RISK-FACTORS FOR SECONDARY HYPERPARATHYROIDISM IN A NURSING-HOME POPULATION

OBJECTIVE Secondary hyperparathyroidism may cause bone loss and struct ural deterioration of bone and may thus be a cause of fracture in the elderly. Vitamin D deficiency, renal impairment and medications are po tential causes of hyperparathyroidism and may also directly predispose to fracture. We present the first findings of an ongoing study of hip fracture, vitamin D deficiency and hyperparathyroidism in a large Aus tralian nursing home. DESIGN Descriptive prevalence study. PATIENTS Tw o hundred and fifty-one nursing home residents were eligible for inclu sion. Informed consent and successful venepuncture were obtained for 9 9. Residents were of median age 83 years with interquartile range (IR) 77-89 years.MEASUREMENTS 25-Hydroxyvitamin D (25OHD), intact parathyr oid hormone (PTH), creatinine and biochemistry, demographic data and c urrent medications. RESULTS Fifty-two per cent of 99 subjects had 25OH D below the reference range of 26-165 nmol/l and 96.5% were below the reference range mean. Those with low 25OHD had lower plasma calcium co rrected for albumin than those with normal 25OHD (medians 2.34 vs 2.41 mmol/l, 95% confidence interval for the difference between medians (C I) -0.10 to -0.04 mmol/l, P=0.0001) and higher PTH (medians 5.8 vs 3.9 pmol/l, CI 0.10-2.6pmol/l, P=0.0360). Twenty-eight per cent of 97 res idents had PTH above the upper reference range limit of 6.5 pmol/l. Re sidents receiving frusemide had higher PTH than other residents (media ns 6.95 vs 3.45 pmol/l, CI 1.9-4.2 pmol/l, P<0.0001). In linear modell ing, the most important predictor of the natural logarithm of PTH was daily frusemide dose, adjusted R(2) (Ra-2)=31.8%, F=39.3, P<0.001. Cre atinine and the reciprocal of 25OHD were other significant predictors with the final Ra-2=39.4%, F=17.7, P<0.001. CONCLUSIONS Vitamin D defi ciency is a common risk factor for secondary hyperparathyroidism in nu rsing home residents despite a climate in which vitamin D nutrition is thought to be ample. However, the daily frusemide dose is a more impo rtant predictor of PTH in this population.