Implantable loop recorders (ILRs) enable continuous rhythm monitoring; they are generally used in patients with unexplained syncope or with presumed infrequent atrial arrhythmias, particularly after stroke of an unknown cause. These devices have also been used in patients with presumed borderline risk for life threatening arrhythmic events.

Cardiomyopathies comprise a heterogeneous group of diseases that often include patients at risk for arrhythmic events. In fact, many of the individuals will be, at some point, subject to arrhythmic risk stratification.

According to the 2023 European Society of Cardiology (ESC) guidelines for the management of cardiomyopathies, in patients with dilated cardiomyopathy (DCM), an implantable cardioverter-defibrillator (ICD) is a Class IIa recommendation for primary prevention of sudden cardiac death (SCD) in patients with left ventricle ejection fraction (LVEF) ≤35%. Although LVEF ≤35% is an established independent risk marker of all-cause and cardiac death in DCM, it has also shown only modest ability in identifying high-risk DCM patients, suggesting that additional factors should be taken into consideration when deciding on ICD implantation in this etiologically heterogeneous disease. Recent studies suggest that DCM patients carrying pathogenic variants in high-risk genes should be regarded as having a genetic predisposition to SCD. In such cases, primary prevention with ICD therapy may be justified even when LVEF is above 35%, particularly in the presence of additional risk markers, one of which is late gadolinium enhancement (LGE) on cardiac magnetic resonance (CMR). This is especially relevant in the setting of a left ventricular (LV) ring-like scar pattern, defined as ≥3 contiguous segments with subepicardial or midwall LGE within the same slice.1–3

The non-dilated LV cardiomyopathy (NDLVC) phenotype is defined by the presence of non-ischemic scarring or fatty replacement in the absence of LV dilatation, with or without global or regional wall motion abnormalities, or isolated global LV hypokinesia without scarring, that is unexplained solely by abnormal loading conditions or coronary artery disease.2 Similar to patients with DCM, an ICD should be considered for primary prevention to reduce the risk of SCD in patients with heart failure (HF) symptoms and LVEF ≤35%. In patients with LVEF >35%, implantation of an ICD for primary prevention may be recommended as Class IIa or IIb depending on the genotype associated with a high risk of SCD and the presence of additional risk factors.2

According to the 2022 ESC guidelines for the management of patients with ventricular arrhythmias and the prevention of SCD, in both DCM/NDLVC patients, cases of LVEF between 36% and 50% and unexplained syncope, with no other arrhythmic risk factors, ILRs are a Class I recommendation.1 In adult patients, hypertrophic cardiomyopathy (HCM) is defined by LV wall thickness ≥15 mm in any myocardial segment that is not explained solely by loading conditions, or by lesser degrees of wall thickening (≥13 mm) in the presence of other features, including family history or pathogenic/likely pathogenic variants. In these patients, ICDs for primary prevention should be considered when the estimated five-year HCM Risk-SCD score is ≥6%. In patients with an intermediate five-year risk of SCD (≥4% to <6%), or even in those at low risk (<4%) who have clinical risk factors, such as the presence of extensive LGE (defined as ≥15% of LV mass) on cardiac magnetic resonance, ICD implantation may be considered, following comprehensive clinical assessment and shared decision making with the patient.1,2

In patients with HCM and recurrent episodes of unexplained syncope, who are at low risk of sudden SCD, an ILR should be considered.4 Due to its prolonged monitoring capacity, the ILR has a higher diagnostic yield for detecting ventricular tachycardia (VT). The detection of VT can influence risk stratification for SCD and the decision to implant an ICD.5

Transthyretin amyloid cardiomyopathy (ATTR-CM) is caused by the accumulation of different misfolded precursor proteins, which form deposits in tissues and organs. It is an increasingly recognized cardiomyopathy associated with a high risk of congestive HF and bradyarrhythmias and/or tachyarrhythmias. The advent of new therapies has improved patient survival, raising concerns about the long-term risk of ventricular arrhythmias. Currently, there is no consensus on the absolute benefit that ICDs may provide for these patients. However, documenting arrhythmias with ILR might be useful for guiding clinical decisions.6

In clinical practice, the use of ILRs in cardiomyopathies often extends beyond current recommendations, particularly in borderline risk cases, where evidence for risk stratification is scarce.7 The role of these devices in improving the detection of significant arrhythmias requiring changes in clinical management has yet to be established. Our purpose was to evaluate the diagnostic yield, types of arrhythmia, and subsequent management in patients with cardiomyopathies receiving ILRs.

A total of 45 patients diagnosed with cardiomyopathy were included. Twenty-three patients (51%) were male, and median age was 62 (48–71) years. The underlying disease was HCM in 31 patients (69%), DCM/NDLVC in 12 patients (26%) and ATTR-CM in two patients (4%).

The baseline clinical characteristics of the patients are described in Table 1. The most frequent risk markers were brief run (3–5 beats) of NSVT on 24-hour ECG monitoring in the absence of other risk markers (42% of patients), followed by unexplained syncope or presyncope (36%), family history of premature SCD in a first-degree relative (36%), and palpitations suspicious of arrhythmic origin (18%).

The implanted ILR device was a Reveal LINQ II™ in 22 patients (49%), a Confirm Rx™ in 17 patients (38%) and a Reveal LINQ™ in six patients (13%).

During the mean follow-up of 19±13 months, 44% of patients had at least one ILR-guided diagnosis and 36% had an ILR-based therapy. ILR-guided diagnosis and therapies are depicted in Tables 2 and 3, respectively.

Patients with transthyretin amyloid cardiomyopathy

The two patients with ATTR-CM exhibited mildly reduced LVEF (42% and 40%) and diastolic dysfunction, with an E/e′ ratio of 22 and 16, as well as pulmonary hypertension (systolic pulmonary artery pressure of 50 and 45 mmHg). Both patients were in NYHA Class II. The indication for ILRs was a brief run of NSVT in one patient and recurrent episodes of presyncope in the other. Dysautonomia was excluded as the cause of presyncope. No evidence of AF or conduction disease was previously described in either patient.

The main ILR-guided diagnosis in this group of patients was de novo AF, which occurred in both patients (100%) – Table 2. This finding led to initiation of anticoagulation and antiarrhythmic therapy – Table 3. Neither NSVT nor VT episodes occurred in these patients. However, in one patient, due to worsening biventricular function (LVEF of 35% and right ventricular ejection fraction of 33%) together with extensive LGE on CMR, primary prevention ICD implantation was performed. Conduction disease was not detected during follow-up in these patients.

While ILR-guided diagnosis was established in 44% of the study population, the incidence of the primary endpoint of meaningful arrhythmic events leading to a change in clinical management was 36%; the most common change was initiation of anticoagulation (22%). With regard to ILR-related complications, the only event was a device infection in one patient (2%) 17 days after device implantation, requiring ILR explantation in the group of patients with DCM/NDLVC. Subsequently, a new device was implanted without complications, and the patient was included in the analysis. Exploratory univariate analyses were planned to identify potential predictors of any arrhythmia detected by the ILR and of changes in therapeutic management; however, no significant predictors were identified.

Figures 1 and 2 present the cumulative event-free survival for arrhythmic episodes detected by ILR using the Kaplan–Meier method in patients with HCM and DCM/NDLVC, respectively. During follow-up, event-free survival progressively declined in both patient groups monitored with an ILR. In patients with HCM, the estimated mean event-free survival was 31.0 months (95% CI: 26.5–35.6), with a median event-free survival of 37.0 months (95% CI: 14.4–59.6) (Figure 1). In contrast, patients with DCM/NDLVC showed a more gradual decline, with an estimated mean event-free survival of 39.9 months (95% CI: 26.1–53.7); the median event-free survival was not reached during the follow-up period (Figure 2).

Figure 1.

Kaplan–Meier analysis of event-free survival during follow-up in patients with HCM with ILR-detected AF/AT or VT events.

Figure 2.

Kaplan–Meier analysis of event-free survival during follow-up in patients with DCM/NDLVC with ILR-detected AF/AT or VT events.

The main findings of our study were: (a) ILR-guided diagnosis was established in nearly half of the cohort (44%); (b) ILR-guided clinical intervention was verified in 36% of patients; (c) the most frequent finding was asymptomatic AF (24% of patients); (d) VT/NSVT were found in 22% of patients, leading to ICD implantation in 20% of the cases.

The ILR-guided diagnosis of AF led to anticoagulation therapy in 22% of the cases. This is a clinically relevant observation, especially in both cardiac amyloidosis and HCM, since these two entities are associated with an increased thromboembolic risk, with an annual incidence of stroke or transient ischemic attack approximately three times higher in patients with AF. In addition to thromboembolic complications, these patients may experience impaired quality of life, worsening functional capacity, and increased mortality.7 Accordingly, oral anticoagulation in patients with HCM or ATTR-CM and any form of AF or atrial flutter is a Class I recommendation.2 For these reasons, early detection of AF is crucial, as it may prevent thromboembolic events and facilitate the timely initiation of rhythm control therapy. Current recommendations support catheter ablation after the failure or intolerance of at least one Class I or III antiarrhythmic drug in patients with paroxysmal (PAF) or persistent AF (PersAF). In selected patients with cardiomyopathy and PAF or PersAF, and without major risk factors for recurrence, catheter ablation may also be considered as a first-line rhythm control strategy to improve symptoms, as an alternative to antiarrhythmic drug therapy.2

In our study, de novo AF was diagnosed in seven patients with HCM (23%) and in both patients with ATTR-CM (100%). This finding resulted in the initiation of anticoagulation. It also prompted the initiation of antiarrhythmic drug therapy in 7% of HCM patients and in the two ATTR-CM patients, as well as catheter ablation in 7% of HCM patients. Patients in whom no changes in antiarrhythmic therapy were made had asymptomatic, less frequent, and shorter episodes of AF.

In a previous study designed to assess the impact of ILR in HCM and its association with clinical outcomes, an ILR-based strategy in patients with HCM at elevated risk for AF was associated with a 3.5-fold higher rate of new AF diagnosis compared with patients managed without ILR.8

Furthermore, we highlight the potential role of ILR in stratifying the arrhythmic risk of these patients, since all of them had previously undergone 24-hour ECG monitoring without AF detection, underscoring the added value of ILR in this setting.

Risk stratification for SCD in patients with HCM, the major group in our cohort, has demonstrated satisfactory progress with the advent of risk stratifying scores (HCM SCD-Risk Score) and the integration of additional high-risk features, such as extensive LGE or LVEF<50%, which together have improved the process of clinical decision making.1 Although current ESC guidelines recommend considering ILR implantation in HCM patients at low risk for SCD with recurrent syncope, and in patients with frequent unexplained palpitations, the precise role of ILRs in this population remains uncertain, as the recommendations are based on limited evidence.5 A considerable proportion of patients is not yet represented in the available data, warranting individualized, case-by-case evaluation within multidisciplinary teams.

In our cohort, 23% of the HCM patients who had an ILR were subject to an ILR-guided therapy, of which 16% implanted an ICD due to ILR-diagnosis of VT. It should be noted that the only patient who received appropriate ICD therapies was a patient with non-obstructive HCM, with a low HCM SCD-Risk Score, non-sarcomeric variant, and 10% LGE on CMR. This finding strengthens the role of ILR as a valuable tool for individualized diagnostic decision making in low-risk patients with unexplained symptoms.

Regarding previous studies addressing this topic, a meta-analysis comprising 4381 patients undergoing ILR insertion for unexplained syncope reported an ILR-based diagnosis in 43.9% of patients, of which 2.7% were ventricular arrhythmias. Notably, most studies excluded patients with suspected VT, although SCD remains the most feared complication in patients with cardiomyopathy.9,10

The ELUCIDATE study, published in 2020, evaluated ILRs in a low-risk HCM population and detected NSVT in seven out of 30 patients (23%), including five with newly identified NSVT undetected on 24-hour ECG monitoring, findings that are comparable to our study.11

In the study mentioned above, which aimed to investigate the impact of ILR versus non-ILR strategy for HCM patients, patients with ILR had a 2.5-fold higher detection rate of ventricular arrhythmias. However, there was no significant difference in rates of ICD implantation or mortality over a 2.3 years’ follow-up period.8

According to the guidelines, in patients with DCM/NDLVC with LVEF between 36% and 50% and unexplained syncope, with no other arrhythmic risk factors, ILR implantation is a Class I recommendation.1 The current available data suggest that genotype is a major determinant of SCD risk, with patients harboring variants in PLN, TMEM43, DES, DSP, LMNA, FLNC (truncating variants), and RBM20 having a substantially higher rate of major arrhythmic events, regardless of LVEF. Accordingly, specific risk-prediction models have been recently developed for certain high-risk genotypes and gene-specific variants. In the former, the presence of NSVT is a key parameter, whereas in the latter, the burden of premature ventricular beats on 24-hour ECG monitoring is taken into account.2 In patients without a high-risk genotype and LVEF >35%, the presence and extent of myocardial scarring determined by LGE on CMR imaging can be helpful in risk stratification in these patients. Additional risk factors, such as syncope or presence of NSVT and burden of ventricular ectopy, may also help guide ICD implantation – however, currently there are no data to support a specific threshold for ventricular ectopy burden, and this will depend on the underlying genotype and other clinical factors.2

Different guidelines on primary prevention for this patient group have been published in consecutive years. The 2022 ESC guidelines recommend a higher five-year risk threshold of 10% to guide ICD implantation in patients with DCM/NDLVC and LMNA variants, whereas the 2023 ESC guidelines on the management of cardiomyopathies, acknowledging the difficulty of defining universal risk thresholds across cardiomyopathy phenotypes, support an approach to these patients similar to that used for HCM risk stratification.1,2 ILR may play an important role in patients with DCM/NDLVC, particularly in gene-elusive patients, in whom risk stratification is more challenging due to the absence of genetic evidence of risk. Furthermore, until recently no specific risk score was available for these patients, as this study was initiated prior to the publication of the new risk scores.12

Our cohort of patients with DCM/NDLVC had a 42% rate of ILR-guided diagnosis, and an ICD was implanted for primary prevention in 25% of patients. The decision not to implant an ICD in the remaining patients was based on the fact that the detected arrhythmias consisted solely of brief runs of NSVT, in combination with other clinical considerations indicating a lower risk of malignant arrhythmias, underscoring that these devices may be useful on an individualized basis.

In our research, we did not identify any previous studies addressing the role of ILR in predicting arrhythmic risk in patients with DCM/NDLVC.

The natural history of cardiac amyloidosis associates electrical conduction disease of the heart with symptomatic bradycardia and conduction abnormalities, for which the threshold to consider pacemaker implantation should be low. In contrast, the role of ICD in ATTR-CM for SCD prevention is not clearly known. Although NSVT and VT are frequently observed in these patients, its predictive value for subsequent SCD has not been clearly established. Current guidelines therefore conclude that evidence is insufficient to support ICD implantation for primary prevention beyond standard indications, emphasizing the need for individualized decision making.2

In our cohort, the two patients with ATTR-CM, had a ILR-diagnosis of AF, which subsequently led to therapeutic decisions, including the initiation of oral anticoagulation and antiarrhythmic drug therapy.

A study including 112 ATTR-CM patients, randomized to either standard clinical follow-up or ILR, showed that intensive monitoring of arrhythmias in the ILR group lead to an accurate and earlier detection of ventricular arrhythmias, leading to timely initiation of ICD therapy for SCD prevention. Documentation of ventricular arrhythmias in these patients was significantly higher by ILR and was a clear predictor of patient survival (HR 2.51).6

In another study including 24 patients with wild type ATTR-CM, who underwent monitoring with an ILR for a mean duration of 10.3±6.6 months, 10 patients experienced arrhythmic events – nine patients with AF, one patient with VT. Conduction disease was diagnosed by the ILR in three of the patients with AF, leading to pacemaker in two patients and a cardiac resynchronization therapy pacemaker in the other patient.13

In another study including 53 patients with various cardiomyopathies, ILR monitoring identified clinically relevant arrhythmic events in almost one-fifth of patients (17%), specifically, NSVT and VT episodes, leading to an ICD implantation in all cases.14 This is comparable to our findings, with 22% of clinically relevant arrhythmias across all cardiomyopathies, leading to ICD placement in 18% of patients.

Given the challenges in defining universal thresholds for acceptable risk across different cardiomyopathy phenotypes, shared decision making based on real-world data, along with individual preferences, beliefs, circumstances, and values, is essential. Since the decision to implant an ICD for primary prevention in patients with cardiomyopathies often falls within a gray zone, gaps in the evidence should be shared with patients, and competing risks related to the disease, age, comorbidities, and potential device-related complications should be thoroughly discussed.

Current guidelines for managing patients with cardiomyopathies recommend evaluation of arrhythmic risk using 24-hour ECG monitoring. However, our results suggest that this may be insufficient for at-risk patients, and that ILR could be a valuable tool to better redefine a patient's arrhythmic risk. Nonetheless, more robust evidence from larger cohorts is needed to confirm these findings.

In this study, we found that ILRs may have a role in the risk stratification of ventricular arrhythmias in patients with cardiomyopathies considered to be borderline arrhythmic risk. Furthermore, ILRs may also help in the detection of subclinical AF in this population.

This study was approved by the local Ethics Committee (Hospital de Santa Marta, Unidade Local de Saúde de São José, Lisboa, Portugal).

The authors declare that they have not used any type of generative artificial intelligence for the writing of this manuscript, nor for the creation of images or videos.

This work did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

None declared.