Nine RCTs (n=463) were included in the review. Study quality varied: two studies scored 1, one study scored 2, five studies scored 3 and one study scored 4. Only two studies reported the randomisation method in sufficient detail. Recruitment was unclear in five studies. Details of exercise adherence were not reported in six studies. In three studies changes in drug therapy were stable.
Exercise training significantly lowered BNP (MD -79pg/mL, 95% CI -141 to - 17, I2=73%; five studies) and NT-pro-BNP (MD -621pg/mL, 95% CI -844 to -398, I2=75%; six studies); significant heterogeneity was noted for both comparisons.
Sensitivity analysis showed changes in BNP and NT-pro-BNP (four studies) remained statistically significant. Significant heterogeneity was noted for both comparisons (I2=77% and I2=84%).