Introduction: In daily practice the diagnostic process for osteoporosis in elderly patients should also include physical assessment. The aim of the study was to verify the hypothesis that height loss (HL) predicts fracture incidence.
Material and methods: The study was performed in an epidemiological sample of postmenopausal women recruited in the RAC-OST-POL study. At baseline, data were collected in 978 postmenopausal women at a mean age of 66.48±7.6 years, and at 10-year follow-up 640 patients remained, with a mean age of 75.04 ± 6.95 years. Current height and HL were established in regard to maximal life height. Data on fracture incidence were gathered throughout the period of observation.
Results: During the follow-up period 190 osteoporotic fractures were noted. Ninety-one women had one fracture, and in 38 women, multiple fractures occurred. In the fractured and unfractured subgroups, HL was 5.45 ± 3.28 and 4.8 ± 3.56 cm, respectively, and differed significantly (p < 0.05). HL in subjects without fracture did not differ from those with one fracture (HL 4.8 ± 3.56 vs. 4.8 ± 2.66 cm, respectively). For patients with more than one fracture HL was 7.03 ± 4.06 cm and was significantly higher than in subjects with one or without any fracture (p < 0.01). Based on receiver operating characteristic (ROC) analysis, HL of 6 cm was identified as the cut-off point for high risk of multiple fractures.
Conclusion: HL of at least 6 cm is the predictor of multiple fractures in a prospective observation of a representative epidemiological female sample. Therefore, the measurement of HL should always be included in patients' assessments.
Keywords: epidemiological sample; fracture prediction; height loss; women.