Two methods of evaluating mandibular trabecular pattern in intraoral radiographs and the association to fragility fractures during a 47‐year follow up

• Published in: European journal of oral sciences Vol. 129; no. 5; pp. e12801 - n/a
• Main Authors: Elleby, Charlotta, Skott, Pia, Jonasson, Grethe, Theobald, Holger, Nyrén, Sven, Salminen, Helena
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 01.10.2021
• Subjects: Automation; Biomedical materials; bone mineral density; Cancellous bone; Dentistry; Female; Follow-Up Studies; Fractures; Fragility; Humans; Longitudinal Studies; longitudinal study; Mandible; Medicin och hälsovetenskap; Odontologi; Osteoporosis; Radiographs; Radiography; bone mineral density; longitudinal study; cancellous bone; osteoporosis
• ISSN: 0909-8836; 1600-0722; 1600-0722
• DOI: 10.1111/eos.12801

In this longitudinal cohort study, we explored the association of fragility fractures to sparse trabecular bone pattern in intraoral radiographs using two methods, a visual and a semi‐automated. Our aim was to study both sexes and to include younger age‐groups, during a follow‐up time of 47 years. The cohort consisted of 837 men and women aged 18–65 years, with intraoral radiographs from 1970–1971. The trabecular pattern was assessed in the mandibular premolar region with a visual and a semi‐automated method. Data on fragility fractures were acquired from the Swedish National Patient Register for 47 years of follow‐up time. Sparse trabecular pattern was found in 2.2% of the cohort using the visual method, and 18% were deemed at ‘risk of osteoporosis’ using the semi‐automated method. A total of 132 individuals suffered at least one fragility fracture during the follow‐up period. We found no significant association between fractures and sparse trabecular pattern using either method. This study shows that visual assessment, as a predictor of future fractures, may not be a suitable method for individuals of all ages and sexes. As for the semi‐automated method, there is still very limited evidence for its fracture predictive ability.