Comparison of Different Centrifugation Protocols for the Best Yield of Adipose-Derived Stromal Cells from Lipoaspirates

Sir: Since 1893,1 autologous fat transplantation has been used for various applications in surgery.2 It was initially used as filler material for simple soft-tissue augmentation, but the capacity of transplanted fat to provide long-term tissue repair and regeneration has recently emerged.3 The extraordinary plasticity of fat is attributable to the presence of adipose-derived stromal cells endowed with an immunophenotypical profile and multilineage plasticity analogous to those of bone marrow–derived mesenchymal stem cells.4 Centrifugation entered clinical practice as an empirically accepted step of pretransplantation fat treatment5 that will preserve the viability and self-renewing capacity of adipose-derived stromal cells, but the rationale for this procedure has not yet been investigated. Here, we comparatively quantified the viability and self-renewal of adipose-derived stromal cells from human fat samples centrifuged at 3000, 6000, or 10,000 rpm, corresponding to 1126, 12,519, and 4507 g, respectively. Centrifugation at 3000, 6000, or 10,000 rpm for 3 minutes reduced the volume of harvested fat to 51.55 ± 6.95 percent, 44.58 ± 5.05 percent, or 40.46 ± 7.99 percent of its initial volume, respectively. The difference in volume reduction between 3000- and 10,000-rpm protocols proved to be statistically significant (n = 7) (p = 0.01). After collagenase digestion of lipoaspirate samples, we counted the total number of cells composing the stromal vascular fraction. Samples centrifuged at 10,000 rpm (n = 7) provided 75.76 ± 13.84 percent of the number of stromal vascular fraction cells in fat samples centrifuged at 3000 rpm. In the two noncentrifuged samples, we counted on average 0.85 × 106 cells, which was also lower than the number obtained from fat samples centrifuged at 3000 rpm (1.01 × 106 ± 0.68 × 106). The samples centrifuged at 6000 rpm gave variable results: in one case, the 6000 rpm–centrifuged sample provided more stromal vascular fraction cells than the 3000 rpm–centrifuged sample (0.57 × 106 and 0.47 × 106, respectively); in two cases, we extracted more stromal vascular fraction cells from samples centrifuged at 3000 rpm than from those centrifuged at 6000 rpm. After counting, three aliquots of 1 × 105 stromal vascular fraction cells for each sample were seeded in three 10-mm cell culture dishes and the numbers of viable and self-renewing adipose-derived stromal cells were estimated by counting adherent cells (Fig. 1) or colony-forming units (Fig. 2) (n = 7) on each dish after 10 days of culture.Fig. 1.: Adipose-derived stromal cell viability after centrifugation. Ten days after stromal vascular fraction seeding, the total number of cells in selected areas was counted.Fig. 2.: Ten days after stromal vascular fraction seeding, the fibroblastic colony-forming unit number was counted. The arrows indicated the magnification of a single colony.Both methods demonstrated a significantly higher number of adherent cells or colonies in samples centrifuged at 3000 rpm than in those centrifuged at 10,000 rpm (Table 1). The yield of adherent cells proved to be lower in 10,000 rpm–centrifuged than in 3000 rpm–centrifuged samples in all cases (80.5 ± 12.2 percent). The number of fibroblastic colony-forming units was also lower in all cases (60.59 ± 21.09 percent). The yield of fibroblastic colony-forming units from the two samples that were not centrifuged was 70 ± 23.9 percent of the yield of the sample from the same patient that was centrifuged at 3000 rpm. The yield of the three samples centrifuged at 6000 rpm appeared similar (105.92 ± 42.69 percent) to that of samples centrifuged at 3000 rpm, but the estimate was seriously affected by the high variability of the few (n = 3) samples analyzed.Table 1: Adipose-Derived Stromal Cell Viability after CentrifugationIn conclusion, we demonstrated that pretransplantation fat sample centrifugation at 3000 rpm allows the best compromise between fat volume reduction and a good yield of viable and self-renewing adipose-derived stromal cells. M. Galiè, Ph.D. Department of Morphological and Biomedical Sciences Anatomy and Histology Section University of Verona Marco Pignatti, M.D. Plastic and Reconstructive Surgery (Second Division) City Hospital Verona, Italy Ilaria Scambi, Ph.D. Andrea Sbarbati, M.D., Ph.D. Department of Morphological and Biomedical Sciences Anatomy and Histology Section University of Verona Gino Rigotti, M.D. Plastic and Reconstructive Surgery (Second Division) City Hospital Verona, Italy