Abstract Introduction and purpose Cardiac cell therapy to patients with heart failure after myocardial infarction (MI), has shown little improvement in cardiac function. The rate of engraftment after cell transplantation (TX) to the ischemic heart is low because of the leakage of transplanted cells without scaffold, the friction between the graft and thorax, and poor vascularization to the graft. We created a novel TX method (Figure 1) to transplant adipose-derived stromal cells (ASCs) with scaffold into the pericardial space where the pericardium protected the graft. We evaluated the therapeutic efficacy of pericardial TX of ASCs on the cardiac function and remodeling after MI. Methods We isolated ASCs from the brown adipose tissue of donor mice (C57BL6-Tg (CAG-EGFP), five weeks). ASCs increased and differentiated into spontaneously beating myocytes, endothelial cells, and pericytes three weeks after ex vivo culture. Cells were trypsinized and mixed with self-assembling peptides scaffold. Each graft (100μl gel scaffold) had 1×106 ASCs. To make recipient MI mice (C57BL6, 12 weeks), we ligated the left coronary artery through small chest incision without cutting rib bone to avoid postoperative adhesions. We confirmed the deterioration of cardiac function by echocardiography (n=21) and MRI (n=4) three weeks after MI. Then, recipient MI mice had TX (syngeneic, no immunosuppression). We opened the pericardium and transplanted the graft into the pericardial space. By suturing the pericardium, the graft was fixed on the MI scar area (Figure 1). We labeled donor cells with GFP and the scaffold with biotin. We evaluated cardiac function of TX group (n=10) and control group (MI with sham TX, n=11) by echocardiography. Results In the scaffold, donor cells increased three days to two weeks after TX, and slightly decreased four weeks after TX. The graft thickness was 0.9±0.2mm (two weeks after TX) and 0.7±0.2mm (four weeks after TX). There were many vWF positive vessels in the scaffold and some of which were GFP positive. Echocardiography showed that left ventricular diastolic dimension (LVDd) of TX group, did not increase four weeks after TX (ΔLVDd = −0.02mm, P=0.80). While, LVDd of control group significantly increased (ΔLVDd = +0.23mm, P=0.02) due to cardiac remodeling after MI. MRI confirmed the increase in left ventricular wall thickness in the MI scar area up to 4 weeks in TX group (n=2). Conclusions Pericardial transplantation of ASCs prevents cardiac remodeling. Its beneficial effect might be mediated by improved rate of engraftment, neovascularization, and increased ventricular wall thickness in the MI scar area. Acknowledgement/Funding Grants-in-Aid for Scientific Research