Development of a histomorphologic scale to quantify cutaneous scars after burns.

OBJECTIVE: Cutaneous wound healing in adults invariably results in scarring; however, there are few scales to quantify the degree of such scarring. The authors developed a histomorphologic scale for quantifying scarring after cutaneous burn injury.
METHODS: As part of a randomized trial comparing a variety of burn therapies, 40 partial-thickness burns were created on the backs and flanks of anesthetized pigs and treated with a tissue adhesive, antibiotic ointment, occlusive dressing, or dry gauze. Gross scar appearance was independently assessed by two investigators at 90 days on a 100-mm visual analog scale (VAS) marked "best appearance" at the high end. One of the investigators repeated the observation 30 days later. Full-thickness biopsies were taken 90 days after injury and evaluated histologically by a dermatopathologist for the presence of hyperkeratosis, epidermal hyperplasia, presence and depth of scar (defined as abnormally oriented collagen under polarized light), fibroplasia, vascular proliferation, and absence of adnexa, including hair follicles, apocrine glands, and smooth muscles. One point was assigned for each category in the presence of a normal finding, whereas an abnormal finding was assigned a score of zero. The normal dermis (absence of abnormal collagen) was given a score of 3, while decreasing scores of 2 to 0 were given for progressively deeper scars (i.e., 2 for papillary dermis, 1 for upper half of reticular dermis, and 0 for deep dermal lower half). The total histomorphologic score was derived by adding the scores on the individual items. The score ranges from 0 to 10 from worst scarring to absence of scarring, respectively. A subset of observations was evaluated a second time by one of the observers one month later. Intraobserver reliability of the histomorphologic scale was assessed with Spearman's correlation. Inter- and intraobserver Pearson's correlations for the gross scar VAS were calculated, and the correlation between gross and histomorphologic scores was assessed.
RESULTS: Intraobserver correlation for individual histomorphologic categories ranged from 0.19 to 1.00. Intraobserver correlation for the total histologic score was 0.95. Inter- and intraobserver correlations for the gross scar VAS were 0.8 each. Correlation between the histomorphologic scale and the gross scar VAS was 0.38.
CONCLUSIONS: A new reliable histomorphologic method for quantifying and scoring cutaneous scars is described together with a reliable scar VAS. However, these two scales are not highly correlated.