Detecting women at risk for osteoporosis using dental radiographs
Two researchers at the University of Manchester, UK, School of Dentistry, discuss how analyzing panoramic dental radiographs for early signs of osteoporosis and combining this information with a few simple clinical questions can help clinicians identify at-risk patients, leading to early medical intervention and possible fracture prevention.
By Hugh Devlin, PhD, MSc., BSc., BDS, and Keith Horner, BChD, MSc., PhD, FDSRCPSGlasg, FRCR, DDR, University of Manchester, UK, School of Dentistry
Osteoporosis is an insidious disease that mainly afflicts older women, resulting in bone fractures, a loss of height, and a disfiguring stooping appearance. Unfortunately, affected individuals are often unaware that they have osteoporosis until they sustain a bone fracture following mild trauma. Most authorities agree that mass population screening would not be cost-effective, and identifying and treating those most at risk is the best strategy. Research at the University of Manchester has tried to come up with a solution by analysing panoramic dental radiographs (see Fig. 1) for early signs of the disease, and combining this information with a few simple clinical questions. Using the radiograph, software is used to semi-automatically identify and measure the thickness of the lower border of the mandible (the cortical width). The limits of “normal” thickness are known. In previous publications we have used the term “Osteodent index” to describe the value of the combined radiographic and clinical measurements. Once patients at risk of osteoporosis have been identified, they can be treated by lifestyle advice or other appropriate medical intervention so that fractures can be prevented.
Fig. 1 — This is a typical panoramic dental radiograph where the thickness of the lower border of the mandible is reduced as a result of osteoporosis. The measurement can be made automatically using software or manually. The manual measurement of cortical width is made at a reference point at right angles to the lower border of the mandible below the mental foramen (circled in the figure). This patient’s mandibular cortical width (2.8 mm) is thinner than the mean width for females her age. She has osteoporosis.
Previously, there was great emphasis on measurement of bone mineral density at fracture-prone sites such as the hip and lumbar spine. However, fractures can occur in those people whose bone density lies within the normal range. The emphasis has now shifted toward identifying an individual’s clinical risk of fracture, in recognition that risk factors other than the density of a bone are important in determining fracture risk. Our research has used a simple questionnaire to estimate the osteoporosis risk of our participants, based on the person’s race and age, weight, whether there is a history of rheumatoid arthritis or a previous non-traumatic fracture, and whether the patient has received estrogen therapy. Other indices are available and vary in complexity. Most of these questionnaires are powerful indicators of a person’s risk of osteoporosis, but have not been adopted widely.
Worldwide, dentists take many millions of panoramic dental radiographs annually. When discussing our research, we have been careful to mention that patients should not receive a radiograph just for the purposes of osteoporosis diagnosis. Our osteoporosis assessment should be made only on patients who need to have a panoramic radiograph taken for other diagnostic reasons. This avoids unnecessary patient exposure to radiation.
So how good is this diagnostic test? There is some overlap of osteoporotic and non-osteoporotic clinical scores and cortical width values, which makes it difficult to identify a specific threshold value for referral. In a previous publication, a threshold value of the Osteodent index giving a false positive rate of 10% resulted in an ability to detect 69% of patients with osteoporosis.1 Even though statistical techniques on large samples of patients show the methodology works quite well, there will always be those patients given a false diagnosis of osteoporosis who are unnecessarily worried by the “high risk” label. Similarly, an incorrect diagnosis of the patient being unaffected by osteoporosis, when in reality she has osteoporosis, is called a false negative diagnosis. Both types of false diagnoses will limit accuracy and should be discussed openly with patients.
When comparing any new test, the diagnostic accuracy should be compared with the established “best practice.” FRAX is the established case-finding strategy for osteoporosis detection. This method uses a series of clinical questions (with the inclusion of femoral neck bone mineral density), which allows the 10-year probability of hip fracture to be calculated (http://www.shef.ac.uk/FRAX/). The FRAX Web site is linked to the NOGG management guidance tool (http://www.shef.ac.uk/NOGG), allowing a classification of our subjects into either “treat” or “lifestyle advice and reassurance” categories. We compared our diagnostic method with this gold standard diagnostic method and found that there was a similar diagnostic performance between our index and FRAX.2 Our methodology has the advantage of not requiring a bone density measurement of the femoral neck, although FRAX can be calculated without this figure. The results of our research suggest that the Osteodent index has value in prediction of hip fracture risk.
A dental “checkup” may be different in the future. Many dentists already provide oral cancer checks, tobacco smoking cessation advice, and routinely advise patients to contact their doctor when they suspect disease. We believe that dentists can play a role in the detection of those at risk of osteoporosis by identifying and referring these patients to their general medical practitioner for further management advice. How patients might react to being told by their dentist that they may have osteoporosis has not yet been assessed in a detailed way, but their reaction to this new diagnostic methodology is crucial to whether it is accepted into clinical practice. The referral route of an “at risk” osteoporotic patient from general dental practitioner to medical doctor and then onward to specialist physicians has not been tested for cost-effectiveness. The health economic arguments would be important to health authorities in helping them determine whether they should spend their scarce resources on financing this technology.
Dr. Hugh Devlin Dr. Keith Horner
Dr. Hugh Devlin is a professor of restorative dentistry and Dr. Keith Horner is a professor of oral and maxillofacial imaging, both at the University of Manchester, UK, School of Dentistry. Their research seeks to prevent painful osteoporotic bone fractures and the loss of independence caused by a hip fracture. Many osteoporotic women may also suffer stooping disfigurement. Hip fractures are serious incidents, as they carry an increased risk of mortality. Dentists can detect osteoporosis on dental radiographs at an early stage and can then arrange early preventive treatment. Their research group has developed an automated method of computer assessment, which automatically alerts the dentist when a patient has high-risk features of osteoporosis.
1. Devlin H, Allen PD, Graham J, Jacobs R, Karayianni K, Lindh C, van der Stelt PF, Marjanovic E, Adams JE, Pavitt S, Horner K. The role of the dental surgeon in detecting osteoporosis: the OSTEODENT study. British Dental Journal. 2008; 204:E16. DOI: 10.1038/sj.bdj.2008.317.
2. Horner K, Allen PD, Graham J, Jacobs R, Boonen S, Pavitt S, Nackaerts O, Marjanovic E, Adams JE, Karayianni K, Lindh C, van der Stelt PF, Devlin H. The relationship between the OSTEODENT index and hip fracture risk assessment using FRAX. Oral Surg, Oral Med Oral Pathol Oral Radiol Endod. In Press.