Reduced bone density of the hip in elderly patients with Parkinson's disease

Age and Ageing, July, 1995 by H. Taggart, V. Crawford

Introduction

Parkinson's disease (PD) is an important cause of falls and reduced mobility in elderly people. Many of the more severely affected patients have frequent falls, the main cause of which is postural instability [1]. Not infrequently, such falls result in fractures [2, 3]. One retrospective study of fractures in PD [4] found that sufferers were 66% more likely to sustain a hip fracture than controls. This was felt to be more probably due to the high risk of falling that to osteoporosis. There have been no clinical studies of bone mineral density in PD. We have carried out a prospective 2-year study of 29 elderly women and 26 elderly men with PD and compared them with an equal number of age- and sex-matched controls.

Methods

PD was diagnosed clinically when one or more of the following criteria were met: typical resting tremor, bradykinesia, rigidity and postural instability. Those who were chair-or bed-fast, or had dementia, were excluded. Twenty-nine women (mean age 76.8 years) and 26 men (mean age 74.3 years) were studied.

Details of duration of disease, drug therapy and fractures were obtained. A past history of fractures and fractures occurring during follow-up were recorded. Spinal radiography was not performed and thus there watiprobably an underestimate of vertebral fractures.

Control subjects: Were healthy elderly people who had been recruited for a study of bone density in old age. They were attending day hospital, clinics or day centres or were recruited through retirement clubs. They were fairly representative of the ambulant elderly population and were matched for age and sex with the PD patients. Those with known osteoporotic fractures, osteoarthritis, immobility, dementia and steroid therapy were excluded. Twenty-nine women (mean age 76.7 years) and 26 men (mean age 74.7 years) were studied.

Ethical committee approval was obtained for the study and all subjects gave informed consent.

Bone mineral density (BMD) was assessed by dual-energy X-ray absorptiometry (g/[cm.sup.2]) at the lumbar spine (L2-4) and the total hip and neck of femur using a Hologic 1000 bone densitobieter. The coefficient of variance on the spine phantom was 0.3%. Within-patient coefficient of variance was 1.07% for the spine and 2.1% for the hip.

Subjects and controls were followed up for a maximum of 2 years. BMD was assessed at 6-12-monthly intervals and patients were asked about fractures. Details of fractures were also taken from control subjects.

Variable distributions were assessed for normality and appropriate statistical tests applied. Descriptive statistics were employed throughout. BMD was compared between PD patients and controls using paired t tests.

The comparison of case-control differences between sexes was not significant. Thus since there was no evidence of heterogeneity, the male and female data could be pooled. Differences in initial BMD according to whether or not female, patients suffered fractures were examined using independent t tests. All statistics were obtained using SPSS for windows [5].

Results

The men with PD had had their disease for 10.5 years on average compared with 6.2 years for the women. Twenty-three (88%) of the men and 26 (90%) of the women were taking L-dopa preparations.

During the mean follow-up period of 17 months for the men, five patients (20%) died. Nine women (31%) with PD died during the mean follow-up period of 14 months. The commonest cause of death was bronchopneumonia, but information was not available in all patients. In no case was the death directly attributable to a recent fracture. None of the control subjects died because they had been chosen to have follow-up periods matching those of the patients.

Only two (8%) men with PD had a fracture at any time. In contrast 11 (38%) women sustained 17 fractures. Six were hip fractures, four distal radius, two pelvis, two humerus, two ankle and one nose. Six occurred during the course of the study. None of the control subjects sustained a fracture.

BMD for the PD patients and controls is shown in the Table. In both male and female PD patients there is a differential reduction in hip BMD over spine BMD compared with controls. When male and female groups are combined, there is a 10% lower BMD for total hip (p = 0.014) and a 12% lower BMD for the neck of femur (p < 0.004) in PD patients compared with controls. This contrasts with a 2% difference in mean spine BMD between PD patients and controls.

[TABULAR DATA OMITTED]

The mean BMD in PD women with fractures was 10% lower than controls in the total hip and neck of femur regions (0.61 vs. 0.67g/[cm.sup.2] and 0.53 and 0.50 g/[cm.sup.2] respectively) but this did not reach statistical significance.

Discussion

Several recent studies have highlighted the considerable morbidity and mortality due to falls and fractures in elderly PD patients [1, 6]. A recent large Australian population study [2] of 3851 people over 60 years of age found that postural sway and reduced bone mineral density were the most significant risk factors for fracture. Nineteen per cent of the community sample fractured their hip and 18.5% fractured their distal radius (compared with 33% and 20% respectively in our study).