Margins of oral leukoplakia: autofluorescence and histopathology.

Autofluorescence devices are widely used to examine oral lesions. The aim of this study was to see whether there were any signs of dysplasia, parakeratosis, or mucosal inflammation in the borders of homogeneous oral leukoplakia using autofluorescence, and we also compared clinically visible extensions with those detected by autofluorescence. Twenty patients with 26 homogeneous areas of oral leukoplakia were included in the study. After the clinically visible extensions of the lesion had been marked, we took a photograph through the autofluorescence device, which showed both borders in one picture. We then used photo-editing software to measure the size of the area of leukoplakia together with the area with loss of autofluorescence. We took 3 punch biopsy specimens: one from the leukoplakia, one 2.5mm from its marked borders, and one from healthy mucosa. Seventy-eight biopsy specimens were examined by an experienced pathologist, and 95% CI calculated to assess the amount of parakeratosis. Spearman's rank correlation was used to assess the association with mucosal inflammation. Ten areas of leukoplakia were surrounded by normal green autofluorescence, and 16 were consistent with loss of autofluorescence with a mean size of 66%, which exceeded the clinically visible size of the area of leukoplakia. We calculated that there was a strong association between these entities and their surrounding areas, with loss of autofluorescence for parakeratosis. Some leukoplakias showed clinically invisible extensions during histopathological examination and autofluorescence. The technique described enables clinicians to measure the extent of these lesions beyond their visible margins. We found no dysplasia, which emphasises that autofluorescence detects non-dysplastic lesions caused by mucosal inflammation and parakeratosis.