Tissue engineering, a field that combines polymer scaffolds with isolated cell populations to create new tissue, may be applied to soft-tissue augmentation-an area in which polymers and cell populations have been injected independently. We have developed an inbred rat model in which the subcutaneous injection of a hydrogel, a form of polymer, under vacuum permits direct comparison of different materials in terms of both histologic behavior and their ability to maintain the specific shape and volume of a construct. Using this model, we compared three forms of calcium alginate, a synthetic hydrogel, over an 8-week period-standard alginate that was gelled following injection into animals (alginate post-gel), standard alginate that was gelled before injection into animals (alginate pre-gel) and alginate-RGD, to which the cell adhesion tripeptide RGD was linked covalently (RGD post-gel). Parallel groups that included cultured syngeneic fibroblasts suspended within each of these three gels were also evaluated (alginate post-gel plus cells, alginate pre-gel plus cells, and RGD post-gel plus cells). The study used 54 inbred Lewis rats (n = 9 for each of the six groups). Construct geometry was optimally maintained in the alginate post-gel group in which 58 percent of the original volume was preserved at 8 weeks and increased to 88 percent at 8 weeks when syngeneic fibroblasts were included within the gel. Volume was not as well preserved in the RGD post-gel group (25 percent of original volume at 8 weeks), but again increased when syngeneic fibroblasts were included (41 percent of original volume at 8 weeks). Maintenance of volume was poorest in the alginate pre-gel group (31 percent of original volume at 8 weeks) and failed to be augmented by the addition of fibroblasts (19 percent of original volume at 8 weeks). Histologically, the gel remained a uniform sheet surrounded by a fibrous capsule in the alginate post-gel groups. In the alginate pre-gel and RGD post-gel groups, there was significant ingrowth of a fibrovascular stroma into the gel with fragmentation of the construct. In constructs in which syngeneic fibroblasts were included, cells were visualized throughout the gel but did not extend processes or appear to contribute to new tissue formation. Material compression testing indicated that the alginate and RGD post-gel constructs became stiffer over a 12-week period, particularly in the cell-containing groups. Our results suggest that calcium alginate could be a suitable agent for soft-tissue augmentation when gelled subcutaneously following injection. The addition of syngeneic fibroblasts enhanced the ability of the gel to maintain the volume of a construct; this seems to be mediated by increased gel stiffness rather than by de novo tissue formation. Our animal model, in combination with material testing data, permits rigorous comparison of different materials used for soft-tissue augmentation.