 
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
21-September-2007, Vol 120 No 1262
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Vitamin D and muscle strength in patients with
previous fractures
Charles A Inderjeeth, Denise Glennon, Anthony Petta,
Jessamine Soderstrom, Irene Boyatzis, Jeffrey Tapper
Low trauma fracture is a major health problem for
postmenopausal women and to a lesser extent older men.
The two major causes of fracture in these groups are osteoporosis
and falls. There are established pharmacological treatments for
osteoporosis,3,4 however treatment of falls is
much more difficult.5
In the setting of osteoporosis, higher body sway, quadriceps
weakness, and conditions linked with increased risk of falling are associated
with a significantly increased risk of
fractures.6,7 Studies have also suggested an
association between bone mineral density and muscle
strength.8
Muscle strength has been shown to decline with
age9 and has been
shown to be associated with the functional status of older
people.10 Frail older
people11 and patients with hip
fracture12 are reported to have a high
prevalence of vitamin D deficiency due to reduced sunlight exposure, reduced
synthesis of vitamin D, low vitamin D dietary intake, poor absorption, and
hepato-renal disease.11,12
Receptors for vitamin D metabolites that are functionally
responsive to vitamin D have been identified in human skeletal
muscle.13,14 Vitamin
D deficiency and osteomalacia have been shown
to be associated with myopathy,15,16 and have
in some studies17,18 been shown to be
reversible with vitamin D treatment.
However, other studies looking at vitamin D supplementation
in unselected or vitamin D replete patients have shown no benefit with
vitamin D supplementation19 or no association
between serum vitamin D level and muscle
strength.16.
Fractures caused by falls occur in about 5% of older persons
each year.1,20 It has
been suggested that vitamin D deficient patients are at higher risk of falls due
to increased postural sway.21,22 In at least
one study, supplementation with vitamin D and calcium reduced the risk of falls
in recurrent fallers.20
However, evidence for the association between vitamin D levels and
muscle strength on objective testing has not been conclusively
proven.16
A possible reason for the inability to show a direct
association between vitamin D levels and muscle strength, and the failure to
show any improvement with vitamin D
supplementation,19 may be due to poor patient
selection. It is postulated that vitamin D replete patients will not
benefit from further supplementation.
Hence in this study we looked at a frail, at risk population
of patients who had previously been treated for hip or other long-bone fractures
following a fall. We assessed muscle strength as the primary measure rather than
using an indirect measure like falls.
The main aims of the study were to assess the vitamin D
status of this group of patients and to assess whether there is an association
between their serum 25 hydroxyvitamin D level and objective muscle strength
measurements. The secondary aim was to determine whether tests of balance and
functional status are valid predictors of muscle strength and whether they
correlate with serum 25 hydroxyvitamin D levels.
MethodsSubjects—Patients were selected
from a tertiary hospital database in Western Australia with a primary admission
diagnosis of a long-bone fracture as described previously by Inderjeeth et al in
Internal Medicine Journal (2006).23
Only subjects who were identified from the case notes
as suffering a low impact fracture of the hip, humerus or forearm were invited
to participate. Subjects had to be female, aged >60 years, at least 6 months
post fracture and be able to walk independently with or without walking aids and
participate in muscle strength measurements. 365 patients were identified as
suitable from case records over a 12-month period and invited to participate in
the study. Patients were mailed an osteoporosis questionnaire and invited to
attend a Fracture, Falls and Balance Clinic for assessment of their osteoporosis
and offered participation in this study.
Exclusion criteria were male gender, poor cognition
(defined as mini mental state examination <20), significant systemic disease
limiting their ability to participate, hypercalcaemia, primary
hyperparathyroidism, or a long-bone fracture within 6 months to allow for
fracture healing and muscle recovery. Patients on vitamin D supplements were
allowed in the study. Enrolment occurred throughout the year and was not
seasonally based. The protocol was approved by the Sir Charles Gairdner Hospital
Ethics Committee. All subjects gave informed consent.
Study design—All patients had
demographic data collected and a research nurse and doctor collected information
using validated instruments. Instruments selected include the Berg Balance Scale
(BBS),24 which was used to determine the
subject’s balance in a variety of spheres. The subject’s daily
functional status was assessed using the Frenchay Activity Index
(FAI).25
The Modified Bartel Index
(MBI)26 was used to assess functional
independence in relation to personal care and mobility. The Falls Efficacy Scale
(FES)27 was used to assess patient’s
confidence and fear of falling in carrying out routine tasks. RPH (Royal Perth
Hospital) outdoor score11,12 was used to
measure sunlight exposure [range 0(no exposure) to 7(regular adequate
exposure)].
Timed “Up & Go”
(TUAG)28 was used as a measure of functional
mobility. Cognitive function was assessed using the mini mental state
examination (MMSE)29 and mood was assessed
using the brief assessment of depression
(BASDEC).30
Muscle strength was measured in the left leg using a
Keylink Kinitech Dynamometer.31 Left leg power
and torque were measured in flexion and extension. Left leg flexion and
extension was measured in a standard seating position as recommended by the
manufacturer of the Keylink Kinitech Dynamometer.
Seating position was adjusted for height and the angle
of the hip and knee were standardised. The highest reading of three measures for
each were taken. Left Leg Extensor maximum power (LLExtmp) and Left Leg
Extension peak Torque (LLExtpt) and Left Leg Flexor maximum power (LLFxmp) and
Left Leg Flexor peak torque (LLFlxpt) were measured.
Blood samples were collected to measure serum calcium,
phosphate, creatinine, albumin, alkaline phosphatase, intact parathyroid hormone
(PTH), and 25 hydroxyvitamin D (25OHD). The blood samples were taken in
standardised tubes as recommended by the local pathology centre and the
reference limits were as standardised for the local population.
Blood samples were collected in the morning on the day
of the muscle strength assessment. All blood tests were performed at the one
pathology centre. 25OHD was determined by radioimmunoassay using the Diasorin
kit. The confidence values at 37 and 135 nmol/L were 10.8% and 8.1%
respectively.
Statistical analyses—The data
were entered into an SPSS software database and analysed by a biostatistician.
Descriptive statistics were performed including mean, standard deviation (SD),
and range. Univariate correlations (Pearson’s) for normally distributed
variables and Spearman’s for non normally distributed variables were
performed looking for association between 25OHD, functional and biochemical
parameters, and between muscle strength, functional and biochemical parameters.
Stepwise multivariate analysis was performed looking
for significant associations between muscle strength and age, biochemical, and
functional parameters. The t-test and one-way ANOVA were used to assess
differences between groups. A p value of <0.05 is considered significant.
ResultsOf the 365 patients mailed questionnaires, 105 subjects
agreed to participate in this cross sectional study and were enrolled. A total
of 99 subjects adequately completed all assessments and were included in the
muscle strength analysis. Of the 6 excluded subjects, 3 had primary
hyperparathyroidism, 2 had no serum 25OHD measurement, and 1 had difficulty
completing the muscle strength assessment.
All subjects were female with a mean age of 79.5 with a
standard deviation of 7.9 and a range of 61 to 95 years. All patients enrolled
had sustained a previous long-bone fracture. Thirty-four percent had sustained a
lower limb fracture (hip) only, and 53% an upper limb fracture (wrist or
humerus) only. Thirteen percent of subjects had sustained both upper and lower
limb fractures.
The mean 25OHD level of the cohort was 52.0 nmol/L (SD 22.3)
with a range of 16–159 (reference limits >50 nmol/L). Only 10% were
taking vitamin D either as simple vitamin D or as multivitamin supplements. The
values of the functional and biochemical parameters assessed are reported in
Table 1.
Table 2 describes the univariate associations between 25OHD
and age, functional, and biochemical assessments. All functional assessments
were significantly associated with measured serum 25OHD apart from the
patients’ age, Berg Balance Scale, and the depression scale.
Table 1. Mean (SD), minimum, and maximum values
for age, functional, and biochemical measures of patients
SD: Standard Deviation;
MMSE: Mini Mental State Examination; FAI:
Frenchay Activities Index; BASDEC: Brief Assessment Schedule
Depression Cards FES: Falls Efficacy Scale;
RPH: Royal Perth Hospital; TUAG: Timed
“Up & Go” Score; BBS: Berg Balance Scale;
ALP: Alkaline Phosphatase; PTH: Parathyroid
Hormone; 25OHD: 25 hydroxyvitamin D; g/L:
grams per litre; mmol/L: Millimole per litre;
umol/L: Micromole per litre; pmol/L: Picomole
per litre U/L: Units per litre.
Table 2. Association between 25 hydroxyvitamin
D and age, functional, and biochemical measures
MMSE: Mini Mental State Examination;
BASDEC: Brief Assessment Schedule Depression Cards
FAI: Frenchay Activities Index; RPH: Royal
Perth Hospital; FES: Falls Efficacy Scale;
TUAG: Timed “Up & Go” Score;
BBS: Berg Balance Scale; PTH: Parathyroid
Hormone; *p<0.05; **p<0.01.
Albumin and PTH were the only biochemical assessments
associated with measured serum 25OHD. There was a positive association with
albumin and a negative association with PTH. The two factors most strongly
associated with 25OHD were serum albumin (r=0.32, p=0.001) and the RPH outdoor
score (r=0.30, p=0.002).
Table 3 describes the significant univariate associations
between muscle strength and age, functional and biochemical assessments.
Although most of the functional assessments were associated with muscle
strength, serum 25OHD level was the only biochemical assessment associated with
muscle strength.
Table 3. Association between muscle strength
and age, biochemical, and functional measures
LLExtpt: Left Leg Extensor Peak
Torque; LLExtmp: Left Leg Extensor Max Power;
LLFlxpt: Left Leg Flexor Peak Torque; LLFlxmp:
Left Leg Flexor Max Power; MMSE: Mini Mental State Examination;
BASDEC: Brief Assessment Schedule Depression Cards
FAI: Frenchay Activities Index; RPH: Royal
Perth Hospital; FES: Falls Efficacy Scale;
TUAG: Timed “Up & Go” Score;
BBS: Berg Balance Scale; 25OHD: 25
hydroxyvitamin D; *p<0.05; **p<0.01.
Table 4. Multivariable associations between
muscle strength and age, biochemical, and functional measures
Model includes age and all biochemical and functional
measures. Only those measures with significant associations are shown.
LLExtpt: Left Leg Extensor Peak Torque;
LLExtmp: Left Leg Extensor Max Power; LLFlxpt:
Left Leg Flexor Peak Torque; LLFlxmp: Left Leg Flexor Max
Power; 25OHD: 25 hydroxyvitamin D; BBS: Berg
Balance Scale; TUAG: Timed “Up & Go” Score;
*p<0.05; **p<0.01.
Table 4 describes the significant multivariate associations
between muscle strength and all other parameters measured (age, biochemical, and
functional). In the multiple regression model, 25OHD was significantly
associated with both extensor assessments (LL Extpt r=0.489, p<0.001 and LL
Extmp r=0.476, p=0.001) as well as the flexor PT (r=0.367; p=0.019) assessment,
and showed a strong trend to an association with flexor MP (r=0.259,
p=0.055).
Only two other factors (both functional) showed any
association with muscle strength in the multiple regression model. Berg Balance
was associated with extensor strength while TUAG was inversely associated with
left leg flexor PT only. Site (upper or lower limb) and side of fracture (left
or right) did not change the association between muscle strength and 25OHD and
were non significant associations in the multiple regression model.
The functional assessments with no significant association
in the regression model included MMSE, BASDEC, FAI, RPH outdoor score, and FES.
The biochemical assessments with no significant association in the regression
model included calcium, phosphate, albumin, alkaline phosphatase, and
parathyroid hormone.
Forty-seven patients had a 25OHD level <50 nmol/L and 14
patients had a 25OHD <30 nmol/L. The association between muscle strength and
25OHD was strongest in the sub-group with 25OHD levels >50 nmol/L (p<0.01
for all four muscle strength assessments with the r value ranging between 0.46
and 0.51).
The association between muscle strength and 25OHD levels was
not significant in subjects with 25OHD levels <50 nmol/L (p>0.05). Using
one-way ANOVA, the quartile with the highest 25OHD level (>60 nmol/L)
demonstrated the highest mean muscle strength (Figure 1). However, this
association was only significant for LL Extpt (p=0.035) but not LL Extmp
(p=0.065), LL Flxpt (p=0.11), or LL Flxmp (p=0.25).
Figure 1. Mean and 95% confidence intervals of
muscle strength (left leg extensor peak torque) for quartiles of 25
hydroxyvitamin D
 Subgroup analysis was undertaken to compare muscle strength
and 25OHD between the group with left versus right-hip fracture and those with
upper-limb and lower-limb fractures. There was no significant difference in
left-leg muscle strength based on side of hip fracture (p>0.05). However,
those with upper-limb fracture only had higher left-leg muscle strength than
those with lower-limb fracture (p<0.05). There was no significant difference
in 25OHD levels in these subgroups (p>0.05).
DiscussionVitamin D deficiency is being increasingly identified as a
significant problem in older patients.11,12,32
Its impact on bone has been extensively investigated. However, its impact on
muscle strength and falls is less well understood.
Given the high incidence of falls, osteoporosis, and vitamin
D deficiency in older individuals, establishing an association and correcting
vitamin D deficiency may be potentially beneficial in improving muscle strength,
reducing falls, and hence reducing fractures.
This cross sectional study looked at the potential
association between muscle strength and 25OHD deficiency in patients previously
admitted with fracture, including 46% with a previous hip fracture.
We found a moderate and significant positive correlation
between muscle strength and 25OHD. Although a number of factors appear to be
associated with muscle strength on univariate analysis, the only factors
independently associated after correction for other factors in a multiple
stepwise regression analysis were 25OHD, the Berg Balance Scale, and the Timed
“Up & Go” score.
The 25OHD level appears to be the only consistent
significant association with left-leg muscle strength with a moderate positive
correlation. The association with Berg Balance and Timed “Up &
Go” is weaker and only associated with some parameters of muscle strength.
Although side of fracture does not appear to influence muscle strength, the
presence of lower limb fracture does appear to result in worse muscle
strength.
The main limitation of this study is the relatively low
response rate. In addition, as with other similar studies, frailer subjects and
those with cognitive impairment were excluded on the basis that they could not
perform the muscle strength assessment.
These are both groups that are more likely to have been
vitamin D deficient. The relatively high mean 25OHD level of 52 nmol/L reflects
this and unfortunately the relatively small sample size does not allow for
adequate subgroup analysis.
The stronger association between 25OHD level and muscle
strength parameters in the sub-group with levels >50 nmol/L is unexpected, as
we would have predicted a stronger association in the group deemed to be vitamin
D deficient i.e. <50 nmol/L. This may reflect a threshold level of 25OHD
which must be reached before vitamin D starts to affect muscle strength in
patients.
It would also strongly support the argument that 25OHD
levels need to be maintained well above 50 nmol/L for adequate replacement and
benefit in terms of muscle strength. This is an area that certainly requires
further study especially with regards to the impact of replacement of vitamin D
in subjects with levels below 50 nmol/L and the most appropriate level desirable
for greatest benefit in terms of muscle strength.
Higher levels may be the goal rather than aiming for low
normal levels for greater benefit in terms of muscle strength and possibly bone.
The strength of this study is that it looked at a high-risk
group who are more susceptible to vitamin D deficiency due to reasons of
relative frailty, previous falls, and fractures. This is the population that
would need to be targeted as a priority in terms of vitamin D
replacement.12
In patients with osteoporosis, the goal should be to reduce
fracture risk. To achieve this we need to look beyond the improvement of bone
strength and quality. Falls prevention is another obvious objective.
As muscle weakness is a major cause of falls, improving
muscle strength is an important component of this strategy. Hence identifying
and treating conditions likely to be associated with reduced muscle strength
(such as vitamin D deficiency) is an area that warrants further investigation.
And to confirm the causal association between vitamin D and muscle strength, a
large randomised controlled intervention trial is needed.
Competing interests: None.
Author information: Charles A Inderjeeth,
Geriatrician/Rheumatologist; Denise Glennon, Geriatrician; Anthony Petta,
Senior Physiotherapist; Irene Boyatzis, Geriatric Medicine Registrar; Jessamine
Soderstrom, Geriatric Medicine Registrar; Jeffrey Tapper, Head of Physiotherapy
and Adjunct Associate Professor;
Sir Charles Gairdner Hospital, Nedlands, Western Australia (WA), Australia Acknowledgements: The Sir Charles Gairdner
Hospital Clinical Research Fund provided an unconditional research grant for
this study. We also thank Ian Cooper for assistance with muscle strength
assessments as well as Rebecca Mackesey, Caroline Reberger, Sister Jacqui
Bates, and DRAC Staff (medical and non medical) for their assistance with
patient recruitment, data collection, and management.
Correspondence: A/Professor Charles A
Inderjeeth, Director of Clinical Training, Research and Programs, Area
Rehabilitation and Aged Care, Osborne Park Hospital Program, Sir Charles
Gairdner Hospital Campus, Hospital Avenue, Nedlands, WA 6009, Australia.
Email: Charles.Inderjeeth@health.wa.gov.au
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|