MRC National Survey of Health and Development: hip and spine morphologies from DXA images related to osteoporosis, osteoarthritis and low back pain

The shapes of our hips and the lower part of our back largely determine our flexibility and how easily we can move. Diseases such as osteoarthritis not only destroy the tissues in our joints but also change the shape of the joint, which is at least part of the reason sufferers become stiff and movement becomes painful. In other situations, the shape of the joint may make it more likely that falling leads to a fracture of the hip, for instance, when bones are weakened by osteoporosis. We have also recently shown that we each have our own distinctive shape to our lower backs and we are currently exploring whether the particular shape might make back pain or injury more likely. What we don't know, however, is what determines the shapes of our joints. How much is inherited? How much depends on what we eat; or our lifestyle - active or sedentary, or other diseases from which we may suffer as we get older?

A wonderful resource is available to help us start to tackle these questions: the National Survey of Health and Development. This is a group of individuals who have been studied at intervals from when they were born in 1946. They have recently had x-ray images taken of their hips and their spine and we want to make measurements of the shape of the hip and the curviness of the spine. We do this using a computer program we have written to perform a process called 'Shape Modelling'. This software is trained to identify the shape of interest (hip or spine) and from lots of images uses a mathematical technique to look for features that describe the shape. This gives a numerical description of the shape that we can then compare with other measurements. In the National Survey of Health and Disease a wealth of social, behavioural and biomedical data have been collected over the years, although this is the first time that x-ray pictures of the joints have been taken. By looking to see if individuals with particular features to the shape of their hips, for instance, have other things in common (for example: male or female, low birth weight, obesity) we can begin to see whether these things might have affected the way the joints have acquired the shape they have. This, in turn, might help us to understand why some people are more prone to osteoarthritis, osteoporosis or low back pain.

At present, we do not understand what starts off these disorders. The results from this study might give us some clues to this and from there we can begin to work out if it is possible to avoid some of the 'risk factors'. This might be possible using existing or new drugs but, more importantly, might equally be a matter of diet or exercise.

This project will develop and apply Active Shape Models to dual-energy x-ray absorptiometry (DXA) images of the hip and spine recently acquired for the MRC National Survey of Health and Developmental (NSHD). This cohort was recruited in 1946 and has been followed-up regularly since. They are now approaching the age where diseases associated with ageing are becoming increasingly important. Our specific interests are musculoskeletal disorders, especially osteoarthritis, osteoporosis and low back pain. From the images, mode scores will be derived using Active Shape Modelling, which includes principal components analysis, to provide a quantitative description of the morphology of each joint. Mode scores will be calculated for each image and submitted to the NSHD database. These scores will provide a valuable description of joint shapes in a cross-section of the community at age 65 that can be related to social, behavioural and biomedical data collected over the previous 67 years and will form a useful baseline for future recall of these individuals. Associations between joint shape and life course factors that might affect the shape will be explored using adaptations of existing life course models. These will explore factors such as those related to birth, early and teenage development, body size, activity and strength.

This research programme is aimed at identifying inherited, developmental and lifestyle factors affecting the morphology of hips and the lumbar spine. These are sites of highly prevalent morbidities; osteoarthritis, osteoporosis and low back pain. As such this research will have long-term impact on a variety of different groups including: project staff, other academic beneficiaries, every individual going about their daily activities, patients and clinicians and wider society. Outcomes may inform an assessment of diet and lifestyle as environmental modifiers of joint shape. For prevention of spinal injuries and back pain, a better understanding of the role of lumbar morphology in methods of lifting is expected to influence workplace manual handling guidelines.

Our wider goal is to develop imaging biomarkers that can identify individuals at greatest risk of disorders of the musculoskeletal system. These biomarkers would enable early identification of risk and assessment of avoidance measures to reduce that risk. There is an urgent need for biomarkers for the assessment of prognosis and disease progression for those affected, especially for OA, and early therapeutic intervention where this is available. We will ensure that potentially commercially valuable data are protected whilst maintaining a policy of data sharing and publication in line with MRC best practice which will beneficially impact the wider research community.

One of our key areas of societal impact that we will achieve within the 3 year programme centres on our communications plans for public engagement for which both centres have an extensive track record. Dissemination in Aberdeen will be primarily through our dynamic presentation 'The secret life of the skeleton', and public seminars. We will especially try to inform those volunteers within the NSHD cohort that have given so much over many years to make this whole study the success that it is.

Within the grant period we will also provide beneficial impact on staff associated with the research; the PDRAs, the investigators and members of their respective research groups. In particular this will involve multidisciplinary, internationally leading research training for the PDRAs. This training will provide long term benefit to the research community and the staff themselves will acquire skills that can be applied in other employment sectors a well as academia.