ABSTRACT
BACKGROUND: The diagnosis of actinic keratosis (AK) is based on clinical evaluation and confirmed by histopathological analysis (HA). The challenge is to establish the correct diagnosis with a minimally invasive assessment. The aim of this study is to validate the analysis of AK by reflectance confocal microscopy (RCM), a cellular resolution, noninvasive imaging method and to determine the relevant parameters for diagnosis, compared to HA, by calculating the sensitivity (S), specificity (E), positive predictive value (PPV), and negative predictive value (NPV) of each criterion. MATERIALS AND METHODS: Through clinical examination, 25 AKs were selected for dermoscopy and RCM evaluation followed by shaving excision for HA. Statistical analysis was done by hypothesis tests (McNemar for binary and Wilcoxon for continuous variables). RESULTS: There was no significant difference between RCM and HA for 5 of the 6 parameters analyzed. The criteria that were statistically relevant were as follows: parakeratosis (p-value 0.449690; S 90%; PPV 78.26%), hyperkeratosis (p-value 0.248213; S 87.5%; E 100%; PPV 100%; NPV 25%), dyskeratosis (p-value 0.617075; S 85.71%; E 75%; PPV 94.74%; NPV 50%), spinous layer keratinocyte atypia classified as mild, moderate or severe (P-value 0.145032) and inflammation in epidermis (P-value 1.000000; S 75%; E 20%; PPV 78.95%; NPV 16.67%). RCM could not adequately measure inflammation in dermis (P-value 0.013328), despite good sensitivity (68%) and PPV (100%). CONCLUSION: RCM proved to be an effective method for the diagnosis of AK, contributing to the selection of the most appropriate treatment option.
