Accuracy of the Panoramic Radiograph in Diagnosing Condylar Asymmetries

Document Type




Format of Original

26 cm

Publication Date



SAGE Publications

Source Publication

Journal of Dental Research

Source ISSN



Objectives: The purpose of this craniometric study was to analyze the accuracy of panoramic radiograph in diagnosing mandibular ramal and condlyar vertical asymmetries.

Methods: In this study the accuracy of vertical condylar measurements is assessed on the OP1OO using human skulls. Three different head positions are used: normal (according to manufacturers instructions), tilted 10°, and rotated 10°. Condylar ratios are measured from the skulls and compared to the commensurate measurements on the radiographs according to the method of Kjellberg. The asymmetry between left and right condyles is determined by the following formula: [(Ratio R-Ratio L)/(Ratio R + Ratio L)] X 100. The hypothesis tested was that the use of condylar ratios and asymmetry index has no clarifying effect on the magnification and distortion effects of poor head position when evaluating the vertical height of the condyle as shown on the OP100. Multivariate tests of variance (MANOVA) were used: Pillai's Trace, Wilks' Lambda Test, Hotelling's Trace, and Roy's Largest Root. In addition, a Factor Analysis between each of the head position groups was assessed.

Results: The results of this study revealed that, 1) The vertical symmetry relationship of condyle to ramus is not reproduced correctly on the OPG in any of the head positions used (p = 0.375 for Normal, p = 0.810 for Rotated, and p = 0.531 for Tilted). 2) Due to the inaccuracy in determining condylar ratios, it is concluded that the formula [(Ratio R-Ratio L)/(Ratio R+Ratio L)] X 100 is not appropriate for assessing left-right condylar height asymmetry.

Conclusion: Many false positive and false negative results were found in this study, thus one can conclude that the validity of detecting vertical asymmetries of the condyle using current panoramic head positioning techniques in the OP1OO is low.


Journal of Dental Research, Vol. 81, No. 1 (2002): A91. DOI.