ABC | Volume 110, Nº6, June 2018

Original Article Reis et al Hypertrophic Cardiomyopathy Arq Bras Cardiol. 2018; 110(6):524-531 Probability of SCD at 5 years = 1 – 0.998 exp(prognostic index) Prognostic index = [0.15939858 x maximal wall thickness (mm)] - 0.00294271 x maximal wall thickness 2 (mm 2 )] + [0.0259082 x left atrial diameter (mm)] + [0.00446131 x maximal LV outflow tract gradient (rest/Valsalva - mm Hg)] + [0.4583082 x family history of SCD] + [0.82639195 x NSVT] + [0.71650361 x unexplained syncope] - [0.01799934 x age on clinical assessment (years)]. those authors have concluded that the limited power in risk stratification results from the fact that the algorithm is based on a dichotomous classification of the risk variables. 11 Thus, the risk factors are recognized to be non-static and to have a cumulative evolutionary potential, with corresponding increase in the likelihood of SCD. 12 In 2013, a new mathematical model was proposed to estimate the individual risk of SCD at 5 years. 13,14 That model, based on a retrospective study of a population of 3675 patients from six centers, comprises some classical risk factors combined with LV outflow tract gradient, left atrial diameter, and age, which are considered continuous variables. 13 The following formula is used: neuromuscular causes (2 patients with cardiac amyloidosis, 1 patient with Noonan syndrome and 1 patient with Anderson‑Fabry disease) were excluded. The total sample of this study comprised 105 index patients diagnosed with HCM. The indication for an ICD implantation was based on the 2011 ACCF/AHA recommendations, and the patients received an ICD when they had at least one risk factor for SCD, according to the 2011 guidelines. Later, a new analysis was performed based on the current recommendations (2014 ESC), using the data of the patients at the time of the diagnosis. The current model of risk for SCD due to HCM is part of a predefined set of 7 potentially prognostic variables. 1 By using an online calculator, a predictive risk score of SCD due to HCM at 5 years was generated. According to that value, patients were stratified into three risk categories for ICD implantation: < 4%/5 years (ICD usually not considered); 4% to 6%/5 years (ICD can be considered); > 6%/5 years (ICD should be considered). 1 Characteristics of the population base and complementary study The following baseline characteristics were collected at the time of diagnosis: age, sex, arterial hypertension, diabetes mellitus , atrial fibrillation, unexplained syncope, history of SCD in a first-degree relative (< 40 years), New York Heart Association (NYHA) functional class. All patients underwent initial 12-lead electrocardiography, with assessment of LVH voltage criteria, Q waves, left axis deviation and atrioventricular conduction disorders. All patients underwent transthoracic echocardiography. The following parameters were recorded: LV diastolic diameter, LV wall thickness from base to apex, presence of LV outflow tract gradient at rest and after the Valsalva maneuver, left atrial diameter, classification of LV systolic (LV ejection fraction) and diastolic function. The LV outflow tract obstruction caused by the systolic anterior motion (SAM) of the mitral valve leaflets was defined as a peak pressure gradient at the LV outflow tract ≥ 30 mm Hg at rest or during physiological challenge. 1 Twenty-five patients (23.8%) with no gradient at rest underwent exercise echocardiography to assess the presence of gradient during exercise. All patients underwent 24-hour Holter at the initial assessment or during clinical follow-up, allowing the identification of ventricular extrasystoles and/or NSVT episodes, defined as the presence of at least three consecutive ventricular complexes, lasting less than 30 seconds and without hemodynamic impairment. All patients underwent exercise test according to the Bruce protocol to assess blood pressure response during exercise. Anomalous response was defined as the lack of blood pressure increase by 20 mmHg or a decrease of at least 20 mmHg during exertion. Cardiac magnetic resonance imaging was performed in 85 (80.2%) patients who had access to a magnetic resonance scanner 1.5 Tesla (Phillips®). The following parameters were recorded for analysis: left atrial area, greater LV wall thickness, LV ejection fraction and presence of late enhancement after intravenous gadolinium administration. According to the literature, that score is more accurate to differentiate patients at low risk from those at high risk, 13 and was incorporated into the most recent European Society of Cardiology (ESC) recommendations published in 2014 as a valid and independent method for risk stratification. 1 The direct comparison of the discriminative value of the two risk score systems to identify patients requiring an ICD in a non-selected population with HCM has not been performed in Portugal. This study aimed at comparing the risk stratification of SCD in a population of patients with HCM, according to the 2011 and 2014 recommendations, and at characterizing the clinical performance of the risk model of SCD due to HCM individually in a Portuguese population with HCM. Methods Population Retrospective single-center analysis of patients diagnosed with HCM and regularly followed up at a cardiology outpatient clinic of one single tertiary center for 6 years. The definition of HCM was based on a wall thickness ≥ 15 mm in one or more LV myocardial segments, which was not explained only by LV overload, and measured by use of any imaging technique [echocardiography, cardiac magnetic resonance imaging (CMRI) or computerized tomography (CT)]. The clinical diagnosis of HCM in a first-degree relative of a patient with unequivocal disease (LVH ≥ 15 mm) is based on the presence of unexplained LV wall thickening ≥ 13 mm in one or more myocardial segments, measured by use of cardiac imaging techniques. 1-3,15,16 This study included 109 patients with LVH. Those whose complementary study revealed hereditary metabolic and 525