Clinical implication of the patient's disease awareness and adherence to medications in patients undergoing atrial fibrillation ablation

Abstract Background The effects of the patient's disease awareness on the management of postablation of atrial fibrillation (AF) are unknown. Methods One hundred thirty‐three AF patients undergoing an initial ablation were given a disease awareness questionnaire with a score of 16 points (8 points about AF in general and 8 points about oral anticoagulants) for the Jessa Atrial Fibrillation Knowledge Questionnaire (JAKQ) before and 1‐year‐after ablation. We divided them into the poor disease awareness group and good disease awareness group according to the median value (75%) of the total JAKQ score about AF in general, and compared the baseline patient characteristics and the 1‐year changes in the JAKQ score, medication adherence, blood pressure, laboratory data, echocardiographic parameters, and AF/atrial tachycardia (AT) recurrence rate between the two groups. Results Forty‐two (31.6%) patients were classified as having poor disease awareness (<75% of the total JAKQ score), which was closely associated with poor medication adherence, hypertension, diabetes, dyslipidemia, and greater left atrial volume (LAV). These trends in the poor disease awareness group remained unchanged 1 year after the ablation. During the 25.3‐month follow‐up, the AF/AT recurrence rate was significantly higher in the poor disease awareness than the good disease awareness group (23.8% vs. 7.7%; p = .003 by the log‐rank test). Conclusions Poor disease awareness was linked to poor medication adherence, lifestyle‐related diseases, and greater LAV before and even 1 year after the ablation, making it a potential surrogate marker for AF/AT recurrence. These findings highlight the clinical significance of disease awareness in AF management.


| INTRODUC TI ON
Atrial fibrillation (AF) is the most common arrhythmia in adults and increases the risk of a stroke.4][5] Even after AF ablation, an integrated and holistic approach to the management of AF is also important to prevent not only AF recurrence but also further clinical adverse events. 6We hypothesized that the patient's disease awareness regarding AF itself, AF-related events, and its treatment, may also be one of the important elements to adhering to medications, managing cardiovascular risk factors, and preventing future clinical events because there has been a large gap reported between the patient's disease awareness and AF treatment. 7A valid AF-specific knowledge questionnaire called the Jessa Atrial Fibrillation Knowledge Questionnaire (JAKQ) is known to be used in routine practice to assess the patients' insight into their condition. 8Nevertheless, there are no data on the relationship between the disease awareness and the time course changes in the medication adherence, comorbidities, and AF/atrial tachycardia (AT) recurrence after ablation.Therefore, we aimed to investigate the clinical effect of the patient's disease awareness on medication adherence and cardiovascular risk factors for the management after AF ablation.

| Study design
This study was a single-centre prospective observational study.
The study participants included 156 consecutive patients who had undergone catheter ablation of paroxysmal AF (PAF; defined as AF returning to sinus rhythm within 7 days) and persistent AF (PerAF; defined as AF lasting ≥7 days) at Nihon University Itabashi Hospital between July 2019 and March 2020.The inclusion criteria were (1) patients who had initially undergone ablation of PAF or PerAF and (2) those who had been given questionnaires on their disease awareness and medication adherence before and 1-year-after ablation.
Exclusion criteria were (1) patients who had undergone ≥2 sessions of ablation, (2) hemodialysis patients, (3) active cancer patients, (4)   patients not cooperating with the study, (5) patients with incomplete questionnaires, and (6) warfarin users after ablation.Of a total of 156 patients, we excluded 8 who had ≥2 sessions of ablation, 1 hemodialysis patient, 3 who had active cancer, 8 not cooperate with the study, 2 with incomplete questionnaires, and 1 warfarin user.As a result, 133 patients (89 men and 44 women; aged 65 ± 11 years) were enrolled in this study for the final analysis (Figure 1).This study conformed to the Declaration of Helsinki and the Ethical Guidelines for Clinical Studies issued by the Ministry of Health, Labour, and Welfare, Japan.All participants provided written informed consent and could withdraw their consent at any time.
This study protocol was approved by the Institutional Review Board of Nihon University Itabashi Hospital, Clinical Research Judging Committee.

| Clinical data collection
The patient information including the age, male sex, height, and body weight was collected before the ablation.The systolic and diastolic blood pressure (SBP and DBP) at baseline were collected in the morning during admission before the ablation.The blood pressure 1-year postablation was obtained from the average weekly home BP recordings.Laboratory data included the hemoglobin A1c (HbA1c), triglycerides (TG), total cholesterol, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol.The echocardiographic parameters included the left ventricular ejection fraction (LVEF), left atrial diameter (LAD), LA volume (LAV), E/A, and E/e'.The medication information including that of the antiarrhythmic drugs (AADs) and oral anticoagulation (OAC) therapy before and 1 year after ablation was collected.

| Questionnaire
All patients watched an educational video on AF and received an educational explanation of the AF etiology, management, and ablation for informed consent of the AF ablation from each ablation physician.At 3-8 weeks later, all patients were given a questionnaire for their disease awareness and medication adherence within 1 week before and 1 year after the ablation.The questionnaires were based on the JAKQ for disease awareness and the 8-item Morisky Medication Adherence Scale (MMAS-8) for medication adherence, respectively. 8,9The JAKQ questionnaire on disease The patient flow of this study.For each question, there was an "I do not know" option to avoid guessing.A correct answer was scored as 1 point and an incorrect or I do not know answer was scored as 0 points.The total score was calculated from the completed questions and separately displayed as the percentage for the questions about AF in general and those about OAC therapy.The MMAS-8 questionnaire on adherence to medications also included 8 questions (Tables S1   and S2).The MMAS-8 for medication adherence was scored on an 8-point scale.The questionnaires were collected by the doctors.
Data verification was performed subsequently by another person.
In this study, the JAKQ about AF in general was used to investigate the association between disease awareness and AF/AT recurrence after ablation.That was because the 8 DOAC questions might have affected the clinical events including strokes and bleeding but might not have been directly associated with the postablation AF/AT recurrence.The median percentage of the JAKQ about AF in general, was 75% and the median MMAS-8 score was 6.
Therefore, we defined ≥75% as good disease awareness, ≥6 points as good medication adherence, <75% as poor disease awareness, and <6 points as poor disease awareness and medication adherence, respectively.

| Ablation methods
For all patients, AADs were discontinued for at least five half-lives prior to the ablation procedure, and OACs were generally discontinued on the day of ablation.

| Postablation follow-up and endpoints
On the day after the ablation procedure, all antiarrhythmic drugs previously prescribed were resumed, at the individual operator's discretion.Routine follow-up was performed at the hospitals' respective outpatient clinics, where physical examinations and 12lead electrocardiography were performed at 2 weeks, 1 month, and every 3 months thereafter.Twenty-four-hour Holter recordings were obtained at 3, 6, and 12 months, and every 1 year thereafter.Any symptomatic or documented atrial arrhythmias of ≥30 s after a 3-month blanking period were taken as a recurrence of the AF/AT.During the follow-up, all patients received a face-to-face standard interview regarding the patient's AF-related symptoms and conditions which were generally conducted at the outpatient clinic.

| Study endpoint
The primary endpoint was the annual AF/AT recurrence rate after ablation between poor and good disease awareness.The secondary endpoint was the late-phase AF/AT recurrence rate at 2 years post-ablation between the two groups.The tertiary endpoint was the time-course change in the parameters such as the patient background, medication adherence, laboratory data, and transthoracic echocardiography between the two groups.Asymptomatic AF was significantly more prevalent in the poor disease awareness group (45.2% vs. 24.2%;p = .015).Hypertension was more likely to be observed (61.9% vs. 44.0%;p = .05)and DM and dyslipidemia were significantly more prevalent in the poor disease awareness group than in the good disease awareness group (DM: 31.0%vs. 11.0%;p = .005:dyslipidemia 45.2% vs. 24.2%;p = .015).There were no differences in prior strokes and vascular disease between the two groups.Regarding the laboratory data, there were no differences in any of the parameters between the two groups.As for the echocardiographic variables, the LAD and LAV were greater in the poor disease awareness group than good disease awareness group (LAD 42.1 ± 7.0 mm vs. 39.8 ± 6.1 mm; p = .05:LAV 56.8 ± 23.0 mL vs. 49.3 ± 19.0 mL; p = .049).There were no differences in the LVEF, E/A, and E/e' between the two groups.Medication adherence was also significantly lower in the poor disease awareness group (6.0 ± 1.7 vs. 7.0 ± 1.

| The serial changes in the parameters potentially related to AF progression before and 1-year after ablation
The changes in various parameters before and 1 year after ablation between poor and good disease awareness are provided in Table 2.
In both the poor and good disease awareness groups, the BMI and HbA1c tended to decrease.The SBP was significantly elevated after ablation in the poor disease awareness group (from 129.6 ± 15.1 at baseline to 135.3 ± 15.5 mmHg after ablation; p = .042),whereas it was not in the good disease awareness group (from 127.0 ± 13.9 to 129.8 ± 13.6 mmHg, p = .09).The HbA1c tended to decrease after ablation in the poor disease awareness group, but it decreased significantly in the good disease awareness group (from 6.1 ± 0.7 to 5.9 ± 0.7%; p = .06and from 6.0 ± 0.6 to 5.8 ± 0.5%; p < .001).The TG levels were significantly increased in the poor awareness group, while they did not change in the good awareness group.As for echocardiographic parameters, both the poor and good disease awareness groups had a significant reduction in the LAD and LAV (p < .001for all).In the poor disease awareness group, the mean score for disease awareness about AF in general improved from 56.5 ± 8.8 to 68.8 ± 13.3% (p < .001),while no significant change was observed in the good disease awareness group (from 85.9 ± 9.5 to 85.4 ± 9.4%; p = .62).In the poor disease awareness group, the mean score for disease awareness about DOACs decreased from 68.2 ± 18.6 to 66.4 ± 14.7% (p = .45),while an improvement was observed in the good disease awareness group (from 78.7 ± 14.6 to 80.1 ± 10.9%; p = .27),but both differences were not significant.In both groups, the score of medication adherence improved from 6.0 ± 1.7 to 6.5 ± 1.3 (p = .002)and from 7.0 ± 1.2 to 7.2 ± 1.0 (p = .007)before and 1 year after ablation, respectively.Despite the significant improvement in disease awareness and medication adherence in the poor disease awareness group, those scores remained significantly lower than in the good disease awareness group (p < .001and p = .001,respectively).

| AF/AT recurrence rate at 1 year and thereafter between the patients with poor disease awareness and good disease awareness
The Kaplan-Meier freedom rate from AF/AT recurrence showing a comparison of the endpoints in the two groups is provided in Figure 3.
The use of antiarrhythmic drugs at baseline was significantly lower (28.6% vs. 57.1%;p = .002),but that at 1 year and 2 years after the ablation tended to be higher in poor disease awareness group than

| DISCUSS ION
This study had three major findings: (1) poor disease awareness was associated with a higher BMI, higher prevalence of comorbidities and asymptomatic AF, greater LAV, and poor medication adherence, (2) poor disease awareness can be a surrogate marker for the recurrence of AF/AT after ablation even after multivariate adjustment for significant variables such as the BMI, LAV, or DM, and (3) regardless of poor disease awareness or good disease awareness, most parameters related to AF progression improved, but an improvement in the SBP, HbA1c, and TG was less in those with poor disease awareness.
Despite the significant improvement in disease awareness and medication adherence, those scores remained significantly lower in those with poor disease awareness.

| Clinical significance of poor disease awareness
This study disclosed the relatively high score on the JAKQ in patients who planned to undergo AF ablation.The mean score on the JAKQ was 76.0 ± 13.5%, including 76.6 ± 16.5% for 8 questions about AF in general and 75.4 ± 1.4% for 8 questions about OAC therapy with DOACs, which was higher than that in those who did not have any education as reported previously. 8In the original study, the mean score on the JAKQ was 55.8 ± 18.6%, including 51.6% for 8 questions about AF in general and 61.9% for 8 questions about OAC therapy with DOACs.In their study, as compared  to the initial score on the JAKQ , 20 hospitalized AF patients scored significantly better about 2 days after they had received an individualized education (60.9 ± 16.6% vs. 78.8± 14.8%; p = .001),and a longer time span of 1 month after the initial completion of the JAKQ followed by targeted education also improved the scores in a different population of 20 AF patients (61.6 ± 14.5% vs. 76.9 ± 13.8%; p = .001).A randomized controlled trial also showed a significant improvement in the JAKQ score by a targeted educational session (62.5% to 87.5% 1 year) compared with standard care (56.3% to 62.5% 1 year). 134][15] We also characterized the patients who were diagnosed with poor disease awareness assessed by the JAKQ about AF in general.Patients with poor disease awareness tended to have a greater BMI, and significantly higher prevalence of hypertension, DM, and a larger TA B L E 2 The changes in the various parameters before and 1-year after ablation between the poor and good disease awareness groups.F I G U R E 3 Kaplan-Meier curve of the AF/AT recurrence rate at 1-year and during the late phase after ablation.The median (25th, 75th percentile) length of the patient follow-up was 25.3 (15.7, 34.9) months.AF, atrial fibrillation; AT, atrial tachycardia.LAD and LAV.Poor disease awareness was closely associated with a poor medication adherence.This study disclosed novel insights into understanding the changes in lifestyle diseases driven by routine follow-up after ablation.A previous study showed that among patients with intensive risk factor management, the most marked improvements are seen with respect to hypertension and glycemic control, despite improvements across the board regarding lifestyle-related diseases. 16Reverse atrial remodeling in response to weight loss optimizes the results following ablation. 17We did not intervene with any specific education during the follow-up, but the disease awareness and medication adherence scores improved, and the BMI, HbA1c, LAD, and LAV all became reduced 1-year postablation, regardless of a poor or good disease awareness.That suggested an intervention by ablation followed by routine post-ablation follow-up may also provide a favorable effect on the lifestyle disease to some extent.However, only the poor disease awareness group had an elevated TG level at 1-year after ablation.The SBP increased significantly at 1-year after ablation in both the good and poor disease awareness groups, possibly due to sinus restoration after ablation.Nonetheless, its extent was significantly greater in the poor disease awareness group.The serial improvement in the HbA1c was statistically lesser in the poor disease awareness group.As a result, those scores remained lower and the BMI, HbA1c, and LAD were larger at 1-year after ablation in the poor disease awareness group.Those findings not only highlight the importance of a baseline poor disease awareness but also might be a hindering factor for modifiable lifestyle diseases during a routine follow-up after ablation.The reason why poor disease awareness had lesser serial favorable effects on lifestyle diseases remains unclear.The poor disease awareness group might be associated with asymptomatic AF, which might also have acted to hinder their disease awareness and the importance of the management of lifestyle diseases.Our study demonstrated that poor disease awareness potentially led to a late AF/AT recurrence.Poor disease awareness in this study had more lifestyle diseases including metabolic syndrome.8][19] Chang and colleagues 19 reported that patients with metabolic syndrome had a larger LAD, shorter fractionated intervals, and higher dominant frequencies as compared to those without.Furthermore, patients with metabolic syndrome have been found to experience more frequent recurrent AF. 18,19 It has been reported that asymptomatic AF is associated with more comorbidities, a high thromboembolic risk, and higher 1-year mortality than symptomatic patients. 20Another previous study using the JAKQ scores in anticoagulated patients with AF also showed that lower baseline JAKQ scores had a higher incidence of the composite of ischemic cerebrovascular events, major or nonmajor clinically relevant bleeding, and death. 14The study also reported that intensive risk factor management, resulting in improved anthro-morphometric profiles, cardiac structure, and symptom scores, led to a higher freedom from arrhythmia recurrence. 16Therefore, it was not clear whether the recurrence of AF/AT was due to poor disease awareness, poor serial changes in modifiable risk factors, or the original patient background as multiple cofounders.However, since poor disease awareness remains an independent factor, it can at least serve as a surrogate marker to identify patients at high risk for AF/ AT recurrence.In light of these findings, enhancing disease awareness and emphasizing the importance of anticoagulant therapy may be modifiable factors for reducing the risk of future AF/AT recurrences and related clinical events, particularly in patients with limited disease awareness.

| Study limitations
This study had several limitations that should be considered.First, this study was a single-center observational study, so no causal relationships could be established.Educated AF ablation patients were included, so it cannot generalized to all AF patients.To extend the applicability of our findings, additional research based on a multicenter study is warranted.Second, there are currently no clinically relevant evidence to support using a cut-off of below 75% of the JAKQ about AF in general as an indicator of the "poor disease awareness."It is important to note that this value was based on the median percentage of the scale, as described in the Section 2. As such, this study serves as an initial pilot study to explore the clinical significance of this threshold on outcomes.Further research and increasing evidence will be required to substantiate its validity.Third, identification of a postablation recurrence during routine follow-up may underestimate actual recurrences.Fourth, the use of antiarrhythmic drugs 1 year and 2 years after ablation might have influenced our results.
Nonetheless, the effects of the higher success rate in those with a good disease awareness than in those with poor disease awareness might have been small because the use of antiarrhythmic drugs at 1 year and 2 years after the ablation were even lower in good disease awareness groups.Finally, the routine follow-up after ablation regarding the face-to-face standard interview was dependent on the physicians' discretion, which may have affected our results.However, it is worth noting that the physicians at the outpatient clinic were the same for both the poor disease awareness and good disease awareness groups.

| CON CLUS IONS
Poor disease awareness was associated with hypertension, DM, a large LAD and LAV, and poor medication adherence.Regardless of whether a poor or good disease awareness, the LAD and LAV decreased 1-year after ablation, but patients with poor disease awareness remained to have a larger LAD and LAV and a lower disease awareness and medication adherence than those with a good disease awareness.Poor disease awareness was also associated with AF/AT recurrences after ablation.These findings suggest that disease awareness could be an additive factor that influences clinical outcomes in the context of postablation AF management, especially for patients with a poor understanding of AF.

AUTH O R CO NTR I B UTI O N S
Frequency distribution of the scores on the JAKQ.(A) The scores of the patients who completed the entire questionnaire of 16 questions (n = 133).(B) Scores for the first 8 questions dealing with AF in general (n = 133).(C) Scores for the 8 questions about DOAC therapy including the 8 general questions (n = 133).AF, atrial fibrillation; JAKQ, Jessa Atrial fibrillation Knowledge Questionnaire; DOAC, direct oral anticoagulation.

Follow
after ablation: 7.1% vs. 4.4%, p=0.48 by the log-rank late phase after ablation: 23.8% vs. 7.7%, p=0.003 by the log- Sawada and Yasuo Okumura wrote the first draft of the protocol manuscript and carried the overall responsibility for the full study and the study protocol.Yasuo Okumura, Naoto Otsuka, and Koichi Nagashima were substantial contributors to the study concept and design, manuscript drafting, and critical review of the manuscript and will contribute to the acquisition, analysis, and interpretation of the data.Masanaru Sawada, Naoto Otsuka, Koichi Nagashima, Ryuta Watanabe, Yuji Wakamatsu, Satoshi Hayashida, Moyuru Hirata, Shu Hirata, and Sayaka Kurokawa collected the data and conducted the study and have approved the final version of this manuscript.Yasuo Okumura gave us critical comments on the statistical methods and contributed to the analysis and interpretation of the data.
Univariate and multivariate Cox hazard models for the relationship between AF/AT recurrence and the various parameters.
TA B L E 3Note: Mean ± SD or median are shown.The abbreviations are shown in Table1.