STRUCTURE.

Study characteristics
Methods	Study design: two‐arm, individual, RCT Centres: Poland, "of each centre" suggests multicentre trial, no details Start of enrolment: Novemer 2011 End of enrolment: February 2015 Mean follow‐up: 6 months Run‐in period: not reported
Participants	Inclusion criteria: "Patients who presented with signs or symptoms of HF (dyspnea, fatigue, and exercise intolerance) consistent with New York Heart Association functional class II or III, with preserved LV ejection fraction (> 50%), and with evidence of diastolic dysfunction, were considered suitable for screening." Exclusion criteria: "Exclusion criteria were: Atrial fibrillation or flutter Resting heart rate > 90 beats/min Ischemic heart disease (defined by a positive coronary angiogram or inducible ischemia during exercise testing) Moderate or worse valvular heart disease Primary myocardial diseases Established or suspected pulmonary diseases (spirometry results < 80% of age‐ and sex‐specific reference values) Hemoglobin ≤ 11 g/dl Adrenocortical, hepatic, rheumatic, neoplastic, skeletal, thyroid, and renal diseases (including renal insufficiency with serum creatinine > 1.5 mg/dl [132 mmol/l]) Hyperkalemia > 5.0 mmol/l Known intolerance or treatment with an MRA within the last 3 months Concomitant therapy with a potassium‐sparing agent Current lithium use Pregnancy" Randomised (N): 150 (75 intervention, 75 control) Withdrawn (N): for reasons other than death 12 (7 intervention, 5 control) Lost to follow‐up (N): 7 (4 intervention, 3 control) Analysed (N): 131 (64 intervention, 67 control) Age (years, mean, SD): intervention: 66.3, 7.7; control: 67.6, 9.1 Sex (% men): intervention: 12; control: 19 Ethnicity (%): not reported Systolic blood pressure (mmHg, mean, SD): intervention: 131, 15; control: 130, 18 Heart rate (beats/min, mean, SD): intervention: 72, 10; control: 73, 10 BMI (mean, SD): intervention: 30.7, 4.5; control: 29.7, 4.6 Serum creatinine (mg/dL, mean, SD): intervention: 0.99, 0.20; control: 1.03, 0.24 B‐type natriuretic peptide (pg/mL, median, IQR): intervention: 40 (26‐63); control: 54 (27‐99) NT pro B‐type natriuretic peptide (pg/mL): not reported LVEF (%, median, IQR): intervention: 72.6 (70.4–74.8); control: 71.4 (69.2–73.5) NYHA class I (%): 0 NYHA class II (%): intervention: 78; control: 79 NYHA class III (%): intervention: 22; control: 21 NYHA class IV (%): 0 Hypertension (%): intervention: 92; control: 91 Diabetes (%): intervention: 39; control: 40 Atrial fibrillation (%): not reported Hospitalisation for heart failure (%): intervention: 17; control: 21 Coronary heart disease (%): significant CAD excluded Stroke (%): not reported Diuretic (%); intervention: thiazides 54, loop 13; control: thiazides 46, loop 18 Digoxin (%): not reported Beta‐blocker (%): intervention: 78; control: 72 ACEI/ARB (%): intervention: 97; control: 95 MRA (%): study drug
Interventions	Intervention: spironolactone, 25mg/day Comparator: matching placebo (120 mg/day of microcellulose) Concomitant medication: "Enrollees continued to receive other prescribed treatments throughout the study period."
Outcomes	Planned: unclear Reported: "Coprimary outcomes were change at 6 months in exercise capacity (assessed by peak VO2) and exertional E/e' (reflecting LVFP). The secondary outcomes included change at follow‐up in exercise blood pressure (BP) response and post‐treatment global longitudinal myocardial deformation (GLS) measured by 2‐dimensional strain."
Notes	Emailed investigator to ask for additional outcome data relevant to this review. No response.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"The study coordinator, who was not involved in study procedures, was responsible for drug randomization and dispensing"
Allocation concealment (selection bias)	Low risk	"sequentially‐numbered, opaque, sealed envelopes"
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"Patients and investigators performing the assessments and data analysis were blinded to group assignment."
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"core laboratory in Hobart, Australia, for independent adjudication of the primary endpoint"
Incomplete outcome data (attrition bias) All outcomes	Low risk	ITT not used, withdrawals reported with reasons, lost to follow‐up reported, similar numbers for treatment arms
Selective reporting (reporting bias)	Unclear risk	unclear, clinical trial registration was post‐hoc
Other bias	Low risk	"The authors have reported that they have no relationships relevant to the contents of this paper to disclose" "This study was funded by grants ST‐678 from Wroclaw Medical University and 13‐024 from the Royal Hobart Hospital Foundation."