From THE DEPARTMENT OF NEUROBIOLOGY, CARE SCIENCES AND SOCIETY, CENTRE FOR FAMILY AND COMMUNITY MEDICINE Karolinska Institutet, Stockholm, Sweden OSTEOPOROSIS IN ELDERLY WOMEN IN PRIMARY HEALTH CARE

Objective: All the studies are parts of the PRIMOS (Primary Health Care and Osteoporosis) project. Study 1 investigates the relationship between central Dual X-ray Absorptiometry (DXA) measurements of the hip and spine and peripheral measurements of the calcaneus using Dual X-ray and Laser (DXL) technique. Study II investigates the association between the nutritional status of elderly free-living women, as determined by the Mini Nutritional Assessment (MNA) method, and the women's bone mineral density measured with DXA. Study III investigates the relationship to osteoporosis of calcium-regulating hormones and the IGF-I and IGFBP-1 status in the cohort. Study IV is an RCT evaluating the effect of ultra-low dose of estradiol on bone mineral density. Methods: Study population: The participants in all studies come from the same population of approximately 940 women born between 1920 and 1930 living in the same primary care region in the southern part of Stockholm. Study I has 393 participants (388 included in the statistical analysis). Of these women, 351 were recruited by first inviting a random sample of 300, and then inviting all the rest of the women born 1926 and 1930. These 351 women are the population in study II and (with the exception of one woman excluded) in study III. The remaining 42 participants in study I and all 115 participants of study IV were included if eligible for a randomised controlled clinical trial (RCT) with estradiol. The design of the RCT was an open-label, randomised, parallel-group study with two treatment arms, one arm treated with a vaginal ring releasing 17 β-estradiol (average dose 7.5 μg/day) and a daily tablet containing 500 mg of calcium and 400 IU of vitamin D3, the other arm receiving treatment with 500 mg of calcium and 400 IU of vitamin D3. Bone mineral density measurements: The bone mineral densities (BMD) of the hip and lumbar spine (L1–L4) were determined using Hologic QDR 4500 equipment for DXA. The peripheral measurements on the calcaneus were performed with Calscan DEXA-T. Mini Nutritional Assessment (MNA): The nutritional status was determined with the MNA test consisting of 18 questions in four categories: anthropometric measurements, clinical and functional evaluations, assessment of dietary intake and self-assessment of health. The maximum score obtainable is 30 points, a score of <17 indicates malnutrition, 17–23.5 a risk of malnutrition and > 24 adequate nutritional status. Laboratory measurements: Parathyroid hormone (PTH), 25-hydroxy vitamin D, IGF-I, IGFBP-1, glucose and calcium-status were measured in all participants. Estradiol, SHBG, CTx, U-Dpd and other markers were followed in the RCT. Results: Study I showed that measurements of the heel bone with DXL technique correlated fairly well to central measurements of the hip and spine on the group level. The same WHO cut-off point, –2.5 SD, was also applicable for the heel BMD when comparing with most central sites or combinations of sites with the exception of total hip. The change of reference population had a great influence on the amount of subjects classified as osteoporotic, which varied between 7% and 53% depending on the chosen reference population and site. Study II showed that women with an MNA score under the median score of 27 points had a twofold increased risk of having osteoporosis compared to women with MNA scores above the median. Very few women (7.4%) were assessed as at risk of malnutrition and only one woman was classified as malnourished. Study III showed a significant inverse relation of IGFBP-1 to the BMD values and a significant positive relation of IGF-I values to the BMD values at all sites with the exception of the lumbar spine. The use of loop diuretics was a more important cause of secondary hyperparathyroidism than the vitamin D status of the women. Study IV showed a small but significant effect on BMD of 7.5 μg/day estradiol administered through a vaginal ring during a follow-up of two years. Conclusions: Bone mineral density measurements of the calcaneus with DXL technique correlate fairly well with central measurements. Adequate reference populations are important for T-scores. Elderly women with only a slight deterioration in their nutritional status have an increased risk of osteoporosis. IGF-I and IGFBP-1 are related to the BMD values. Secondary hyperparathyroidism may have other more clinically important causes than the vitamin D status in elderly women, i.e. treatment with loop diuretics. Estradiol doses of 7.5 μg/day seem to have a small but significant effect on BMD of elderly women.