IJCS | Volume 32, Nº2, March/April 2019

136 Morais and Rassi Six-minute walk test in heart failure Int J Cardiovasc Sci. 2019;32(2)134-142 Original Article semiautomated sphygmomanometer (OMRON ® HEM 711). Mean arterial pressure (MAP) was calculated using the formula: (systolic arterial pressure + 2 times diastolic pressure) divided by 3. These measures were taken during the period of rest from the 6MWT. Height (m) and weight (kg) were measured with patients standing barefoot, using aWelmy ® scale (model W300) (Sao Paulo, Brazil), and the values were used for body mass index (BMI) calculation (weight in kilograms divided by the square of height in meters). Health-related quality of life (HRQOL) was assessed by the Minnesota Living with Heart Failure Questionnaire (MLHFQ), composed of 21 questions, that provides a total score that may range from 0 to 105, from best to worst quality of life. 7 The total score (0-105), physical dimension (0-45) and emotional dimension (0-25) scores were used for analysis. Dyspnea while performing activities of daily living (ADLs) was assessed by the Medical Research Council modified dyspnea scale (mMRC); scores of the items ranged from 0 to 4, in which 4 indicated the strongest limitation of ADLs due to dyspnea. 8 Physical activity level was assessed using the short version of the International Physical Activity Questionnaire (IPAQ), validated in Brazil by Matsudo et al., 9 In the presente research, the participants were grouped into two groups – partially active or sedentary in one group, and active or very active in the other. Handgrip strength (HGS) was measured using a hydraulic dynamometer (Saehan ® ), a validated instrument to measure isometric HGS, 10 according to the American Society of HandTherapists recommendations. 11 HGS was measured in the dominant hand; the mean of three measures was used for analysis. Values were expressed as percentage of predicted. 12 Pulmonary function was measured using a portable spirometer (One Flow ® ; Clement Clark, UnitedKingdom), following the Brazilian Society of Pneumology and Phthisiology’s recommendations. 13 The following parameters were measured – forced vital capacity (FVC) in liters, forced expiratory volume in 1 second (FEV1) in liters, FEV1/FVC ratio (%). Values were expressed as percentage of predicted. 14 Respiratory muscle strength was measured using a calibrated hand-held respiratory pressure meter (Globalmed ® MVD300). Maximal inspiratory pressure (MIP) and maximal expiratory pressure (MEP) were measured based on the residual volume and total pulmonary capacity, respectively. Three reproducible measurements (less than 10% variation) were obtained; a mean of 3 to 5 measurements were taken, and the highest value was considered for analysis. The values were compared with those predicted. 15 Symptoms of depression and anxiety were assessed by the Beck depression inventory (BDI) 16 and Beck anxiety inventory (BAI), 17 respectively. These are self- scored instruments composed of 21 items that evaluate depression and anxiety symptoms. Each item has four possible answers with scores ranging from 0 to 3 points, from absent (0) to most severe (3). The total score ranges from 0 to 63, with higher scores indicating more severe symptoms. 16 Analysis of data Statistical analysis was performed using the Statistical Package of Social Sciences (SPSS), version 23.0. Normally distributed continuous variables were presented as absolute numbers and percentages. Normality of distribution was verified by the Kolmogorov Smirnov test. The t-test for independent samples and the Mann- Whitney test was used for comparisons of normally distributed continuous variables and those without normal distribution, respectively. Bivariate correlations analysis was performed by Pearson’s correlation (normally distributed variables) and Spearman’s correlation (variables without normal distribution). Multiple linear regression analysis was used to determine the predictive value of the variables on 6MWT, and the stepwise method was used for selection of the variables. All assumptions made by the multiple linear regression analysis - linear relationship, homoscedasticity, no or little multicollinearity between exposure variables were considered for selection of the model with the highest predictive value. Variables with p < 0.10 were excluded from the model. In addition, hierarchical multiple linear regression analysis was performed to verify the predictive value of sociodemographic, clinical, physical functional and emotional data combined. The level of significance adopted in all tests was 5%. Results Mean age of participants was 56.7 ± 12.4 years; 58% were younger than 60 years, 65.4% were male (Table 1). Table 2 describes emotional, quality of life, and physical functional characteristics, Borg scale and 6MWD.