Upadhya 2017.

Methods	Study design: parallel, individual, RCT Centres: not reported Start of enrolment: not reported End of enrolment: not reported Follow‐up: 9 months Run‐in period: not reported
Participants	Inclusion criteria: "HFpEF was defined as history, symptoms, and signs of HF, a preserved LVEF of 50% or greater and no evidence of other medical condition that could mimic HF symptoms" Exclusion criteria: "Coronary disease was excluded according to history, medical record, electrocardiogram, and rest and exercise echocardiogram" "Exclusions included aldosterone antagonist use within the previous 3 months, a known contraindication, concomitant therapy with a potassium‐sparing diuretic or potassium supplementation, baseline serum potassium level greater than 5.0 mEq/L, or serum creatinine level of 2.5 mg/dL or greater." Randomised (N): 80 (42 intervention, 38 control) Withdrawn (N): for reasons other than death 9 (5 intervention (adverse event N = 1, patient choice N = 4), 4 control (patient choice N = 3, death N = 1)) Lost to follow‐up (N): not reported Analysed (N): 71 (37 intervention, 34 control) Age (years, mean, SD): intervention: 70.0, 1.1; control: 72.0, 1.2 Sex (% men): intervention: 19; control: 21 Ethnicity (%): African American: intervention: 21 , control: 37 Systolic blood pressure (mmHg, mean, SD): intervention: 139, 2.7; control: 143, 3.2 Heart rate not reported BMI (mean, SD): intervention: 31.5, 0.8; control: 32.4, 1.2 Serum creatinine not reported B‐type natriuretic peptide (unit not reported): intervention: 55, 46; control: 61, 50 NT pro B‐type natriuretic peptide (pg/mL): not reported LVEF (%, mean, SD): intervention: 62.6, 1.1; control: 62.0, 1.1 NYHA class I (%): 0 NYHA class II (%): intervention: 29; control: 26 NYHA class III (%): intervention: 64; control: 63 NYHA class IV (%): 0 Hypertension (%): intervention: 83; control: 92 Diabetes (%): intervention: 17; control: 29 Atrial fibrillation (%): not reported Hospitalisation for heart failure: not reported Coronary heart disease (%): 0 Stroke (%): not reported Diuretic (%); intervention: 74; control: 71 Digoxin (%): intervention: 2; control: 0 Beta‐blocker (%): intervention: 31; control: 32 ACEI (%): not reported ARB (%): not reported MRA (%): study drug
Interventions	Intervention: spironolactone. "The starting dose of spironolactone was 12.5 mg/d in individuals with baseline creatinine of 2.0 mg/dL or greater or potassium greater than 4.5 mEq/L; in all other participants, the starting dose was 25 mg/d. In participants who initiated therapy with the 12.5‐mg/d dose, the dose was increased to 25 mg/d once creatinine fell below 2.5 mg/dL and potassium fell below 5.0 mEq/L and maintained at that dosage as long as those levels were maintained. Spironolactone was discontinued if 1‐week creatinine was 2.5 mg/ dL or higher or potassium was 5.0 mEq/L or higher. " "The mean daily dose of spironolactone was 24.3 2.9 mg/d." Comparator: matching placebo Concomitant medication: not reported
Outcomes	Planned: July 2005, NCT record: primary outcomes: exercise intolerance, quality of life Reported: exercise performance, aortic distensibility and LV structure and function, carotid artery stiffness, pulse wave velocity, LV diastolic filling, QoL
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"randomly assigned"
Allocation concealment (selection bias)	Unclear risk	not reported
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"The research pharmacy prepared and distributed placebo and active drug using a secure methodology. All investigators, staff, and participants were fully blinded to treatment group assignment throughout the study period"
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	"The research pharmacy prepared and distributed placebo and active drug using a secure methodology. All investigators, staff, and participants were fully blinded to treatment group assignment throughout the study period"
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	unable to assess
Selective reporting (reporting bias)	Unclear risk	posthoc clinical trial registration
Other bias	High risk	"This study was funded by the National Institutes of Health (NIH; R01AG18915), the Claude D. Pepper Older Americans Independence Center, Wake Forest University (P30AG21332), the Clinical and Translational Science Institute, Wake Forest School of Medicine (NIH UL1TR001420), and the Kermit G. Phillips Chair in Cardiovascular Medicine of Wake Forest School of Medicine" Published as conference abstract only.