Sürder 2013.

Methods	Type of study: parallel RCT Type of publication: full Source of funding: funded by Fondazione Cardiocentro Ticino, Lugano, Switzerland; Zurich Heart House‐Foundation for Cardiovascular Research, Zurich, Switzerland; Bern University Hospital, Bern, Switzerland; Cardiovascular Research Foundation, Zurich, Switzerland, and an unrestricted grant from Abbott Vascular Country of origin: Switzerland Number of centres: 4 Dates of trial enrolment: 10/06 to 01/12 Length of follow‐up: 4 months Number (N) of participants randomised to each arm: 66 in the early cell therapy arm, 67 in the late cell therapy arm, 67 in the control arm Number (N) of participants analysed (primary outcome) in each arm: 58 in the early cell therapy arm, 49 in the late cell therapy arm, 60 in the control arm
Participants	Population: STEMI with PCI in 24 hours and EF ≤ 45% Age, mean (SD) each arm: median 55 (IQR 15) years (early cells), 62 (IQR 15) years (late cells), 56 (IQR 14.5) years (controls) Sex, % male in each arm: 86.2% (early cells), 82.5% (late cells), 83.6% (controls) Number of diseased vessels: 1 (54%), 2 (32%), 3 (14%) (early cells), (57%), 2 (27%), 3 (16%) (late cells), 1 (64%), 2 (21%), 3 (15%) (controls) Number of stunned hyperkinetic, etc segments: not reported Time from symptom onset to initial treatment: 6 (2) days (early cells) or 24 (7) days (late cells) after AMI Statistically significant baseline imbalances between the groups? Higher age in the late treatment group compared with controls (median 62 years versus 56 years; P value = 0.06); lower percentage of smokers in the late treatment group compared with controls (40.3% versus 62.7%; P value = 0.01); higher baseline LVEF in the control group compared with the treatment group (median 39.6% versus 35.6%, P value = 0.03)
Interventions	Intervention arm: BMMNC Type of stem cells: bone marrow‐derived stem cells (mononuclear cells‐MNC) Summary of how stem cells were isolated and type and route of delivery: bone marrow aspiration was performed 5 to 7 days after AMI. Between 60 and 80 mL of bone marrow was collected from the iliac crest under local anaesthesia. Then 1 mL of a solution containing 1000 IU heparin was added to each 10 mL of bone marrow aspirate to prevent clotting. Then the aspirate and 20 mL of the patient's serum were sent at room temperature by courier to the cell‐processing centre. The BM‐MNC cell suspension was shipped back to the participating hospital within 24 hours. Briefly, with the use of density gradient centrifugation, the mononuclear cell fraction was re‐suspended in 10 mL of serum‐free medium with 20% of autologous serum added without any additional heparin. An aliquot of cell suspension was utilised for fluorescence‐activated cell sorting analysis with the use of fluorochrome conjugated antibodies against anti‐human CD34 and CD133; cell viability was assessed by 7‐AAD cell uptake, and sterility was assessed by the Bact/Alert rapid method. Release criteria of the BMMNC were product sterility, a cell count between 5 × 107 and 5 × 108, and cell viability of ≥ 80% Dose of stem cells: 1.59 (± 1.25) x 108 cells (early cells); 1.39 (± 1.20) x 108 cells (late cells) Timing of stem cell procedure: 5 to 7 days post‐AMI (early cells); 3 to 4 weeks post‐AMI (late cells) Comparator arm: no additional therapy (control)
Outcomes	Primary outcomes: absolute change in global LVEF from baseline to 4 months Secondary outcomes: change in LVEF, LVESV, LVEDV infarct size proportion of scar mass to total LV mass, global and regional myocardial thickening, major adverse events Outcome assessment points: 4 and 12 months Method(s): MRI
Notes	Data from the 2 active intervention arms of the trial are pooled in this review. There is a discrepancy between the absolute change LVEF values and baseline/endpoint values reported. The authors were contacted to request clarification on this discrepancy but none was forthcoming
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation was performed using closed envelopes in a 1:1:1 pattern
Allocation concealment (selection bias)	Low risk	Closed envelopes were used
Blinding (performance bias and detection bias) All outcomes	High risk	The trial was described as "open label"; controls did not undergo bone marrow aspiration and no placebo was administered; neither participants nor patients were blinded. However, it is reported that "the entire analysis was performed in a CMR core laboratory, blinded to the treatment assignment of the patients enrolled."
Incomplete outcome data (attrition bias) All outcomes	High risk	In the analysis of clinical outcomes, the number of withdrawals and exclusions was unbalanced between groups (early cells: 11/66 versus late cells: 15/67 versus control: 7/67). Although reasons for withdrawals were given (withdrawal of informed consent or death in all missing patients), these do not fully explain the sample sizes described in individual analyses. In the analysis of scientific outcomes by MRI analysis at 4 months, 8 additional patients were missing in the BMSC arm due to the lack of paired MRI data
Selective reporting (reporting bias)	Low risk	All outcomes described in the trial protocol (www.clinicaltrials.gov: NCT00355186) were reported
Other bias	Low risk	None reported or identified