Diurnal, Week-to-Week, and Long-Term Variation in Urine Deoxypyridinoline Cross-Link Excretion in Healthy Older Women


  • This work was presented at the British Geriatric Society spring meeting, Cork, Ireland, 1999.

  • The project was financially supported by the British Geriatric Society and the King's Osteoporosis Support Group.

Address correspondence to Dr. Theresa Allain, Department of Care of the Elderly, Elgar House, Southmead Hospital, North Bristol NHS Trust, Bristol, BS10 5NB, United Kingdom. E-mail: theresa.allain@blueyonder.co.uk


OBJECTIVES: To establish a reference range for morning and afternoon excretion of urinary deoxypyridinoline (DPD) in apparently healthy older women selected from a volunteer database. To assess the extent of diurnal variation and short and long-term within-subject longitudinal variation.

DESIGN: Prospective, observational, cohort study.

SETTING: Clinical Age Research Unit, King's College School of Medicine, London, United Kingdom.

PARTICIPANTS: Forty-two women aged 68 to 89 (median age 75) selected from a volunteer database.

METHODS: Subjects completed an osteoporosis risk factor questionnaire and a physical examination and had a measurement of the broadband ultrasound attenuation and speed of sound of their right heel. Subjects provided six urine samples: morning and afternoon at baseline and 1 week and 60 weeks later for measurement of DPD.

RESULTS: The mean baseline values for DPD of morning and afternoon samples were 7.2 nM/mM and 6.0 nM/mM creatinine, respectively. The majority of subjects showed diurnal variation, with mean afternoon values 15% lower than morning values (P < .0001 for afternoon vs morning values). The mean difference in DPD after 60 weeks was 1.67 nM/mM for morning and 1.34 nM/mM for afternoon creatinine. This difference was not significant. Some individuals displayed marked changes in DPD excretion with no change in health status or treatment. DPD excretion in a nonfasting afternoon sample showed similar characteristics to morning void samples in terms of scatter, week-to-week variation, and long-term reproducibility.

CONCLUSIONS: The study was set up to provide background data to assist the development of a clinical osteoporosis service for older women. Further studies are needed to determine whether these measurements predict fracture risk and respond to treatment changes in this age group.