ICD indication in hypertrophic cardiomyopathy: which algorithm to use?

SUMMARY

OBJECTIVE:

This study aimed to evaluate the agreement in the indication of implantable cardioverter-defibrillators in patients with Hypertrophic cardiomyopathy, as per the 2014 European Society of Cardiology and 2020 American Heart Association recommendations, and evaluate fragmented QRS as a predictor of cardiovascular outcome.

METHODS:

Retrospective cohort with 81 patients was evaluated between 2019 and 2021. Patients with hypertrophic cardiomyopathy ≥16 years old were included. Exclusion criteria include secondary myocardiopathy and follow-up <1 year. Kappa coefficient was used to determine the agreement. Survival and incidence curves were determined by Kaplan-Meier method. A p<0.05 was considered significant.

RESULTS:

The fragmented QRS was identified in 44.4% of patients. There were no differences between patients with and without fragmented QRS regarding clinical parameters, echocardiography, fibrosis, and sudden cardiac death risk. During follow-up of 4.8±3.4 years, there was no sudden cardiac death, but 20.6% patients with implantable cardioverter-defibrillator had at least one appropriate shock. Three of the seven appropriate shocks occurred in European Society of Cardiology low- to moderate-risk patients. Three shocks occurred in moderate-risk patients and four in American Heart Association high-risk patients. Overall recommendations agreement was 64% with a kappa of 0.270 (p=0.007). C-statistic showed no differences regarding the incidence of appropriate shock (p=0.644).

CONCLUSION:

sudden cardiac death risk stratification algorithms present discrepancies in implantable cardioverter-defibrillator indication, both with low accuracy.

KEYWORDS:
Sudden cardiac death; Hypertrophic cardiomyopathy; Implantable cardioverter-defibrillator; Cardiac arrhythmia

INTRODUCTION

Hypertrophic cardiomyopathy (HCM) is the most common genetic heart disease, with an estimated prevalence of 1–167 individuals¹. It is recognized as the main cause of sudden death (SD) in young people². It is commonly asymptomatic. When present, the main symptoms are chest pain, dyspnea, palpitation, and syncope³.

Through risk stratification strategies and prophylactic implantable cardioverter-defibrillator (ICD) indication, the mortality of patients with HCM has been reducing from 6 to <1% per year¹,⁴. However, recommendations for implantation of ICD are divergent and tend to over- or underestimate the real risk of SD, increasing the risk of unnecessary intervention in low-risk patients or nonindication in high-risk patients¹,⁴,⁵.

Traditionally, the main risk factors for MS are age, report of syncope, family history of multiple sclerosis (MS), evidence of ventricular arrhythmia, and left ventricular hypertrophy (LVH) ≥30 mm⁴,⁶. Although widely validated, these parameters have low accuracy in predicting MS in low- and medium-risk patients, which correspond to the majority of patients with HCM⁷.

Some studies have shown that fragmented QRS (fQRS) on electrocardiogram (ECG) correlates with myocardial fibrosis and represents a potential precursor of heart failure (HF) and arrhythmic events⁷. Despite this, the relevance of the fQRS in HCM is limited and its role in the prediction of SD is controversial⁸.

The aim of this study was to evaluate the agreement in the indication of ICD as primary prophylaxis of SD in HCM patients, according to the 2014 European Society of Cardiology (ESC) and 2020 American Heart Association (AHA) recommendations, and to evaluate the fQRS as a predictor of cardiovascular outcome.

METHODS

Type of study and population

This is a retrospective cohort study carried out in a university cardiology outpatient clinic specialized in HCM.

Inclusion criteria were diagnosis of HCM and age ≥16 years, while exclusion criteria were indication of ICD as secondary prophylaxis, follow-up time <1 year, and incomplete medical records.

The clinical variables collected were age, sex, clinical data (family history of SD, symptoms, ICD implantation), data from complementary examinations (ECG, transthoracic echocardiogram [ECOTT], cardiac magnetic resonance [CMR], 24 h Holter), clinical outcomes, and follow-up time.

The diagnosis of HCM was defined as LVH ≥15 mm in the largest segment (or ≥13 mm in those with a family history of HCM) in the absence of cardiac or systemic diseases that would justify ventricular overload⁴.

The definition of SD was any sudden-witnessed death with or without documented ventricular fibrillation (VF), death within 1 h of the onset of new symptoms, or nocturnal deaths without prior history of worsening symptoms⁹. Time of follow-up was determined by the difference in years between the initial assessment and the last visit or outcome. The functional class was determined by the New York Heart Association (NYHA).

In the case of shock administration by the ICD, the electrograms recorded by the device were retrieved and analyzed. Shocks were considered appropriate in the event of sustained VT and VF¹⁰.

The primary end point was the composite of SD or equivalent SD (SDE), namely, aborted MS and/or appropriate ICD shock. The secondary end point, acronym SEHS, was composed of SDE, hospitalization for decompensated HF, and fatal or nonfatal stroke.

The techniques used to perform the ECOTT and CMR were described previously¹¹.

Electrocardiographic analysis

The duration of QRS complex was manually determined in long lead II. In patients with narrow QRS (<120 ms), fQRS was defined as the presence of an additional R wave or notch in the R or S wave; in the case of wide QRS (≥120 ms), 2 notches or higher of R or S were considered¹².

Risk stratification for cdi implantation

The probability of MS in 5 years was calculated using the mathematical model validated by the ESC⁴.

Each patient had its indication for ICD determined according to the recommendations of each guideline, grouped according to the level of clinical evidence:

• ESC 2014⁴ – using estimated risk (ER) of SD in 5 years

  • ○

    Class IIa – ER ≥6%

  • ○

    Class IIb – ER <6 and ≥4%

  • ○

    Class III – ER <4%

• AHA 20201

  • ○

    Class IIa – at least one of the following: family history of MS, maximal LVH ≥30 mm, unexplained syncope; apical aneurysm; left ventricular ejection fraction (LVEF) ≤50%

  • ○

    Class IIb – nonsustained ventricular tachycardia (NSVT) or myocardial fibrosis on CMR ≥15%

  • ○

    Class III – absence of the aforementioned factors

Statistical analysis

Continuous variables were expressed as mean±standard deviation or as median and range of 25th and 75th quartiles, as appropriate, and categorical variables as absolute and proportional values. The Shapiro-Wilk test was used to determine normality.

Unpaired Student's t-test or Mann-Whitney U test was used in the analysis of continuous variables, while the χ² test or Fisher's exact test was used for categorical variables.

Survival curves, together with the p-value of the log-rank test, were calculated using the Kaplan-Meier method. The analysis was adjusted for age, sex, family history of MS, NYHA (I–II vs. III–IV), maximal LVH, gradient in left ventricular outflow tract (LVOT), syncope, and the presence of VT or NSVT on 24 h Holter. Receiver operating characteristics curve (ROC), area under the curve (AUC), and Harrell C-statistics analyses were used to assess the accuracy of the guidelines for predicting appropriate ICD shock. The linearly weighted kappa coefficient was calculated to determine the degree of agreement between the recommendations of the 2014 ESC and 2020 AHA 2020. For all the analyses, p<0.05 was considered significant.

Ethical aspects

The study was submitted and approved by the Research Ethics Committee. All Brazilian legal norms and Helsinki Declaration principles were observed. Since it was a retrospective study, the collection of the informed consent form was waived.

RESULTS

From March 2019 to February 2021, 96 patients with HCM were identified, of which 15 were excluded from the study due to not meeting the established criteria; therefore, only 81 patients were included in the study.

The fQRS was diagnosed in 36 (44.4%) patients. There was no statistically significant difference between patients with and without fQRS regarding clinical, echocardiographic, fibrosis, and estimated risk of MS (Table 1).

Table 1
Basic characteristics of the study population.

		fragmented QRS (n=36)	No fragmented QRS (n=45)	p-value
Age, years		42.8±15.6	48.1±15.8	0.137a
Gender male		17 (47.2)	23 (51.1)	0.728b
SD family history		21 (58.3)	23 (51.1)	0.564b
NYHA III/IV		3 (8.3)	2 (4.4)	0.470c
Syncope history		13 (36.1)	19 (42.2)	0.168b
ICD implant		14 (38.9%)	20 (44.5)	0.615b
TTE
	LVEF (%)	67 (62–72)	69 (63–74)	0.284d
	Max LVH (mm)	20 (16.5–27.6)	20 (17–24)	0.668d
	LA (mm)	39 (34–46)	39 (38–44)	0.647d
	LVOT (mmHg)	0 (0–31)	0 (0–30)	0.668d
	Myocardial fibrosis (%)χ	3.5 (2.3–7.5)	3.5 (1.7–13.8)	0.542d
ECG
	AFib	3 (8.3)	6 (13.3)	0.724c
	LVH	18 (50.0)	17 (37.8)	0.399d
	LAO	10 (27.8)	9 (20.0)	0.515d
	Stroke/TIA	2 (5.6)	6 (13.3)	0.284c
	SD risk in 5 years (%)	4.6 (2.7–7.3)	3.8 (2.1–6.1)	0.541d

• Values expressed as n (%), mean±standard deviation, or median (p25–75%). SD: sudden death; NYHA: New York Heart Association functional class; ICD: implantable cardioverter-defibrillator; TTE: transthoracic echocardiogram; LVEF: left ventricular ejection fraction; Max LVH: maximal left ventricular hypertrophy; LA: left atrium; LVOT: left ventricular outflow tract; AF: atrial fibrillation; LVH: left ventricular hypertrophy; LAO: left atrial overload; Stroke/TIA: stroke or transient ischemic stroke.
• ^χ
  Only 55 (67.9%) patients underwent CMR;
• ^a
  Student's t-test;
• ^b
  χ² test;
• ^c
  Fisher's exact test;
• ^d
  Mann-Whitney U test.

During a mean follow-up of 4.8±3.4 years, no SD occurred, but 7 (20.6%) of 34 patients with ICD had at least one appropriate shock, 4 (4.9%) hospitalizations for decompensated HF, and 6 (7.4%) nonfatal cerebrovascular events. Three of the seven appropriate shocks occurred in patients considered to be at low or moderate risk by the 2014 ESC guidelines. In the case of the 2020 AHA guidelines, three of the appropriate shocks occurred in patients at moderate risk and four shocks in patients at high risk. The incidence of SDE was 10.2% and that of SEHS was 21.6%.

The agreement between the indications for ICD implantation according to the 2014 ESC and 2020 AHA guidelines was 64% (kappa 0.270; p=0.007) (Table 2).

Table 2
Implantable cardioverter-defibrillator indication agreement according to the degree of evidence.

	ESC 2014
AHA 2020	III	IIb	IIa
III	12	0	1
IIb	w6	1	5
IIa	16	13	27

• General agreement: 64.2%, kappa 0.270 (95%CI 0.118–0.422; p=0.003)
• ESC: European Society of Cardiology; AHA: American Heart Association. IIa, IIb, III: degrees of scientific evidence for implantable cardioverter-defibrillator; implantation according to each guideline.

The Kaplan-Meier curve showed a trend toward lower outcome-free survival in patients with fQRS (71.3 vs. 82.6%; p=0.515, Figure 1A). There was no statistical difference regarding the cumulative incidence of appropriate shock (10.5 vs. 16%, with and without fQRS; p=0.598, Figure 1B).

Figure 1
Survival and incidence curves for the primary outcome and appropriate shocks.

Considering the indications for ICD implantation as a binary outcome (implant [evidence IIa/IIb] or not implant [evidence III]), C-statistics analysis did not show differences regarding the incidence of appropriate shock (p=0.644). The AUC was 0.557 for the 2014 ESC recommendations (95%CI 0.406–0.707) and 0.548 for the 2020 AHA (95%CI 0.548–0.636) Figure 2.

Figure 2
Receiver operating characteristic curve comparing the incidence of appropriate shocks according to European Society of Cardiology 2014 and American Heart Association 2020 guidelines implantable cardioverter-defibrillator indication.

DISCUSSION

In this study, we evaluated the agreement between the 2014 ESC and 2020 AHA guidelines in the indication of ICD as primary prophylaxis of MS in patients with HCM. Our results show significant divergence in the indication of ICD, with an overall agreement of 64%. The 2020 AHA algorithm indicated class IIa ICD in 69% of patients, compared to 40.7% by 2014 ESC. Two patients classified as low risk by ESC had appropriate shocks 1 and 5 years after ICD implantation. Of the 13 patients classified by the AHA as low risk, there was a divergence from the ESC in only one case.

The analysis of the agreement of indications resulted in a kappa of 0.270. Kappa coefficient between 0.21 and 0.39 represents minimal agreement, implying that only 4–15% of the indications analyzed between both guidelines are, in fact, reliable¹³.

Mattos et al. demonstrated that in relation to the 2011 AHA, the ESC algorithm also had low agreement and would leave all patients (8/90) unprotected with appropriate shock⁵. Other studies have also demonstrated low sensitivity of the algorithm, especially in patients considered to be at low risk¹⁴,¹⁵. Our results showed low accuracy of the algorithm to predict MS, especially in the group considered low risk.

The 2020 AHA recommendations showed high sensitivity (100%), but their low specificity (17.6%) implies unnecessary indication of ICD in low-risk patients. ICD implantation is related to complications such as infection and inappropriate shocks, with an incidence of 2.1% per year¹⁶. In the C-statistics analysis, both guidelines showed similar discrimination in predicting appropriate shock.

In HCM studies, a good correlation has been shown between the presence of fQRS and fibrosis estimated by CMR and histology¹⁷. In our sample, it was not possible to demonstrate the association between fQRS and fibrosis. There was a trend toward a greater outcome-free survival in patients without fQRS, but this difference was not significant. There was no statistical difference between the cumulative incidence of appropriate shocks between patients with or without fQRS, despite a trend toward more shocks in the fQRS group (10 vs. 16%).

Few studies have evaluated the direct link between MS risk and appropriate ICD shock in patients with HCM and the presence of fQRS. One study evaluated the calculated risk of SD in 5 years of the 2014 ESC and showed that the presence of fQRS was related to a risk of SD >4%¹⁸. In the SHIFT study, the fQRS was included as a risk predictor with a hazard ratio of 3.6⁷. However, the study included only patients at low and moderate risk for MS, compromising its practical applicability.

The spectrum of clinical presentation of HCM is quite heterogeneous. The mechanisms underlying the occurrence of fibrosis and arrhythmia are not fully understood and appear to be influenced by epigenetic factors¹⁹. Rigid predictor models are unable to represent the complexity of individual risk, which reinforces the role of specialist experience in risk stratification and individualized indication of primary ICD prophylaxis.

Some limitations of this study were the sample size, the retrospective design, and the factor of being unicentric. The occurrence of appropriate shocks would not necessarily represent life-threatening events and the patients treated tended to present later diagnoses and greater severity.

CONCLUSIONS

The 2014 ESC and 2020 AHA MS risk stratification algorithms for HCM patients present discrepancies in the indication of ICD implantation, both with low accuracy. The European guideline showed better specificity, while the American guideline showed excellent sensitivity, despite similar discrimination using C-statistics.

• Funding: none.

REFERENCES

• ¹
  Ommen SR, Mital S, Burke MA, Day SM, Deswal A, Elliott P, et al. 2020 AHA/ACC guideline for the diagnosis and treatment of patients with hypertrophic cardiomyopathy: a report of the American college of cardiology/American heart association joint committee on clinical practice guidelines. Circulation. 2020;142(25):e558-631. https://doi.org/10.1161/CIR.0000000000000937
  » https://doi.org/10.1161/CIR.0000000000000937
• ²
  Maron BJ, Maron MS. Hypertrophic cardiomyopathy. Lancet. 2013;381(9862):242-55. https://doi.org/10.1016/S0140-6736(12)60397-3
  » https://doi.org/10.1016/S0140-6736(12)60397-3
• ³
  Adamczak DM, Oko-Sarnowska Z. Sudden cardiac death in hypertrophic cardiomyopathy. Cardiol Rev. 2018;26(3):145-51. https://doi.org/10.1097/CRD.0000000000000184
  » https://doi.org/10.1097/CRD.0000000000000184
• ⁴
  Authors/Task Force members, Elliott PM, Anastasakis A, Borger MA, Borggrefe M, Cecchi F, et al. 2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC). 2014;35(39):2733-79. https://doi.org/10.1093/eurheartj/ehu284
  » https://doi.org/10.1093/eurheartj/ehu284
• ⁵
  Mattos BPE, Scolari FL, Garbin HI. Discrepancy between International Guidelines on the Criteria for Primary Prevention of Sudden Cardiac Death in Hypertrophic Cardiomyopathy. Arq Bras Cardiol. 2020;115(2):197-204. https://doi.org/10.36660/abc.20190161
  » https://doi.org/10.36660/abc.20190161
• ⁶
  Steriotis AK, Sharma S. Risk stratification in hypertrophic cardiomyopathy. Eur Cardiol. 2015;10(1):31-6. https://doi.org/10.15420/ecr.2015.10.01.31
  » https://doi.org/10.15420/ecr.2015.10.01.31
• ⁷
  Ruivo C, Montenegro Sá F, Correia J, Belo A, Loureiro MF, Morais J, et al. The SHIFT model combines clinical, electrocardiographic and echocardiographic parameters to predict sudden cardiac death in hypertrophic cardiomyopathy. Rev Port Cardiol (Engl Ed). 2019;38(12):847-53. https://doi.org/10.1016/j.repc.2019.05.012
  » https://doi.org/10.1016/j.repc.2019.05.012
• ⁸
  Bi X, Yang C, Song Y, Yuan J, Cui J, Hu F, et al. Quantitative fragmented QRS has a good diagnostic value on myocardial fibrosis in hypertrophic obstructive cardiomyopathy based on clinical-pathological study. BMC Cardiovasc Disord. 2020;20(1):298. https://doi.org/10.1186/s12872-020-01590-2
  » https://doi.org/10.1186/s12872-020-01590-2
• ⁹
  O’Mahony C, Jichi F, Ommen SR, Christiaans I, Arbustini E, Garcia-Pavia P, et al. International external validation study of the 2014 European Society of Cardiology Guidelines on Sudden Cardiac Death Prevention in Hypertrophic Cardiomyopathy (EVIDENCE-HCM). Circulation. 2018;137(10):1015-23. https://doi.org/10.1161/CIRCULATIONAHA.117.030437
  » https://doi.org/10.1161/CIRCULATIONAHA.117.030437
• ¹⁰
  Magnusson P, Gadler F, Liv P, Mörner S. Risk markers and appropriate implantable defibrillator therapy in hypertrophic cardiomyopathy. Pacing Clin Electrophysiol. 2016;39(3):291-301. https://doi.org/10.1111/pace.12801
  » https://doi.org/10.1111/pace.12801
• ¹¹
  Chaves AVF, Bezerra LS, Santos-Veloso MAO, Calado EB, Markman M, Lordsleem ABMS, et al. Apical hypertrophic cardiomyopathy: A case series at a Brazilian referral center with a maximal follow-up of 15 years. Ann Cardiol Angeiol (Paris). 2021;70:215-9. https://doi.org/10.1016/j.ancard.2021.06.005
  » https://doi.org/10.1016/j.ancard.2021.06.005
• ¹²
  Das MK, Saha C, El Masry H, Peng J, Dandamudi G, Mahenthiran J, et al. Fragmented QRS on a 12-lead ECG: a predictor of mortality and cardiac events in patients with coronary artery disease. Heart Rhythm. 2007;4(11):1385-92. https://doi.org/10.1016/j.hrthm.2007.06.024
  » https://doi.org/10.1016/j.hrthm.2007.06.024
• ¹³
  Dettori JR, Norvell DC. Kappa and Beyond: Is There Agreement? Global Spine J. 2020;10(4):499-501. https://doi.org/10.1177/2192568220911648
  » https://doi.org/10.1177/2192568220911648
• ¹⁴
  Desai MY, Smedira NG, Dhillon A, Masri A, Wazni O, Kanj M, et al. Prediction of sudden death risk in obstructive hypertrophic cardiomyopathy: potential for refinement of current criteria. J Thorac Cardiovasc Surg. 2018;156(2):750-9.e3. https://doi.org/10.1016/j.jtcvs.2018.03.150
  » https://doi.org/10.1016/j.jtcvs.2018.03.150
• ¹⁵
  O’Mahony C, Akhtar MM, Anastasiou Z, Guttmann OP, Vriesendorp PA, Michels M, et al. Effectiveness of the 2014 European Society of Cardiology guideline on sudden cardiac death in hypertrophic cardiomyopathy: a systematic review and meta-analysis. Heart. 2019;105(8):623-31. https://doi.org/10.1136/heartjnl-2018-313700
  » https://doi.org/10.1136/heartjnl-2018-313700
• ¹⁶
  Maron MS, Rowin EJ, Wessler BS, Mooney PJ, Fatima A, Patel P, et al. Enhanced American College of Cardiology/American Heart Association Strategy for Prevention of sudden cardiac death in high-risk patients with hypertrophic cardiomyopathy. JAMA Cardiol. 2019;4(7):644-57. https://doi.org/1010.1001/jamacardio.2019.1391
  » https://doi.org/1010.1001/jamacardio.2019.1391
• ¹⁷
  Ratheendran AC, Subramanian M, Bhanu DK, Prabhu MA, Kannan R, Natarajan KU, et al. Fragmented QRS on electrocardiography as a predictor of myocardial scar in patients with hypertrophic cardiomyopathy. Acta Cardiol. 2020;75(1):42-6. https://doi.org/10.1080/00015385.2018.1547355
  » https://doi.org/10.1080/00015385.2018.1547355
• ¹⁸
  Özyılmaz S, Akgül Ö, Uyarel H, Pusuroğlu H, Karayakalı M, Gül M, et al. Assessment of the association between the presence of fragmented QRS and the predicted risk score of sudden cardiac death at 5 years in patients with hypertrophic cardiomyopathy. Anatol J Cardiol. 2017;18(1):54-61. https://doi.org/10.14744/AnatolJCardiol.2017.7593
  » https://doi.org/10.14744/AnatolJCardiol.2017.7593
• ¹⁹
  Tobita T, Nomura S, Fujita T, Morita H, Asano Y, Onoue K, et al. Genetic basis of cardiomyopathy and the genotypes involved in prognosis and left ventricular reverse remodeling. Sci Rep. 2018;8(1):1998. https://doi.org/10.1038/s41598-018-20114-9
  » https://doi.org/10.1038/s41598-018-20114-9

Publication Dates

• Publication in this collection
  19 Sept 2022
• Date of issue
  Aug 2022

History

• Received
  11 Feb 2022
• Accepted
  05 Apr 2022