Background: Rare monogenic disorders such as sclerosteosis have been reported to lead to an increase in bone mass, in association with abnormal facial features such as mandibular overgrowth (1). We have recently observed a similar association in adults identified within the population as having elevated bone mineral density (BMD) (2). More detailed analysis of this cohort of high bone mass subjects revealed increased thickness and circumference of cortical bone, which might conceivably also be related to growth of skull bones. Since facial growth is largely complete by age 15, particularly in girls, analysing associations between indices of face shape obtained in the ALSPAC cohort at age 15, and bone parameters ascertained at the same age, may shed important light on relationships between skeletal development and facial appearance.

Exposure variables: The primary exposure variables will be indices obtained from pQCT of the mid-tibia at age 15, particularly cortical thickness, cortical density, and periosteal circumference. Further exposure variables will include bone measures derived from total body DXA scans performed at the same time, including total body bone area and BMD, and skull area and BMD. Whether bone measures at earlier time points predict subsequent facial features will also be explored, based on total body DXA scans obtained at earlier time points.

Outcome measures: Facial shells, consisting of in excess of 40,000 points, have been generated in 4747 children in ALSPAC at 15 years of age, from which outcome variables for this study have been derived. These comprise 22 distinct facial landmarks, including the relative prominence (z-distance) of the pogonion, nasion and glabella, and thewidth (x-distance) of the chin and bridge of the nose.