ABC | Volume 110, Nº6, June 2018

Original Article Reis et al Hypertrophic Cardiomyopathy Arq Bras Cardiol. 2018; 110(6):524-531 The gold-standard treatment for primary and secondary prevention of SCD in patients with HCM is ICD implantation, which proved effective in interrupting potentially lethal ventricular tachyarrhythmias, altering the disease’s natural history. 1,7 The efficacy of that therapy has been consolidated since 2000, and has been recently reinforced in a meta-analysis examining the results of 16 studies published between 1998 and 2012, regarding ICD interventions and complications in primary and secondary preventions. 17-22 The risk stratification of SCD in patients with HCM according to the 2011 ACCF/AHA recommendations was effective in identifying many patients who could benefit from ICD implantation. However, the method proved to be incomplete and some patients without the conventional risk factors were excluded and remained at risk for SCD. 23,24 Thus, the development of new SCD markers for risk stratification is required. 11 In 2013, a group of English researchers suggested a new risk score of SCD due to HCM at 5 years. It is a mathematical and statistically complex model. 13 That score has been rapidly incorporated into the 2014 ESC recommendations as the valid and independent method to select/exclude patients for ICD implantation in primary prevention. 1 The major objective of any stratification method is its reliability to identify patients at major risk for events, being thus candidates for ICD implantation in primary prevention of SCD. It is worth noting that the new SCD risk model has incorporated arbitrarily two new risk markers (LV outflow tract gradient and left atrial diameter), which had not previously shown to be independent predictors of SCD due to HCM and are not included as risk markers for patients’ assessment. 2,10,18 This study was not aimed at validating (or invalidating) the risk score of SCD due to HCM, but at characterizing the clinical performance of that model individually in a population of Portuguese patients with HCM. It is worth noting that this analysis showed that the risk model seems to have little sensitivity to identify patients at elevated risk for arrhythmic events and SCD, who, according to the conventional criteria, would be candidates for prophylactic ICD implantation. For example, in the sample of 28 patients with complex dysrhythmic events during the 6-year clinical follow-up, only 4.7% had a risk score > 6%/5 years, which would have justified ICD implantation in primary prevention. In addition, most patients had a score <4%/5 years, that is, no indication for treatment with ICD. It is worth noting that HCM is a complex pathology, with a spectrum of histological findings and varied and unpredictable clinical manifestations, and a relatively low percentage of SCD. 1,2,10,22,24-29 Thus, intuitively it would not be expected that the clinical decision individualized for each patient could be based only on a complex mathematical formula, minimizing the fundamental clinical reasoning when facing a patient with HCM. Being a genetic pathology, some specific mutations might pose a higher risk for SCD. However, it is difficult to determine the existence of a consistent genotype/phenotype correlation, explaining the inability to establish an accurate prognosis based on specific mutations. 4 Thus, given the inconsistency, they were not included as markers in the current risk model. However, an important omission in this model is that of quantified late enhancement on CMRI, which several studies have shown to be an independent marker of adverse arrhythmic events (NSVT, VT, ventricular fibrillation) and SCD, 30-34 even in patients without the conventional risk factors. Some individuals with HCM can develop LV apical aneurysms, associated with local healing and greater propensity to potentially lethal arrhythmias and SCD, 35 in addition to heart failure with systolic dysfunction 36 and coronary atherosclerotic disease, 37 which are not contemplated in the risk score of SCD. Some prediction inconsistency of the new risk model might be related to the inclusion of some variables, such as syncope, NSVT, left atrial diameter and LV outflow tract obstruction gradient (non-static variables). 11,24,38,39 The strategy of conventional risk stratification prioritizes SCD prevention in patients with HCM versus excessive ICD implantation. On the contrary, the new risk score seems to identifymany patients at low risk, who are not candidates for ICD implantation. There is, thus, a significant reduction in the number of devices implanted, but it seems at the cost of misclassifying some patients at high risk for arrhythmic events and SCD. Study limitations Our study has some limitations, because it is based on a single center, with a reduced number of patients and events. However, calculating the sample power ensured that the number of patients was sufficient to draw conclusions. As in any retrospective study, we were limited by the information available in the patients’ medical records. Conclusion Hypertrophic cardiomyopathy is a complex pathology, with a wide and unpredictable clinical spectrum. According to our data, the current risk stratification model seems to reduce the proportion of patients with indication for ICD implantation. It is worth noting that the decision based on a mathematical model that minimizes the individual clinical reasoning seems a little reliable strategy to identify patients at risk for events due to HCM. Author contributions Conception and design of the research and writing of the manuscript: Reis L; Acquisition of data: Reis L, Silva J; Analysis and interpretation of the data: Reis L, Teixeira R, Fernandes A, Almeida I, Madeira M, Silva J, Botelho A; Statistical analysis: Reis L, Teixeira R, Fernandes A, Almeida I, Madeira M, Silva J; Critical revision of the manuscript for intellectual content: Reis L, Teixeira R, Fernandes A, Almeida I, Madeira M, Silva J, Botelho A, Pais J, Nascimento J, Gonçalves L. Potential Conflict of Interest No potential conflict of interest relevant to this article was reported. Sources of Funding There were no external funding sources for this study. 529