Background The objective of the study was to determine whether the effects of infarct-related artery (IRA) infusion of autologous bone marrow–derived CD34+ cells after ST elevation myocardial infarction (STEMI) are dependent on the dose (quantity and mobility) of the cells infused. Beneficial effects of IRA infusion of mononuclear cells after STEMI have been inconsistent, possibly because of differences in timing, cell type, quantity, and mobility of infused cells. Methods Patients were randomized to bone marrow harvest (n = 16) or control (n = 15). At a median of 8.3 days after coronary stenting for STEMI, CD34+ cells were infused via the IRA at 3 dose levels (5, 10, and 15 × 106 ) in cohorts of 5 patients each. Baseline and follow-up imaging and ex vivo CD34+ cell mobility were performed. Results Cell harvest and infusion were safe. Quantitative rest hypoperfusion score measured by single-photon emission computed tomography improved at 6 months in the ≥10 million cohorts compared with controls (−256 vs +14, P = .02). There was a trend toward improved ejection fraction at 6 months (+4.5%) in the ≥10 million cohorts compared with no change in the controls and 5 million cohort (+0.7%). Improved perfusion and infarct size reduction correlated with the quantity and mobility of the infused CD34+ cells. Conclusions The effects of CD34+ cell IRA infusion during the repair phase after STEMI are dose dependent and, at a threshold dose of 10 million CD34+ cells, associated with a significant improvement in perfusion that may limit deterioration in cardiac function (IRA infusion of CD34+ cells in patients with acute myocardial infarction AMR-01 NCT00313339).