Kosmala 2013.

Study characteristics
Methods	Study design: RCT Unit of randomisation: No information Total duration of study: Screening from December 2011 to December 2012 Run‐in period: No information Intervention time: 7 days Follow‐up: No information Setting: Wroclaw Medical University, Wroclaw, Poland University of Queensland, Department of Medicine, Brisbane, Australia University of Tasmania, Menzies Research Institute Tasmania, Hobart, Australia
Participants	Type of heart failure: CHF N = 61 (ivabradine and BB: 30; placebo: 31) Mean age: 67 ± 8 years Gender: 11 (18%) male, 50 (82%) female Severity of condition: HFpEF ≥ 50% Inclusion criteria: Met the exercise capacity and post‐exercise LV filling pressure ratio criteria Categorised in NYHA functional II or III Exclusion criteria: Absence of stable sinus rhythm Ischaemic heart disease (excluded on the basis of the absence of significant atherosclerotic lesions on coronary angiography and no evidence of inducible ischaemia during exercise testing) Moderate and severe valvular heart disease Heart rate < 60 bpm Sick sinus syndrome Second‐degree and third‐degree atrioventricular block Severe obesity (body mass index > 36 kg/m2) Established or suspected pulmonary diseases (vital capacity < 80% or forced expiratory volume in 1 second < 80% of age‐specific and sex‐specific reference values) Haemoglobin 11 g/dL Treatment with non‐dihydropyridine calcium‐channel blockers, class I antiarrhythmic agents, strong inhibitors of cytochrome P450 3A4, and QT interval–prolonging medications Withdrawals: None
Interventions	Intervention: Ivabradine 5 mg twice a day Comparison: Placebo Concomitant medications: BB Excluded medications: Non‐dihydropyridine calcium‐channel blockers Class I antiarrhythmic agents Strong inhibitors of cytochrome P450 3A4 QT interval–prolonging medications
Outcomes	Outcomes and time points measured in the study: [0d, 7d] Exercise capacity Ratio of peak early diastolic mitral flow velocity to peak early diastolic mitral annular velocity HR pVO2 Postexercise LV filling pressure Alterations in myocardial deformation LV systolic and diastolic function Plasma BNP Conclusion: In participants with HFpEF, short‐term treatment with ivabradine increased exercise capacity, with a contribution from improved LV filling pressure response to exercise as reflected by the ratio of peak early diastolic mitral flow velocity to peak early diastolic mitral annular velocity. Because this patient population is symptomatic on exertion, therapeutic treatments targeting abnormal exercise haemodynamic status may prove useful. Ivabradine therapy is an effective therapy to increase exercise tolerance in patients with HFpEF. This beneficial effect is potentially mediated by the improved LV filling pressure response to exercise.
Notes	Funding for trial: Paid with internal funds from Wroclaw Medical University and Brisbane University. (W Kosmala via email on 22 November 2018) Notable conflicts of interest of authors: No information Contact to authors/unpublished data: We contacted W Kosmala via email on 22 November 2018 to ask for funding and missing data. W Kosmala answered on 22 November 2018, providing information about funding and that no other outcome data were available.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"The procedure of randomization to receive either ivabradine 5 mg or placebo twice daily was performed by computerized sequence generation."
Allocation concealment (selection bias)	Unclear risk	Insufficient information to base judgement
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"The present study was designed as a prospective, blinded, parallel‐group, placebo‐controlled trial." "The hospital pharmacies were responsible for drug randomization and dispensing, and both the investigators and patients were blinded to the treatment option."
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Insufficient information to base judgement
Incomplete outcome data (attrition bias) All outcomes	Low risk	Less than 20% missing data. Outcomes reported for 61 of 61 participants (100%).
Selective reporting (reporting bias)	Low risk	All outcomes stated in the methods section were adequately reported or explained in the results.
Other bias	Low risk	Paid with internal funds from Wroclaw Medical University and Brisbane University