OBJECTIVES: To determine the diagnostic accuracy of mandibular and maxillary bone density in detecting osteoporosis using receiver operating characteristic (ROC) analysis. METHODS: 671 women between 45 years and 70 years of age underwent dual energy X-ray absorptiometry (DXA) of the hip and lumbar spine. This was the gold standard for diagnosing osteoporosis. Intraoral radiography of the upper and lower right premolar region was performed, using an aluminium wedge as a densitometric reference. Jaw bone density was determined using dedicated software. Observer differences and ROC curves were analysed. RESULTS: For detecting osteoporosis using jaw bone density, the area under the ROC curve (A(z)) was 0.705. For separate analysis of mandibular and maxillary films, sensitivity varied from 33.9% to 38.7% and specificity from 83.5% to 85.3% when using a threshold of 4.3 mm Al equivalent. CONCLUSIONS: Density of the premolar region reaches a fair diagnostic accuracy, which might improve when including additional factors in the analysis and refining the densitometric tool.
OBJECTIVES: To determine the diagnostic accuracy of mandibular and maxillary bone density in detecting osteoporosis using receiver operating characteristic (ROC) analysis. METHODS: 671 women between 45 years and 70 years of age underwent dual energy X-ray absorptiometry (DXA) of the hip and lumbar spine. This was the gold standard for diagnosing osteoporosis. Intraoral radiography of the upper and lower right premolar region was performed, using an aluminium wedge as a densitometric reference. Jaw bone density was determined using dedicated software. Observer differences and ROC curves were analysed. RESULTS: For detecting osteoporosis using jaw bone density, the area under the ROC curve (A(z)) was 0.705. For separate analysis of mandibular and maxillary films, sensitivity varied from 33.9% to 38.7% and specificity from 83.5% to 85.3% when using a threshold of 4.3 mm Al equivalent. CONCLUSIONS: Density of the premolar region reaches a fair diagnostic accuracy, which might improve when including additional factors in the analysis and refining the densitometric tool.
Authors: Eliete N S Guerra; Fabiana T Almeida; Fernanda V Bezerra; Paulo T D S Figueiredo; Maria A G Silva; Graziela De Luca Canto; Camila Pachêco-Pereira; André F Leite Journal: Dentomaxillofac Radiol Date: 2017-07-03 Impact factor: 2.419