ABC | Volume 113, Nº4, October 2019

Original Article Gil et al. LV function in HCM by Strain technique Arq Bras Cardiol. 2019; 113(4):677-684 Methods Study participants A retrospective cross-sectional study was conducted between September 2014 and April 2016 in patients followed up at the cardiology outpatient clinic of Hospital Universitário Pedro Ernesto – UERJ – diagnosed with HCM. This study was approved by the Research Ethics Committee with Certificate of Presentation for Ethics Appreciation (CAAE) number 23561113.2.0000.5259, and it is according to the Helsinki Declaration of 1975 updated in 2013. All patients who accepted to participate in the study have read and signed an informed consent form. Inclusion criteria were: diagnosis of HCM confirmed by magnetic resonance imaging (MRI), age over 18 years, preserved left ventricular ejection fraction (EF) (>55%), no interventions for septal reduction and no pacemaker or defibrillator. Patients with atrial fibrillation and known coronary artery disease were excluded. Diagnostic confirmation by MRI with gadoliniumwas based on the distribution of hypertrophy and late enhancement pattern. 13 The convenience sample included 45 patients, including 22 (48.9%) males withmean age of 45.1±13.9 years. In this group, 14 (31.1%) had hypertension previously diagnosed according to the Brazilian guidelines for SAH. 8 The flowchart with the selection of patients is shown in Figure 1. Echocardiographic analysis Transthoracic echocardiographic test was performed by an experienced echocardiographer on a Philips® iE33 Matrix equipment using 3-1MHzmatrix transducer. One‑dimensional, two-dimensional and Doppler echocardiography analysis was performed following the recommendations of the American Society of Echocardiography. 14 The echocardiographic classification of Maron et al., 15 which divides hypertrophy into types I, II, III, and IV (Figure 2) was used to define the type of left ventricular hypertrophy. The obstructive pattern was considered for left ventricular outflow tract gradients greater than 30 mmHg, measured by continuous Doppler, at rest and after Valsalva maneuver. 5 The approach of diastolic function and ventricular filling pressures followed the recommendations of the American Society of Echocardiography for patients with HCM. 16 In the myocardial strain analysis, we used speckle tracking. Strain is calculated for each left ventricular segment as the relative mean strain between two speckles. As a measure of strain, it is expressed in negative percentages (-%); the closer to 0, the lower the strain. Values lower than -18%were considered normal strain. Only GLS was analyzed because it is more widely used and considered a robust index for clinical studies. Also, GLS is the first to be impaired in most heart diseases, including HCM, when ejection fraction is still preserved. 12 Echocardiographic protocol for GLS included 4-chamber, 3-chamber and 2-chamber apical views. GLS analysis was processed offline using the software Philips® QLab 9.0. These results were translated into curves, one for each ventricular segment, and the overview, with the quantification of velocities, was expressed on a bull’s eye map, exemplified in Figure 3. Echocardiographic scans were stored and the images revised. Examiner and reviewer are the authors of the study. Strain analysis was repeated by the reviewer on all scans. Intraobserver and interobserver variability was evaluated using the coefficient of variation (CV = 100 (s/x) (%)). We obtained good agreement and the coefficients were considered low (<10%). Statistical analysis The collected data were entered in a Microsoft Excel™ spreadsheet, and later analyzed in R Studio, version 1.0.143. Continuous variable distributions were expressed usingmean and standard deviation asmeasures of central tendency and dispersion for each of the groups analyzed. To assess whether there was a difference between the groups, unpaired Student’s t -test was used after Levene’s test for equality of variances. For categorical variables, the nonparametric approach was chosen, where the difference between proportions was evaluated by the X 2 test (with Yates correction) and Fisher’s exact test. In cases where therewere more than two categories, the Kruskal‑Wallis test was used. In this study, p < 0.05 was considered statistically significant. Results Of 55 initially eligible patients, 10 were excluded: 6 by atrial fibrillation, which impairs GLS analysis, 2 by known coronary artery disease, which also interferes with strain analysis, and 2 whose HCM diagnosis had not been confirmed by MRI. The patients’ general characteristics are shown in Table 1. Figure 4 shows a set of charts with the main results. Mean age was higher in the hypertensive group, and body mass index (BMI) and mean systolic and diastolic pressures were higher in this group. No significant differences were observed regarding gender and functional class between the groups. Regarding the echocardiographic findings, there was less strain in hypertensive patients (–10.29% ± 2.46) than in normotensive patients (–12.35% ± 3.55), indicating greater impairment of ventricular function in this group (p = 0.0303). Although all patients had preserved EF, mean left ventricular systolic diameter (LVSD) was higher in hypertensive patients, but still within normal limits. Diastolic dysfunction was more evident in hypertensive patients (p = 0.0242), with a lower number of undetermined cases. In hypertensive patients, longer isovolumetric relaxation time (IVRT), lower E/A ratio in mitral flow, as well as a lower septal E/e’ ratio were observed on mitral annulus tissue Doppler. Mean left atrial volume was increased in both groups, but without any significant difference between them (Table 2). In the hypertrophy type analysis, in the general sample, type III was the most frequent one (40%), followed by type II (31%), I (15.7%) and IV (13.3%), but no significant difference was observed between the groups regarding the type of hypertrophy. In this sample, there were no cases of concentric hypertrophy. Also, there was no significant difference between the groups regarding left ventricular outflow tract obstruction, with a higher percentage of the nonobstructive form in the overall sample (66.7%). 678