After a decade of controversial results substantial clinical evidence has become available to evaluate the long-term efficacy of autologous stem cell therapy following acute myocardial infarction (AMI) and to conduct extensive sensitivity analyses to determine sources of bias in the current clinical evidence. This meta-analysis is an update of a previous Cochrane Systematic Review (published in 2008) where thirteen randomized controlled trials (RCTs) were eligible for inclusion.Method
MEDLINE (1950-January 2011), EMBASE (1974-January 2011), CENTRAL (The Cochrane Library 2011, Issue 1), CINAHL (1982-January 2011), and ongoing trials registers were searched for randomised trials of stem cells as treatment for AMI. Additionally, hand searching was used to screen relevant conference proceedings (2005-2011). Randomized trials (RCTs) where participants with AMI received autologous bone marrow stem cells (BMSC)where included. Three independent reviewers undertook study screening, selection, quality assessment and data abstraction. Disagreements were solved by consensus and the advice of a fourth reviewer. Meta-analyses using random-effects models and pre-planned sub-group analyses were conducted.Results
Thirty-three trials (thirty-nine comparisons) with a total of 1,765 participants were included. Overall, this meta-analysis showed no clinical efficacy of BMSC treatment but the intervention significantly improved left ventricular volumes and ejection fraction and significantly reduced infarct size and wall motion score in long-term follow-up. There was no evidence of bias due to publication date or time-lag, methodological quality of included studies, participant drop-out, duration of follow-up or date of the first disclosure of results. However, the treatment seemed more effective when administered at doses greater than 10e8 cells and to patients with more severe heart dysfunction (LVEF ≤ 40%).Conclusions
BMSC treatment achieved a significant long-term improvement of heart contractility but no clinical efficacy in patients with AMI. No significant sources of bias were identified.Funding
NIHR programme grant to NHS Blood and Transplant (RP-PG-0310-1001). The views expressed in this publication are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.