ABC | Volume 110, Nº5, May 2018

Original Article Oliveira et al Exercise in acute heart failure Arq Bras Cardiol. 2018; 110(5):467-475 Table 1 – Baseline characteristics of hospitalized acute heart failure patients allocated into one of the three groups – exercise training + non‑invasive ventilation (ET+NIV), ET + Sham or Control group Control (n = 9) ET+Sham (n = 9) ET+NIV (n = 11) Anthropometrics/Demographics Male, n (%) 7 (78%) 8 (89%) 7 (64%) Age, years 58 ± 7 57 ± 5 56 ± 8 Weight, kg 65.3 ± 14.8 74.0 ± 13.5 66.4 ± 10.8 Height, m 1.60 ± 0.71 1.68 ± 0.10 1.64 ± 0.40 BMI, kg/m 2 24.2 ± 5.0 26.9 ± 4.6 24.8 ± 4.0 LVEF, % 23.8 ± 4.9 25.4 ± 6.7 26.0 ± 4.8 NTpro-BNP, ρg/mL 2467 ± 547 2331 ± 429 2594 ± 633 hs-CRP, mg/L 8 ± 3 9 ± 4 9 ± 5 Length of stay, days 39 ± 15 23 ± 8* 17 ± 10* † Main comorbidities Hypertension, n (%) 5 (56%) 3 (33%) 5 (54%) Dyslipidemia, n (%) 4 (44%) 1 (11%) 1 (9%) Diabetes mellitus, n (%) 2 (22%) 2 (22%) 1 (9%) Etiology Ischemic, n (%) 6 (67%) 7 (80%) 7 (44%) Main medications β-blocker, n (%) 7 (78%) 6 (67%) 8 (73%) ACE inhibitors or ARBs, n (%) 4 (43%) 6 (63%) 7 (64%) Diuretics, n (%) 9 (100%) 9 (100%) 11 (100%) Definition of abbreviations: BMI: body mass index; LVEF: left ventricular ejection fraction; NTpro-BNP: brain natriuretic peptide; hs-CPR: high sensitive C-reactive protein; ACE: angiotensin conversor enzyme; ARBs: angiotensin II receptor blockers. Values expressed as mean ± standard deviation or frequency (n and %). Repeated-measures ANOVA with appropriate Bonferroni corrections was applied to variables described as mean ± standard deviation and the chi-square test was used to assess differences in categorical data. * p < 0.05 vs. Control; † p < 0.05 vs. ET+Sham Statistical analysis Statistical analysis was carried out with the SPSS software (version 20.0, SPSS Inc., USA). Data were expressed as mean ± standard deviation or as median and interquartile range, as appropriate, and categorical data are expressed as frequency (n and %). The normality of data distribution was determined by Shapiro-Wilk test. The chi-square test was used to assess differences between categorical data, and repeated-measures ANOVA followed by Bonferroni corrections were used for multiple comparisons. Pearson’s correlation was used for parametric correlations. For all analyses, statistical significance was set at 5% (p < 0.05). Results Baseline measures Twenty-nine patients who fulfilled all the inclusion criteria were enrolled in the study and randomized into three groups: Control (n = 9, 58 ± 7 years of age), ET+Sham (n = 9, 57 ± 5 years) and ET+NIV (n = 11, 56 ± 8 years). All patients had diagnosis of acute HF. There were no differences in anthropometric and demographic variables, cause of HF, LVEF, main comorbidities, medications, and NT‑proBNP or hs-CRP plasma levels among groups (Table 1). The functional class, exercise tolerance and pulmonary function were not different among groups (Table 2). Effects of exercise training associated with NIV and sham ventilation None of the patients of group ET+NIV or ET+Sham had any criteria for exercise interruption. Total exercise time was shorter in the ET+Sham group (~30% lower compared to ET+NIV) (Table 2). On D10, the ET+NIV and ET+Sham groups had a greater walking distance compared to the control group (Table 2). In addition, ∆6MWT distance on D10 was greater in the ET+NIV group (Figure 1, Panel C) than in the ET+Sham group (Figure 1, Panel B) and the control group (Figure 1, Panel A). There were no differences in blood pressure, HR and SpO 2 during 6MWT between the groups (data not shown). Dyspnea score at rest was higher at baseline (D1) and decreased over time in all three groups. Moreover, ET+NIV group had the lowest dyspnea value on D10 (Figure 2). The number of patients receiving dobutamine infusion at D1 was similar among groups; however, on D10 the exercise 469