Background  Slow breathing (SLOWB) alleviates symptoms of chronic heart failure (HF) but its long-term effects are unknown. We examined the acute and long-term impact of device-guided breathing on hemodynamics and prognostic parameters in HF patients with reduced ejection fraction (HFrEF).Methods and results  Twenty-one patients with HFrEF (23.9 ± 5.8%, SD ± mean) on optimal medical therapy underwent blood pressure (BP), heart rate (HR), HR variability, 6-min walk test (6MWT), cardiopulmonary exercise testing (CPET), and echocardiography measurements before and 3 months after SLOWB home training (30 min daily). After 3 months, all patients were assigned to continue SLOWB (Group 1) or no-SLOWB (Group 2). All tests were repeated after 6 months. Acute SLOWB (18 ± 5 vs 8 ± 2 breaths/min, P < 0.001) had no influence on BP and HR but improved saturation (97 ± 2 vs 98 ± 2%, P = 0.01). Long-term SLOWB reduced office systolic BP (P < 0.001) but not central or ambulatory systolic BP. SLOWB reduced SDNN/RMSSD ratio (P < 0.05) after 3 months. One-way repeated measures of ANOVA revealed a significant increase in 6MWT and peak RER (respiratory exchange ratio) from baseline to 6-month follow-up in group 1 (P < 0.05) but not group 2 (P = 0.85 for 6MWT, P = 0.69 for RER). No significant changes in echocardiography were noted at follow-up. No HF worsening, rehospitalisation or death occurred in group 1 out to 6-month follow-up. Two hospitalizations for HF decompensation and two deaths ensued in group 2 between 3- and 6-month follow-up. Conclusions  SLOWB training improves cardiorespiratory capacity and appears to slow the progression of HFrEF. Further long-term outcome studies are required to confirm the benefits of paced breathing in HFrEF.