Characterisation of patients who develop atrial fibrillation-induced cardiomyopathy

Abstract

Introduction

Atrial fibrillation (AF)-induced cardiomyopathy (AIC) is retrospectively defined after normalisation of left ventricular ejection fraction (LVEF) in sinus rhythm. It is unclear why some patients develop AIC.

Hypothesis

Patients with AIC have a subtle cardiomyopathic process that precedes their AF-mediated LVEF reduction. Detailed assessment of cardiac function after successful catheter ablation will reveal this.

Objective

To evaluate the utility of measures to identify cardiomyopathic features that persist after LVEF normalisation in AIC.

Methods

Patients with rate-controlled persistent AF and LVEF<50% undergoing catheter ablation (CA) were prospectively evaluated using echocardiography, cardio-pulmonary exercise testing and serum N-terminal pro b-type natriuretic peptide (NT-proBNP) at baseline and 6 months after CA. Participants with AIC, (LVEF recovery (≥50%) and no other cause for cardiac dysfunction) were evaluated using left ventricular (LV) longitudinal strain and left atrial (LA) reservoir strain (LARS). Changes in peak oxygen consumption and the minute ventilation/carbon dioxide production slope were measured as markers of functional capacity and ventilatory inefficiency. A control group of patients with persistent AF with preserved LVEF were also enrolled.

Results

34/41 (82.9%) participants recovered LVEF in sinus rhythm; defined as AIC. NT-proBNP levels were elevated in 18 (52.9%), and 16 reported ongoing heart failure (HF) symptoms. 10 (29.4%) had no improvement in functional capacity, and seven (20.6%) showed persistent ventilatory inefficiency. 20 (58.8%) had impaired global LV longitudinal strain with a relative apical sparing pattern. Nine (26.5%) had impaired LARS. There was an overlap of these abnormalities. 32 (94.1%) demonstrated at least one, 17 (50.0%) having no cardiovascular risk factors. Patients with preserved LVEF during persistent AF had similar demographics but a lower burden of short R-R intervals (<660 ms) on Holter monitoring.

Discussion

Abnormal structural, metabolic and HF biomarkers are seen in patients with AIC in sinus rhythm. These features may represent a precedent subtle cardiomyopathic process predisposing them to left ventricular systolic dysfunction in AF.

Trial registration number

NCT04987723.

Correspondence to Prof Richard J Schilling; [email protected]

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Atrial fibrillation (AF) can reduce the left ventricular ejection fraction (LVEF), which may normalise after sinus rhythm is successfully restored using catheter ablation. Why some patients develop this complication in AF whereas others preserve their LVEF is unknown.

WHAT THIS STUDY ADDS

• This is the first study to characterise cardiac function in patients with recovered AF-induced cardiomyopathy using strain parameters, biomarkers and cardiopulmonary exercise testing. Although normalisation of LVEF is common in sinus rhythm, alternative parameters of cardiac function remain abnormal. These may reflect an underlying cardiomyopathic processes that were unmasked by incident AF.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• The prognostic significance of these persisting abnormal features should be studied, including the impact of pharmacological heart failure therapies.

Introduction

Atrial fibrillation (AF) can cause cardiomyopathy with a reduction in left ventricular ejection fraction (LVEF), resulting in worse outcomes compared with patients who do not develop this complication.¹ Catheter ablation (CA) of AF can restore sinus rhythm and normalise LVEF, leading to a retrospective diagnosis of AF-induced cardiomyopathy (AIC) in these patients.² Although LVEF recovery is associated with an improvement in heart failure (HF) markers, continue to have a greater risk of HF hospitalisation than patients who do not experience LVEF reduction with AF.^3–5 Only a proportion of patients with AF will develop AIC and the underlying factors that make some individuals susceptible or resistant to AIC are unknown. Precedent cardiac imaging during sinus rhythm to be able to attribute new LVEF reduction to incident AF prospectively is rare. Re-evaluation of patients with AIC after treatment of their AF and reversal of the attributable LVEF reduction may enable post hoc identification of these underlying factors, as they would not be reversed by AF treatment. The prevalence of abnormal HF biomarkers, subtle ventricular dysfunction or impaired functional capacity after LVEF recovery in AIC is unknown.

Speckle-tracking imaging is a quantitative ultrasound method to measure myocardial shortening. When used to measure left ventricular (LV) longitudinal strain, it can detect subclinical systolic dysfunction when the LVEF is normal.⁶ The peak longitudinal strain of the left atrial (LA) during ventricular systole reflects reservoir phase function. This is an integrated measure reflecting passive LA capacitance but also LV longitudinal function and diastolic performance.^{6 7}

This study aims to characterise the cardiac function of patients with recovered AIC in sinus rhythm using strain parameters of ventricular dysfunction, cardiopulmonary exercise testing (CPET) and HF biomarkers to identify features of an underlying cardiomyopathic process.

Methods

Study population

Patients with persistent AF (more than 7 days of continuous AF) and left ventricular systolic dysfunction (LVSD) (LVEF≤50%) undergoing first-time CA at the study institution were invited to enrol in this prospective observational study. Ventricular rate control during AF (mean heart rate (HR)<110 bpm) was evaluated using Holter monitoring at baseline.⁸ The study was registered on ClinicalTrials.gov. The Arrhythmia Alliance UK charity kindly provided patient and public involvement in the study’s design, conduct and reporting. Participants with known ischaemic or non-ischaemic cardiomyopathy were excluded from this analysis. Patients with a reversible cause or change in LVSD treatment or left atrial indexed volume (LAVi) greater than 50 mL/m² were excluded. Echocardiography, serum NT-proBNP measurement and CPET were undertaken at baseline and six months after CA. A control group of patients with persistent AF with preserved LVEF (>50%) undergoing first-time CA were also enrolled in the study. Baseline characteristics and the incidence of cardiovascular comorbidities; hypertension, ischaemic heart disease, diabetes, hypercholesterolaemia and cerebrovascular events and the CHADS2VASc score were recorded. The mean heart rate was determined from the baseline Holter monitoring during AF, and the burden of short R-R intervals (<660 ms) during AF was derived from a supervised 10-min period while supine at rest.

Study design

Transthoracic echocardiography was performed using a GE Vivid E95 system (GE HealthCare, Wauwatosa, Wisconsin, USA) with a 4-MHz transducer. All measurements were made according to the British Society of Echocardiography recommendations.⁹ A standard hierarchical sequence was applied for LVEF determination; Simpson’s biplane approach was preferred, followed by a two-dimensional method and, finally, the visually estimated LVEF.

AIC was diagnosed if participants with rate-controlled AF with LVEF<50%, recovered their LVEF to ≥50% in sinus rhythm 6 months after CA.² Dedicated images optimising the left ventricle and left atrium were acquired, and analysis was performed offline. LV volumes and LVEF were averaged over three cycles in AF, where image quality sufficed. Myocardium speckle tracking of the global and per-segment LV longitudinal strain and LA using customised software (EchoPoint V.6, GE HealthCare). Strain measurements were expressed as scalar values. The LV and LA myocardium were contoured on two-dimensional echocardiographic images of the LV and LA. Left atrial reservoir strain (LARS) was the peak positive deflection in the strain curve during systole. Global LV strain was not calculated if >2 segments could not be processed and LARS was excluded if the wall did not track after three-point annotation. The vendor-specific threshold of >18% was used to define normal global LV strain in sinus rhythm.¹⁰ An LARS cut-off of 18% is associated with elevated LV filling pressures in sinus rhythm (SR) and used in the absence of an agreed normal range.⁷

Conventional diastolic dysfunction parameters were also evaluated in sinus rhythm, and detection was based on mitral inflow and annular velocities, LA volume and the tricuspid regurgitant jet velocity.¹¹ In the context of AF and reduced LVEF, the assessment was limited to a shortened mitral deceleration time during early filling (≤160 ms).¹¹

CPET was performed on a semi-recumbent cycle ergometer (ERG 911 S/L, Schiller, Switzerland) to evaluate metabolic capacity using compatible software (Quark, Cosmed, Italy). Minute ventilation, oxygen saturation and carbon dioxide production were monitored throughout. A 3-min warm-up was followed by exercise with the work rate incremented by 10–20 watts/min until exhaustion, aiming for 8–10 min of exercise and a respiratory exchange ratio of 1.0 or greater.

The minute ventilation/carbon dioxide production (VE/VCO₂) slope gradient, an effort-independent composite assessment of cardiopulmonary physiology and the peak oxygen consumption (VO₂) consumption, an objective measure of functional aerobic capacity, have the greatest evidence-base of prognostic value in systolic HF and were selected as representative parameters.¹² Peak VO₂ was averaged over the final 30 s of exercise with significant improvement defined as ≥6% improvement at follow-up or if greater than predicted VO₂ max. This increment is associated with a corresponding reduction in major adverse cardiac events and mortality rates in the HF population.¹³ The VE/VCO₂ slope was calculated as the linear relationship from start to peak.¹² A gradient steeper than 30 was in keeping with ventilatory inefficiency.¹²

An ambulatory NT-proBNP threshold of >125 mL was used to define abnormality in sinus rhythm based on contemporary HF guidelines.¹² The Minnesota living with heart failure questionnaire (MLWHQ) survey was completed at baseline and follow-up to characterise symptom burden. The 21-question survey provides a total score (range 0–105, from best to worst health-related quality of life), with specific dimensions to characterise physical (eight items, range 0–40) and emotional (five items, range 0–25) symptoms.¹⁴

Catheter ablation

The ablation strategy was at the discretion of the clinical operator, with wide-area circumferential radio-frequency ablation of the pulmonary veins with LA electro-anatomical mapping (EnSite X, Abbott, USA) as a minimum. Antiarrhythmic medication was routinely discontinued at 3 months. Oral anticoagulation and guideline-directed HF medications were routinely continued beyond primary endpoint evaluation. AF recurrence was defined as any documented atrial arrhythmia on a 12-lead ECG or >30 s on a single-lead ECG technology detected after the 90-day blanking period. Repeat CA was at the discretion of the referring clinician and restarted the follow-up assessment time points within this study.

Statistics

The Shapiro-Wilk test was used to determine whether data was normally distributed. Continuous variables were analysed using a two-tailed independent t-test for normally distributed data or the Mann-Whitney U test for non-normally distributed data. The χ² test was used for categorical variables. Normally distributed data was presented as mean±SD, and non-normally distributed data as median (IQR). The Pearson correlation coefficient was calculated for linear relationships between normally distributed continuous variables, and the Spearman correlation coefficient for non-linear relationships and non-normal variables. A sensitivity analysis excluding participants with cardiovascular risk factors was also performed. A p value<0.05 was used to determine significance.

Results

Patient characteristics

43 participants with no known alternative cause for reduced LVEF aside from the AF, met the study criteria and were enrolled between January 2022 and September 2023. The mean age of participants was 58.9±12.1 years and eight (18.6%) were women. The mean HR during AF was 87±14 bpm and the median duration of persistent AF was 8 (5–14) months with 16 (37.2%) having long-standing persistent AF. The mean LVEF was 35±10%, and 14 (32.6%) patients had severe LVSD. The mean LV strain was 10.7±3.1% and abnormal (<16.0%) in all participants. Mean LARS was 8.0±5.3%, and the mean E-wave deceleration time (DT) during AF was 181±51 ms. Indexed mean peak VO₂ on exercise testing was 19.3±6.7 mL/kg/min^–1. Of the 43 study patients, 16 (37.2%) had a steep VE/VCO₂ slope in keeping with ventilatory inefficiency. Ambulatory NT-proBNP was 823 (511–1496) (median, IQR) pg/mL.

Catheter ablation outcome

All participants underwent CA, with seven (16.3%) undergoing repeat CA procedures for AF or atrial tachycardia (AT) recurrence before the 6-month re-evaluation time point. Pulmonary vein isolation (PVI) was achieved in 42 (97.7%) participants, and extra-PVI ablation was performed in 10 (23.3%).

Recovered AIC

One participant was lost to follow-up, and one died 5 months after CA due to HF. Of the 41 participants, re-evaluated after CA, the mean LVEF at follow-up was 55±7%. 38/41 (92.7%) participants were in sinus rhythm at follow-up evaluation with no evidence of AF on ECG or Holter monitoring or based on symptoms. 34 (79.0%) participants recovered their LVEF to ≥50% in sinus rhythm and were retrospectively diagnosed as having AIC. The absolute LVEF improvement in patients with AIC was 22±9% to 57±4% in sinus rhythm. 17 (50.0%) participants with AIC had one or more cardiovascular risk factors. Four (10.5%) participants did not normalise their LVEF in sinus rhythm, and three participants did not achieve freedom from AF (7.3%).

AF with preserved LVEF

51 participants with persistent AF and preserved LVEF were enrolled before CA. There was no significant difference in age, sex or cardiovascular comorbidities compared with 34 participants with AIC (table 1). Although there was no significant difference in mean HR, the burden of short R-R intervals during AF was significantly higher among participants with AIC (58.8+21.1% vs 39.5±26.0%, p=0.01).

Table 1

Baseline characteristics in participants with proven atrial fibrillation (AF)-induced cardiomyopathy versus participants with AF with preserved left ventricular ejection fraction (LVEF)

Parameter	AF-induced cardiomyopathy (n=34)	AF with preserved LVEF (n=51)	P value
Age (years)	58.2+12.3	60.8±11.1	0.43
Female n (%)	6 (17.7)	10 (19.6)	0.77
BMI kg/m2	30.0+5.1	29.7±5.5	0.81
Hypertension, n (%)	13 (38.2)	21 (41.2)	0.47
IHD, n (%)	1 (2.9)	1 (2.0)	0.77
CVA, n (%)	2 (5.9)	5 (9.8)	0.67
Diabetes, n (%)	5 (14.7)	5 (9.8)	0.62
AF duration (months)	8.0 (5.0, 14.0)	9.0 (6.0, 13.0)	0.54
Mean HR (bpm)	87.5+13.2	83.5±12.5	0.70
Short R-R interval burden (%)	58.8+21.1	39.5±26.0	0.009

BMI, body mass index; CVA, Cerebrovascular accident; HR, heart rate; IHD, Ischaemic heart disease.

HF characterisation

NT-proBNP levels significantly reduced after CA in participants with AIC, however levels remained elevated in 18 (52.9%) participants in sinus rhythm (figure 1). The median value was 138.0 (58.0, 205.0) pg/ml. The MLWHQ score also improved in participants with AIC after CA (p<0.001); however, 16/18 (88.9%) participants with persistently elevated NT-proBNP reported concurrent physical HF symptoms, with 12 (66.6%) reporting accompanying emotional symptoms. Difficulty walking or climbing stairs and persisting shortness of breath were the most frequently reported symptoms by 13 (72.2%) participants.

Figure 1. Distribution of serum NT-proBNP level at baseline and follow-up after catheter ablation. The red dashed line is at 125 pg/mL.

Functional capacity

Mean peak VO₂ improved in participants with recovered AIC to 22.0±7.5 mL/kg/min⁻¹ (p=0.02). However, no significant improvement in peak VO₂ was seen in 10 (29.4%) participants despite sinus rhythm restoration (figure 2). Ventilatory inefficiency persisted in seven (20.6%). Higher median NT-proBNP levels were seen in all patients with ventilatory inefficiency despite LVEF recovery (208 pg/mL (172–310) vs 96.5 pg/mL (49.8–152.0)), p=0.005).

Figure 2. Distribution of (a) minute ventilation/carbon dioxide production (VE/VCO 2 ) slope gradient at follow-up after catheter ablation. The red dashed line is at a slope gradient of 30. (b) The peak VO 2 as a percentage of predicted VO 2 max at baseline. The red dashed line is at 100%. VO 2 , oxygen consumption.

LV strain

The mean global longitudinal strain improved in participants with recovered AIC 6 months after CA (16.6±2.2%, p<0.001). However, it remained abnormal in 20/34 (58.8%) participants with AIC despite sinus rhythm restoration (figure 3). Per-segment analysis showed improved longitudinal strain in the basal, mid and apical segments after CA (basal: 7.8±3.4% vs 15.0±3.6%, mid: 10.6±3.8% vs 16.3±3.2%, apical 13.8±4.5% vs 18.6±3.1%; p<0.001 for all). A relative apical sparing pattern was seen at follow-up with a greater longitudinal strain of the apical segments than the mid or basal segments (18.6±3.1% vs 15.7±3.2%, p<0.001).

Figure 3. Distribution of left ventricular global longitudinal strain measurements at baseline and at follow-up after catheter ablation. The red dashed line is at -18%. LV, left ventricular.

LA reservoir strain and diastolic function

The LARS at follow-up was 19.3±6.0% and remained impaired in nine (26.5%) participants with recovered AIC. Impaired LARS was not associated with LAVi (p=0.537) or the baseline LARS in AF (p=0.817). Participants with AIC and impaired LARS performed significantly worse on follow-up CPET than participants with normal LARS. They had a lower Peak VO₂ (p=0.011), oxygen uptake efficiency slope (p=0.040) and steeper VE/VCO₂ slope gradient (p=0.013).

Three (8.8%) participants with AIC met conventional criteria for diastolic dysfunction at follow-up, and it was indeterminate in six (17.6%). Eight participants had a short mitral DT at baseline during AF but had no evidence of diastolic dysfunction after restoration of sinus rhythm. No participant with a normal mitral DT at baseline developed diastolic dysfunction at follow-up. No difference in LARS between participants who had PVI only versus extra-PVI CA was observed (p=0.30).

Parameter correlation

The LARS, peak VO₂ and ambulatory NT-proBNP at follow-up were all significantly correlated (figure 4). All participants showing ventilatory inefficiency had a raised NT-proBNP. Five (14.7%) participants with AIC had raised NT-proBNP, ventilatory inefficiency and abnormal strain (figure 5). Of the 17 participants with no known cardiovascular risk factors and AIC, eight (47.1%) had a raised NT-proBNP and five (29.4%) also had an abnormal strain. Four (11.7%) participants with AIC had a normal NT-proBNP, normal strain parameters and a normal VE/VCO₂ slope.

Figure 4. Pair plot to show the correlations between left atrial reservoir strain, peak VO 2 and serum NT-proBNP (log-transformed) levels during sinus rhythm in patients with atrial fibrillation-induced cardiomyopathy. Individual values (blue dots), the best-fit linear regression line (red) and the 95% CI (pink) are shown in the peripheral panels. The diagonal panels show a kernel density distribution plot for each parameter to show normal distribution. LA, left atrial; VO 2 , oxygen consumption.

Figure 5. Venn diagram to show overlapping features in patients with atrial fibrillation-induced cardiomyopathy when in sinus rhythm.

Discussion

CA is the first-line treatment for individuals with suspected AIC.^{15 16} Restoring sinus rhythm improves validated HF surrogate markers, LVEF, peak VO₂ and NT-proBNP, in keeping with previous randomised studies of CA.^{17 18} We have also shown that CA improves LV longitudinal strain, HF symptom severity and ventilatory efficiency in this cohort.

Precedent cardiomyopathy hypothesis

Although these measures of cardiac function improved, normalisation was uncommon, even in participants without cardiovascular risk factors (figure 6). NT-proBNP was elevated in 18 (52.9%) participants, and 21 (61.8%) had abnormal strain parameters in sinus rhythm. Our cohort had no other causes for cardiomyopathy identified, so reverse remodelling of AF-mediated features could reasonably be expected after timely restoration of sinus rhythm. We hypothesise these structural and functional features represent subclinical cardiomyopathic processes unmasked by incident AF. These features coexist, and individual variance may reflect differences in progression and predisposition. Relative apical sparing was seen consistently in the AIC cohort. This pattern has also been reported in cardiomyopathies wherein longitudinally-orientated fibres, that are predominantly expressed in the basal and mid segments are preferentially affected.¹⁹ Compensatory apical torsion gives the impression of relative apical sparing on speckle tracking. An extreme phenotype is associated with cardiac amyloidosis due to the amyloid deposition gradient, which is not suspected here.

Figure 6. Longitudinal strain analysis on apical views of the left ventricle during transthoracic echocardiography in a participant with atrial fibrillation-induced cardiomyopathy after the restoration of normal sinus rhythm and normalisation of left ventricular ejection fraction 6 months after catheter ablation. The strain values demonstrate impaired longitudinal systolic function with relative apical sparing. HR, heart rate.

Abnormal LV strain is associated with early chemotherapy-induced cardiomyopathy and forewarns subsequent LVEF reduction.⁶ Some patients with these features may be in an indolent phase of mild subclinical cardiomyopathy, which increases vulnerability to AF. This precedent cardiomyopathic process may help explain the chicken and egg debate when co-diagnosed together.¹ Alternatively, this could be a more independent process that decompensates after an incident unrelated to AF has reduced the LVEF.

No significant demographic differences were seen in limited comparison with patients who retained a normal LVEF in persistent AF. A difference in the burden of short R-R intervals was seen despite apparent rate control and a similar mean HR. There may be inherent differences in rhythm-level characteristics between patients in AF. AF results in an intrinsically higher ventricular rate than sinus rhythm, and rate control medications are used to address any tachycardia-induced cardiomyopathy phenomenon. Achieving rate control at baseline can mitigate any significant incremental rate control achieved by restoring sinus rhythm. There is no significant difference in reverse remodelling in permanent AF from lenient rate control (mean HR 80–110 bpm) versus strict rate control (mean HR<80 bpm). The threshold for lenient rate control is recommended by contemporary guidelines and was used in our study at enrolment. However, the deleterious effect of tachycardia is more likely to be a spectrum than dichotomised around 110 bpm, and measuring the short R-R interval burden may improve the characterisation of the burden of tachycardia given the non-parametric R-R interval distribution during AF. AIC is likely a complex phenomenon and may result from a combination of rhythm and structural features. A double-hit hypothesis wherein the onset of AF with a high short R-R interval burden in a patient with vulnerable cardiomyopathic features can be speculated as the reason for AF-induced cardiac dysfunction in those patients who develop LVEF reduction during rate-controlled AF adjudicated by the mean HR.

The hypothesis relies on the inherent assumption that the detected abnormalities preceded AF onset. To substantiate or disprove this hypothesis, a pre-emptive study evaluating individuals before they develop AF would be needed. The subtle abnormalities shown in this exploratory study may be helpful during screening.

Alternative hypotheses that assume these features appeared only post-treatment should also be considered; they may represent a residual dysfunction caused by AF onset and incompletely reversed. 27 participants had early persistent AF (of less than 1 year of continuous duration). Studies in dogs and humans suggest rate-controlled AF-mediated remodelling is reversible for the timely restoration of sinus rhythm with normalisation of chamber sizes and ejection fraction.^{20 21} Diffuse LV fibrosis is seen in patients with AF and LVSD, with regression demonstrated 6 months after CA.²² In the absence of precedent imaging before AF onset, it is difficult to determine whether these features resulted from AF or preceded it. The parameters measured in our study have not been previously evaluated in patients after CA. A follow-up echocardiographic assessment was performed 6 months after the final CA procedure. This time point was based on the criteria to diagnose AIC and a plateau in LVEF improvement was demonstrated after this time.^{2 17} Whether the alternative parameters may continue to improve should be considered. Repeat imaging would help to ascertain this, but it was outside the scope of this study. AF recurrence detection was based on interval rhythm assessment and symptom-guided evaluation, keeping with clinical practice. Subclinical AF episodes may have been missed, and continuous monitoring should be considered in future studies.

Prognostic implications

The prognostic implications of these abnormal features require consideration. Improvement of LVEF after CA is associated with prognostic benefits, but the prognosis does not completely regress and normalise.^{4 23 24} As objective features, raised NT-proBNP, abnormal strain parameters and ventilatory inefficiency are associated with hospitalisation risk and cardiovascular events in patients with HF with improved ejection fraction (HFimpEF).²⁵ The LVEF improvement in HFimpEF is commonly attributable to HF pharmacotherapy. Therefore, any residual features of structural abnormality or dysfunction may be attributed to the persistence of the underlying driver that is only countered by effective medications. Decompensation of LVEF is seen after medication withdrawal, suggesting that the cardiomyopathic process is ongoing.²⁶ In contrast, the LVEF improvement in AIC is attributable to the effective resolution of the arrhythmic driver. In the absence of AF recurrence, the effect of withdrawing HF medications has not been studied in these patients. This study raises the question as to whether continuation of this medication is protective in the large subset of patients with apparent normalisation of LVEF and subtle persisting abnormalities. HF is defined as a syndrome of characteristic signs and symptoms with elevated HF biomarkers independent of LVEF.²⁷ 16 (47.1%) participants with AIC had raised ambulatory NT-proBNP levels in the context of physical HF symptoms, fulfilling this criteria. This may be a simple marker determining the continuation of selected pharmacological therapies for patients who usually wish to discontinue medication after CA.

The limitations of this work should be considered. This is the first evaluation and demonstration of these features in a relatively small cohort of patients with AIC. In addition to reproducing these findings in larger cohorts, a comparative assessment of these parameters in patients who have preserved LVEF during AF should be considered. Relative differences would control for the treatment effect and support our hypothesis. Re-evaluation at a later time point would determine whether these abnormalities will progressively resolve over time.

Conclusion

A proportion of patients with AIC continue to have subtle abnormalities of ventricular function, functional capacity and ventilation despite normalising LVEF after CA. These features may represent a precedent cardiomyopathic process that may be arrested by continuing HF medications. Further study is needed to validate these findings in a larger patient cohort.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study involves human participants and was approved by South West - Frenchay Research Ethics Committee 21/SW/0135. Participants gave informed consent to participate in the study before taking part.

Footnote

X @nikhil_ahl

Contributors NA, GL and RS contributed to the study conception and design. Study funding was secured by NA and RS. NA, SH, HA, AWCC, MD, RJH and RS contributed to the study data collection. NA contributed to the data analysis. All authors reviewed, edited and approved the final manuscript. NA and RS are responsible for the manuscript and serve as guarantors.

Funding This study was funded by a Clinical Research Training Fellowship grant from Barts Charity (MGU0592) and a Research Grant from Abbott (#ISS-EP-2021-000180).

Competing interests RJH has received research grants and educational grants from Medtronic and Biosense Webster, and speaker fees and travel grants from Abbott and Biosense Webster and Medtronic. RS has received research grants and educational grants from Abbott and Biosense Webster and Medtronic, and speaker fees and travel grants from Abbott and Biosense Webster and Medtronic. RJH, SH and RS were inventors of the STAR mapping system and are shareholders in Rhythm AI. The remaining authors have nothing to disclose.

Provenance and peer review Not commissioned; internally peer reviewed.

References

¹ Wang TJ, Larson MG, Levy D, et al. Temporal relations of atrial fibrillation and congestive heart failure and their joint influence on mortality: the Framingham Heart Study. Circulation 2003; 107: 2920–5. doi:10.1161/01.CIR.0000072767.89944.6E

² Shoureshi P, Tan AY, Koneru J, et al. Arrhythmia-Induced Cardiomyopathy: JACC State-of-the-Art Review. J Am Coll Cardiol 2024; 83: 2214–32. doi:10.1016/j.jacc.2024.03.416

³ Basuray A, French B, Ky B, et al. Heart Failure With Recovered Ejection Fraction. Circulation 2014; 129: 2380–7. doi:10.1161/CIRCULATIONAHA.113.006855

⁴ Yang P-S, Kim D, Sung J-H, et al. Reduction of mortality by catheter ablation in real-world atrial fibrillation patients with heart failure. Sci Rep 2021; 11: 4694. doi:10.1038/s41598-021-84256-z

⁵ Agostoni P, Pluchinotta FR, Salvioni E, et al. Heart failure patients with improved ejection fraction: Insights from the mecki score database. European J of Heart Fail 2023; 25: 1976–84. doi:10.1002/ejhf.3031

⁶ Oikonomou EK, Kokkinidis DG, Kampaktsis PN, et al. Assessment of Prognostic Value of Left Ventricular Global Longitudinal Strain for Early Prediction of Chemotherapy-Induced Cardiotoxicity: A Systematic Review and Meta-analysis. JAMA Cardiol 2019; 4: 1007–18. doi:10.1001/jamacardio.2019.2952

⁷ Inoue K, Khan FH, Remme EW, et al. Determinants of left atrial reservoir and pump strain and use of atrial strain for evaluation of left ventricular filling pressure. Eur Heart J Cardiovasc Imaging 2021; 23: 61–70. doi:10.1093/ehjci/jeaa415

⁸ Gelder ICV, Rienstra M, Bunting KV, et al. ESC Guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Hear J 2024.

⁹ Robinson S, Rana B, Oxborough D, et al. A practical guideline for performing a comprehensive transthoracic echocardiogram in adults: the British Society of Echocardiography minimum dataset. Echo Res Pract 2020; 7: G59–93. doi:10.1530/ERP-20-0026

¹⁰ Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for Cardiac Chamber Quantification by Echocardiography in Adults: An Update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. Eur Heart J Cardiovasc Imaging 2015; 16: 233–71. doi:10.1093/ehjci/jev014

¹¹ Nagueh SF, Smiseth OA, Appleton CP, et al. Recommendations for the Evaluation of Left Ventricular Diastolic Function by Echocardiography: An Update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr 2016; 29: 277–314. doi:10.1016/j.echo.2016.01.011

¹² Guazzi M, Adams V, Conraads V, et al. EACPR/AHA Scientific Statement. Clinical recommendations for cardiopulmonary exercise testing data assessment in specific patient populations. Circulation 2012; 126: 2261–74. doi:10.1161/CIR.0b013e31826fb946

¹³ Swank AM, Horton J, Fleg JL, et al. Modest Increase in Peak VO 2 Is Related to Better Clinical Outcomes in Chronic Heart Failure Patients. Circ: Heart Failure 2012; 5: 579–85. doi:10.1161/CIRCHEARTFAILURE.111.965186

¹⁴ Bilbao A, Escobar A, García-Perez L, et al. The Minnesota living with heart failure questionnaire: comparison of different factor structures. Health Qual Life Outcomes 2016; 14: 23. doi:10.1186/s12955-016-0425-7

¹⁵ Joglar JA, Chung MK, Armbruster AL, et al. 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 2024; 149: e1–156. doi:10.1161/CIR.0000000000001193

¹⁶ Hindricks G, Potpara T, Dagres N, et al. ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association of Cardio-Thoracic Surgery (EACTS). Eur Heart J 2020; 20: 157.

¹⁷ Hunter RJ, Berriman TJ, Diab I, et al. A randomized controlled trial of catheter ablation versus medical treatment of atrial fibrillation in heart failure (the CAMTAF trial). Circ Arrhythm Electrophysiol 2014; 7: 31–8. doi:10.1161/CIRCEP.113.000806

¹⁸ Jones DG, Haldar SK, Sharma R, et al. A randomised trial to assess catheter ablation versus rate control in the management of persistent atrial fibrillation in chronic heart failure. Circulation 2013; 126: 2788.

¹⁹ Henein MY, Gibson DG. Long axis function in disease. Heart 1999; 81: 229–31. doi:10.1136/hrt.81.3.229

²⁰ Reant P, Lafitte S, Jaïs P, et al. Reverse remodeling of the left cardiac chambers after catheter ablation after 1 year in a series of patients with isolated atrial fibrillation. Circulation 2005; 112: 2896–903. doi:10.1161/CIRCULATIONAHA.104.523928

²¹ Pedro B, Mavropoulou A, Oyama MA, et al. Longitudinal analysis of echocardiographic and cardiac biomarker variables in dogs with atrial fibrillation: The optimal rate control in dogs with atrial fibrillation II study. J Vet Intern Med 2024; 38: 2076–88. doi:10.1111/jvim.17120

²² Prabhu S, Costello BT, Taylor AJ, et al. Regression of Diffuse Ventricular Fibrosis Following Restoration of Sinus Rhythm With Catheter Ablation in Patients With Atrial Fibrillation and Systolic Dysfunction. JACC: Clinical Electrophysiology 2018; 4: 999–1007. doi:10.1016/j.jacep.2018.04.013

²³ Marrouche NF, Brachmann J, Andresen D, et al. Catheter Ablation for Atrial Fibrillation with Heart Failure. N Engl J Med 2018; 378: 417–27. doi:10.1056/NEJMoa1707855

²⁴ Di Biase L, Mohanty P, Mohanty S, et al. Ablation Versus Amiodarone for Treatment of Persistent Atrial Fibrillation in Patients With Congestive Heart Failure and an Implanted Device. Circulation 2016; 133: 1637–44. doi:10.1161/CIRCULATIONAHA.115.019406

²⁵ Janwanishstaporn S, Cho JY, Feng S, et al. Prognostic Value of Global Longitudinal Strain in Patients With Heart Failure With Improved Ejection Fraction. JACC: Heart Failure 2022; 10: 27–37. doi:10.1016/j.jchf.2021.08.007

²⁶ Halliday BP, Wassall R, Lota AS, et al. Withdrawal of pharmacological treatment for heart failure in patients with recovered dilated cardiomyopathy (TRED-HF): an open-label, pilot, randomised trial. Lancet 2019; 393: 61–73. doi:10.1016/S0140-6736(18)32484-X

²⁷ Bozkurt B, Coats AJ, Tsutsui H, et al. Universal Definition and Classification of Heart Failure: A Report of the Heart Failure Society of America, Heart Failure Association of the European Society of Cardiology, Japanese Heart Failure Society and Writing Committee of the Universal Definition of Heart Failure. J Card Fail 2021; 27: 387–413. doi:10.1016/j.cardfail.2021.01.022