Oral cytology shows high sensitivity and specificity for detecting oral epithelial dysplasia in meta-analysis

This meta-analysis of 26 studies evaluated the diagnostic accuracy of oral cytology for detecting oral epithelial dysplasia or worse, analyzing data from 8397 specimens. The analysis used a bivariate random-effects model to pool results across studies, though specific population characteristics and clinical settings were not reported. The primary outcome was diagnostic accuracy for low-grade squamous intraepithelial lesion or worse, with secondary analyses examining accuracy for high-grade lesions and specifically for cytology using Papanicolaou staining.

The pooled sensitivity for detecting oral epithelial dysplasia or worse was 93.8%, with specificity of 87.7%. For high-grade squamous intraepithelial lesion or worse, sensitivity increased to 95.7%. Oral cytology using Papanicolaou staining showed sensitivity of 94.8%. These results represent pooled estimates from multiple studies, though exact numbers of true positives, false positives, and confidence intervals were not reported in the available data.

Safety and tolerability data were not reported in the abstract. Key limitations were not specified, though the authors note that the accuracy of oral cytology remains controversial. The analysis did not report on population demographics, clinical settings, or comparator tests, limiting generalizability to specific patient groups or practice environments.

Oral cytology is used in screening programs for oral cancer, and these findings suggest it has high diagnostic accuracy for detecting epithelial dysplasia. However, clinicians should interpret these results cautiously as they represent pooled estimates from heterogeneous studies with unreported limitations. The evidence supports oral cytology's potential role in detection but does not establish its impact on clinical outcomes or mortality.