Systematic Review and Meta-Analysis of Bone Marrow-Derived Mononuclear Cells in Animal Models of Ischemic Stroke

Background and Purpose - Bone marrow-derived mononuclear cells (BMMNCs) offer the promise of augmenting poststroke recovery. There is mounting evidence of safety and efficacy of BMMNCs from preclinical studies of ischemic stroke; however, their pooled effects have not been described. Methods - Using Preferred Reporting Items for Systematic Review and Meta-Analysis guidelines, we conducted a systematic review of preclinical literature for intravenous use of BMMNCs followed by meta-analyses of histological and behavioral outcomes. Studies were selected based on predefined criteria. Data were abstracted by 2 independent investigators. After quality assessment, the pooled effects were generated using mixed-effect models. Impact of possible biases on estimated effect size was evaluated. Results - Standardized mean difference and 95% confidence interval for reduction in lesion volume was significantly beneficial for BMMNC treatment (standardized mean difference: -3.3; 95% confidence interval, -4.3 to -2.3). n=113 each for BMMNC and controls. BMMNC-treated animals (n=161) also had improved function measured by cylinder test (standardized mean difference: -2.4; 95% confidence interval, -3.1 to -1.6), as compared with controls (n=205). A trend for benefit was observed for adhesive removal test and neurological deficit score. Study quality score (median: 6; Q1-Q3: 5-7) was correlated with year of publication. There was funnel plot asymmetry; however, the pooled effects were robust to the correction of this bias and remained significant in favor of BMMNC treatment. Conclusions - BMMNCs demonstrate beneficial effects across histological and behavioral outcomes in animal ischemic stroke models. Although study quality has improved over time, considerable degree of heterogeneity calls for standardization in the conduct and reporting of experimentation.