Accuracy of Confocal and Dermoscopic Criteria for Predicting Basal Cell Carcinoma Subtype

The correct diagnosis of basal cell carcinoma (BCC) histopathological subtype is crucial for treatment design. Although the diagnostic accuracy of reflectance confocal microscopy (RCM) and dermoscopy for BCC has been studied, the literature of RCM and dermoscopy in predicting BCC subtype remain limited.

To assess the diagnostic accuracy of confocal and dermoscopic criteria for predicting BCC subtype.

Confocal and dermoscopic images of histopathologically confirmed BCC were retrospectively analyzed. Multivariate and adjusted odds ratios were computed, along with sensitivity and specificity. Receiver operating characteristic (ROC) curve was plotted, and the area under the curve (AUC) was calculated.

The study included 103 patients with 124 lesions. And 84 (105 lesions) were assigned to the confocal analysis group and 49 (62 lesions) to the dermoscopic analysis group. Multivariate analysis yielded three confocal models for predicting a diagnosis of nodular BCC (nBCC) with sensitivity of 91.7% and specificity of 80.7% (AUC, 0.913), superficial BCC (sBCC) with sensitivity of 82.3% and specificity of 91.5% (AUC, 0.932), and aggressive BCC (aBCC) with sensitivity of 87.5% and specificity of 97.8% (AUC, 0.945). We developed three dermoscopic models for diagnosing nBCC with sensitivity of 76.5% and specificity of 82.1% (AUC, 0.891), sBCC with sensitivity of 95.0% and specificity of 95.2% (AUC, 0.973), and aBCC with sensitivity of 62.5% and specificity of 90.7% (AUC, 0.849).

Both RCM and dermoscopy are reliable tools for the identification of BCC subtypes. RCM is more accurate for nBCC and aBCC, whereas dermoscopy is more accurate for sBCC.