Background: Rapid and effective closure of full-thickness burn wounds remains a limiting factor in burns of greater than 50% of the total body surface area (TBSA). Hypothetically, cultured skin substitutes (CSS) consisting of autologous cultured keratinocytes and fibroblasts attached to collagen-based sponges may reduce requirements for donor skin, and morbidity from autograft harvesting and widely-meshed skin grafts.
Methods: To test this hypothesis, CSS were prepared from split-thickness skin biopsies collected after enrollment of 40 burn patients by informed consent into a study protocol approved by the local Institutional Review Boards of three participating hospitals. CSS and split-thickness skin autograft (AG) were applied in a matched-pair design to patients with full-thickness burns involving a mean value of 73.4% of the TBSA. Data collection consisted of photographs, area measurements of donor skin and healed wounds after grafting, qualitative outcome by the Vancouver Scale for burn scar, and biopsies of healed skin.
Results: Engraftment at postoperative day (POD) 14 was 81.5 +/- 2.1% for CSS and 94.7 +/- 2.0 for AG. Percentage TBSA closed at POD 28 was 20.5 +/- 2.5% for CSS, and 52.1 +/- 2.0 for AG. The ratio of closed to donor areas at POD 28 was 66.2 +/- 8.4 for CSS, and 4.0 +/- 0.0 for each harvest of AG. Each of these values was significantly different between the graft types. Correlation of percent TBSA closed with CSS at POD 28 with percent TBSA full-thickness burn generated an r value of 0.37 (p < 0.0001). Vancouver Scale scores at 1 year after were not different for erythema, pliability, or scar height, but pigmentation remained deficient in CSS.
Conclusions: These results demonstrate that CSS reduce requirements for donor skin harvesting for grafting of excised, full-thickness burns of greater than 50% TBSA with qualitative outcome that is comparable to meshed AG. Availability of CSS for treatment of extensive, deep burns may reduce time to wound closure, morbidity, and mortality in this patient population.