ABC | Volume 113, Nº4, October 2019

Original Article Rodrigues et al. Myocardial strain measured by speckle tracking Arq Bras Cardiol. 2019; 113(4):737-745 conditions due to opportunistic agents, malnutrition or prolonged immunosuppression has been observed, 5 the incidence of coronary artery disease and peripheral vascular events has increased in HIV-infected patients. 6,7 HIV-infected patients may have specific myocardial abnormalities and conventional two-dimensional tests may fail to detect subtle abnormalities in regional myocardial function. Speckle tracking is an innovative echocardiographic technique that has the capacity to evaluate myocardial strain in order to identify subtle abnormalities in ventricular function. Myocardial strain is a very important mechanical variable in HIV-infected patients, as it shows subclinical left ventricular dysfunction. Unfortunately, the technique of studying cardiac strain is still underused. Global longitudinal strain (GLS) is well correlated with left ventricular ejection fraction (LVEF). Reduced SLG can be found in patients with heart failure with preserved ejection fraction, 8 stable angina, 9 three-vessel coronary artery disease and patients using chemotherapy agents with cardiotoxicity. 10,11 The purpose of this study was to evaluate the presence of subclinical ventricular function abnormalities in HIV-infected patients using or not using ART. Methods Observational cross-sectional study involving 68HIV‑infected patients recruited from the Infectiology Service of Hospital Universitário Antônio Pedro (HUAP), Universidade Federal Fluminense (UFF). Inclusion criteria were: age ≥18 years, HIV infection confirmed by serological tests, no cardiovascular symptoms. Patients were excluded if they were under any therapy with cardiac or neurological medications, if they had any cardiac symptomor history of hypertension, LV ejection fraction <0.55 and pulmonary artery systolic pressure >36 mmHg, stable angina, atrial fibrillation or moderate to severe valvular heart disease. Echocardiography was performed as part of an established research protocol rather than for symptoms or comorbidities. Patients were divided into four groups: 1) HIV‑positive patients not using ART (NT); 2) HIV-positive patients on protease inhibitor therapy for at least 12months (PI); 3) HIV-positive patients on therapy with non-nucleoside reverse transcriptase inhibitors (NNRTI) for at least 12 months and 4) healthy controls. Samples from the NT (n = 11), PI (n = 24) and NNRTI (n = 33) groups were defined by convenience, considering the patients at the time of data collection. For the control group, a sample of size similar to the largest of the study groups (n = 30) was defined. The echocardiographic tests were conducted on an Echo Color Doppler device of the Italian company Esaote Biomédica, model Mylab 30 Gold, with a multi-frequency electronic sectoral transducer (2 to 4 MHz) with continuous electrocardiographic scanning. Traditional measures of left ventricular (LV) systolic function, ejection fraction and systolic shortening, diastolic function indicators, such as mitral flow E/A ratio, myocardial E wave velocity in the septal mitral annulus (septal E’), E/E’ ratio and estimated left atrial pressure were taken. Right ventricular diastolic diameter and two echocardiographic variables that evaluate right ventricular systolic function were determined: tissue Doppler of lateral tricuspid annulus and longitudinal tricuspid annular motion (LTAM). LV ejection fraction was determined by using the Simpson’s technique, on apical four‑chamber and two‑chamber views, on diastole and systole, thus obtaining end diastolic and end systolic volumes. Left atrial volume was obtained from end-systolic four-chamber and two‑chamber views, and the arithmetic mean was then indexed by the body surface area to obtain left atrial volume index. LV mass was obtained from diastolic and systolic LV diameters, as well as from the interventricular septal and inferolateral wall diastolic thickness, following the technical guidelines of the American Society of Echocardiography. 12 Maximum tricuspid regurgitation (TR) rate, an indicator of pulmonary artery pressure, was obtained from apical four-chamber view. LV diastolic and systolic myocardial velocities were obtained by placing the tissue Doppler sample volume in the septal mitral annulus. Digital myocardial strain curves were taken by using the Xstrain software package from scanned cross‑sectional and apical view images. Myocardial strain rate was also evaluated. GLS was obtained by the arithmetic mean of the longitudinal strain values in the seventeen segments, from the four-chamber apical view (Figure 1), three-chamber apical view (Figure 2) and two-chamber apical view (Figure 3). Global circumferential strain (GCS) was obtained by the arithmetic mean of the circumferential strain values in the seventeen segments, from the cross-sectional views at the level of the mitral valve, papillary muscles and tip. Radial global strain (SRG) was obtained from the arithmetic mean of the radial strain values in the seventeen segments, from cross‑sectional views of the mitral valve, papillary muscles and tip. Strain percentage analysis was repeated twice, using the best echocardiographic images. The same echocardiographer conducted transthoracic evaluation, then took the scanned images to calculate the percentages of longitudinal, radial and circumferential strain on an offline workstation. The strain rate in the longitudinal, circumferential and radial planes was also obtained. (Figures 1 and 3). Statistical analysis Statistical analysis of the data was done with IBM SPSS – version 20 for Windows. After analysis of normality of independent variables in the different groups (using the Shapiro-Wilk test) and homogeneity of the variances between the groups (using Levene’s test), it was decided to use Kruskal-Wallis’ non-parametric test followed by Dunn’s multiple comparison tests to test the significance of the differences between the values measured in the study groups. A significance level of 5% was adopted for decision- making on statistical tests. Continuous variables with normal distribution were described as mean and standard deviation and continuous variables with non-normal distribution were described as median and interquartile range. This study was approved by the Research Ethics Committee from Hospital Universitário Antônio Pedro (#HUAP 159/11) and all patients signed an Informed Consent Form. Results The study included 98 individuals: 68 (69.4%) HIV‑infected and 30 (30.6%) healthy controls with negative serology, of which 60 (61.2%) were males and 38 (38.8%) were females. Separately analyzing the groups of HIV-infected 738