Evaluation of a training program for detection of carotid artery calcifications on panoramic radiographs

Objective. Td evaluate the effectiveness of the American Academy of Oral an d Maxillofacial Radiology-sponsored training packet for identification of c arotid artery calcifications on panoramic radiographs. Study design. Two examiners, who completed the training (trainees), examine d 778 panoramic radiographs. The sample included 298 men, with a mean age o f 66, and 480 women, with a mean age of 68. Findings were compared with tho se obtained by an oral and maxillofacial radiologist. A kappa statistic was used to determine agreement between the 2 trainees. The positive predictiv e value (PPV) of the program was estimated by comparing the trainees rating of disease status with an expert in case identification. Results. Examiners 1 and 2 identified 99 and 78 positive cases, respectivel y. A kappa statistic of 0.87 (95% Cl, 0.81-0.92) was obtained, indicating g ood interexaminer agreement. The expert identified 27 positive cases, resul ting in a PPV of 34.6% (95% Cl, 24.4-46.3). Conclusion. Although the training packet offers valuable training, it does not provide a high PPV, suggesting the need to modify it or to seek an expe rt opinion before classification of a patient as having calcification on a panoramic radiograph.