Secondary hyperparathyroidism and mortality in hip fracture patients compared to a control group from general practice

Introduction: Previously, little attention has been paid as to how disturbances in the parathyroid hormone (PTH)-calcium-vitamin D-axis, such as secondary hyperparathyroidism (SHPT), relate to mortality amongst hip fracture patients. This study aimed to (1) determine if SHPT is associated with mortality in this group of patients, (2) investigate the association between serum (s-) PTH, s-total calcium, s-25-hydroxyvitamin D (s-25(OH) D) and mortality and (3) determine the prevalence of SHPT amongst hip fracture patients and a control group. Method: The study included 562 hip fracture patients (HF) (age >= 70 years) admitted to a Danish university hospital. The hip fracture patients were prospectively enrolled in a dedicated hip fracture database. Each hip fracture patient was exactly matched according to age and sex with two controls randomly chosen from a control population of 21,778 subjects who had s-PTH, s-total calcium and s-25(OH)D measured at the Copenhagen General Practitioners Laboratory after referral from their general practitioner. The control group (Con) thus consisted of 1124 subjects. Results: General 1-year mortality: Con-female 8.4%, Con-male 15.3%, HF-female 24.6%, HF-male 33.3%, p < 0.0001 (log rank). SHPT and related 1-year mortality: Con-no SHPT 8.9%, Con-SHPT 16.8%, HF-no SHPT 22.7%, HF-SHPT 34.9%, p < 0.0001 (log rank). The mortality rates were higher for controls with SHPT (OR 2.06, 95% CI: 1.32-3.23), hip fracture patients without SHPT (OR 3.00, 95% CI: 2.14-4.20) and hip fracture patients with SHPT (OR 5.46, 95% CI: 3.32-8.97) compared to the controls without SHPT. Prevalence of SHPT: Con 16%, HF 20%, p = 0.09 (Chi-square). Conclusions: Our study clearly shows that SHPT is significantly associated with mortality in both hip fracture patients and the control group. In the multivariate Cox regression analysis, s-PTH and s-total calcium were both significantly associated with mortality, whereas s-25(OH)D was not associated with mortality in this analysis. Our study furthermore indicates that SHPT is almost equally prevalent amongst the hip fracture patients and the control group. (c) 2011 Elsevier Ltd. All rights reserved.