ABC | Volume 115, Nº2, August 2020

Original Article Mattos BP et al. Sudden death in hypertrophic cardiomyopathy Arq Bras Cardiol. 2020; 115(2):197-204 the 2014 ESC model (p=0.05). Family history of SCD and NSVT associatedwith LVOT obstructionwere the predictors that showed the greatest decrease of ICD class recommendation under the European guideline (Table 6). Discussion In this study, for the first time, we compared the SCD primary prevention criteria established by the 2011 ACCF/AHA and the 2014 ESC guidelines in a Brazilian HCM population based on a non-referred outpatient clinical cohort. Our results demonstrate low agreement between the two systems regarding the recommendations for primary prevention ICD. The ESC/HCM Risk-SCD score has lowered the class of implant recommendation over ACCF/AHA in 36% of the patients. Among those in class IIa in the North American guideline, the ESC risk score decreased the class of device recommendation in 74% of the patients, determined an ICD unwarranted in 56% and maintained the recommendation in only 26%. The European risk score has added recommendation in only 1% of the patients. In almost all cases, in which the implantation was not recommended with the North American guideline, the European criteria reassured the decision. The new model has excluded, from class IIa, the 8 (9%) patients experiencing SCDor ICDappropriate shock along the observation period, although 25% of them remained in class IIb. HCM is an arrhythmogenic heart disease, whose histopathological substrate characterized by hypertrophy, cell disarray, fibrosis and coronary microvascular disease favors the occurrence of lethal ventricular arrhythmias. 5,6,19,20 Risk stratification for SCD is based on observational data obtained in very selected populations. It is considered complex, due to the heterogeneous character of the disease, and imperfect because many deaths occur in the absence of risk predictors. 5-7, 21 The limitations offered by the 2003 and 2011 algorithms have been demonstrated in an international registry showing no difference in appropriate shock rates between patients with one, two, three or more predictors. 22 A posterior validation analysis of these criteria reports that the incidence of SCD and appropriate discharge do not differ between patients with none or only one predictor and that the initial algorithms have limited power to discriminate between high and low risk and could result in unnecessary implants. 23 This study evaluated a HCM cohort with more advanced age and low risk profile: 78% of the patients remained in functional class I/II, 47% presented no risk factors and 35% showed only one. HCM patients aged ≥60 demonstrate reduced morbidity-mortality and SCD rates, even in the presence of risk predictors. 6 Five and ten-year SCD and ICD appropriate shock-free survival rates achieved 93% and 92%, respectively, and only 9% of the patients experienced these events along the period. A multicenter longitudinal study presents similar results and supports that HCM, when conveniently treated, shows reduced mortality in adulthood with a ten-year survival rate similar to that expected in the general population. 3 The mean calculated ESC HCM risk-SCD score of 3.2±2.5 characterized 75% of the patients as low risk. NSVT, syncope and increased MLVWT were more frequent in high-risk patients compared to others. In this study, we determine the percentages achieved by each one of the score predictors in the tree risk categories with the purpose of discriminating those that reached more weight in order to justify the low agreement between the two guidelines. We ascertained that the factors that havemainly contributed to the calculation and reached increased values in the low, intermediate and high-risk levels wereMLVWT, left atrial diameter and age, the latter with a subtractive effect. These findings may justify the low agreement between the two guidelines, considering that MLVWT as a continuous variable, left atrial diameter and age are not included in the North American strategy. Family history of SCDor syncope, both considered as an ACCF/AHA recommendation for ICD therapy, showed lower contribution to the score calculation. The combination of risk factors characterized as ACCF/AHA class IIa recommendation for ICD in the sample, mainly family Table 6 – Sudden cardiac death risk profile in patients with hypertrophic cardiomyopathy in class IIa for implantable cardioverter defibrillator with the 2011 American College of Cardiology Foundation/American Heart Association guideline: restratification with the 2014 European Society of Cardiology model 2011 ACCF/AHA / 2014 ESC IIa2011/IIa2014 IIa2011/IIb2014 IIa2011/III2014 n = 11 (26%) n = 8 (19%) n = 24 (55%) Isolated family history of SCD 2 (12%) 2 (12%) 13 (76%) Isolated syncope 5 (45%) 1 (10%) 5 (45%) Syncope + family history of SCD 2 (40%) 2 (40%) 1 (20%) Family history of SCD + MLVWT ≥30mm 1 (100%) 0 0 NSVT + LVOT obstruction 1 (17%) 3 (50%) 2 (33%) Abnormal BP response + LVOT obstruction + LGE on CMR 0 0 3 (100%) P = 0.05 ACCF: American College of Cardiology Foundation; AHA: American Heart Association; ESC: European Society of Cardiology; SCD: sudden cardiac death; MLVWT: maximal left ventricular wall thickness; NSVT: non-sustained ventricular tachycardia; LVOT: left ventricular outflow tract; BP: blood pressure; LGE: late gadolinium enhancement; CMR: cardiac magnetic resonance. 202