Article Text
PDF
Abstract
Aims Some patients with cardiac dystrophinopathy die suddenly. Whether such deaths are preventable by specific antiarrhythmic management or simply indicate heart failure overwhelming medical therapies is uncertain. The aim of this prospective, cohort study was to describe the occurrence and nature of cardiac arrhythmias recorded during prolonged continuous ECG rhythm surveillance in patients with established cardiac dystrophinopathy and relate them to abnormalities on cardiac MRI.
Methods and results A cohort of 10 patients (36.3 years; 3 female) with LVEF<40% due to Duchenne (3) or Becker muscular (4) dystrophy or Duchenne muscular dystrophy-gene carrying effects in females (3) were recruited, had cardiac MRI, ECG signal-averaging and ECG loop-recorder implants. All were on standard of care heart medications and none had prior history of arrhythmias.
No deaths or brady arrhythmias occurred during median follow-up 30 months (range 13–35). Self-limiting episodes of asymptomatic tachyarrhythmia (range 1–29) were confirmed in 8 (80%) patients (ventricular only 2; ventricular and atrial 6). Higher ventricular arrhythmia burden correlated with extent of myocardial fibrosis (extracellular volume%, p=0.029; native T1, p=0.49; late gadolinium enhancement, p=0.49), but not with LVEF% (p=1.0) on MRI and atrial arrhythmias with left atrial dilatation. Features of VT episodes suggested various underlying arrhythmia mechanisms.
Conclusions The overall prevalence of arrhythmias was low. Even in such a small sample size, higher arrhythmia counts occurred in those with larger scar burden and greater ventricular volume, suggesting key roles for myocardial stretch as well as disease progression in arrhythmogenesis. These features overlap with the stage of left ventricular dysfunction when heart failure also becomes overt. The findings of this pilot study should help inform the design of a definitive study of specific antiarrhythmic management in dystrophinopathy.
Trial registration number ISRCTN15622536.
- ARRHYTHMIAS
- HEART FAILURE
- Magnetic Resonance Imaging
- Cardiomyopathy, Dilated
- Genetic Diseases, Inborn
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
The burden of arrhythmias in patients with established cardiac dystrophinopathy has received little study to date and so, the extent to which premature death could be prevented by defibrillator implantation, for example, is unknown.
WHAT THIS STUDY ADDS
This prospective, observational study defines the nature and frequency of arrhythmias in patients with established cardiac dystrophinopathy over long-term, continuous ECG surveillance. Non-sustained, asymptomatic atrial and ventricular arrhythmias correlated best with increased atrial and ventricular chamber volumes and the extent of ventricular scarring rather than with ejection fraction.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Improving cardiac prognosis in dystrophinopathy is mostly achieved by timely use of drugs in combinations that preserve ventricular function and reduce myocardial fibrosis. Any additive benefits of specific antiarrhythmic measures require further study.
Introduction
A progressive form of cardiomyopathy occurs in almost all males affected by Duchenne muscular dystrophy (DMD), in about 50% of those with Becker muscular dystrophy (BMD) and in up to 17% of females who carry a DMD-gene variation (DMDc).1–3 As survival has improved for patients with DMD as a result of better coordinated, more standardised care delivery, the importance of preserving cardiac function in achieving longer survival in dystrophinopathy is increasingly apparent.4 5 Cardiomyopathy accounts for a greater proportion of deaths in BMD than in DMD, because of its milder skeletal muscle phenotype and longer survival.6 7 The contribution of cardiomyopathy to death in female DMDc is unknown.8
In DMD, the myocardium is abnormal from the beginning and left ventricular (LV) dysfunction progresses silently for most of its course. Consequently, all patients are recommended regular heart assessments by echocardiography (echo) and/or cardiac MRI (CMR) from diagnosis to detect early heart involvement.6 8 9 For the same reasons, the 2018 International Care Standards DMD recommended the initiation of prophylactic ACE inhibitor (ACEi) medication in those with DMD no later than age10 years, or even earlier when myocardial fibrosis or evidence of LV dysfunction is detected by imaging.7 Cardiomyopathy typically occurs at a much older age in patients with BMD and in females carrying the DMD gene. Even in those with heart involvement, its course is more variable. However, detection also requires a similar schedule of heart imaging.
When LV dysfunction is evident in cardiac dystrophinopathy, treatment currently is with combinations of heart medications—an ACEi or angiotensin-receptor blocker and mineralocorticoid receptor blocker. A beta-blocker (BB) is added if heart rate is persistently elevated.6 10 These drugs slow but cannot prevent further decline in LV function, although over a longer time frame in patients with DMD.11 Increasingly, therefore, more potent drugs are being deployed, particularly when there is evidence of progressive LV dysfunction despite standard of care medications.12 The object is to preserve LV function for as long as possible, delay the onset of heart failure symptoms and prolong quality survival.
As LV function deteriorates in cardiac dystrophinopathy, QRS complex duration typically prolongs and both atrial arrhythmias and the frequency and complexity of ventricular premature beats increase on Holter ECG recordings.2 13 14 However, sudden death is quite rare in DMD and the extent to which it occurs due primarily to sustained, correctable, ventricular arrhythmias as opposed to secondary to severe but asymptomatic, LV dysfunction has received little study.15 To date, prophylactic defibrillator implantation (ICD) is used only very selectively in patients with dystrophinopathy, despite evidence from multicentre randomised, controlled trials of the survival benefits of prophylactic ICDs in dilated cardiomyopathy of other aetiologies and the Heart Rhythm Society 2022 consensus statement supporting their use in advanced muscular dystrophy-related cardiomyopathy.16 17 This is because arrhythmia prevalence rates and the risk–benefit ratio of ICD implantation on survival and quality of life are not well established for patients with advanced DMD—a progressive, multisystem disorder in which there are competing causes for death.18
This exploratory prospective, cohort study aimed (1) to define whether and what types of arrhythmias occur spontaneously over prolonged, comprehensive, ECG rhythm surveillance in patients with established cardiac dystrophinopathy (ie, DMD, BMD or DMDc) receiving standard of care medical therapy and (2) to correlate arrhythmia burden with the severity of LV dysfunction on echo, presence of late potentials on signal-averaged ECG and LV-chamber volumes and extent of scarring on CMRI.
Methods
A large cohort of paediatric and adult patients with dystrophinopathy are under regular cardiac review at a dedicated, supra-regional Cardiology-Muscle Clinic at Freeman Hospital, Newcastle upon Tyne, UK. All have previously had their genotype established by the Specialised Neuromuscular Genetics Service at diagnosis and their multidisciplinary care is coordinated by the Muscle Team at the John Walton Muscular Dystrophy Research Centre. Standard assessments include serial ECG, echocardiography performed according to British Society of Echocardiography standards and CMRI when needed to guide clinical decision-making.19
Consecutive patients whose LVEF% was less than 40% at their most recent echo assessment were informed of the study in a patient information document ahead of a scheduled review appointment. Those interested in taking part had a discussion with a member of the research team, during which, they were made aware of the association between LV dysfunction and disturbances of heart rhythm. Those agreeing to participate, gave informed consent and were recruited to the study. To qualify for the study, patients needed to (1) have genetically confirmed DMD, BMD or female DMDc; (2) with LVEF% less than 40% on echocardiography, attributable to dystrophinopathy; (3) be receiving recommended standard of care DMD heart medications; (4) be without symptoms of heart failure (New York Heart Association class 1) or prior evidence of sustained atrial or ventricular arrhythmias. Recruiting patients with DMD, BMD and DMDc females was justified on the basis that, although differing in the severity of skeletal muscle phenotype, they share the same myocardial pathological processes for cardiomyopathy.20
The patients recruited underwent comprehensive assessments of biventricular ventricular function, native and late-Gad enhanced imaging of myocardial tissue by CMR and measures of these were used to explore the relationships between them and the occurrence of cardiac arrhythmias subsequently over prolonged follow-up. Cardiac arrhythmia burden was established by means of implanted ECG-loop recorders (LINQ Insertable Cardiac Monitor, Medtronic, USA).
Echocardiography assessment
Echocardiography was performed according to British Society of Echocardiography standards by experienced echocardiographers in a high-volume quaternary referral cardiac centre as standard of care. LV-systolic function, defined as LVEF% was acquired as standard. LV tissue-Doppler imaging) was acquired as a measure of LV-segmental systolic and diastolic function, as was left atrial volume (LAV).19
Quality of life assessments
To define the overall effect of dystrophinopathy on their lives, participants completed two quality-of-life questionnaires—36-item Short-Form Survey instrument (SF-36) and the Individualised Neuromuscular Quality of Life Questionnaire (Mapi Research Trust, 2005 M R Rose & King’s College National Health Service Trust).21 22
CMRI details
Participants, in whom it was safe to do so (eg, not ventilator dependent; no MR-incompatible spinal rods in situ) then underwent CMR, which included late Gad-enhanced sequences. Consideration was given to the degree of peripheral contractures and comfort in the MRI scanner. All studies were performed on a 1.5T Siemens MAGNETOM Sola system, optimised for cardiac imaging. Imaging was supervised and reported by an CMR experienced cardiologist. The cardiac long and short axes were determined from standard scout views. ECG triggered T1 maps were acquired in the mid-LV before and 15 min after 0.15 mmol/kg Gadobutrol (Gadovist; Bayer) contrast administration with MOLLI sequences optimised for long T1 (native) and short T1 (postcontrast) times. LV mass and volumes, and RV volumes, were obtained from cine imaging covering the ventricles in the short axis. Late gadolinium enhancement images were acquired 8 min after gadolinium administration in three long axis views and covering the left ventricle in the short axis, planned using the short axis stack geometry.
Analysis of all images was performed using cmr42 software (Circle Cardiovascular Imaging, Calgary, AB, Canada) by AKM. Volumetric and mass analysis was performed in the standard manner from the short‐axis stack (LV and RV). T1 values were calculated from source images after manual motion correction with a region of interest in the mid inferior septum—taking care to avoid the blood pool. Extracellular volume (ECV) was calculated precontrast and postcontrast from the mid-inferior septum also by a standard method.23 24 Hyperenhancement abnormalities on late-gadolinium CMRI were reported as either present or absent, and attributed to a myocardial segment as per the 17-segment American Heart Association model. Quantitative assessment was also performed by semiautomated signal intensity analysis at various thresholds for scar detection above normal myocardium with manual correction of blood pool artefact.
ECG loop-recorder details
ECG loop-recorders were implanted (LINQ Insertable Cardiac Monitor, Medtronic, USA) subcutaneously in the left parasternal area under local anaesthesia after CMR, to avoid imaging artefacts.25 26 Devices were programmed to detect pathological brady (ie, heart rate <30/min for >3 s) or tachycardia episodes (ie, heart rate >150/min for >3 beats) automatically. Each patient was provided with a unique telemetry device, allowing full disclosure of all predefined ECG episodes and automatic transmission of all detections to a commercially provided website. Recordings were reviewed remotely, initially by a cardiac physiologist and subsequently by a cardiologist at the implanting centre. This allowed the date, time, nature and a high-quality ECG of each event to be assessed without the patient having to reattend the hospital. Patients were also able to make recordings at time of any heart symptoms using a hand-held device activator. The primary aim was to characterise the spontaneous ‘arrhythmia burden’ and how it varied over the 3-year surveillance made possible by the battery longevity of the device. All episodes of arrhythmia with broad QRS morphology were interpreted as ventricular in origin. Ventricular arrhythmia burden was summarised for each patient using the total number of separate events and the duration of the longest recorded episode.
Signal-averaged ECG recording
Participants had a signal-averaged ECG (sAvECG) recorded using commercial software on a Marquette 3550 machine (Marquette Medical Systems, Florida, USA). Signals were filtered at 40–250 Hz and averaging was based on about 250 QRS complexes. A positive test for ‘late potentials’ was defined by the presence of any two of the following: total filtered QRS duration >114 ms (SAECG-QRS); amplitude of terminal QRS with amplitude <40 uV lasting longer than 38 ms (LAS40); root mean squared signal amplitude of last 40 ms <20 uV.27 ECG late potentials indicate the presence of areas of slow electrical conduction within scared ventricular myocardium and, so, an individual at increased risk of ventricular ectopy inducing sustained re-entrant, ventricular tachycardia.28
Statistical analyses
Results of all variables were first summarised descriptively for each patient. To explore correlations between various CMR measures and occurrence of ventricular arrhythmias, the total number and longest duration of arrhythmia episodes recorded in each patient was used to arrange them hierarchically. Two groups were then created from those with CMR scan results, comprised those with higher or lower ‘arrhythmia burden’ respectively (ie, composite of total event number and duration). Various CMR variables were then compared with determine whether any significant differences or trends were evident between them using SPSS. Significance was defined at the 5% level.
Patient and public involvement and funding
The protocol was discussed during its development with representatives of the Action Duchenne UK (info@actionduchenne.org), patient support charity and funded by Duchenne UK (project number: 2018-01 ‘Arrhythmias’) (https://www.duchenneuk.org) a leading UK-registered (8030768) muscular dystrophy charity. Both indicated high level support for the research.
Study registration
The study is registered with ISRCTN15622536 (https://doi.org/10.1186/ISRCTN15622536) and was portfolio adopted by the National Institute for Health and Care Research (CPMS ID 41597).
Results
10 participants (7 male; DMD 3, BMD 4, DMDc 3) with mean age 36.3 years (range 22.4–56.6 years) and LVEF less than 40% by echo assessment were recruited and underwent CMR and device implants as per protocol in the period December 2019 to December 2021. One patient, known to experience episodes of sustained atrial flutter, was excluded. None were excluded because of prior ventricular arrhythmias. Recruitment was delayed significantly by the general and research-specific restrictions prevailing during the height of the COVID-19 pandemic. The last patient last visit for the study was 8 May 2022.
Participant demographics
Participant ages, gender, genotype, SF-36 quality of life summary scores, medications and level of other supports at time of recruitment are summarised in table 1.
Patient eligibility and dystrophinopathy descriptors at time of recruitment
Supplemental material
SF-36 summary scores were within the normative population range of 50±10, consistent with preserved quality of life and individual patient scores in each of the eight domains are provided in online supplemental table.
Echo-measured LVEF was less than 40% in all 10 patients (mean±SD: 32.9±3.2%; lower limit of normal 55%) and mean LV-fractional shortening (LVFS%) was also severely reduced at 15.2±4.5% (lower limit of normal 28%). LV end-diastolic dimensions were increased on echo in four (40%)—mildly in two and moderately in two, and in one the end-systolic dimensions were also moderately increased. LAVs by echo assessment were enlarged in five and moderately or severely enlarged in three. Tissue Doppler measured E/E’ ratios were more than eight, the upper limit of normal, in five patients. The combination of LA enlargement with abnormally high E/E’ tissue-Doppler measure is indicative of diastolic LV dysfunction. Consistent with optimised BB dosing in those able to receive it, mean heart rate was 66 beats/min (range 45–81). Despite having severely reduced LV systolic function, none had symptoms or signs of heart failure or volume overload and all were on optimised combinations of cardioactive medications as recommended in International Standard of Care DMD guidance.7 10
CMRI findings
CMR was attempted in nine participants, not tolerated in one and contraindicated in one because of need for 24-hour ventilatory support. Mean LVEF% in those who had both assessments was higher by CMR than by echo measurement (LVEF% CMR=42.4%±5.7%; echo=35.3%±3.4%).29 LV-end diastolic and or end-systolic volume index values were enlarged in five (62.5%) and eight (100%) patients, respectively. Myocardial fibrosis was found in all eight patients who underwent CMRI. Fibrosis was transmural in some segments in six and subepicardial only in two. LV myocardial mass index was normal in all. Table 2 summarises and correlates measures from ECG, signal averaged ECG, CMR measures (LVEF%, LV-mass and LV-volume indices; LAi), and extent of LV fibrosis as quantified using the ‘full width half max’ as standard.
Correlation between ECG, CMR function, chamber volumes and tissue measures with all atrial and ventricular arrhythmia episodes
Figure 1A–D shows examples of the location and extent of myocardial scarring in one of the patients from native and late Gad-enhanced image sequences and a short-axis, colour-coded T1-map showing extensive fibrofatty infiltration in the lateral LV wall in the same patient. Figure 2 is an AHA 17-segment, target display summarising the typical pattern in distribution of fibrosis in dystrophinopathy. LV-myocardial fibrosis first becomes evident epicardially and preferentially in the basal and mid antero-lateral and infero-lateral segments. However, scarring had already progressed to become transmural in some segments in six of the patients studied and remained subepicardial in only two. T1 measures were abnormally high (ie, >950±21 ms at 1.5T) in five and ECV was also elevated (ie, >25%) in three of the seven patients who had CMR assessment (table 2).23 24 ECG signal-averaging revealed ‘late potentials’ in four patients, indicating slow intraventricular conduction due myocardial scarring. Myocardial mass index was normal in all patients by CMR assessment.
(A–D) CMR example from one patient, showing presence and extent of fibrosis on late-Gad and ECV imaging. Four-chamber view of the heart showing native (A) and late gadolinium enhanced (B) images of left ventricular myocardial fibrosis (red arrows) in a patient with high arrhythmia burden. (C) Short axis, mid-chamber view showing extensive fibrosis in the lateral segments (red arrows) and lesser fibrosis (red star) in the septum. (D) Short-axis, colour-coded T1-map showing extensive fibrofatty infiltration in the lateral LV-wall in the same patient. CMR, cardiac MR; ECV, extracellular volume; LV, left ventricular.
Preferential distribution of left ventricular myocardial fibrosis in dystrophinopathy. AHA 17-segment target-diagram of the left ventricle, summarising the pattern and extent of fibrosis on late-Gad enhanced CMRI. Numbers indicate how many of the eight patients with CMR images had fibrosis in that segment. CMRI, cardiac MRI.
Arrhythmia detections and descriptors
None of the patients were taking antiarrhythmic therapy other than BBs for LV dysfunction. Automatic device activations due to sinus tachycardia or artefacts were excluded from analysis. None died or required hospital admission for either heart failure or sustained atrial or ventricular arrhythmia during median follow-up 30 months (range 13–35). No episodes of AV block or pathological bradycardia were seen at any time point. Only one patient reported symptoms and activated ECG recordings. These only showed isolated ventricular ectopic beats during sinus rhythm.
ECG recordings at time of confirmed arrhythmia episodes showed asymptomatic regular or irregular broad-QRS tachycardia in 8 (80%) patients lasting for between 5 beats and 45 s. Regular narrow-QRS episodes or atrial fibrillation were recorded in 6 (60%) lasting between 2 and 79 s each. Six patients had both broad and narrow QRS morphology arrhythmias and two had broad-QRS episodes only. Table 2 also summarises all arrhythmia episodes recorded in each patient, categorised simply on the basis of having either narrow or broad QRS morphology along with their duration and cycle length. In this analysis, all broad QRS-tachycardia episodes have been interpreted as ventricular in origin.
Figure 3 shows examples of the types of ventricular arrhythmias recorded. Initiation sequences of the various VT episodes were suggestive of different underlying mechanisms—shorter coupling intervals and regular cycle length within an episode that are more indicative of typical re-entry, long coupling intervals and irregular cycle length more suggestive of abnormal automaticity (Figure 3).30 The frequency and duration of non-sustained VT episodes increased progressively in one patient in the course of the study period, prompting implantation of an ICD. The device has not been required to treat any sustained episodes to date.
Examples of ventricular arrhythmias recorded. Pt 2, 3, 4, 5 and 6=patient numbering as in each of the tables; Pt 4: Long, self-terminating ‘slow VT’ episode (CL 380–420 ms; ‘short-long-short’ induction sequence); Pt 2: Regular 13-beat VT episode (~380 ms initial coupling interval, possible re-entry mechanism); Pt 5: Long coupling interval (640–690 ms; possibly automatic mechanism); Pt 6: Very long initial coupling interval~1000 ms; unlikely re-entrant mechanism); Pt 3: Irregular 5-beat VT episode (‘R-on-T’ induction). VT, ventricular tachycardia.
Figure 4 shows examples of the types of atrial arrhythmias recorded. Atrial volumes were increased in all patients and four also had low amplitude P-waves on standard ECGs. Atrial arrhythmias occurred in five, confirming the validity of these other pointers to atrial electrical instability.
Examples of atrial arrhythmias recorded. Pt 1, 3, 5 and 6=patient numbering as in each of the tables; Pt 1: Non-sustained atrial fibrillation; Pt 5: Longest episode recorded of regular atrial tachycardia; P 6: Artefact resulting in multiple device activations during sinus rhythm.
Relationship between abnormalities in cardiac structure and function and arrhythmia
Even from the findings in such a small number of participants, significant trends were evident between those with higher burden of ventricular arrhythmias (ie, total number of episodes and their maximum duration) and measures of tissue abnormalities (segments with fibrosis and increased ECV) and degree of LV enlargement (LVEDVi mls/m2) although only ECV% difference was statistically significant (p=0.029) (figure 5). Ventricular arrhythmia burden did not correlate with LVEF% (p=1.0).
(A–D) Relationship between various CMRI variables in patients dichotomised into those with ‘higher’ or ‘lower’ ventricular arrhythmia burdens. (A) The significantly higher ECV% found in the patients with the highest arrhythmia burden. (B–D) Non-significant trends for differences between the groups, despite the small patient sample: higher mean mass of fibrosis at the ‘full-width, half-mass’ threshold of late-Gad enhancement (B); higher mean T1 value (C) higher mean LVED-volume index (D). CMRI, cardiac MRI; ECV, extracellular volume; LVED, left ventricular end-diastolic.
ECG and signal averaged ECG findings
All 10 patients recruited were in sinus rhythm at study entry clinically and on 12-lead ECG. PR intervals were >120 ms in seven and short (100–114 ms) in three. None had evidence of AV-conduction delay or abnormal QRS axis. In five, QRS duration was abnormally prolonged (range 129–152 ms). All had prominent R-wave voltages in the right precordial ECG leads and, in eight, small Q-waves, typical of cardiac dystrophinopathy, were evident. Q-waves in them were confined to leads I and aVL in four patients but widespread in the other four. Late potentials were present on four (40%) sAvECGs. However, since only two of those had CMR assessments, it was not meaningful to try to correlate the presence of late potentials with the extent of fibrosis on imaging.31
Discussion
To our knowledge, this is the first prospective, observational study to define the prevalence, nature and frequency of arrhythmias over long-term continuous ECG rhythm surveillance in a cohort of patients with established cardiac dystrophinopathy. Although differing in the severity of their skeletal muscle phenotype (ie, DMD, BMD or DMDc), those recruited shared the same aetiology and severity of cardiomyopathy, as a consequence of inherited DMD-gene variations.20 Therefore, although the number of participants was small, the findings are likely to provide a valid estimate of the arrhythmia burden and CMR features in patients with cardiac dystrophinopathy.
Main findings
Self-limiting episodes of ventricular and/or atrial arrhythmias were confirmed in 80% of those studied during median follow-up of 30 months. However, allowing for the length of follow-up, the overall arrhythmia burden was low. None of the arrhythmia episodes were symptomatic, highlighting the importance of rhythm surveillance, if they are to be detected.32
Even from the results of such a small pilot study, it was evident that a patients’ ventricular arrhythmia burden related to the extent of interstitial fibrosis (ECV) and scar (late-Gad enhancement) on CMR. This is consistent with reports by others showing that the extent and complexity of myocardial scaring is a powerful predictor of ventricular arrhythmias and sudden death in patients with either idiopathic or DMD-related cardiomyopathy.33–36 Patients with the greatest number and longest duration of ventricular arrhythmia episodes, were also those with the largest LV-end diastolic volumes.37 When patients in the higher and lower halves of the group were compared, the trend apparent in figure 5 was not statistically significant due to the small sample size. The association observed between ventricular volume/dilatation and the occurrence of ventricular arrhythmias is underappreciated in the limited literature on the topic to date, although the findings are not unexpected.31 38
Half of those studied also had left atrial dilatation and four had low amplitude P-waves on ECG. These features, interacting with the presence of ventricular scarring and diastolic dysfunction, are already known to favour the development of atrial tachyarrhythmias.39–41 Unsurprisingly, therefore, self-limiting episodes of atrial tachyarrhythmias were seen in 60% of the patients and, interestingly, were of longer duration than any episodes of ventricular arrhythmias. Rapid ventricular response rates to atrial arrhythmias are not benign in the context of significant LV dysfunction since they can cause haemodynamic compromise on their own and/or precipitate ventricular fibrillation. These risks are in addition to their potential to cause stroke and other thromboembolic complications.42 No brady arrhythmias due to either sinus node dysfunction or AV block were recorded. The findings of this pilot study may help inform the planning of future trials of specific antiarrhythmic interventions in patients with cardiac dystrophinopathy.
Consistency of findings with previous publications
The main arrhythmia findings are consistent with previous reports from retrospective studies, based on ‘snap-shot’ Holter ECG recordings, in larger patient cohorts.13 27 31–33 37 However, the length and comprehensive nature of surveillance adds new information, showing how arrhythmia risk can change and evolve over relatively short periods of time, as evidenced by increasingly frequent and longer duration of ventricular tachycardia episodes in one patient, which justified ICD implantation.
The relationship between the extent of myocardial scarring and occurrence of ventricular arrhythmias and sudden death has been recognised previously by others but, this study shows that the strength of this association is such that it is even evident in the results from such a small patient cohort.35–37 43–45 Although measures of ECV correlated with arrhythmia burden, those of late-Gad enhancement did not. This might be explained by LV-wall thinning confounding accurate calculation of fibrosis in the most affected LV-segments or that dense, confluent scar patterns are less arrhythmogenic than more diffuse, patchy ones. The fact that VT burden correlated less well with LVEF% is consistent with previous publications.46
Pointers to various underlying arrhythmia mechanisms
The observation that patients with the largest ventricular volumes were also those with the highest number and longest runs of ventricular arrhythmias is novel and, by extrapolation, could imply that that patients with dystrophinopathy are at greatest risk of experiencing sustained ventricular arrhythmias or sudden death at the stage of ventricular decompensation with progression to overt heart failure. This hypothesis is also supported by the more detailed examination of arrhythmia induction sequences, cycle lengths and whether episodes had regular or irregular rates. These features were consistent with various underlying arrhythmia mechanisms as seen in heart failure—increased automaticity as well as more classical reentry.30 47 48 Multiple, fixed (scar, conduction abnormalities) and dynamic substrates (myocardial stretch, electrolyte disturbance, autonomic imbalance) coexist in patients with advanced cardiomyopathy and, particularly, in the context of incipient heart failure. This range of proarrhythmic factors can all interact to precipitate ventricular fibrillation or sustained arrhythmias, culminating in sudden death.46 49
Clinical significance of the arrhythmias documented?
Previous publications have established that some patients with cardiac dystrophinopathy die suddenly and isolated case reports have confirmed the occurrence of sustained ventricular tachycardia or fibrillation in some.13 50 51 However, sudden deaths, although outwardly indistinguishable, can be due to diverse causes (ie: ventricular tachycardia/fibrillation, profound bradycardia/asystole, electromechanical dissociation (‘pulseless electrical disease’) or large pulmonary embolus).52–55 A fundamental uncertainty, not resolved by this study, because no deaths or sustained ventricular arrhythmia episodes occurred, is whether the finding of infrequent, short arrhythmia episodes identifies an individual likely to benefit from prophylactic antiarrhythmic management or is only a marker of poor cardiac prognosis due mainly to progressive heart failure.56 57
The links observed between increased arrhythmia burden, higher ECV, more extensive fibrosis and greater ventricular dilatation suggest that the occurrence of sustained arrhythmias reflects the presence of severe myocardial damage and dysfunction and so, increased risk of all-cause mortality rather than from arrhythmias preferentially.58 The closer the timing in occurrence of even sustained arrhythmias to the onset of overt heart failure, the less impact specific antiarrhythmic measures would be expected to have on overall survival.51 59 Although the HRS Expert Consensus Statement 2022 suggests a role for ICD implantation under certain circumstances in patients with DMD or BMD to prevent sudden cardiac death, it highlights the current lack of evidence of benefit for their recommendations.16 18 60 61 ICD therapies might simply postpone or change the mode of death without prolonging survival,62
Study limitations
This study has several limitations. First, it was only ever planned as a hypothesis generating study, to describe the typical ‘arrhythmia burden’ in a small cohort of patients with advanced cardiac dystrophinopathy and to correlate event rates with other variables from ECG and heart imaging. Second, because those recruited had the same genetic basis for cardiomyopathy, the assumption that their underlying pathophysiological processes and so, arrhythmia event rates would be the same, may not be correct. However, if recruitment had been restricted to patients with DMD who could undergo CMR safely, it would have led to even greater selection bias. This is because many adult patients with DMD cannot undergo CMR because of having MR incompatible spinal rods implanted, or needing ventilatory support when supine or do not tolerate CMR because of contracture-related discomfort, resulting in poor image quality due to movement artefacts. Excluding patients already known to experience arrhythmia episodes was justified on the basis that most patients with dystrophinopathy are not known to have arrhythmias and so, to include them risked finding unrepresentatively high rates of arrhythmias during follow-up. The fact that all those studied were taking combination heart medications in optimised doses may have reduced the number and duration of arrhythmias observed. Finally, the lack of correlation between the extent of myocardial fibrosis, as measured by late-Gad imaging, and ventricular arrhythmia burden could be incorrect, due simply to the vagaries of the small sample size.
Conclusion
This hypothesis-generating study reports finding only infrequent, short-lived, asymptomatic atrial and ventricular arrhythmias in patients with established dystrophinopathy over prolonged, comprehensive, ECG rhythm surveillance. It was evident, even from the findings from such a small patient cohort that those with highest ventricular arrhythmia burden were also those with largest LV volumes and greatest extent of fibrosis, as measured by abnormally high ECV percentages. Although too small a study to provide definitive conclusions, the types of atrial and ventricular arrhythmias observed are probably better interpreted as indicative of the severity of heart damage, LV-remodelling and dysfunction than as specific identifiers of individuals at high risk of arrhythmic death. The findings could be used to inform the design of a definitive randomised trial of specific, prophylactic, antiarrhythmic management in patients with cardiac dystrophinopathy.
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by North East–Tyne & Wear South Research Ethics Committee (IRAS Project ID: 252541; 29/04/2019). Participants gave informed consent to participate in the study before taking part.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @hanastev
Contributors JB: principle investigator, study concept, design, funding application, patient recruitment and data acquisition, study supervision and data interpretation, drafting manuscript for content, final approval of manuscript, accountable for all aspects of the work as guarantor. MT: device implants and programming, data acquisition and surveillance, manuscript final approval. HS: regulatory and financial management of the study, supervising data collection, manuscript final approval. LB: patient recruitment, data collection, patient follow-up manuscript final approval. OG-F: data analysis and statistical support. AKM: study design, supervision of cardiac MR scan acquisitions, scan analyses and reporting, manuscript preparation, redrafting for content and final approval.
Funding The research team gratefully acknowledge the funding support from Duchenne UK, which made this project possible and Newcastle upon Tyne NHS Hospitals Foundation Trust, who sponsored the research. We also thank sincerely the patients who took part since, without them, the study could never have happened. We are also grateful to the various individuals both at Action Duchenne and Duchenne UK for their helpful feedback at the protocol development stage of this study.
Competing interests JB Research grant from Duchenne UK; Trial Data Monitoring Committee member, Sarepta Therapeutics, Cambridge, Massachusetts, USA; Consulting fees from EspeRare Foundation, Geneva, Switzerland. The coauthors report no conflicts of interest in relation to this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.