients underwent PVAI using one of the two ablation strategies including multipolar duty cycled phased RF ablation (PVAC, Medtronic Inc.) and traditional point-by-point RF catheter ablation. Three patients exited the study due to procedural complications or technical issues related to the anatomy or the catheter (Supplementary material online, S1). Overall, 12 procedural complications occurred.¹³ A total of 217 patients were followed for 12 months post-procedure (see consort flow diagram, Supplementary material online, S1). During 12 months of follow-up post-ablation, freedom from any atrial tachyarrhythmia more than 30 s was 41.70%. The detailed primary results have been published.¹³ Amongst patients with recurrence, only 2 (1.60%) patients did not have symptoms and were diagnosed solely by ambulatory monitoring. A repeat ablation was performed in 29 patients within 12-months follow-up.

During the initial 3 month blanking period, 149 (64.8%) of patients received antiarrhythmic drugs. At 6 months, 82 (35.7%) of patients were on antiarrhythmics, and at 12 months, 62 (27%) of patients were on antiarrhythmic drugs. Of the 62 patients on drugs at 12 months, 54 had experienced the primary endpoint of AF recurrence, while 8 did not.

Effect of ablation on quality of life

There was a significant increase in the global and subscale scores for both AFEQT and SF12 questionnaires and a significant decrease in CCS-SAF score post-AF ablation that peaked at the 3 month visit and stayed constant at the 6, 9, and 12-month visits (P < 0.001) (Figure 1, Table 2).

Change in quality of life according to procedural success

An improvement in AFEQT, SF-12 subdomains, and CCS-SAF scores occurred in both patients with and without recurrences of atrial arrhythmia >30 s at 12 months post-ablation (Table 2, Figure 2A). However, the magnitude of improvement was greater in patients without recurrence compared to those with recurrence with the exception of the mental component summary (MCS). (Table 2, Figure 2B).

The QOL score that best differentiated patients with and without recurrence was AFEQT. ROC analysis showed a cut-off point of 91.20 (sensitivity = 70%, specificity = 74%) for AFEQT to differentiate recurrence. The CCS-SAF was the most specific score which showed a specificity of 96.30%. The SF12 subdomains were less differentiating of AF recurrence, particularly the MCS (Figure 3).

ROC analysis demonstrated that an increase more than 22.16 [(sensitivity = 70%, specificity = 53%, area under curve (AUC) = 0.61, P = 0.009)] and 4.30 (sensitivity = 60%, specificity = 56%, AUC = 0.61, P = 0.009) in the AFEQT and PCS scores, and a reduction of more than 2.50 (sensitivity = 74.30%, specificity = 48.10%, AUC = 0.66, P < 0.001) in the CCS-SAF score could differentiate patients with and without recurrence. The change in MCS score was not a significant predictor (P = 0.18).

Change in quality of life according to atrial fibrillation burden

Median AFB 12 months post-ablation was 0 [interquartile range (IQR: 0–2)] h/month in all patients, as well as, 0 (IQR: 0–0) h/month, and 0 (IQR: 0–5) h/month in RF, and PVAC groups, respectively. Median AFB change and rate of change from baseline to 12 months post-ablation were −19 (IQR: −40 to −4) h/month and −100% (IQR: −100 to −95.27%) in all patients, respectively. There was no significant difference in median AFB change and percentage of change between PVAC and RF group (median change: RF: −24 (IQR: −45 to −6) vs. PVAC: −15(IQR: −40 to −30), P = 0.24; median percentage of change: RF: −100 (IQR: −100 to −100), PVAC: −100 (IQR: −100 to −91.31), P = 0.08 (Supplementary material online, S3).

Data were classified into three categories of change in AFB based on median change = −19 and 90-percentile change = 0: including ≥19 h decrease, 0–19 h decrease, or an increase ≥0 h. Patients with AFB decrease equal or greater than 19 h had a significantly greater change in the magnitude of AFEQT, and CCS-SAF scores compared to those with a decrease in burden between 0 and 19 h (AFEQT: P_LSD = 0.03, CCS_SAF: P_LSD = 0.04) or an increase in AFB (AFEQT: P_LSD = 0.008, CCS_SAF: P_LSD < 0.001) (Figure 3). Those who had an increase in AFB experienced deterioration or no change in QOL as measured by all scales (Figure 4).

By categorizing patients according to AFEQT MCID = a 19-point change, AFB at 12 months was significantly different in patients with MCID ≥ 19 vs. <19 (Supplementary material online, S4). Median AFB at 12 months in patients with equal or greater than 19-point change was 0 (IQR = 0–1) h/month, while those with MCID < 19 points, had median AFB of 1 (IQR = 0–23.50) h/month at 12-month follow-up (P < 0.001). Moreover, patients with MCID equal or greater than 19 points had mean Δ %AFB [(12 months-baseline/Baseline) *100] of −88.10 ± 39.20 h/month, while those with <19-point change, had mean Δ%AFB of +60.30 ± 690.20 h/month (P = 0.01).

Predictors of change in quality of life scores (Δ_{12-month-baseline})

In univariate linear regression analysis (Table 3), recurrence >30 s and AFB <24 h/month at 12 months post-ablation were the most significant predictors of change in QOL and symptom severity scores. Recurrence >30 s significantly predicted an increase in ΔCCS-SAF score by 0.80 points and a decrease in AFEQT and PCS by 9.40 and 4.80 points, respectively. This is while AFB <24 h/month at 12 months post-ablation significantly predicted a decrease in CCS-SAF by 1.90 points and an increase in AFEQT and PSC scores by 19.50 and 5.50 points, respectively. Female gender was a significant predictor of change in AFEQT score, which predicted an increase of ΔAFEQT by 8.90 points. Atrial fibrillation burden <24 h/month at 12-months post-ablation was associated with significant changes in QOL and severity scale measures when adjusting for baseline scores and other covariates in multivariate regression analysis (Table 4).

Effect of ablation strategy on quality of life

There was no significant difference in QOL (ΔAFEQT: PVAC: 26.77 ± 24.23 vs. control: 29.63 ± 24.01, P = 0.43; ΔMCS: PVAC: 7.01 ± 13.27 vs. control: 6.36 ± 14.63, P = 0.76) and symptom severity measures (ΔCCS_SAF: PVAC: −1.60 ± 1.50 vs. control: −1.70 ± 1.50, P = 0.68) between the two ablation strategies except for PCS score (ΔPCS: PVAC: 4.71 ± 10.9 vs., control: 8.86 ± 10.73, P = 0.01). By categorizing based on type of PVAC catheter, this difference was only significant in PVAC platinum compared to control (Table 3).

Discussion

Main findings

The present evaluation of QOL from a multicentre prospective cohort study demonstrated significant improvement in QOL scores (including AFEQT, SF12, and CCS-SAF) in all patients during the 12 months following ablation. Patients who remained free of recurrence and those with greater decrease in AFB showed significantly more improvement. The AFEQT was best in differentiating those with and without recurrence. Atrial fibrillation burden <24 h/month at 12-months post-ablation was associated with significant changes in QOL and severity scale measures.

Measuring quality of life in atrial fibrillation ablation patients

Quality of life can be influenced by an individual’s general state of health which is impacted by comorbidities, life experience, beliefs, social and cultural differences, rather than only AF. Thus, there is a need to understand how much outcomes post-AF ablation can affect changes in standard QOL measures. One of the most commonly employed health-related QOL scores is the SF36 that was subsequently downsized to a subset of 12 items to improve patient compliance in larger sample sized studies.^7,9 Some studies have demonstrated an inability of SF-based measures to show a correlation between ablation outcome, AFB, and QOL perhaps due to general nature of the SF QOL scale. To better define changes in QOL for AF patients specifically, the AFEQT questionnaire was developed.² Studies using this more AF-specific QOL instrument were then able to more consistently demonstrate a correlation between AF reduction and improvement in QoL.^2,4,6,14 In our study, both SF-12 and AFEQT showed improvement in patients without recurrence and with reduced AFB. However, the magnitude of response was greater in AFEQT and we showed that the AFEQT scale had the highest sensitivity and specificity compared to SF-12. The change in AFEQT score was consistent with the result of CABANA trial, which showed 23.5 point change after 12 months post-ablation.¹⁵ Moreover, CAPTAF trial showed a more than 5 point change in each of the components of the SF-36 except for bodily pain which is similar to the 5–8 point change in PCS that we observed.¹⁶

Association of quality of life change with procedural success

Improvement in QOL after AF ablation has not always been shown to be directly correlated to procedural outcome. While some studies do show a significant QoL improvement in patients who maintain sinus rhythm,^1,17,18 others have reported improvement in QoL regardless of ablation success.^4–6 This might be due to a patient response bias or so-called ‘placebo effect’ of ablation or due to transition of symptomatic to asymptomatic AF after ablation. However, the apparent lack of correlation of QOL to procedural success may be due to the strict criteria for recurrence that is employed for AF ablation trials.^1,5 Trials have used a definition of recurrence as any AF episode >30 s, but such small events are unlikely to affect a patient’s sense of well being. In a STAR-AF sub-study, MCS and PCS showed significant increase in both recurrent and non-recurrent patients and the magnitude of change was not significantly different between the two groups.⁴ Moreover, recurrence >30 s did not significantly predict change in these scores, and patients could continue to experience more than 2 h of AF per month and still have an improvement in QOL.⁴ In our current study, we found that AFB reduction has a greater impact on patient QOL compared to success measured only by AF recurrence. Atrial fibrillation burden <24 h/month at 12 months post-ablation was associated with significant improvement in QOL measures in the adjusted model, consistent with the MCID reported in prior studies.^10,11

Other studies have shown that while ablation success is a predictor of QOL change, other factors such as increasing age, sex differences, and baseline QOL scores may also have significant impact.^1,4 Studies have shown that patients with higher baseline QOL score will show less absolute change after AF intervention because they already started with a more positive outlook on their disease.^1,4,6,8 In our study, multivariate analysis adjusting for baseline QOL values and other covariates demonstrated that AFB was the most significant predictor of change in QOL and symptom severity scores post-ablation.

Limitations

A limitation of our study was that we did not have direct continuous assessment of AFB, which could be provided by an insertable cardiac monitor. Therefore, we estimated AF burden based on intermittent monitoring and through patient-reported symptoms. This could introduce a bias since patient-reported episodes of AF are more likely to influence QOL which is also based on patient recall. However, this technique has been used in prior analyses (STAR AF 1, MANTRA PAF^4,19). While it would be interesting to assess QOL changes in patients with continuous monitoring post-ablation to determine more precise cut-off points of AFB that correlate with improvements in QOL, the result of CIRCA-DOSE study, using continuous monitoring revealed similar outcome for AFB change post-ablation.²⁰ Another limitation in assessing changes in QOL related to ablation is the absence of a control group without ablation in our study, preventing the exclusion or quantification of a placebo effect.

Conclusion

Atrial fibrillation ablation procedural outcome is related to significant improvements in QOL when determined by AF recurrence but more so by reductions in AFB despite ongoing brief recurrences of AF. The AFEQT scale may be more specific and sensitive to such changes compared to general scales like the SF-12.

Supplementary material

Supplementary material is available at Europace online.

Acknowledgements

We would like to acknowledge our sites’ coordinators for their dedication; SRHC: Annette Nath, Sherri Patterson, MUHC: Fiorella Rafti and Ida Di Stefano, Sacré-Coeur: Ann Langlois, IUCPQ: Paule Banville and Brigitte Ottinger, HHSC: Ruth Chinchilla and Angela Frechette, LHSC: Lynn Nyman.

Funding

This study was a physician-initiated study and was funded by an unrestricted grant from Medtronic Inc. Clinical Research Scholar Award from the Fonds de recherche du Quebec-Santé (FRQS) to V.E.

Conflict of interest: V.E. has received honoraria from Abbott, Biosense Webster, Boston Scientific and Medtronic. A.V. has received research grants from Bayer, Biotronik, Medtronic and advisory from Medtronic, Biosense, Acutus. All remaining authors have declared no conflicts of interest.

References

Author notes