Correlation between WHO and Modified WHO Classification Systems in the Histopathologic Diagnosis of Oral Lichen Planus Using Intraobserver and Interobserver Variability

Oral lichen planus (OLP) is a relatively common chronic autoimmune disease. In the present study, we tried to correlate the histopathological criteria of WHO and modified WHO (mod.WHO) classification systems using two methods, namely, intraobserver and interobserver observations in these samples.

This cross‑sectional study was performed on 64 microscopic slides with the diagnosis of the OLP lesions, based on both clinical and histopathological features. At first, each pathologist individually (as intraobserver) examined microscopic slides based on both histopathologic diagnostic criteria. Later, three pathologists in a group (as interobserver) reevaluated microscopic slides 2 months later in the second phase of the study, based on both systems. Eventually, the findings were statistically analyzed with Cohen’s kappa coefficient (κ) and reported.

According to the results, the lichen planus was detected in 8 cases using the WHO method, and in 41 cases using the Mod.WHO method. Intrarater Kappa coefficients were κ = 0.114, P = 0.299; κ = 0.181, P = 0.012; and κ = 0.062, P = 0.424 for three pathologists, respectively. The findings showed no reproducibility (κ = 0.148, P = 0.024) and there was no correlation between the two systems. Statistical analysis revealed that the histopathological criteria of the WHO classification for detecting the lichen planus microscopy were more sensitive but the Mod.WHO classification criteria were more specific for detecting the lichen planus.

Due to the higher specificity of the histopathological criteria of Mod.WHO classification rather than WHO classification, it seems that Mod.WHO classification has more important and useful criteria for histopathological diagnosis. Finally, we can conclude that the use histopathologic criteria of the Mod.WHO classification is more useful in the diagnosis of lichen planus, although it should be in combination with clinical information.