Purpose: Major drawbacks of conventional bone marrow stromal cells (BSCs) transplantation method are mainly caused by direct transplanted cell to host cell interactions. We hypothesized that separation of the transplanted cells by a microporous membrane might inhibit most of the potential adverse effects and induce superior effect. The purpose of the study is to determine the optimal condition of the microporous membrane. Methods: First, BSCs were placed in polyethylene terephthalate (PET) transwell inserts with 3, 8, or $12{mu}m$ pore size, and cultured in 24 well culture plates. After 5 days, bottoms of the plates were observed for presence of attached BSCs in monolayer and cell numbers were evaluated. Second, BSCs were placed PET, polycarbonate (PCT), and mixed cellulose esters (MCE) transwell inserts with 3 and $8{mu}m$ pore size, and cultured in 24 well culture plates. After 3 days, the supernatants of the media left in culture plate were analyzed for collagen, vascular endothelial growth factor (VEGF), platelet derived growth factor BB (PDGF-BB), and basic fibroblast growth factor (bFGF). Third, BSCs were placed in 15% and 70% of the PET membrane with $3{mu}m$ pore size. All the experimental conditions and methods were same as the second study. Results: The optimal pore sizes to prevent BSC leakage were $3{mu}m$ and $8{mu}m$. The amounts of type I collagen and three growth factors tested did not show significant differences among PET, PCT, and MCE groups. However, the collagen, VEGF, and bFGF levels were much higher in the high (70%) density group than in the low (15%) density group. Conclusion: This study revealed that the optimal pore size of membrane to prevent direct BSC to recipient cell contact is in between $3{mu}m$ and $8{mu}m$. Membrane materials and pore sizes do not influence the collagen and growth factor passage through the membrane. The most striking factor for collagen and growth factor transport is pore density of the membrane.