Comparison of radiography and scintigraphy in the diagnosis of dental disorders in the horse.

Abstract: Scintigraphy, with 99mTechnetium methylenediphosphonate (99mTc-MDP) and 99mTc-labelled leucocytes, was compared to radiography in the diagnosis of dental disease in the horse in a prospective case-controlled study, comprising 30 horses with clinical signs of dental disease and 30 control horses. In each case, right and left lateral, ventral and dorsal soft tissue and bone phase scintigraphic images were obtained after i.v. injection of 1 GBq/100 kg bwt 99mTc-MDP, using a gamma camera. The same views were acquired in 10 horses with clinical signs of dental disease and 12 control horses after injection of 99mTc-labelled leucocytes. Standard radiographic projections of the paranasal sinuses and of the apices of the maxillary and mandibular teeth were obtained. The scintigraphs and radiographs were assessed subjectively by 2 board-certified surgeons and one board-certified radiologist, with extensive experience of equine radiology, from who the clinical history was withheld. Sensitivity, specificity and kappa, as a measure of agreement, were calculated for the different methods. Bone phase images were also scored subjectively on a scale from 0 to 3 on the basis of isotope uptake over the teeth. Regions of interest were defined over the teeth, and normal teeth compared to diseased counterparts. Total scintigraphic counts were related to the age of the animal and to the disease process. Differences in density ratios between left and right teeth were evaluated using the Mann-Whitney Test. Dental disease was confirmed in 22 horses at surgery or postmortem examination. Horses with dental disease showed a significant increase in scintigraphic activity over the affected tooth compared to the contralateral tooth, with a typical pattern for different diseases. The sensitivity of scintigraphy with 99mTc-MDP proved to be excellent (95.5%), whereas the specificity was moderate (86.4%). In contrast, radiography had excellent specificity (95.0%) and a low sensitivity (51.5%). The greatest sensitivity and specificity were achieved by evaluating radiographs and scintigrams together. The objective scintigraphic density ratios were found to be significantly different between diseased and control horses. The results of this study suggest that, if a density ratio of 1.5 or greater between a suspected diseased tooth and its contralateral number is regarded as abnormal, only 1% false positive diagnoses and 20% false negative diagnoses will occur. In contrast, scintigraphy with 99mTc-labelled leucocytes was not very successful, due to the lack of anatomical detail provided by this technique, which made identification of the diseased tooth impossible. Accurate radiographic interpretation of dental disease presents difficulties, both in terms of missed diagnoses and mistaken diagnoses. Scintigraphy complements radiographic examination of dental structures by providing information important for accurate diagnosis and is, therefore, conceived to be essential for selection of the appropriate treatment for dental disease.