Toluidine blue staining for detection of oral premalignant lesions and carcinomas.

The discriminative value of toluidine blue vital staining for detecting oral premalignant lesions and carcinomas was tested in the hamster DMBA buccal pouch model. The stain technique was further tested for specificity using the turpentine liquid petrolatum (TLP) hamster cheek pouch model for hyperplasia. Although carcinomas frequently exhibited toluidine blue staining, a 27.8% false negative result was noted. Premalignant lesions (dysplasia and carcinoma in situ) seldom retained toluidine blue stain and yielded a histologically confirmed (95.2%) false negative result. DMBA induced papillomas yielded a 5% false positive result. The TLP model failed to produce either clinically or histologically evident hyperplastic/hyperkeratotic lesions.