Previous studies have shown that addition of transforming growth factor-beta3 (TGF-beta3) is capable of reducing scar tissue formation in skin defects. Therefore, we examined whether TGF-beta3 can also influence the organization of a fibrous capsule around implants in vivo. For this reason, 24 silicone implants with microgrooves with a groove depth of 1.0 microm and a ridge and groove width of 10.0 microm were made and loaded with human recombinant TGF-beta3 (0, 5, 50, and 250 ng). An in vitro release enzyme-linked immunosorbent assay (ELISA) test was done with another 10 implants to estimate the amount of TGF released from the implants. The implants were inserted subcutaneously in the backs of 6 guinea pigs. Each animal received four implants, which were left in place for 10 weeks. At the end of the implantation time, the implants were retrieved, embedded, and processed for histology. Histomorphometrical measurements were done on the capsule formation and the implant-cell interface quality and quantity. The results showed a fibrous capsule of 15 microm up to 50 microm thickness around all implants. There were no significant differences between the TGF-beta3-loaded implants or the controls. Frequently, inflammatory cells were present in the capsule. The implant-tissue interface was on average between 5 and 15 microm thick and consisted mostly out of one or two layers of macrophages or foreign body giant cells. Statistical analysis again showed no significant differences between the various TGF-beta3-coated implants and controls. Finally, we concluded that a microtextured surface can indeed be used for the release of TGF-beta3. On the other hand, this did not result in major differences in wound healing between implants loaded with 5, 50, or 250 ng of TGF-beta3 and controls.