Clinical Spectrum and Outcomes in Hypertrophic Cardiomyopathy With Apical Aneurysms

Abstract

Background

Apical aneurysms in hypertrophic cardiomyopathy (HCM) have been linked to sudden cardiac death (SCD) and a nidus for thromboembolism. Uncertainty remains regarding the level of risk and significance of aneurysm size.

Objectives

The objective of the study was to determine the rate of SCD events and prevalence of apical thrombus or embolic events by size (maximum transverse dimension).

Methods

Apical aneurysms were identified in 510 patients from 10 centers, followed a median of 4.1 years for SCD events (SCD, appropriate implantable cardioverter defibrillator therapy, and resuscitated SCD) or development of apical thrombus/thromboembolism. Relationship between size and SCD events was analyzed using multivariable Cox proportional hazard models.

Results

In 510 HCM patients: 19% had small aneurysms (<10 mm), 39% medium (10-19 mm), 39% large (20-39 mm), and 3% very large (≥40 mm). SCD event rate was 2.1%/year, with risk increasing with increasing aneurysm size: 0.2%/year in small, 2.3%/year in medium, 3.0%/year in large and 8.2%/year in very large (P < 0.001). On multivariable analysis, greater size was associated with SCD events, independent of other risk markers or European Society of Cardiology-SCD score. Either an embolic event (3.6% of patients) or apical thrombus (10% of patients) occurred in 13% of patients and was independently associated with greater aneurysm size, 3% in small to 25% in very large aneurysms (P < 0.001).

Conclusions

In a large cohort of HCM patients with apical aneurysms, rates of SCD events were high, with continuous relationship between size and risk. Small aneurysms (<10 mm) were associated with low risk for SCD events (0.2%/year), whereas aneurysms ≥10 mm with high risk (>2%/year). Although embolic events were uncommon, increasing aneurysm size was associated with the prevalence of apical thrombi.

Central Illustration

Scarred and thin walled akinetic/dyskinetic left ventricular (LV) apical aneurysms in patients with hypertrophic cardiomyopathy (HCM) have been associated with an increased risk for ventricular tachyarrhythmia and sudden cardiac death (SCD), albeit with heterogeneous morphology and clinical profile.1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 However, most available apical aneurysm data have been assembled in relatively small single-center HCM cohorts, in some instances reporting inconsistencies regarding the attributable level of arrhythmic risk.^13,14 In this regard, although the American Heart Association/American College of Cardiology (AHA/ACC) HCM guidelines consider LV apical aneurysm to be a major risk marker for sudden death and a reasonable consideration for primary prevention implantable cardioverter defibrillator (ICD), the 2023 European Society of Cardiology (ESC) HCM guidelines consider the evidence insufficient to recommend prophylactic ICDs for patients with LV apical aneurysm.^14,15 Further confounding this issue is uncertainty related to arrhythmic risk and the variable size of the LV apical aneurysms. Some studies suggest similar risk across the spectrum of aneurysm size,¹ whereas other studies conclude that risk is increased only in patients with larger aneurysms.^2,3,5 In addition, although some analyses have raised concern that the akinetic aneurysm is a nidus for thrombus formation resulting in risk for embolic events and deserving of prophylactic anticoagulation,^1,3,5 others have found a low risk for embolic events.² Therefore, we sought to clarify in this particularly large and diverse multicenter international cohort of >500 patients with LV apical aneurysms the risk of life-threatening arrhythmic events, thrombus formation/embolic events, and their relationship to aneurysm size.

Methods

Demographics

For this retrospective cohort study of HCM patients with LV apical aneurysms, databases of 10 HCM centers in United States (n = 6), Canada (n = 1), United Kingdom (n = 1), India (n = 1), and Italy (n = 1), consisting of a total of 19,132 consecutive HCM patients seen and evaluated from 2000 to 2023 were accessed. Given the importance of LV apical aneurysm as disease feature, each participating center systematically evaluates the presence of aneurysm in all patients at baseline and at each follow-up visit; the presence/absence of an aneurysm is recorded as a discrete identifying field within the participating centers HCM database. In total, 510 of the 19,132 patients (2.7%) were identified with LV apical aneurysms (Supplemental Table 1). Study entry was defined as the initial clinical assessment during which an apical aneurysm was detected by imaging. Patients were followed for a median of 4.1 years (IQR: 1.5, 7.1 years) from initial aneurysm detection to the most recent outpatient visit/telephone contact or death.

Decisions to implant a primary prevention ICD in 272 aneurysm patients was predicated on contemporaneous guideline-based risk stratification¹⁵ and the clinical judgment of the managing cardiologist and mutually shared agreement with patients. This included 182 patients implanted solely due to the presence of the aneurysm, and 90 in which the aneurysm was associated with ≥1 other major marker as part of the patients clinical profile (including: family history of HCM-related SCD in 27, unexplained syncope in 25, massive LV hypertrophy in 16 and/or end stage HCM with ejection fraction <50% in 16 patients).

Of the 510 patients, 288 were administered prophylactic anticoagulation against thromboembolic events: 130 based on the aneurysm alone, whereas 158 had associated atrial fibrillation (AF) as an indication for anticoagulation.

Definitions

Diagnosis of HCM was based on imaging documentation of a hypertrophied nondilated LV with wall thickness of ≥15 mm (or ≥13 mm in setting of family history of HCM and/or with a pathogenic/likely pathogenic sarcomere gene variant), in the absence of another cardiac or systemic disease capable of producing a similar magnitude of hypertrophy.^14,15 LV apical aneurysm was defined as a discrete thin-walled dyskinetic/akinetic segment of the most distal LV chamber in the absence of obstructive atherosclerotic coronary artery disease in the relevant vascular territory.⁷ LV aneurysm was identified by noncontrast echocardiography in 179 (35%), contrast echocardiography in 153 (30%), and only by cardiac magnetic resonance imaging (CMR) in 168 (33%). In 10 other patients, aneurysm was detected only by computed tomography (n = 8) or invasive LV angiogram (n = 2).

Aneurysm size was measured by experienced cardiac imagers at each HCM center as the maximum transverse dimension: by CMR (n = 385); contrast echo (n = 63); noncontrast echocardiography (n = 52); or computed tomography (n = 8) most frequently in the long axis/apical 4 chamber view, or in the right anterior oblique projection when only identified by angiography (n = 2). Aneurysm size was characterized nominally as small (<10 mm), medium (10-19 mm), large (20-39 mm), and very large (≥40 mm).^1,3, 4, 5, 6 For patients with multiple imaging modalities documenting an aneurysm, preference was given to CMR measures, followed by contrast echocardiogram, and noncontrast echocardiogram, given the higher spatial resolution and reproducibility of measurements in the respective imaging modalities (ie, CMR > contrast echocardiogram > noncontrast echocardiogram).¹⁶ In 168 patients with paired CMR and echocardiogram within 6-months of initial aneurysm detection, aneurysm size was similar between modalities (median difference of 1 mm; IQR: 0, 4), including 85% of patients with a discrepancy between measurements of ≤5 mm, whereas 15% had greater discrepancies (>5 mm). The measurement of discrepancies of >5 mm increased with increasing aneurysm size (0% with small, 10% with medium, 23% with large, and 40% with very large aneurysms).

Other major AHA/ACC risk markers (2a) included: family history of HCM-related SCD in ≥1 relatives; massive LV hypertrophy ≥30 mm; unexplained syncope; and LV systolic dysfunction (ejection fraction <50%). Nonsustained ventricular tachycardia (VT) episodes on ambulatory monitoring or extensive late gadolinium enhancement on CMR were selectively considered risk modifiers (2b).¹⁵ ESC-SCD 5-year risk score was calculated for each patient.^14,17

Endpoints

Primary endpoint was life-threatening ventricular arrhythmic events over follow-up and included: sudden death; resuscitated cardiac arrest; or appropriate ICD interventions for VT/ventricular fibrillation (VF) (≥180 bpm) terminated by defibrillation shocks or antitachycardia pacing. Events were adjudicated by expert electrophysiologists at participating centers.

Secondary endpoints included: 1) HCM-related mortality defined as death directly related to the disease or its treatment modalities; and 2) embolic event in absence of AF history and/or presence of thrombus within the apical aneurysm typically identified by contrast echocardiogram or CMR. In 41 patients diagnosis of stroke was based on current criteria requiring evidence of symptomatic cerebral or retinal focal ischemic injury causing symptoms and signs that persist ≥24 hours, and with imaging evidence of central nervous system focal ischemic injury and classified as cardioembolic (n = 27) vs noncardioembolic when occurring in the presence of other potential noncardiac etiology (n = 14).¹⁸ In addition, we classified cardioembolic events based on the presence (n = 9) or absence of AF (n = 18) as alternative potential etiology. Evaluation for occult AF in these patients was at discretion of treating cardiologist, albeit all undergoing ambulatory monitoring of variable duration as part of evaluation.² Transient ischemic events without imaging evidence of infarction were not included.

As an exploratory analysis evaluated development of drug refractory advanced heart failure symptoms (NYHA functional class III or IV) and the association with either midventricular/apical obstruction, LV outflow obstruction or systolic dysfunction (ejection fraction <50%) as previously described.¹⁹ Midventricular or apical muscular obstruction was defined as complete systolic muscular opposition of the hypertrophied mid/distal ventricular septum and LV free wall invariably including hypertrophied papillary muscles regardless as to the presence of intracavity gradients (≥30 mmHg) on continuous wave Doppler as previously described19, 20, 21 and LV outflow obstruction as due to typical mitral valve contact with outflow gradient of ≥30 mmHg at rest or with provocation by continuous wave Doppler.¹⁹

Statistical methods

Statistical analyses were performed with SAS for Windows (version 9.3; SAS Institute). Continuous variables are expressed using mean and SD, or median and IQR if not normally distributed, and categorical data using frequency and percentage. Differences between groups were compared using Fisher exact test or chi-square test with statistical significance assumed at P < 0.05. Annualized event rates and 5-year arrhythmic event rates were estimated using the Kaplan-Meier method. Event-free survivor curves comparing patients with different aneurysm sizes were constructed by the Kaplan-Meier method, and differences between groups were examined by use of a log-rank test for equality of survivor functions. The relationship between the aneurysm size and likelihood of life-threatening arrhythmic events was further evaluated with univariable and multivariable Cox proportional hazards model. The proportional hazards assumption was assessed using Schoenfeld residuals. Multivariable models were constructed using either predefined variables (AHA/ACC risk makers or ESC-SCD risk score) and/or via stepwise selection using entrance criteria of P < 0.20 and retained if P < 0.05. Data were missing for ≥1 variable in ESC-SCD risk score for 57 patients, most commonly nonsustained VT (n = 48, 9%) with data imputed as described within the ESC-risk score development model.¹⁷ The relationship between the prevalence of apical thrombus or embolic event was evaluated using univariable and multivariable logistic regression models. This study was reviewed and approved by Institutional Review Board of the participating institutions.

Results

Clinical profile

At study entry, the 510 patients with LV apical aneurysms were 59 ± 14 years of age including most, n = 267 (52%), ≥60 years old (Table 1). Aneurysms were identified at a median 5.7 years from initial HCM diagnosis (IQR: 0.4, 12.5), including 152 (30%) at ≥ 10 years. In 89% of patients, aneurysm was identified after the year 2010, whereas aneurysms were identified before the year 2010 in remaining 11% (Supplemental Table 1B). Patients were mostly asymptomatic (NYHA functional class I, n = 278 [55%]) or mildly symptomatic (NYHA functional class II, n = 162 [32%]) and 36% developed AF during clinical course.

Table 1.

Baseline Demographics and Outcomes of 510 HCM Patients With LV Apical Aneurysm According to Aneurysm Size

Variable	Entire Cohort (N = 510)	Very Large (≥40 mm) (n = 16, 3%)	Large (20-39 mm) (n = 197, 39%)	Medium (10-19 mm) (n = 198, 39%)	Small (<10 mm) (n = 99, 19%)
Age at aneurysm, y	59.1 ± 13.6	54 ± 17	59 ± 13	60 ± 13	59 ± 14
Male, n (%)	346 (68%)	10 (63%)	139 (71%)	131 (66%)	66 (67%)
Median time, HCM diagnosis to LV apical aneurysm identification	5.7 (0.4,12.5)	8.2 (2.6, 16.4)	6.2 (0.4, 14.9)	4.9 (0.3, 11.5)	4.9 (0.6, 10.9)
Aneurysm transverse dimension, mm	16 (10, 23)	45 ± 5	26 ± 5	15 ± 3	7 ± 2
Maximum LV wall thickness	18 ± 4	17 ± 4	19 ± 4	18 ± 4	17 ± 4
LV ejection fraction, %	60 (55, 68)	42 (35, 60)	60 (55, 65)	65 (60, 70)	61 (60, 68)
Left atrial transverse dimension, mm	42 ± 7	45 ± 9	43 ± 8	43 ± 7	42 ± 7
Rest outflow obstruction, (≥30 mm Hg)	43 (8%)	2 (13%)	9 (6%)	16 (8%)	14 (14%)
Intracavity gradient (≥30 mm Hg)	161 (32%)	7 (44%)	81 (41%)	58 (29%)	15 (15%)
Complete systolic emptyinga	304 (70%)	5 (42%)	121 (75%)	123 (73%)	55 (60%)
Contrast enhanced CMR	410 (80%)	11 (69%)	154 (78%)	159 (80%)	86 (86%)
LGE, extent, %	9 (4, 16)	11 (9, 24)	12 (5, 19)	8 (5, 13)	6 (3, 15)
NYHA functional class, at aneurysm detection
I	278 (55%)	7 (44%)	106 (54%)	111 (56%)	54 (54%)
II	162 (32%)	7 (44%)	64 (33%)	60 (30%)	31 (31%)
III/IV	67 (13%)	2 (13%)	24 (12%)	27 (14%)	14 (14%)
Family history HCM-sudden death	37 (7%)	1 (6%)	16 (8%)	11 (6%)	9 (9%)
Unexplained syncope	42 (8%)	4 (25%)	17 (9%)	16 (8%)	5 (5%)
Massive LV hypertrophy (≥30 mm)	24 (5%)	0	14 (7%)	8 (4%)	2 (2%)
Nonsustained VT (ambulatory monitoring)	160 (31%)	2 (13%)	88 (45%)	56 (28%)	14 (14%)
End-stage HCM (ejection fraction <50%)	53 (10%)	9 (56%)	31 (15%)	10 (5%)	3 (3%)
Extensive LGE (>15% LV mass)	114 (22%)	4 (25%)	56 (28%)	37 (19%)	17 (17%)
History of atrial fibrillation	183 (36%)	5 (31%)	73 (37%)	72 (36%)	33 (33%)
Myectomy/alcohol septal ablation prior to aneurysm detectiond	20 (4%)	1 (6%)	3 (1.5%)	10 (5%)	6 (6%)
Medications
Beta-blocker	356 (70%)	10 (63%)	141 (72%)	139 (70%)	66 (67%)
Calcium channel blockers	110 (22%)	3 (19%)	36 (18%)	44 (22%)	27 (27%)
Disopyramide	31 (6%)	2 (13%)	9 (5%)	12 (6%)	8 (8%)
Amiodarone	46 (9%)	5 (31%)	26 (13%)	11 (5.6%)	4 (4%)
Sotalol	11 (2%)	1 (6%)	6 (3%)	3 (1.3%)	1 (1%)
Warfarin	76 (15%)	9 (56%)	41 (21%)	15 (8%)	11 (11%)
Direct oral anticoagulant	212 (42%)	2 (13%)	93 (47%)	84 (42%)	33 (33%)
Myosin inhibitors	0	0	0	0	0
NYHA functional class last follow-up
I	270 (53%)	8 (50%)	96 (49%)	109 (55%)	57 (58%)
II	173 (34%)	6 (38%)	72 (37%)	64 (32%)	31 (31%)
III/IV	56 (11%)	2 (13%)	22 (11%)	21 (11%)	11 (11%)
Potentially life-threatening arrhythmic eventsb	82 (16%)	9 (56%)	47 (24%)	22 (11%)	4 (4%)
Appropriate ICD intervention for VT/VF	68 (14%)	8 (50%)	41 (21%)	16 (8%)	2 (2%)
Out of hospital cardiac arrestc	18 (4%)	2 (13%)	9 (5%)	5 (2.5%)	2 (2%)
Sudden death	4 (1%)	0	1 (0.5%)	3 (2%)	0
Thromboembolic event without atrial fibrillation	18 (4%)	1 (6%)	10 (5%)	5 (3%)	2 (2%)
LV thrombus without embolic event	50 (10%)	3 (19%)	32 (15%)	14 (7%)	1 (1%)
Either Embolic event or LV thrombus	68 (13%)	4 (25%)	42 (21%)	19 (10%)	3 (3%)
Death	32 (7%)	2 (13%)	15 (9%)	10 (6%)	5 (5%)
Age at death	73.1 ± 13	73 ± 17	72 ± 13	72 ± 14	79 ± 13
HCM related	11 (34%)	0	6 (40%)	4 (40%)	1 (20%)
Heart failure-death	5	0	3	1	1
Stroke death	1	0	1	0	0
Postoperative death	1	0	1	0	0
Sudden death	4	0	1	3	0
Non-HCM related	19 (59%)	2 (100%)	7 (47%)	6 (60%)	4 (80%)
Unknown	2 (6%)	0	2 (13%)	0	0

Value are n (%), mean ± SD, or median (IQR).

CMR = cardiac magnetic resonance imaging; HCM = hypertrophic cardiomyopathy; ICD = implantable cardioverter defibrillator; LGE = late gadolinium enhancement; LV = left ventricular; VT = ventricular tachycardia; VF = ventricular fibrillation.

^aAvailable in 434 patients; defined as complete systolic muscular opposition of the hypertrophied mid/distal LV invariably including hypertrophied papillary muscles.

^bLV apical aneurysm identified prior to event in 67.

^c8 patients with out of hospital cardiac arrest had subsequent appropriate ICD intervention for VT/VF a median of 1.3 (0.4, 2.4) years after.

^dMyectomy via subaortic approach in 19; Alcohol septal ablation in 1.

Transverse aneurysm dimension ranged from 3 mm to 56 mm (mean of 19 ± 9), including small in 99 (19%; <10 mm), medium in 198 (39%; 10-19 mm), large in 197 (39%; 20-39 mm), and very large in 16 (3%; ≥40 mm).

Maximum LV wall thickness was 18 ± 4 mm, including 24 (4%) patients with massive LV hypertrophy (thickness ≥30 mm). The median global LV ejection fraction was 60% (IQR: 55, 68). LV obstruction (resting or Valsalva gradients ≥30 mm Hg) occurred in 2 mechanisms: 1) due to typical mitral valve septal contact in 43 (8%; peak gradient of 65 ± 23 mm Hg); and 2) intracavity gradient from muscular apposition of the ventricular septum and LV free wall in the mid or apical LV, invariably including hypertrophied papillary muscles in 161 (32%; peak gradient of 47 ± 22 mm Hg); 14 patients had both mechanisms of obstruction.¹⁹ Notably, 304 (70%) of 434 patients had complete systolic muscular opposition of the hypertrophied mid/distal LV, consistent with mid-ventricular or apical muscular obstruction,19, 20, 21 although only 120 (39%) of these had an intracavity gradient of ≥30 mm Hg measured with continuous wave Doppler. In 53 other patients, there was global systolic dysfunction (ejection fraction <50%) without muscular obstruction.

Serial observations of aneurysm size

Paired imaging studies were assessed in 261 patients, with a median interval of 4.3 years (IQR: 2.5, 5.6 years), including 57 with paired CMRs and 204 with paired echocardiograms. Transverse dimension was 17 ± 9 mm initially and 20 ± 10 mm at follow-up (P < 0.001) increasing by a median of 2.0 mm (IQR: 0, 5.7 mm), 0.6 mm/year. In 71 patients (27%), aneurysm size increase would result in size reclassification: 28 from small to medium, 37 medium to large, and 6 large to very large. Over follow-up, each aneurysm remained intact without evidence of rupture.

Potentially life-threatening arrhythmic events

Of the 510 patients, 82 (16%) patients had ventricular tachyarrhythmic events, including 60 with appropriate ICD intervention for VT/VF, 4 sudden deaths, and 18 resuscitated cardiac arrests. The aneurysm had been detected 4.0 ± 3.0 years before event in 67 patients; annualized arrhythmic event rate was 2.1%/year and 12.4% ± 1.9% over 5-years.

Age at arrhythmic event was 54 ± 14, including 42 (51%) ≥60 years old, with similar proportion in patients with aneurysms detected at 60 to 74 years (16.4%) vs <60 years (18.9%, P = 0.49), albeit with significantly lower proportion at ≥75 years (1.9%, P = 0.005) (Supplemental Figure 1).

Multiple episodes of VT/VF (5.9 ± 9.0; range, 2 to ≥40) occurred in 41 of the aneurysm patients, including 8 with ICD interventions after resuscitated cardiac arrest, 29 with multiple discrete ICD interventions, and 4 with VT storms. Notably, in 25 of these 41 patients, recurrent VT episodes led to radiofrequency ablation of the arrhythmia focus in the area of scar contiguous with the aneurysm rim; 21 had a single ablation procedure while 4 required multiple procedures (up to 3). After the most recent ablation, 18 of the 25 patients (72%) have experienced no further recurrence of VT over a median of 2.4 years (IQR: 0.4, 4.7).

In the 67 patients with apical aneurysm detected before a ventricular tachyarrhythmic event, the aneurysm was the only major AHA/ACC risk marker (2a) in 44 (66%) whereas the other 23 patients (34%) had other risk markers: unexplained syncope (n = 10), family history of HCM-SCD (n = 10), massive LV hypertrophy (n = 5), and/or LV systolic dysfunction (n = 6).

When the ESC-SCD risk score was applied to these 67 patients, only 7 (10%) were classified as high risk sufficient to recommend an ICD (≥6%/5-years), whereas 51 (77%) were considered to be at the lowest risk (<4%/5-years), inconsistent with an ICD recommendation (Supplemental Figure 2). The remaining 9 (13%) were considered intermediate risk (≥4 and <6%/5-years).

Aneurysm size and arrhythmic events

In the 67 patients with arrhythmic events after aneurysm detection, risk of potentially life-threatening ventricular arrhythmic events increased linearly with respect to aneurysm size (HR: 1.06 per 1-mm increase; 95% CI: 1.03-1.08; P < 0.001), with event rates: 0.2%/year with small aneurysms, 2.3%/year with medium, 3.0%/year with large, and 8.2%/year with very large (Figure 1).

Figure 1.

Life-Threatening Ventricular Tachyarrhythmias in Hypertrophic CardiomyopathyPatients With Apical Aneurysm

(A) Kaplan Meier analysis showing survival free from arrhythmic events in 510 HCM patients with apical aneurysms according to aneurysm size demonstrating incremental increase in risk with increasing size (P < 0.001). (B) Event rates (%/year) for life-threatening ventricular arrhythmias by aneurysm size. Life-threatening ventricular tachyarrhythmias include: appropriate ICD interventions for VT/VF, resuscitated cardiac arrest, or sudden death.

On multivariable analysis, aneurysm size was independently associated with ventricular arrhythmic events, independent of nonsustained VT on ambulatory monitoring or other AHA/ACC risk markers (HR: 1.06 per 1-mm; 95% CI: 1.04-1.10; P < 0.0001) (Table 2), and also independent of ESC-SCD risk scores (HR: 1.05 per 1-mm; 95% CI: 1.03-1.08; P < 0.001). Aneurysm size ≥10 mm was associated with increased risk for life-threatening ventricular arrhythmic events in the absence of other AHA/ACC risk markers and low ESC-SCD risk scores (Table 3).

Table 2.

Univariableand Multivariable Analysis for Predictors of Potentially Life-Threatening Arrhythmic Events (SCD, Cardiac Arrest, or Appropriate ICD Discharges for VT/VF) in HCM Patients With LV Apical Aneurysms

	Univariable Analysis		Multivariable Analysis
	HR (95% CI)	P Value	HR (95% CI)	P Value
Univariable and multivariable analysis including LV apical aneurysm size with age and AHA/ACC risk markers
Aneurysm size, mm	1.06 (1.03-1.08)	<0.001	1.06 (1.03-1.08)	<0.001
Age, years	1.00 (0.98-1.01)	0.59
Family history sudden death from HCM	2.59 (1.31-5.10)	0.006	2.28 (1.12-4.62)	0.02
Unexplained syncope	2.97 (1.56-6.06)	0.003	2.31 (1.12-4.77)	0.03
LV wall thickness ≥30 mm	1.20 (0.47-3.09)	0.71
LV ejection fraction <50%	1.72 (0.91-3.24)	0.09
NSVT on ambulatory monitor	1.74 (1.03-2.92)	0.04	1.78 (1.07-2.95)	0.03
Extensive late gadolinium enhancement (≥15% of LV mass)	1.35 (0.78-2.34)	0.29

	Bivariate Analysis
	HR (95% CI)	P Value
Bivariate analysis evaluating LV aneurysm size and ESC risk score
Aneurysm size, mm	1.05 (1.03-1.08)	<0.001
ESC risk score (per 1% increase)	1.12 (1.06-1.31)	0.003

AHA/ACC = American Heart Association/American College of Cardiology; ESC = European Society of Cardiology; NSVT = nonsustained ventricular tachycardia; SCD = sudden cardiac death; other abbreviations as in Table 1.

Table 3.

Estimated 5-Year Risk for Life-Threatening Arrhythmic Events in the Absence of Other AHA/ACC Risk Markers or With Otherwise Low ESC-SCD Risk Scores (<4%/5-Years)

Aneurysm Transverse Dimension	Estimated 5-Year Risk for Life-Threatening Arrhythmic Eventsa, %/5-ys (95% CI)	95% CI
Absence of other AHA/ACC risk markersb
<10 mm	3.8%	0.2-7.3
10 mm	5.6%	1.5-9.5
15 mm	6.7%	2.5-10.7
20 mm	8.0%	3.4-12.4
25 mm	9.6%	4.0-14.9
30 mm	11.5%	4.0-18.4
35 mm	13.7%	3.1-23.1
40 mm	16.3%	1.3-29.0
Low ESC-SCD scorec
<10 mm	3.75%	1.3-6.1
10 mm	6.04%	3.1-8.9
15 mm	7.84%	4.6-11.0
20 mm	10.15%	6.5-13.6
25 mm	13.09%	8.7-17.3
30 mm	16.81%	11.0-22.3
35 mm	21.44%	13.1-28.9
40 mm	27.12%	15.0-37.5

Abbreviations as in Table 1 and 2.

^aIncludes appropriate ICD interventions for VT/VF, resuscitated cardiac arrest, or sudden death.

^bIncludes other 2a and 2b AHA/ACC risk markers besides apical aneurysms: family history of HCM related SCD in ≥1 relatives; massive LV hypertrophy ≥30 mm; unexplained syncope; LV systolic dysfunction (EF <50%); nonsustained ventricular tachycardia (NSVT) episodes on ambulatory monitoring; or extensive late gadolinium enhancement (LGE).

^cESC-SCD risk score of <4%/5-years.

Incidence of embolic events and prevalence of apical thrombus

Embolic events (n = 27) occurred following aneurysm detection in 18 patients without a history of AF (3.5%, rate 0.86%/year), and in 9 patients with AF history. Of the 18 patients without AF, 10 had an LV apical thrombus identified at time of embolic event (56%): 15 stroke, 2 renal infarct, and 1 coronary thrombus. Each of these 18 patients was not on prophylactic anticoagulation at the time of event. There was no difference in embolic event rates between patients with small (<10 mm) and larger aneurysms (≥10 mm; P = 0.94) (Figure 2).

Figure 2.

Thromboembolic Events and Apical Thrombus in Patients With LV Apical Aneurysms

(A) Kaplan Meier survival analysis showing no difference in rates of thromboembolism between small (<10 mm) and larger aneurysms (≥10 mm) (P = 0.94). (B) Prevalence of LV thrombus and/or embolic events increasing according to aneurysm size. Imbedded figure demonstrates an apical aneurysm with transmural scarring (arrow-heads) and large apical thrombus (arrow). LV = left ventricle.

In 50 other patients (10%), LV thrombus was identified in the apical aneurysm, unassociated with embolic events, 25 at initial detection of the aneurysm (50%), and 25 at a median of 5.7 years after aneurysm identification (IQR: 2.2, 8.3). A total of 47 were not on anticoagulation at the time the thrombus was recognized. The prevalence of LV thrombus increased with size (OR: 1.97 per 10 mm; 95% CI: 1.48-2.59; P < 0.0001), 1% in patients with small, 7% with medium, 15% with large, and 19% in patients with very large aneurysms (P < 0.0001).

Aneurysm size and occurrence of apical thrombus/embolic event

Overall, either recognition of an apical thrombus or a thromboembolic event occurred in 68 patients (25%), increasing with respect to aneurysm size (Figure 2) (P < 0.001): infrequent in patients with small (3%) and more frequent in patients with medium (10%), large (21%), and very large aneurysms (25%) These emboli or LV thrombi occurred in just 2% of patients on prophylactic anticoagulation (predominantly with a direct oral anticoagulant) (Table 1) whether for AF or for the aneurysm per se, as compared to 25% of patients not on prophylactic anticoagulation (OR: 0.14; 95% CI: 0.07-0.28; P < 0.001). On multivariate analysis, aneurysm size was associated with prevalence of apical thrombus or embolic events (OR: 1.05 per 1-mm; 95% CI: 1.02-1.08; P = 0.003), after adjustment for AF and LV ejection fraction (Table 4); anticoagulation was associated with lower prevalence of apical thrombus or embolic events (OR: 0.09; 95% CI: 0.04-0.20).

Table 4.

Univariable and Multivariable Analyses for Occurrence of Either an Embolic Events or Apical Thrombus in HCM Patients With LV Apical Aneurysms During Clinical Course

	Univariable Analysis		Multivariate Analysis
	OR (95% CI)	P Value	OR (95% CI)	P Value
Aneurysm size, mm	1.06 (1.03-1.09)	<0.001	1.05 (1.02-1.08)	0.003
age	1.00 (0.98-1.02)	0.76
Male	1.25 (0.71-2.20)	0.44
History of atrial fibrillation	0.85 (0.50-1.45)	0.56	2.22 (1.11-4.43)	0.02
LA diameter, mm	1.03 (0.99-1.07)	0.07
Maximal wall thickness	0.99 (0.94-1.06)	0.85
Prophylactic anticoagulation	0.14 (0.07-0.28)	<0.0001	0.09 (0.04-0.20)	<0.0001
LV ejection fraction, per 10% decrease	1.42 (1.22-1.67)	<0.0001	1.31 (1.07-1.55)	0.006

LA = left atrium; other abbreviations as in Table 1.

Heart failure

Medication refractory NYHA functional class III/IV symptoms developed in 86 patients (17%), with similar incidence across aneurysm size (range 18%-25%) including 36% with systolic dysfunction (ejection fraction <50%), 30% with subaortic obstruction due to typical mitral valve contact, and 11% with midventricular or distal muscular obstruction due to complete systolic emptying.19, 20, 21 Nine patients underwent apical aneurysm resection and apical myectomy to relieve obstruction resulting in improvement in symptoms in 7.

Mortality

Over a median of 4.1 years of follow-up (IQR: 1.5, 7.1 years), 478 of the 510 study patients (94%) have survived, and 32 (6%) died. All-cause (total) mortality was 1.4%/year, with rates similar across aneurysm size (P = 0.91). In 11 patients (2%; rate 0.5%/year), death was attributable to HCM at 68 ± 14 years of age (range 39-86): 4 sudden deaths, 5 of advanced heart failure, 1 from stroke, and 1 after transaortic myectomy. There was no difference in HCM-related mortality rates in patients with small (<10 mm) vs larger aneurysms (≥10 mm; P = 0.61). In 19 other patients, death was unrelated to HCM (4%, rate: 0.9%/year) at 74 ± 13 years (range 43-98): most commonly cancer (n = 7), pulmonary disease (n = 4), sepsis (n = 2), and 2 unknown at 71 and 83 years.

Discussion

In this largest and most diverse multicenter international data set of >500 consecutive HCM patients with LV apical aneurysms of various sizes, representing 2.7% of the overall HCM cohort, we present evidence to support the principle that apical aneurysms are associated with an increased risk of adverse disease-related events (Central Illustration). This assertion is supported by the proportion of the study group (16%) that incurred 1 or more life-threatening ventricular tachyarrhythmic events (rate >2%/year) terminated reliably by primary prevention ICDs. Patients’ high-risk status was identified by the aneurysm alone (in 66%) or in combination with other major risk markers (34%), and included a subset with recurrent ventricular arrhythmias terminated by ICDs and/or controlled with radiofrequency catheter ablation.22, 23, 24 Indeed, this high rate of life-threatening ventricular tachyarrhythmic events is comparable in frequency to other major AHA/ACC risk markers in HCM, is above the threshold in which ICD is typically recommended for primary prevention in HCM (∼1%/year), and is nearly 4-fold higher than the rate of these events (0.5%/year) observed in the general HCM population.^14,15,17,25

Central Illustration.

Management of HCM Patients With Apical Aneurysms Based on Size

In 510 patients followed for a median of 4.1 years (IQR: 1.5, 7.1), those with medium or large aneurysms (≥10 mm in transverse dimension) were at an increased risk for life-threatening arrhythmic events with risk increasing incrementally according to size, supporting the AHA/ACC recommendation for primary prevention ICDs. In addition, prevalence of LV thrombus and/or embolic events increased in relation to size raising selective decision-making for prophylactic anticoagulation (typically in larger aneurysms ≥20 mm). In contrast, patients with small apical aneurysms were at substantially lower risk for life-threatening arrhythmic events (for aneurysms <10 mm) and apical thrombi/stroke (for aneurysms <20 mm), albeit with need for longitudinal surveillance for increase in size. Symbols: ∗Includes appropriate ICD interventions for VT/VF, resuscitated SCD, or SCD. †Selectively considered for aneurysms ≥20 mm, with aneurysms <20 mm at lower risk for LV thrombus or embolic events. ‡: Typically performed annually for LV thrombus formation or increase in size. HCM = hypertrophic cardiomyopathy; ICD = implantable cardioverter defibrillator; LV = left ventricle.

Notably, LV apical aneurysms have been recognized as a major independent risk marker for HCM patients in AHA/ACC guidelines since 2020,²⁶ but are excluded from the most recent 2023 ESC cardiomyopathy guidelines, generating controversy regarding management considerations.¹⁴ It has been assumed by some investigators that SCD risk in patients with HCM and apical aneurysms is primarily attributable to other concurrent risk markers (particularly nonsustained VT).¹³ However, our data, as well as that of other investigators1, 2, 3 demonstrate that apical aneurysms are associated with an increased risk of potentially life-threatening arrhythmic events, even after adjusting for other risk markers, including nonsustained VT and/or the ESC-SCD risk scores. Indeed, >75% of our HCM-aneurysm patients who experienced life-threatening arrhythmic events had low ESC-SCD risk scores (<4%/5-years), inconsistent with recommendation for primary prevention implants, whereas only 10% has sufficiently high ESC-SCD risk scores (≥6%/5-years) to justify an ICD recommendation. To this point, given the large cohort of patients, we were able to quantify 5-year risk for arrhythmic events based on the aneurysm size in the absence of AHA/ACC risk markers and/or with low-risk ESC-SCD risk scores <4%/5-years (Table 3). This aids in understanding the magnitude of individualized risk for arrhythmic events to assist in personalized ICD decision-making in patients with aneurysm as sole risk marker.¹⁵

Patient age at the first life-threatening arrhythmic event in patients with aneurysms is notable for occurrence at much more advanced ages than many other high risk HCM patients (59 years vs ∼45 years).^15,27 This difference in patient age is likely attributable to the progressive remodeling and structural evolution that is presumably ongoing throughout life and at more advanced ages.28, 29, 30 Indeed, in the present cohort, most events occurred after 60 years of age, with similar event rates in patients 60 to 74 years compared to <60 years, albeit with substantially lower risk in patients ≥75-years. Therefore, assessment of risk specifically in LV apical aneurysms deviates from a paradigm previously advanced for the overall HCM population in which SCD events are particularly uncommon for individuals >60-years.²⁷ Consequently, although primary prevention ICD implants are generally not recommended in HCM patients >60 years when associated with major risk markers,¹⁵ our present data support LV apical aneurysms to be an exception for patients 60-74 years of age.

Aneurysms vary widely in size, with controversy as to the level of risk for patients with the morphologically smallest of these abnormalities.1, 2, 3 In this regard, there has been conflicting findings throughout the HCM literature as to if risk is similar across spectrum of aneurysm size, with some studies suggesting aneurysms ≥20 mm associated with an increased risk^3,5 and others finding no difference in risk for arrhythmic events by size.¹ Therefore, a major impetus of the present analysis was to clarify the clinical significance attached to aneurysm size. Indeed, our large data set identified a significant linear relationship between maximal transverse dimension of the aneurysm, establishing that the greatest risk was associated with aneurysms ≥10 mm (>2.0%/year), which was >10-fold higher than the negligible risk associated with the smallest aneurysms <10 mm (ie, 0.2%/year).²⁵ Consequently, small LV apical aneurysms as a sole risk marker due not appear to rise to the risk threshold in HCM that would routinely justify primary prevention ICDs.^14,15 Notably, with serial imaging, we also show the potential for some LV apical aneurysms to increase in size, underscoring the importance of serial surveillance in this population with imaging.29, 30, 31

There has been uncertainty regarding the risk of apical thrombus formation and embolic events in HCM patients with apical aneurysms also resulting from conflicting data.1, 2, 3^,5 To examine this relationship, and given the potential link to an LV apical thrombus formation with a subsequent embolic event, we opted to examine a combined end-point of either an embolic event or the development of apical thrombus. We observed that about 15% of patients in this large data set had either an embolic event or were found to have an apical thrombus, driven by the discovery of an apical thrombus (in ∼10% of patients). In contrast, the overall rate for an embolic event was particularly low in the absence of documented AF (3.5% of patients; 0.8%/year; and could potentially be even lower given our inability to completely exclude occult AF as a cause for these events). The explanation for these discrepant findings is unclear. The low number of embolic events could be related to the development of layered organized thrombus in the distal aneurysm over prolonged periods of time and pose lower risk for thromboembolism.³² On the other hand, identification of thrombi is unlikely completely benign as over 50% of patients with embolic events did in fact have an apical thrombus identified during the resulting hospitalization. Therefore, the low rates of thromboembolism may reflect judicious anticoagulation after the identification of an LV thrombus for secondary prevention and/or the utilization in some centers for prophylactic anticoagulation use (based solely on presence of the LV apical aneurysm). Indeed, occurrence of our combined endpoint was substantially lower in patients on prophylactic anticoagulation (OR: 0.09).

The linear relationship between aneurysm size and prevalence of an apical thrombus formation or embolic event allow for individualized decision-making regarding select consideration for prophylactic anticoagulation. Indeed, the graded prevalence levels allows for patient-centered shared decisions regarding anticoagulation treatment weighing the risk of major bleeding (rates up to 2% to 4%/year) compared to the combined risk conveyed by thrombus formation or embolism.³³ The risk for events in smaller aneurysms (<10 mm) was particularly low (<5% of patients), such that the benefit of prophylactic anticoagulation would not appear to justify the bleeding risks associated with lifelong therapy.34, 35, 36 Alternatively, it may be reasonable to consider prophylactic anticoagulation on a case-by-case basis for larger aneurysms ≥20 given the substantially higher prevalence of thrombus formation in these patients. Ultimate decisions would appear to be best guided by shared decision-making in the absence of randomized trials to demonstrate a clear benefit to this strategy. If decision is to ultimately forgo prophylactic anticoagulation, continued imaging surveillance for clot formation within the aneurysm would appear prudent.

The overall role and significance of muscular and mid-cavity obstruction in the cause of apical aneurysms has recently been investigated and in virtually all cases mid or apical LV obstruction is an associated abnormality^6,19,20 with the putative mechanism being supply-demand ischemia and concomitant small vessel narrowing. In our cohort, muscular mid-cavity obstruction does not appear to be a strong marker for symptom development as 70% of patients had this phenotype, yet only about 10% of these patients developed drug refractory heart failure. Notably, a minority of these patients underwent apical myectomy (and aneurysm resection) resulting in normalization of distal LV systolic pressures and improvement in symptoms.^2,37,38 Ongoing phase 3 trials of myosin inhibitors in this subset of patients will inform if these novel drugs are a potential effective alternative treatment for such patients.³⁹ In contrast, adverse LV remodeling from extensive LV replacement fibrosis resulting in systolic dysfunction was not uncommon (10% of patients) with over one-third of these patients developing advanced symptoms, raising priority to identify therapies to prevent or reverse this form of adverse disease progression.

LV apical aneurysms have proved to be a highly treatable subset within the broad HCM clinical spectrum. With contemporary treatments including ICD and anticoagulation, both all-cause and HCM-related mortality rates were low for these patients, that is, HCM-mortality rates of 0.5%/year, all-cause of 1.4%/year (predominantly of non-HCM related causes), and ∼95% of patients surviving to end of follow-up, similar to that observed in other contemporarily treated HCM populations.^15,25

Study Limitations

The prevalence of apical aneurysms of 2.7% in this study cohort is likely an underestimate given not all patients evaluated were imaged with contrast echocardiogram or CMR.³¹ Similarly, there was a skewed distribution in the year of apical aneurysm detection, with almost 90% of aneurysms detected after 2010, which may reflect increased utilization of these imaging modalities.³¹ In addition, increased recognition of this novel phenotype as part of the HCM spectrum probably resulted in an increased referral of patients with aneurysms to HCM centers. Notably, echocardiographic and CMR analysis was performed by experienced cardiac imagers at each HCM center and not adjudicated within a core lab, potentially resulting in misclassification of some aneurysms, albeit significant measurement discrepancies between modalities occurred less frequently in aneurysms of smaller size. Because apical thrombi were detected at time of aneurysm identification in a large proportion of patients, the annual incidence of thrombus formation could not be reliably quantified, and resulted in our description of prevalence of these events.

Conclusions

LV apical aneurysms constitute a major and not uncommon risk marker for life-threatening ventricular arrhythmias in HCM. The greater aneurysm size itself modulates risk for these events, with aneurysms of medium and larger size (≥10 mm) at higher risk for arrhythmic events and justifying the AHA/ACC HCM guidelines consideration of aneurysms as a major risk marker independent of age.¹⁵ In contrast, our data identify that particularly small aneurysms (<10 mm in transverse dimension) convey low risk and below that generally considered for primary prevention ICDs in HCM, albeit warranting long-term surveillance for possible increase in size. A linear relationship was observed between aneurysm size and risk for apical thrombus providing the opportunity to guide individualized patient decision-making for prophylactic anticoagulation. This large multicenter data set informs treatment decisions for HCM patients with apical aneurysms and demonstrates the potential for low HCM-related mortality and extended survival for these high-risk patients with contemporary treatment.

Perspectives.

COMPETENCY IN MEDICAL KNOWLEDGE: Patients with HCM and apical aneurysms are at an increased risk for potentially life-threatening ventricular arrhythmic events, with increasing risk by aneurysm size, and justifying consideration for primary prevention ICDs for patients with aneurysms ≥10 mm. The prevalence of apical thrombi also increased by aneurysm size raising consideration for prophylactic anticoagulation in select patients with larger aneurysms.

TRANSLATIONAL OUTLOOK: Future studies should focus on identifying additional risk modifiers in this unique subset of patients to improve specificity for ICD implantation. Additional studies are also needed to clarify degree that the akinetic aneurysm plays on risk for embolic events in the absence of AF.

Funding support and author disclosures

The authors acknowledge the generous support of Lucille Protas, who donated personal funds to this project. Drs Carrick and Madrazo received funding from the Talles Family HCM Research Foundation. Dr Rowin has served as a consultant for iRhythm and Cytokinetics; Dr Sherrid has received consultant fees from Cytokinetics, Edgewise Therapeutics, and Pfizer; Dr Tower-Rader; and research funds from Bristol Myers Squibb, Cytokinetics, and Edgewise Therapeutics. Dr Adler was funded by the Tiffin Trust and an unrestricted research grant from Bristol Myers Squibb Canada; and has served as a consultant to Bristol Myers Squibb. Dr Fifer has received consulting fees from Bristol Myers Squibb, Cytokinetics, Edgewise Therapeutics, and Viz.ai. Dr Olivotto reports research grants for Respect, StratifyHF, and SmashHCM; grant support from Bayer, MyoKardia (a wholly owned subsidiary of BristolMyers Squibb), Sanofi Genzyme, and Shire (now part of Takeda); personal fees from Bayer, Sanofi Genzyme, and Shire; and consultant fees from MyoKardia. Dr Massera has received consultant fees from Chiesi Pharmaceuticals, Cytokinetics, Rocket Pharmaceuticals, and Sanofi. Dr Maron has received consultant/advisor fees from Imbria, Edgewise, and BioMarin; and Steering Committee fees for SEQUOIA-HCM from Cytokinetics. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.