Observer performance in the assessment of periapical pathology: a comparison of panoramic with periapical radiography

Observer performance in the assessment of the periapical pathology from panoramic and periapical radiography was examined. Five endodontists, five general practitioners and five oral radiologists were asked to assess the periapical status of 117 teeth. The observers assessed the panoramic and periapical radiographs of the teeth, which were evenly distributed throughout the jaws with a 50% probability that either an osteolytic or sclerotic lesion was present. The results of the comparison between panoramic and periapical radiography were influenced by the selection of observers. When the oral radiologists acted as observers, the mean P(A) value for periapical radiography was higher than for panoramic radiography (P less than 0.001), resulting in periapical radiography presenting a higher overall diagnostic accuracy than panoramic radiography for all 15 observers (P less than 0.01). There was, however, no difference between panoramic and periapical radiography when the two groups of endodontists and general practitioners acted as observers. The comparison of the three groups of observers showed no difference between their diagnostic accuracy when assessing panoramic radiographs. With periapical radiography, the oral radiologists demonstrated a higher diagnostic accuracy than the endodontists (P less than 0.05). The observers in each group with the highest diagnostic accuracy also had the highest intra-observer agreement. The mean intra-observer agreement of the five general practitioners was higher than those of the other two groups of observers for panoramic radiography. For periapical radiography, the mean agreement rates of the groups were comparable.