| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Chemical Pathology (D.R., W.L., G.W., B.M., P.E.H.) and Medicine (R.C.C.) and the Spinal Injuries Unit (J.W.), Princess Alexandra Hospital, Woolloongabba, Queensland 4102; and Sullivan and Nicolaides Laboratories (R.F.), Taringa, Queensland 4068, Australia
Address all correspondence and requests for reprints to: Dr. David Roberts, Department of Chemical Pathology, Princess Alexandra Hospital, Ipswich Road, Woolloongabba, Queensland 4102, Australia. E-mail: davidr{at}australiamail.com
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Loss of biomechanical stress on the skeleton is the most obvious cause of osteoporosis in SCI. However, other factors have been implicated, including neurovascular changes secondary to the modification of the autonomic nervous system (5), resistance to insulin-like growth factor I (8), and decreased levels of insulin-like growth factor I (9). It has been suggested that SCI may cause structural changes in collagen, inducing increased resorption (10, 11).
In recent years, new biochemical markers of bone turnover [deoxypyridinoline (Dpd), pyridinoline (Pyd), N-telopeptide (NTx), and osteocalcin (OC)] have been described and have largely replaced older markers such as hydroxyproline. Previous studies of bone metabolism in SCI have generally used the older markers and/or have been performed cross-sectionally. The older markers of bone metabolism (e.g. hydroxyproline) are relatively nonspecific, and we believed that we had the opportunity to examine reportedly specific markers in a well defined clinical population. Recently, two groups (12, 13) have published longitudinal data on modern bone markers in patients with acute SCI, each following six patients. One study (13) followed patients monthly between the first and sixth months of injury, and the other (12) followed patients for 2–3 months after injury. Neither study examined bone mineral density (BMD). We aimed to examine the changes in bone metabolism markers and BMD in a prospective, longitudinal study of a larger group of patients for 6 months. Our other reason for assessing the changes after SCI in modern markers of bone metabolism was that we intend in future studies to assess the changes in these markers in response to therapeutic intervention.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study was a prospective study of bone markers and bone
densitometry in patients with acute SCI. Between March 1, 1995, and
March 30, 1996, 30 subjects were recruited from the Spinal Injuries
Unit, Princess Alexandra Hospital (the unit serves a population of
3.5 million and a geographical area equivalent in size to the
British Isles and Western Europe). Inclusion criteria included informed
consent and acute, traumatic SCI with neurological involvement.
Potential subjects were excluded if they refused consent, had SCI
without neurological involvement, had SCI due to nontraumatic cause
(e.g. malignancy), or had preexisting disease or took
medications known to influence bone metabolism. The exception was the
acute administration of methylprednisolone, a standard therapy for
acute SCI. The study was approved by the Princess Alexandra Hospital
ethics committee.
Biochemical data
After SCI, early morning, nonfasting plasma was collected at weeks 1, 2, 3, 4, 6, 8, 10, 12, 16, 20, and 24 h for determination of multiple biochemical analytes, ionized calcium, intact PTH, and intact OC. At the same time, 24-h urine and early morning nonfasting urine specimens were collected for determination of clearance of creatinine, calcium, Dpd, Pyd, and NTx. In the subgroup of eight quadriplegic subjects who completed the 6 months of the study, 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D were measured at weeks 1, 12, and 24. After collection, specimens for PTH, OC, NTx, Dpd, Pyd, and vitamin D studies were immediately placed in a -20 C freezer until analysis. Specimens from the one patient were always analyzed within one run. Ionized calcium was measured by an ion-selective electrode (Radiometer Copenhagen ICA 2, Brisbane, Australia) on blood drawn into heparinized vacuum tubes without the use of an arm cuff. Total alkaline phosphatase (tAP) was measured spectophotometrically (BM/Hitachi 747, Boehringer Mannheim, Sydney, Australia). 1,25-Dihydroxyvitamin D was measured by a RRA (Incstar Corp., Stillwater, MN), and 25-hydroxyvitamin D was determined by RIA (Incstar Corp.). NTx was measured by enzyme-linked immunosorbent assay [ELISA; Osteomark kit, Ostex International, Seattle WA; in-house interassay coefficient of variation (CV), 8%], free Dpd was measured by ELISA (Pyrilinks-D kit, Metra Biosystems, Mountain View CA; in-house interassay CV, 10%), intact PTH was determined by RIA (Incstar PTH-sp kit; in-house interassay CV, <6%), and OC was determined by immunoradiometric assay (Human Osteocalcin Kit, Nichols Institute, San Juan Capistrano, CA; in-house interassay CV, <7%).
Total Dpd and total Pyd were measured by an in-house high performance liquid chromatography (HPLC) method, involving overnight hydrolysis in concentrated hydrochloric acid at 110 C, sample clean-up on a cellulose slurry column, and separation and quantification with reverse phase isocratic HPLC on a Symmetry C18 column (Waters Associates, Millford, MA) with fluorescence detection. Isodesmosine was used as an internal standard. This method is a modification of that described by Randall and co-workers (14). Our interrun CVs for total Pyd are 7.8% (at a concentration of 66 nmol/L), 4.1% (287 nmol/L), and 3.9% (2387 nmol/L), whereas the CVs for total Dpd are 5.1% (19 nmol/L), 4.8% (67 nmol/L), and 5.0% (480 nmol/L). The reference ranges we have quoted apply to adult males and adult premenopausal females, and those for free Dpd, NTx, and OC were supplied by the product manufacturers. For total Dpd and total Pyd, reference values were derived from assessment in our laboratory of 80 apparently healthy, adult male and adult premenopausal female Red Cross blood donors and laboratory staff.
Bone mineral densitometry
Paired measurements of BMD were made in 11 subjects at weeks 8 (median, 9.5 weeks; range, 6–13 weeks) and 24 (median, 23.3 weeks; range, 20–27 weeks). Clinical and logistic considerations precluded paired measurements in the remaining subjects. Measurements of BMD were made by dual energy x-ray absorptiometry at the distal third of radius, L2-L4 spine, and femoral neck (DPX-L, Lunar Corp., Madison, WI). In 3 subjects, alternative areas of the lumbar spine were used because of the presence of metallic prosthetic material. In a post-hoc analysis, BMD of the entire legs (n = 10) was calculated using the Manual Analysis capacity of the Lunar software. Lower limb measurements were defined as the femora and lower legs, but excluding the pelvic bones. One of the 11 subjects could not be included in this analysis as she had not had a whole body examination on 1 occasion.
Statistical analysis
Statistical analysis was performed using Microsoft Excel (version 7.0), Lotus 1-2-3 (version 2.4, after data conversion from Excel), and DataQuest (QC Division, U.S. Environmental Protection Agency). Data sets were tested for significant deviations from statistical normality using the Shapiro-Wilk test (P = 0.05, DataQuest software), and no such deviations were demonstrated.
Two-tailed paired t tests (P = 0.05) were performed on BMD data (raw and corrected) to test for significant differences over the course of the study (average interval between paired BMD, 14 weeks). Average bone marker data for quadriplegic and paraplegic subjects were tested for significant difference on each week of data collection using t tests (P = 0.05). Except where stated otherwise, statistical significance was accepted at P < 0.05. Data are presented as mean ± SEM unless otherwise stated.
Subjects were divided into paraplegics and quadriplegics. As there were few females (n = 6), no division was made according to gender.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
During a 13-month period, 30 subjects were enrolled in the study
from a potential 55 eligible patients. The reasons for nonenrollment
included refusal (n = 16) and inability to obtain informed consent
(n = 9), generally because patients were too ill (e.g.
mechanically ventilated). Figure 1
shows
the number of subjects at each week of follow-up. The commonest reason
for withdrawal from the study was discharge from hospital. Our Spinal
Injuries Unit serves a vast geographic area (1.7 million
km2), and only 10% of patients live within a 65-km
radius of the hospital. Thus, nearly all patients were discharged home
to remote areas and were unable to continue participation in the study.
There were more males than females (24 and 6, respectively), and the
average age at time of SCI was 28.6 yr (range, 13.8–66.3 yr),
reflecting the typical gender and age of patients with SCI.
Methylprednisolone (standard protocol of 30 mg/kg over 15 min, followed
by 5.4 mg/kg·h infused for 23 h) was administered to all
patients within 24 h, with the exception of 2 paraplegic and 3
quadriplegic subjects. Apart from SCI, other significant injuries were
sustained by 9 patients. Among the quadriplegic subjects these included
2 subjects with pneumonia, 1 subject with a pulmonary embolus, and 1
subject with pneumothorax and fractured tibia. Among the paraplegic
subjects these included 2 subjects with pneumothoraxes, 1 subject with
fractured wrist, 1 subject with diaphragmatic trauma, and 1 subject
with an infected bone plate.
|
tAP (Fig. 2) rose slightly with
duration of study, and only at weeks 3, 4, 6, 12, and 16 were the
values significantly greater than those at baseline (by t
test, P < 0.05). OC (Fig. 2) similarly showed a minor
rise, from the bottom to the middle of the reference range, with this
rise not being statistically significant. There was no difference
between paraplegics and quadriplegics in markers of bone formation.
|
), where it remained for the duration of
study. Coincident with the rise in ionized calcium, PTH rapidly dropped
and remained low for the duration of study (Fig. 3). Urinary 24-h
calcium excretion rose in parallel with the serum ionized calcium, but
never exceeded 10 mmol/day. In the eight quadriplegics who completed
the full study, 25-hydroxyvitamin D remained within the reference range
of 25–150 nmol/L (mean values at weeks 1 and 26 were 95 and 55 nmol/L,
respectively; P = 0.01), whereas 1,25-dihydroxyvitamin
D fell below the reference range of 40–150 pmol/L (mean values at
weeks 1 and 26 were 83 and 29 pmol/L, respectively; P =
0.004).
|
), illustrating total Dpd, free Dpd, and
NTx measured from random, early morning urine collections.
|
).
However, with percent total Dpd and percent NTx, there was no
difference between paraplegic and quadriplegic subjects (figures not
shown).
|
Paired BMD measurements at the hip, lumbar spine, and
distal radius were available in 11 of the 30 subjects. Figure 6 shows paired BMDs at the neck of femur,
visually demonstrating overall little change in BMD for most subjects.
There was no statistically significant difference between paired BMDs
at the neck of femur (mean difference, -0.045 g/cm-2;
P = 0.15). Similarly, there was no difference in paired
BMDs at the radius (mean difference, +0.034 g/cm-2;
P = 0.41) or lumbar spine (mean difference, -0.007
g/cm-2; P = 0.69). However, paired
measurements of BMD from the entire lower limbs (Fig. 6) showed a
significant fall (mean difference, -0.131 g/cm-2;
P < 0.001).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We did not measure the bone-specific isoenzyme of alkaline phosphatase, which we recognize is a more specific and sensitive marker of bone formation than tAP. However, bone-specific isoenzyme of alkaline phosphatase constitutes approximately 50% of tAP, and we believed that in the absence of liver disease, tAP would reasonably reflect bone formation in our subject population.
After week 3, the cohort became mildly hypercalcemic, and PTH levels fell, reflecting the mobilization of skeletal calcium due to immobility. Pietschmann et al. (13) similarly showed that PTH was suppressed for 1–4 months post-SCI and thereafter returned toward the normal range. Vaziri et al. (16), in a cross-sectional study of patients with SCI of 3- to 5-yr duration, showed persistent suppression of PTH, suggesting that net bone resorption continues for many years. Only one subject in our study developed clinically important hypercalcemia, although drug treatment was not required. This was a 16-yr-old male with C5 quadriplegia, and his peak ionized calcium was 1.47 mmol/L (reference range, 1.16–1.26 mmol/L). Vitamin D studies were performed in a subgroup judged to have been the most likely to have abnormalities, the eight quadriplegics who completed the full study. We demonstrated a significant decrease in both 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D, although no subject had subnormal values for 25-hydroxyvitamin D. The fall in 1,25-dihydroxyvitamin D parallels the decrease in PTH. Dietary details were not obtained, and we are unable to comment on the dietary intake of vitamin D. In our Spinal Injuries Unit, patients spend time outdoors on a covered verandah even when they are in traction, and this may help to prevent hypovitaminosis D. Vaziri et al. (16) found that in subjects with chronic SCI, 25-hydroxyvitamin D levels did not differ from reference data, whereas 1,25-dihydroxyvitamin D was significantly lower, related to lowered PTH. Because the secretion of OC is vitamin D dependent (17), the changes we observed in vitamin D status may have influenced our OC results. Of interest, in elderly subjects it has been suggested that vitamin D deficiency may lead to undercarboxylation of OC, increasing the risk of fracture (18).
We found no significant difference in absolute values of bone turnover markers between paraplegic and quadriplegic subjects. However, when total Pyd was expressed as a percentage of baseline (percent total Pyd), there was a suggestion of greater bone resorption in quadriplegics. With percent total Dpd and percent NTx, there was no difference between the two groups. Whereas Pyd is found mainly in bone, a small amount is also found in connective tissue, and it is possible that the difference we demonstrated between quadriplegics and paraplegics in percent total Pyd was due to differences in connective tissue turnover, as quadriplegic subjects may have experienced greater muscle atrophy of the upper limbs and had greater soft tissue turnover. In contrast, Dpd and NTx are more bone specific. The difference seen with percent total Pyd between paraplegics and quadriplegics was not the result of enrollment at different times during the first week of injury (which would have influenced baseline values because the bone markers increase from the time of SCI). The mean times at which week 1 values were obtained in paraplegics and quadriplegics were 5.7 and 5.2 days, respectively (P = 0.69). The number of subjects with available week 1 values was small (9 paraplegic and 10 quadriplegic subjects), and this, too, may have influenced the analysis. Thus, we emphasize that there was a suggestion and not conclusive proof that quadriplegics may have had greater bone turnover than paraplegics.
The literature relating to potential differences in bone turnover between quadriplegics and paraplegics is conflicting. Vaziri et al. (16) found 1,25-dihydroxyvitamin D was lower and calcitonin was higher in quadriplegics, but could not demonstrate differences with PTH or ionized calcium, whereas in another cross-sectional study (13), OC, PTH, calcitonin, and hydroxyproline did not differ between quadriplegics and paraplegics. Densitometry studies have shown a difference in BMD at the arms and trunk (1) and femoral shaft (19), but not the arms (20). The failure to clearly demonstrate differences between quadriplegics and paraplegics may reflect heterogeneity of these populations; patients with low quadriplegia can have sufficient use of their arms to propel themselves in a wheelchair, as do paraplegics. Thus, functionally, some paraplegics and quadriplegics may not be very different, and their bone turnover would not be expected to be dissimilar. Future studies may overcome this problem by using larger numbers of subjects, allowing stratification by exact level of neurological impairment, degree of immobility, presence or absence of spasticity, etc.
The rise in resorption markers was equivalent in both 24-h urine collections and nonfasting, early morning collections (data not shown). As random morning specimens are easier to collect, we recommend their use in subjects with acute SCI.
We examined whether the pattern of changes in serum and urinary markers of bone metabolism was influenced by the changing patient population, as the markers represent all 30 subjects at the beginning but 12 subjects at the end of the study. Graphs of markers for only the finishing subjects showed no major differences, and statistical analysis of just these subjects did not reveal any new information, although sample sizes were small, making detection of statistical differences difficult. These difficulties were compounded by further study of subset patient groups.
The routine administration of methylprednisolone within the first 24 h of acute SCI may have influenced the biochemical markers of bone metabolism. The present study had insufficient numbers to investigate the effects of the methylprednisolone pulse therapy, which was administered to 25 of the 30 subjects. There are published reports regarding the effects of methylprednisolone on bone markers in other disease states. In 7 patients with severe rheumatic diseases treated with 1 g methylprednisolone, OC decreased significantly, as early as 6 h after the pulse therapy, reaching a nadir at 24 h and returning to pretreatment values by 72 h (21). In 47 patients with active rheumatoid arthritis, a single dose of methylprednisolone (15 mg/kg BW) produced a significant decrease from baseline after 7 days in total Pyd (median decrease, 1.9 nmol/mmol creatinine), but not in Dpd, although the differences between the placebo (n = 39) and methylprednisolone groups in the changes in Pyd and Dpd were not significant (22). In these 2 studies, the modest influence of methylprednisolone on bone markers suggests that the single dose of methylprednisolone administered after acute SCI in this study had minimal effect on the changes in bone metabolism markers.
After SCI, our paired measurements of BMD at the lumbar spine, hip, and distal radius did not show, overall, any change. The literature suggests that BMD is preserved at the lumbar spine due to mechanical loading from being seated in a wheelchair (19, 23), whereas BMD in the distal femur is depleted by 22% at 3 months, 27% at 4 months, and 32% at 14 months (1). Whether BMD at the radius decreases with time is not clear, with some studies showing a fall in BMD (1, 20), whereas others did not observe any long term changes (24, 25). We believe that the reason that we failed to demonstrate a decrement in paired hip BMD was the short interval between measurements (average, 13.8 weeks), coupled with the relatively higher error in measuring BMD at the hip compared with that at other sites (26). BMD could not be performed any closer to the time of injury due to logistic difficulties (e.g. traction devices). Measurement of BMD in the entire lower limbs decreases the error because of the larger area measured, and we were able to show a decrease in paired measurements of this area. Another possible explanation for failure to detect a difference between paired BMDs may be that most of the bone loss had already occurred by the time of the first BMD. However, this is unlikely, as markers of bone resorption in this study peaked between weeks 10 and 16, suggesting that most of the bone loss was occurring during the interval of the paired BMD scans. We did not have sufficient numbers to compare changes in BMD between paraplegics and quadriplegics.
Possible bias may have arisen at recruitment, as only 30 of the eligible 55 patients were enrolled. Data were not available for patients refusing participation to ascertain whether they differed in any way from participants. Another source of potential bias in this study is loss to follow-up (after discharge from hospital to remote communities), as those subjects lost may have had lesser degrees of injury and disability. By stratifying subjects into paraplegic and quadriplegic, this bias may have been reduced.
To our knowledge, only one study has looked at pharmacological intervention in subjects with SCI (27) using bone biopsy parameters as the outcome measure of treatment with tiludronate. It would be instructive to examine the response of the biochemical markers to drug treatment.
In conclusion, we have followed longitudinally the changes in bone mineral density and modern biochemical markers of bone turnover in a cohort of patients with acute spinal cord injury, demonstrating a dramatic rise in resorption markers but only a minor rise in formation markers. Bone resorption markers show similar changes when measured from either early morning, nonfasting urine or 24-h urine collections. Paired BMD measurements at the hip, lumbar spine, and radius, performed, on the average, 14 weeks apart, was relatively insensitive in reflecting the changes in bone status. However, paired BMD of the entire lower limb showed a decrement between measurements performed, on the average, 14 weeks apart, probably reflecting the lower error associated with scanning a larger area. We were unable to unequivocally demonstrate a difference in bone markers between paraplegics and quadriplegics, although on one analysis (percent total Pyd), there was a suggestion that bone loss was greater in quadriplegics. Fractures are clinically relevant in people with SCI, and we suggest that future studies examine the response of modern markers of bone metabolism to therapeutic intervention.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received May 7, 1997.
Revised October 29, 1997.
Accepted November 3, 1997.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. Imberti, P. Pricca, L. Pietrobono, and M. Langer Extensive Endobronchial Calcification Contributing to Hypoxia in a Quadriplegic Patient Anesth. Analg., August 1, 2007; 105(2): 551 - 552. [Full Text] [PDF]
|
|
|
|
 |
|
 N. L. Gilchrist, C. M. Frampton, R. H. Acland, M. G. Nicholls, R. L. March, P. Maguire, A. Heard, P. Reilly, and K. Marshall Alendronate Prevents Bone Loss in Patients with Acute Spinal Cord Injury: A Randomized, Double-Blind, Placebo-Controlled Study J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1385 - 1390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. W. Vesper, L. M. Demers, R. Eastell, P. Garnero, M. Kleerekoper, S. P. Robins, A. K. Srivastava, G. R. Warnick, N. B. Watts, and G. L. Myers Assessment and Recommendations on Factors Contributing to Preanalytical Variability of Urinary Pyridinoline and Deoxypyridinoline Clin. Chem., February 1, 2002; 48(2): 220 - 235. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
