Intravenous pamidronate attenuates bone density loss after acute spinal cord injury

243

Intravenous Pamidronate Attenuates Bone Density Loss After Acute Spinal Cord Injury Patricia W. Name, MD, Orpha Schryvers, BN, William Leslie, MD, Sora Ludwig, MD, John Krahn, PhD, Daniel Uebelhart,‘MD ’ ABSTRACT. Nance PW, Schryvers 0, Leslie W, Ludwig S, Krahn J, Uebelhart D. Intravenous pamidronate attenuates bone density loss after acute spinal cord injury. Arch Phys Med Rehabil 1999;80:243-5 1. Objective: To compare the effects of a 6-month treatment with intravenous pamidronate (30-mg infusion once per month) to conventional rehabilitation without pamidronate on bone density of the spine and leg bones and on the excretion rate of N-telopeptide, a urinary marker of bone catabolism, in acutely spinal cord injured patients. Design: A nonrandomized control trial in which 24 spinal cord injured subjects entered the study within 6 weeks of their injury. Fourteen subjects received pamidronate; 10 did not. Outcome Measures: Bone density measurements by dual x-ray absorptiometry were performed before the initial treatment (within 6 weeks of the injury) and at 3, 6, and 12 months postinjury and was the primary efficacy parameter. Urine for N-telopeptide levels was the secondary efficacy parameter. Results: After acute spinal cord injury, patients treated with intravenous pamidronate had significantly less bone density loss compared with those who did not receive pamidronate (parametric ANOVA, p < .02). Also, ambulatory subjects had significantly less bone density loss over the study period (p < .05) than nonambulatory subjects. In general, a high excretion level of the urinary bone-breakdown product Ntelopeptide was found before intravenous pamidronate treatment, followed by a dramatic reduction in excretion after pamidronate treatment. Ambulatory subjects excreted significantly less N-telopeptide than motor-complete subjects at all time points. Conclusion: Intravenous pamidronate treatment and ambulatory ability in the first 6 months after an acute spinal cord injury prevents bone density loss. o 1999 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation From the Spinal Cord Research Centre, Health Sciences Centre (Dr. Nance, Ms. Schryvers), the Section of Physical Medicine and Rehabilitation (Dr. Nance, Ms. Schryvers), the Section of Nuclear Medicine (Dr. Leslie). the Section of Endocrinolozv (Dr. Ludwig). the DeDartment of Biochemistrv (Dr. Krahn). St. Boniface Hosuital (Es: Leslie, Ludwig, Krihn), the Department of Internal Medicine (Dr. Nance,‘Ms. Schryvers, Dr. Leslie. Dr. Ludwig), University of Manitoba (Dr. Nance, Ms. Schryvers, Dr. Leslie, Dr. Ludwig, Dr. Krahn), Winnipeg, Manitoba. Canada; and the Division of Physical Medicine and Rehabilitation and Spinal Cord Injury Center. Department of Clinical Neurosciences, University Hospital. Geneva, Switzerland, and Department of Biochemistry. Rush Presbyterian-St. Luke’s Medical Center. Chicago. IL (Dr. Uebelhalt). Submitted for publication June 10, 1998. Accepted in revised form September 24, 199x. Supported by a grant from the Paralyzed Veterans of America Spinal Cord Injury Research Foundation and in part by SCORE grant 32-355X2.92 from the Swiss National Foundation for Scientific Research. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Patricia W. Nance, MD. X00 Sherbrook Street, Winnipeg, Manitoba. Canada R3A lM4. D 1999 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/99/X003-5065$3.00/0

tine and the American Academy of Physical Medicine and Rehabilitation IGNIFICANT BONE LOSS of the lower extremities in patients with spinal cord injury (SCI) was first described by S Albright and colleagues1 Although the terms immobilization and disuse have been applied to describe these changes in bone density, the pathophysiology is likely to be complex. In the acute injury period, there is an increased urinary excretion of calcium and hydroxyproline and accelerated bone remodeling.2-11The increase in urinary calcium cannot be attributed solely to lack of exercise or to prolonged inactivity because a greater degree of hypercalciuria is found in SC1patients than in age-matched controls after prolonged bed rest.12zi3 A rapid and clinically significant loss of bone mass after spinal cord injury has been shown using a variety of methods. Histomorphic techniques, based on iliac crest biopsy, showed a predominate loss of trabecular bone within the first 4 to 6 weeks after SCI.4,‘4 Similarly, the bone mineral density loss is detectable by less invasive methods within the first 4 weeks of injury. The greatest bone density loss occurs in the lower extremities in SC1 patients, a distribution unique to this type of osteoporosis, as shown by dual-photon and dual-energy x-ray absorptiometry and ultrasound bone densitometry. 15-20These observations would lead to the conclusion that an intervention to prevent the occurrence of osteoporosis should be initiated as soon as possible after injury, before bone density loss is detectable. The mechanism of bone density loss in patients with SC1 remains controversial. It is a long-established observation that bone remodels along lines of mechanical strain.21-24Since the discovery that mechanical perturbations on bone produce a piezoelectric current,25 electrical and electromagnetic fields induced by implanted electrodes have been used to promote osteogenesis in the repair process of fracture healing.26%27 Although mechanical stress on bone is normally provided by physical activity, ambulatory ability alone or verticalization therapy does not prevent bone density loss after SCI.17.2x,29 Chantraine7 has suggested that an autonomic alteration due to the neurologic injury is related to loss of bone density after SCI. The following observations lend support to the notion that the neurologic damage due to the SC1per se may be a causal factor in the ensuing loss of bone density rather than the immobilization factor alone: (1) the degree of increased calcium excretion of SC1 patients is much greater than in immobilized healthy subjects with intact spinal cord12,13;(2) exercise shortens the duration of hypercalciuria after SC1 but not the peak excretion of calciumi2; and (3) physical activity does not prevent bone loss in SC1patients.28 Treatments of well-established osteoporosis include the use of calcitonin, fluoride, steroids, parathyroid hormone, and bisphosphonates.29-32Although injectable salmon calcitonin has been reported to retard bone density loss in SC1 patients,33 a high incidence of nausea limits compliance.34 The side effects of fluoride32,35and anabolic steroids36preclude their use. The use of parathyroid hormone is still investigational.37 BisphosArch

Phys

Med

Rehabil

Vol 80, March

1999

244

PAMIDRONATE

phonates, including pamidronate, are known to be powerful inhibitors of bone resorption by osteoclasts.38These medications are adsorbed onto the calcified bone matrix3g and are resistant to enzymatic hydrolysis. In tissue culture, they inhibit bone resorption and prevent osteolysis induced by parathyroid hormone.40,41Some bisphosphonates, such as pamidronate, are used for the treatment of hypercalcemia.42 Etidronate, given as an intermittent therapy, was reported to be useful in the treatment of postmenopausal osteoporosis.43Although a study in SC1patients has shown that clodronate attenuated bone loss, unusually high doses of the drug were required (1,600mg per day for 3 months).44A study comparing etidronate and pamidronate in the treatment of hypercalcemia showed that a single 60-mg infusion of pamidronate was safer and more effective in normalizing serum calcium levels than etidronate given as a daily dose of 7Smg/kg for 3 days.45 In comparative studies, pamidronate was found to be more potent than clodronate or etidronate in inhibiting osteoclastic resorption of calcified cartilage, alleviating bone pain, and lowering serum calcium.46-4gReid and associates50~51 have shown, using quantitative computed tomography, that pamidronate produces an early significant increase in both the vertebral mineral density and the hand bone cortical area to total bone area in patients with steroid-induced osteoporosis, compared to patients taking placebo. The increase in bone mass occurs between 0 and 6 months posttreatment, reaches a plateau by 6 months, and is maintained thereafter for up to 12 months of treatment. Gallacher and coworkers5* have also shown that intravenous pamidronate prevents bone loss in glucocorticoid-treated patients over the course of 1 year with a 1.8% increase in femoral neck bone mineral density in treated patients. A single infusion of 30mg pamidronate is effective in normalizing serum calcium levels within 48 hours in immobilization hypercalcemia and lasts for several weeksi Before 1992, the most commonly cited indices of bone metabolism had been the urinary excretion of calcium, phosphorus, and hydroxyproline. These are nonspecific indicators of bone metabolism due to a number of factors unrelated to bone turnover, such as dietary intake of minerals and protein, intestinal transit time and absorption, degradation of nonbone collagen, and fecal excretion of calcium. The N-terminal telopeptide cross-linking domain of bone type I collagen (NTX) has been identified and characterized in urine by specific immunoassay.53 The unique amino acid sequence of this peptide provides a sensitive and specific marker for bone resorption.54-56Urinary excretion of NTX is significantly increased after acute SCI, as shown by Roberts and colleagues.57 In the course of the treatment of postmenopausal osteoporotic women with alendronate, NTX was shown to be elevated before treatment. A response to treatment was detected that correlated significantly with bone mineral density measurements.58 Similarly, pamidronate treatment in healthy volunteers decreases NTX excretion and prevents elevation of the urinary excretion after an g-day course of treatment with triiodothyronine.54 The purpose of this study was to test the effects of intravenous pamidronate treatment administered in patients with acute SCI, a clinical condition known to present with dramatically increased bone catabolism. The efficacy of treatment was assessedby the prospective measurement of both bone mineral density by dual energy x-ray absorptionmeter (DEXA) and by the urinary excretion of NTX. The hypothesis was that pamidronate treatment would decrease the rate of bone density loss in these patients. Arch

Phys

Med

Rehabil

Vol 80, March

1999

AFTER

SCI. Nance

METHODS Subject Inclusion Criteria All patients admitted to our rehabilitation hospital with the diagnosis of SC1 regardless of level or extent were advised of the nature of this study and given the opportunity to participate. Once able to tolerate oral feeding, all subjects were given a diet containing at least 1,OOOmgof calcium. The American Spinal Injury Association (ASIA) Impairment Scale classification was used to grade the completeness of injury. Serum levels of calcium, albumin, phosphate, and alkaline phosphatase and urinary excretion of creatinine were determined preinfusion, on 2 consecutive days after each infusion, and at 12 months postinjury. Urine was collected, frozen, and storeda for determination of creatinine-corrected N-telopeptide. For comparison, urine was collected at a minimum of 4-week intervals after injury from 10 patients with SC1 who were not treated with pamidronate. The control data from SC1 subjects not treated with pamidronate were made available, and the control samples were not taken at all time points indicated for the treated patients. Where data points were missing from the control database, a mathematical interpolation was used to supply the missing points, which did not alter the mean values significantly. Subjects were excluded if they met any of the following criteria: ventilator dependency, bed traction duration of more than 1 month, systemic anticoagulation therapy with heparin (excluding low-dosage thromboembolic prophylaxis), pregnancy, family history of idiopathic osteoporosis, or treatment with a bisphosphonate other than pamidronate. Fourteen subjects were assigned to receive 30mg pamidronate administered intravenously via 4-hour infusions (rate of 7Smgk) every 4 weeks for six treatments, a total of 18Omg. Two subjects did not receive all planned treatments. Heart rate, body temperatures, and adverse events (if any occurred) were recorded after each infusion. Data from 10 control subjects, recruited before those who received pamidronate treatment, were collected, but three subjects were excluded due to treatment with etidronate, a first-generation bisphosphonate. A Lunar DPXb DEXA was used in this study for all bone mass measurements. The DEXA is precise and accurate for measuring bone mineral density, reported to be 0.5% to 2% precise and 3% to 5% accurate.5gA reproducibility study using the machine used in this study and able-bodied volunteers resulted in an average variance of .0003. The initial bone density assessments for each subject were performed upon entrance into the study, which was within 6 weeks of the SCI. Repeat bone density measurements were carried out at 3,6, and 12 months postinjury. The specific regions of interest quantified for each subject were as follows: spine (Ll-L4), the right hip (femoral neck, Ward triangle, and trochanter), right femoral and tibia1 diaphyses, and right femoral and tibia1 epiphyses. Statistical Analysis An analysis of variance (ANOVA) for repeated measure@O using a CLRANOVA softwarec and a Macintosh computer were used to analyze the bone density and NTX data generated from this study. The ANOVA for the bone density data used a 2-by-2 analysis with two between factors and two within factors. The two between-group factors were pamidronate treatment and ambulatory status. The bone density data for pamidronatetreated subjects were compared to data of control subjects not treated with pamidronate. The two within-group factors were bone density location (spine, hip, and femoral and tibia1

PAMIDRONATE

AFTER

SCI,

245

Nance

epiphyses) and time. The measured bone mineral density in g/cm2 was converted to a percent change from the baseline value for each subject. The NTX data were analyzed in two ways. The first analysis compared the pretreatment excretion levels to the immediate posttreatment levels. In a separate analysis, the pretreatment excretion levels were compared to the control subjects at comparable time points after SCI. Similar to the bone density data comparison, the analysis was stratified for motor-incomplete versus motor-complete subjects. The second NTX data analysis is the result of an ANOVA for repeated measures with two between factors and one within factor.

pco.05

RESULTS Patient Characteristics The average age of the pamidronate-treated group was 30.8 t 8.3 years (range 20 to 45); the average age in the control group was 35.1 2 10 years (range 25 to 57) (table 1). There were two women in the pamidronate-treated group, but none in the control group. There were equal proportions of cervical injuries compared to more caudal injury levels in the two groups. There was a higher proportion of ASIA grade D (motor function preserved below the neurologic level of injury, and most key muscles below the neurologic level have a muscle grade greater than or equal to 3/5) compared to ASIA grade A (complete injury with no motor or sensory function preserved in the sacral segments) injuries in the control group (6A/4D) than in the pamidronate-treated group (1OA/4D). Two pamidronatetreated subjects were excluded because one had a family history of idiopathic osteoporosis and the other had a deep-vein thrombosis requiring heparin therapy. Of the pamidronatetreated subjects, two did not receive all six treatments. One was excluded because of a family history of osteoporosis, and the other developed a pruritic rash after the second infusion and did Table Subject

Age

Subjects 1

treated 25

1: Subject

Characteristics

Gender

Level

ASIA Grade

with M

pamidronate T2

A

2 3 4

23 43 20

M M M

T5 Ll C5

A A D

5 6

24 45

F M

TIO C5

A D

7 8

37 29

M M

T8 T8

A A

9 10

31 24

M M

C6 C7

A D

11 12 13

25 40 39

M M F

C4 C6 C6

A D A

14 Subjects

26 M not treated

with

T12 A pamidronate

1 2

28 25

M M

T’l T12

A D

3 4 5

25 57 27

M M M

C6 T12 Ll

A A D

6 7 8

44 35 34

M M M

C8 C6 T12

D D A

9 10

39 37

M M

C6 C8

A A

Exclusion

Familial

Intravenous

Etidronate Etidronate Etidronate

osteoporosis

heparin

treated treated treated

-10

! ASIA

A

ASIA

D

WALK Fig 1. Percent change in bone density (ANOVA) for all locations for pamidronate-treated subjects (a) and the controls who did not receive pamidronate (0). Data are stratified by functional status with the ASIA grade A (complete SCI) compared to the ASIA grade D (motor incomplete with most key muscles below the level of injury at least grade 3 strength). Both between-group comparisonsgroup treatment and ambulatory status-had significant effects, but no additional interaction was observed. By pairwise comparison, the ambulatory, pamidronate-treated subjects had significantly better bone density than the other groups (p < .05).

not receive further pamidronate treatments but was included with the pamidronate-treated group for data analysis. This subject was included in the analysis because of the possibility of experiencing a treatment effect. Bone Density The effect of intravenous pamidronate on bone mineral density loss in patients with SC1 was significant compared with those who did not receive pamidronate (treatment vs controls by ANOVA, F = 6.955, p = .02), with mean overall bone loss of 2.7% in the treated group compared with 8.1% in the control group). The ASIA grades at 3, 6, and 12 months were unchanged. All ASIA grade D subjects were ambulatory by 3 months postinjury and those who walked had significantly less loss of bone density than those with an ASIA grade A injury, none of whom walked (F = 4.957, p = .04). The average loss of bone density in the ASIA D group was 3.1% compared with 7.7% in the ASIA A group. There was no statistical evidence of an interaction between these two factors (p = .26). The ambulatory subjects who received pamidronate treatment showed the best bone density preservation, (p < .05 by pairwise comparison using the Duncan test) (fig 1). The values taken from the ANOVA summary table are illustrated in figure 1. These values illustrate the factors being compared, leaving other factors, such as time and specific bone location, collapsed within the data. Only the specific bone locations listed in the Methods section were included in the analysis. Although the Lunar DEXA does provide a total-body bone mineral density, this was not a measurement used in this study, the analysis, or the illustrations. The effect of time was highly significant (F = 25.9, p < .OOOl). Significant treatment group differences were detected at 6 and 12 months (p = .033 andp = .Oll, respectively) (fig 2). At 12 months postinjury, the control group lost an Arch

Phys

Med

Rehabil

Vol 80, March

1999

246

PAMIDRONATE

0

2

4

6

8

10

12

MONTHS Fig 2. Effect of time was a significant factor on the percent change in bone density from the baseline evaluation (p < .OOl). Specific group differences were apparent at the 6- and 12-month evaluation points where the control subjects showed a significantly greater loss of bone density than the pamidronate-treated subjects (p = .03 and p = .Ol, respectively).

average of 13.5% of bone density compared with 6.8% in the pamidronate-treated group. There was a significant interaction of ambulatory ability by time (p = .0002). The ASIA grade D subjects tended to lose much less bone density over time than the ASIA A subjects, where the simple effects comparison of the 12 months’ time point showed a significant difference (F = 20, p < .OOl), with 15.7% average loss in the ASIA A compared to 4.6% loss in the ASIA D subjects (fig 3). The rate of density loss varied according to bone location (F = 11.917, p < .OOOl). In general, spine density is well maintained whereas the femoral neck, distal femur, and proximal tibia areas of the epiphyses decrease. The bone density loss

AFTER

SCI, Name

from the femur was significantly less in the pamidronate-treated group than in the controls (fig 4). The average loss from the hips of the pamidronate-treated subjects was only 0.9% compared with 8.2% loss in the controls (p = .012). Respective losses from the distal femur were 4.7% and 10.8% (p = ,033). A significant interaction between ambulatory status and bone location was observed (F = 9.957, p < .OOOl),where the spine and hip densities were comparable between the groups but obvious differences were detected about the knees. The loss of density in the proximal knee (distal femur) for ASIA A compared to ASIA D was 12.5% vs 3% (p = .OOl). Loss of density in the distal knee (proximal tibia) was 14.3% vs 3.25% (p < ,001) (fig 5). The treatment effect appears to be independent of the time and location interaction, since there is no significant three-way interaction (F = 1.1, p = .37). A significant three-way interaction was seen with the analysis of walking status, time, and bone location (p < .OOOl),and this interaction was detectable at 6 months and clearly shown at 12 months postinjury (fig 6). Finally, the four factors considered together did not demonstrate a significant interaction (p = 0.2). The average values for individual bone locations across time for control and treated stratified groups are listed in table 2. N-terminal Telopeptide (NTX) As expected, a high excretion level of the urinary bonebreakdown product NTX was found. The average ? SD baseline pretreatment level of this product relative to creatinine excretion in all subjects entered into the trial (within 6 weeks of the spinal cord injury) was 219 t 184 units, which decreased by an average of 61% immediately after the first treatment. High excretion levels recurred before treatments 2 to 6 as follows: 176 ? 137, 177 t 117, 162 2 130, 113 t 86, and 138 +- 148. The final 12-month postinjury average level, which was approximately 5 months after the last infusion, was 106 2 83 units. The NTX data were stratified into four groups based on completeness of spinal cord injury (ASIA grade) and

-15

I

! SPINE

HIP

DISTAL FEMUR

PROX TIBIA

LOCATION

Fig 3. Although ambulatory status and time had significant effects on the percent change in bone density from the baseline time point, a significant interaction was also detected (*p = .002), with a clearly illustrated difference at 1 year after the injury.

Arch

Phys

Med

Rehabil

Vol 80, March

1999

Fig 4. Effect of bone location was significant (p < .OOOl) where the spine density tends to be preserved, but important bone density loss occurs in the lower extremities. Pamidronate-treated subjects showed significantly less bone density loss in the femur (at the hip and distal femur; p = .012 and p = ,033, respectively). A significant treatment effect was not seen in the tibia.

PAMIDRONATE

AFTER

SCI,

Table p=.789

PC.220

p=.oo1

247

Nance

2: Percent

Change in Femoral Neck Density With Baseline Over Time

P<.OOl Area

Control,

neck metaphysis

Tibia1 metaphysis

q

Control, L2-L4

ASIA A ASIAD

-

Femoral neck Femoral metaphysis Tibia1 metaphysis

Femoral Femoral

-HIP

’ DISTAL FEMUR

PROX TIBIA

LOCATION Fig 5. Ambulatory status and bone location showed a significant interaction on the percent change in bone density from the baseline assessment (p < .OOOl) such that ambulatory subjects maintained bone density about the knee (distal femur and proximal tibia) compared to nonambulatory subjects.

treatment (pamidronate) versus control. The database was arranged to conform to an analysis every 4 weeks post-SC1 as follows: 4, 8, 12, 16, 20, and 24 weeks. There were 13 controls and 10 treated complete SC1 subjects, 5 controls and 4 treated incomplete subjects. As illustrated in figure 7, there was a dramatic, short-term suppression of NTX excretion after pamidronate, but the pretreatment scores used in this analysis showed no statistical difference from control; the mean value for treated subjects was 137 vs the control subjects’ mean value of 207 (p = .15). There was a significant effect of completeness of SC1 on NTX excretion (mean complete 232.67 vs incomplete 112.33,

6 Months 1%)

12 Months (%)

-0.3 -3.1

-0.7 -6.2

+0.4 - 14.6

-2.0 -6.8

-8.7 -13.6

-27.4 -27.0

-2.1

-4.2

-5.7

-5.1 -4.6 -3.4

-10.3 -8.9 -6.3

-9.8 -13.4

+1.1

+2.0

+0.3 -1.5

+0.8 -9.6

+3.8 -6.5 -26.0

-1.0

-9.0

-28.5

+2.7 +0.3

-0.7 -0.3

-2.3

+1.2 +1.0

+2.7 -0.4

ASIA-D

Pamidronate-treated, L2-L4

SPINE

3 Months (%i

ASIA-A

L2-L4 Femoral Femoral

0

Compared

neck metaphysis

Tibia1 metaphysis Pamidronate-treated, L2-L4 Femoral

-10.5

ASIA-A

ASIA-D

neck

Femoral metaphysis Tibia1 metaphysis

+0.4 +4.9 -0.03

p = .02). There was no overall effect of time (p = .13), unless one examines the simple effects on the treatment by time interaction. Here, time was significant on the controls (p = .04) with the maximum mean differences at 12 weeks post-SCI. There was a significant interaction between completeness and time as shown by the simple effects, where complete subjects were significantly different from incomplete subjects at 8 weeks (p = .03), 12 weeks (p = ,003); and a trend at 16 weeks (p = .08).

DISCUSSION The results of this experiment are consistent with previous reports that showed a significant decrease of bone mineral

0

Fig 6. Ambulatory status, time, and locations did show a sianificant interaction (p C .OOOi). The data points for the ambulatory subjects are shown as closed shapes, the nonambulatory subjects as open shapes, the hip as squares, the distal femur as circles, the proximal tibia as triangles, across the time points of the study, illustrating the clear loss of bone density in the nonambulatory subjects after the last pamidronate treatment (at 6 months postinjury).

$ _ 6

-20

5 g

-30

L

--c-.+

HIP-D DIST FEMUR PROX TIBIA

-11-o+-

HIP-A DISTAL FEMUR -A PROX TIBIA - A

- D - D

$ TIME (months)

Arch

Phys

Med

Rehabil

Vol 80, March

1999

248

PAMIDRONATE

AFTER

TREATED ma

-

250 -

ol

I 0

, 10

.

, 20

.

, 30

.

, 40

I

, 50

I

TIME (weeks) Fia 7. Urinarv NTX excretion (corrected for creatinine concentration) in 14 SC\ patients treated’with pamidronate (indicated by the lower line graph for those treated) and 18 SCI patients not treated with pamidronate (indicated by the upper liner graph for controls) are shown. The effect of pamidronate can be seen in the decreased NTX excretion after treatment.

density in the lower paralyzed limbs of SC1 patients.14-16The average amount of bone loss during the first year after injury is about 40% to 70%, depending on the site and the age of the patient.61.62 The distribution of bone density loss is unique compared to endocrine causes of osteoporosis1g or spastic cerebral palsy.63In children with SCI, the bone density of the three sites of the hip range from 56% to 65% of able-bodied nonadults, but the bone mineral density increases with growth, and males tend to attain higher bone density, generally.64 Szollar and colleagues65,66observed that significant bone density loss occurs at the three sites of the hip by 1 year postinjury for all persons younger than 60 years and that the decline continues until 19 years after SCI. The loss in bone density increases the risk of SC1patients to incur a leg bone fracture associated with minor trauma. Reviews of lower extremity fractures after SC1 reveal that the most commonly fractured bone is the femur, most often in the supracondylar region (the “paraplegic fracture”); a significant number of femoral neck fractures occur as we11.67-69 The data from the current longitudinal study show that without specific intervention, bone density will be lost in the lower extremities after SCI. The average loss at the end of 1 year, regardless of injury completeness or ambulatory status, was 12.2% at the femoral neck and 18.7% at the proximal tibia. Biering-S@renson and associates? reported the change of bone density in eight subjects who were all wheelchair users over a 53-month period as 60% to 70% loss at the femoral neck and 40% to 50% loss at the proximal tibia. Garland and coworkers18 reported that the average loss of distal femur bone mineral density of 12 ASIA A subjects was 22% in the first 3 months, 27% by 4 months, and 32% at 14 months. Wilmet and colleagues70reported an average loss of trabecular bone density in the legs of 4% per month in the first year after SCI, with ambulatory subjects showing a slightly slower rate of loss. Thus, the control data reported here are a conservative projection of the magnitude of bone density loss after SC1and somewhat lessthan observed in other reports, possibly due to the predominance of ambulatory subjects and Arch

Phys

Med

Rehabil

Vol 80, March

1999

SCI,

Nance

one subject aged 57 years, a point that has been reviewed previously.71 The types of interventions to improve bone density in patients with SC1 that have been reported have involved either electrical stimulation of muscle or pharmacologic targeting of bone metabolism. Rodgers and associates’j2reported that muscle performance improved, but bone density of the tibia did not, using electrical stimulation of the quadriceps muscles in a knee extension exercise system. Sloan’s group72 reported observations in 12 patients with SC1 given electrical stimulation of the quadriceps during cycling on a Kincom isokinetic dynamometer; two with serial bone density assessmentsshowed progressive loss in bone density at the 6- and IZmonth observation periods. Using quantified computerized tomography to measure bone density of the tibia, Hangartner and coworkers73 showed that electrically stimulated knee extension or leg cycle ergometry did not increase bone density but did attenuate the expected rate of trabecular bone loss. Neither BeDell and colleagues74 nor Needham-Shropshire and associates75observed improvement in bone density at the femoral neck, Ward triangle, or the greater trochanter in 12 and 15 ASIA A SC1 subjects, respectively, after 8 weeks of electrically stimulated REGYS I ergometer exercise or 12 weeks of electrically evoked walking using the Parastep 1 ambulation system. Bloomfield and coworkers76 showed that four subjects trained on a REGYS I system at high intensity (more than 18W) had an average increase of 17.8% in bone density over 9 months only at the distal femur. Similarly, Mohr’s group77 observed that electrically stimulated cycling exercise training for 12 months (3 days per week) in 10 motor-complete SC1 subjects increased the bone mineral density of the proximal tibia by IO%, but those improvements were lost after 6 more months of reducedfrequency exercise (once per week). These observations, although somewhat contradictory, could be viewed as consistent with the data presented here. The lack of effect of electrically induced exercise upon the regions about the hip are consistent and the positive effects about the knee are also consistent. The marginal effects reported by Hangarnter73 may be related to insufficient intensity or frequency of the exercise. The present data show that ambulatory status has the most profound effects upon the high-percentage trabecular bone regions about the knee, the distal femur, and proximal tibia, and relatively less effects on the hip, which is consistent with the lack of effect of ambulatory status in the femur, but a positive correlation with an index of mobility in the tibia.78,79 The potential side or adverse effects of pamidronate that were monitored are as follows: postinfusion fever, nausea, hypocalcemia, hypernatemia, hyper/hypokalemia, leukopenia, allergic reaction, and local reactions at the site of infusion. One subject experienced a generalized, red, itchy rash after the second infusion. None of the other potential side or adverse effects were observed. Other medications have been evaluated as potential preventive treatments in post-SCI-related bone mineral density loss. A clinical trial with SC1patients stratified for ambulation capacity failed to show a drug effect with a first-generation bisphosphonate, etidronate, but did reveal a significant interaction between ambulatory ability and drug treatment29; however, inhibitory effects on bone formation may limit its usefulness during recovery from a fracture. A report that tiludronate, a newergeneration bisphosphonate, at the higher dose tested (4OOmg/d), produced a slight increase in the bone volume/total volume of a transiliac bone biopsy, is supporting evidence that a highpotency bisphosphonate would be generally effective in reducing bone density loss after SCI.*O A long-term follow-up of

PAMIDRONATE

women treated for 2 years with pamidronate showed that 1 year after stopping treatment, bone mineral density loss resumes throughout the skeleton except for the lumbar spine, femoral neck, and Ward triangle. 8’ Furthermore, pamidronate has been used successfully to reduce bone mineral density loss in patients with reflex sympathetic dystrophy.82 Finally, a new bisphosphonate that can be administered orally, risedronate, improved spinal bone mineral density by 5.3% in 11 multiple myeloma patients treated for 6 months.x3 As other bisphosphonates, such as with zoledronate, are evaluated, perhaps greater efficacy may be discovered. Active rehabilitation programs often include wheelchair sports and gait retraining, which increase the demand upon the leg bones. It is logical that prevention of osteoporosis will diminish the risk of fractures and reasonably permit the inclusion of walking retraining in general rehabilitation programs. If mobilization strategies such as functional electrical stimulation for walking are to be carried out with maximal safety in SC1 patients, bone density must be maintained at normal or near-normal levels. With the prevention of predictable adverse events, such as low-impact fractures, and the overall improvement in general by improved exercise performance and mobility, the quality of life for SC1patients would be expected to be enhanced. The positive results from this study confirm the original hypothesis that intravenous pamidronate can reduce significantly bone resorption after acute SCI. However, the data are not uniform in that not all bony locations for all patients were protected, suggesting that the original treatment protocol is not optimal. The data indicate that nonambulatory SC1 patients are likely to require a longer duration of treatment. CONCLUSIONS The data reported in this prospective trial of SC1 subjects support the hypothesis that pamidronate treatment during the acute phase of SC1 will retard the development of osteoporosis after SC1 with a low incidence of side or adverse effects. The fact that pamidronate must be given as an intravenous injection however, is a significant limitation. The stratification of subjects based on ambulation capacity shows that nonambulatory subjects are not adequately protected by the treatment protocol described in this report and that more work is needed to find such a protocol. From the NTX excretion data presented, it appears that the dosage frequency is insufficient. However, from the ambulatory stratification, one could speculate that early weight-bearing in combination with longer-term bisphosphonate treatment, or a different bisphosphonate with greater potency and longer duration of action, may improve the protection of bone density in the tibia.

AFTER

5.

6.

I. 8.

9.

10. 11. 12. 13.

14. 15. 16. 17. 18. 19. 20.

21.

23.

Acknowledgment:

25. 26.

References

Albright F, Burnett CH, Gope 0, Parsons W. Acute atrophy of bone (osteoporosis) simulating hyperparathyroidism. J Clin Endocrinol 1941;1:711. Wyse DM, Pattee CJ. Effect of oscillating bed and tilt table on calcium, phosphorus and nitrogen metabolism in paraplegia. Am J Med 1954;645-61. Klein L, Curtis PH. Urinary hydroxyproline as an index of bone metabolism. In: Pearson OH, Joplin GF, editors. Dynamic studies of metabolic bone diseases. Oxford: Blackwell Scientific Publications; 1964. p. 201-44. Minaire P, Meunier P, Edouard C, Bernard J, Courpron P, Bout-ret J. Quantitative histological data on disuse osteoporosis. Comparison with biological data. Calcif Tissue Int 1974;17:57-73.

Claus-Walker J, Spencer WA, Carter RE, Halstead LS, Meier III RH, Campos RJ. Bone metabolism in quadriplegia: dissociation between calciuria and hydroxyprolinuria. Arch Phys Med Rehabil 1975;56:327-32. Kaplan PE, Gandhavadi B, Richards L: Goldschmidt J. Calcium balance in paraplegic patients: influence of injury duration and ambulation, Arch Phvs Med Rehabil 1978:59:447-50. Chantraine A. Actual concept of osteoporosis in paraplegia. Paraplegia 1978-1979;16:51-8. Naftchi NE, Viau AT, Sell GH. Lowman EW. Spinal cord injury: effect of thyrocalcitonin on calcium, magnesium and phosphorus in paraplegic rats. Arch Phys Med Rehabil 1980;61:575-9. Stewart AF, Adler M, Byers CM, Serge GV, Broadus AE. Calcium homeostasis in immobilization: an example of resorptive hypercalciuria. N Engl J Med 1982;306: 1136-40. Maynard FM. Immobilization hypercalcemia following spinal cord iniurv. Arch Phvs Med Rehabil 1986:67:41-4. Char&ink A. Bone>emodeling during the development of osteoporosis in paraplegia. Calcif Tissue Int 1986;38:323-7. Claus-Walker J, Scurry RE, Carter RE, Campos RJ. Steady-state hormonal secretion in traumatic quadriplegia. J Clin Endocrinol Metab 1977;44:530. Gallacher SJ, Ralston SH, Dryburgh FJ, Logue FC, Allam BF, Boyce BF, et al. Immobilization-related hypercalcaemia-a possible novel mechanism and response to pamidronate. Postgrad Med J 1990;66:918-22. Griffiths HJ, Bushueff B, Zimmerman RE. Investigation of the loss of bone mineral in patients with spinal cord injury. Paraplegia 1976;14:207-12. Kiratli BJ: Agre JC. Disuse osteoporosis in spinal cord injured oatients. Rehabil Res Dev Proer Rev 1990:425-6. Tiering-Sorensen F, Bohr HH: Schaadt OP. Bone mineral content in the lumbar spine, the forearm and the lower extremities after spinal cord injury. Paraplegia 1988;26:293-301. Leeds EM. Klose KJ. Ganz W. Serafini A. Green BA. Bone mineral density after bicycle ergometry training. Arch Phys Med Rehabil 1990;71:207-9. Garland DE; Stewart CA, Adkins RH, Hu SS, Rosen C, Liotta FJ, et al. Osteoporosis after spinal cord injury. J Orthop Res 1992;lO: 371-8. Leslie WD, Nance PW. Dissociated hip and spine demineralization: a specific finding in spinal cord injury. Arch Phys Med Rehabil 1993;74:960-4. Chow YW, Inman C, Pollintine P, Sharp CA, Haddaway MJ. Masry WE, et al. Ultrasound bone densitometry and dual energy X-ray absorptiometry in patients with spinal cord injury. Spinal Cord 1996;34:736-41. Wolff JD. Geretz de transformation der knochen. Berlin: Hirchwald: 1892.

22.

24.

The authors would like to thank Dr. David Eyre, University of Washington Medical Center, Seattle, for his participation in the analysis of the NTX data.

249

SCI, Nance

27.

28.

Martin RB. Determinants of the mechanical properties of bones. J Biomech 1991;24(S1):79-88. Lanyon LE. The success and failure of the adaptive response to functional load-bearing in averting bone fracture. Bone 1992;13: S17-22. Goodship AE. Mechanical stimulus to bone. Ann Rheum Dis 1992;5 1:4-6. Yasuda I. Fundamental aspect of fracture treatment, J Kyoto Med Sot 1953;4:395-406. Brighton CT, Friedenberg ZB, Mitchell EL, Booth RE. Treatment of nonunion with constant direct current. Clin Orthop 1977;124: 106-23. Inoue S, Ohashi T, Kajikawa KS Ibaraki K, Sasaki H, Tada M. Application of electric stimulation methods for bone fracture and d&ease. Clin Orthop Surg 1983;18:1291-8. Kunkel CF. Scremin AME. Eisenbem B. Garcia JF. Roberts S. Martinez S. Effect of “standing” onvspasticity, conuacture, and osteoporosis in paralyzed males. Arch Phys Med Rehabil 1993;74: 73-8.

Pearson E, Nance PW, Leslie WD, Ludwig S. Cyclical etidronate: its effect on bone density in patients with acute spinal cord injury. ” Arch Phys Med Rehabil-1997;78:269-72. 30. Avioli LV. Calcitonin therapy in osteoporotic syndromes. J Am Med Womens Assoc 1990;45: 103-7. 29.

Arch

Phvs

Med

Rehabil

Vol 80, March

1999

250

PAMIDRONATE

31. Stevenson JC. Pathogenesis, prevention and treatment of osteoporosis. Obstet Gynecol 1990;75:368-418. 32. Licata AA. Therapies for symptomatic primary osteoporosis. Geriatrics 1991;46:62-7. 33. Minaire P. Immobilization osteoporosis: a review. Clin Rheumatol 1989;s Suppl2:95-103. 34. Pearson E, Name PW, Leslie WD, Ludwig S. Severe adverse effects following the use of calcitonin to prevent osteoporosis in acute spinal cord injury (SCI) [abstract]. Clin Invest Med 1993;16 Suppl4:B112. 35. Stone MD, Hosking DJ. Treatment of osteoporosis: current and future. Ann Rheum Dis 1991;50:663-5. 36. Consensus development conference: prophylaxis and treatment of osteonorosis [conference reoortl. Am J Med 1991;90:107-10. 37. Reeve J, Davies UM, Hesp’ R, I\llcNally E, Katz ‘D. Treatment of osteoporosis with human parathyroid peptide and observations on effect of sodium fluoride. BMJ 1990;301:314-8. 38. Kanis JA, McCloskey EV, Taube T, O’Rourke N. Rationale for the use of bisphosphonates in bone metastases. Bone 1991;12 Suppl l:S13-8. 39. Powell JH, Denmark BR. Clinical pharmacokinetics of diphosphonates. In: Garrafini S, editor, Bone resorption metastasis and diphosphonates. New York Raven Press; 1985. p. 41-9. 40. Russell RGG, Fleisch H. Pyrophosphate and diphosphonates in skeletal metabolism. Physiological, clinical and therapeutic aspects. Clin Orthop 1975;108:241-63. 41. Shinoda H, Ademek G, Felix R, Fleisch H, Schenk P, Hagan P. Structure-activity relationships of various bisphosphenates. Calcif Tissue Int 1983;35:87-99. 42. Elomaa I, Blomqvist C, Porkka L, Lamberg-Allardt C, Borgstrom GH. Treatment of skeletal disease in breast cancer: a controlled clodronate trial. Bone 1987;8 Suppl 1:53-6. 43. Mallette LE, Leblanc AD, Pool JL, Mechanick JI. Cyclic therapy of osteoporosis with neutral phosphonate and brief, high-dose pulses ofetidronate. J Bone Miner Res 1989;4: 143-8. 44. Minaire P. Deuassio J, Berard E. Meunier PJ, Edouard C. Pilonchery’ G, ei al. Effects of clod&ate on immobilization bone loss. Bone 1987;8 Suppl l:S63-8. 45. Gucalp R, Ritch P, Wiernik PH, Sarma PR, Keller A, Richman SP. et al. Comparative study of pamidronate disodium and etidronate disodium in the treatment of cancer-related hypercalcemia. J Clin Oncol 1992;lO: 134-42. 46. Boonekamp PM, van der Wee-Pals LJA, van Wijk-van Lennep MML, Thesing CW, Bijvoet OLM. Two modes of action of bisphosphonates on osteoclastic resorption of mineralized matrix. Bone Miner 1986;1:27-39. 41. Ralston SH. Gallacher SJ. Pate1 U. Drvburnh FJ. Fraser WD. Cowan RA,‘et al. Comparison of three mtra;enous bisphospho: nates in cancer-associated hypercalcaemia. Lancet 1989;2: 1180-2. 48. Carey PO, Lippert MC. Treatment of painful prostatic bone metastases with oral etidronate disodium. Urology 1988;32:403-7. 49. Pelger RCM, Lycklama A, Nijeholt AAB, Papapoulos SE. Shortterm metabolic effects of pamidronate in patients with prostatic carcinoma and bone metasmses. Lancet 1989;2:865. ^ 50. Reid IR. Alexander CJ. King AR. Ibbertson HK. Prevention of steroid-induced osteoporosii with (3-amino-l-hydroxypropylidene)-1, 1-bisphosphonate (APD). Lancet 1988;1:143-6. 51. Reid IR, Schooler BA, Stewart AW. Prevention of glucocorticoidinduced osteoporosis. J Bone Miner Res 1990;5:619-23. 52. Gallacher SJ, Fenner JAK, Anderson K, Bryden FM, Banham SW, Logue FC, et al. Intravenous pamidronate in the treatment of osteoporosis associated with corticosteroid dependent lung disease: an open pilot study. Thorax 1992;47:932-6. 53. Hanson DA, Weis MAE, Bollen AM, Maslan SH, Singer FR, Eyre DR. A specific immunoassay for monitoring human bone resorption: Quantification of type I collagen cross-linked N-telopeptides in urine. J Bone Miner Res 1992;7:1251-8. 54. Ebeling PR, Atley LM, Guthrie JR, Burger HG, Dennerstein L, Hopper JL, et al. Bone turnover markers and bone density across the menopausal transition. J Clin Endocrinol Metab 1996;81:336671. 55. Rosen HN, Dresner-Pollak R, Moses AC, Rosenblatt M, Zeind AJ, Clemens JD, et al. Specificity of urinary excretion of cross-linked Arch

Phys

Med

Rehabil

Vol 80, March

1999

AFTER

56.

57.

58.

59. 60. 61.

62.

63.

64.

65.

66.

67. 68.

SCI, Name

N-telopeptides of type I collagen as markers of bone turnover in humans. Calcif Tissue Int 1994;54:26-9. Gertz BJ, Shao P, Hanson DA, Quan H, Harris ST, Genant HK, et al. Monitoring bone resorption in early post menopausal women by an immunoassay for cross-linked collagen peptides in urine. J Bone Miner Res 1994;9: 135-42. Roberts D, Lee W, Cuneo RC, Wittmann J, Ward G, Flatman R, et al. Longitudinal study of bone turnover after acute spinal cord injury. J Clin Endocrinol Metab 1998;83:415-22. Garner0 P, Shih WJ, Gineyts E, Karpf DB, Delmas PD. Assessment of 8 markers of bone turnover in post menopausal (PMP) osteoporotic women treated with alendronate (ALN) [abstract]. Second Workshop on Bisphosphonates; 1994; Davos, Switzerland. Johnston CC, Slemenda CW, Melton LJ. Clinical use of bone densitometry. N Engl J Med 1991;324:1105-9. Winer BJ. Statistical principles in experimental design. 2nd ed. New York: McGraw-Hill; 197 1. Biering-Sorensen F, Bohr HH, Schaadt OP. Longitudinal study of bone mineral content in the lumbar spine, the forearm and the lower extremities after spinal cord injury. Eur J Clin Invest 1990;20:330-35. Rodgers MM, Glaser RM, Figoni SF, Hooker SP, Ezenwa BN, Collins SR, et al. Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training. J Rehabil Res Dev 1991;28:19-26. Henderson RC. Bone density and other possible predictors of fracture risk in children and adolescents with spastic quadriplegia. Dev Med Child Neurol 1997:39:224-7. Moynaham M, Betz RR, Triolo RJ, Maurer AH. Characterization of the bone mineral density of children with spinal cord injury. J Spinal Cord Med 1996;19:249-54. Szollar SM, Martin EME, Parthemore JG, Sartoris DJ, Deftos LJ. Demineralization in tetraplegic and paraplegic man over time. Spinal Cord 1997;35:223-8. Szollar SM, Martin EME, Parthemore JG, Sartoris DJ, Deftos LJ. Densitometric patterns of spinal cord injury associated bone loss. Spinal Cord 1997;35:374-82. Comarr AE, Hutchinson RH, Bors E. Extremity fractures of patients with spinal cord injuries. Am J Surg 1962;103:732-9. Ragnarsson KT, Sell G. Lower extremity fractures after spinal cord injury: a retrospective study. Arch Phys Med Rehabil 1981;62: 418-23.

Ingram RR, Suman RK, Freeman PA. Lower limb fractures in the chronic spinal cord injured patient. Paraplegia 1989;27: 133-9. 70. Wilmet E, Ismail AA, Heilporn A, Welraeds D, Bergmann P. Longitudinal study of the bone mineral content and of soft tissue comuosition after soinal cord section. Paraoleeia 1995:33:674-7. 71. Uebelhart D, Den&x-Domenech B, RothAM:Chantraine A. Bone metabolism in spinal cord injured individuals and in others who have prolonged immobilisation: a review. Paraplegia 1995;33:

69.

669-73. 72.

73.

14.

15.

76.

77.

Sloan KE, Bremner LA, Byrne J, Day RE, Scull ER. Musculoskeleta1 effects of an electrical stimulation induced cycling program in the spinal injured. Paraplegia 1994;32:407-1.5. Hangartner TN, Rodgers MM, Glaser RM, Barre PS. Tibia1 bone density loss in spinal cord injured patients: effects of FES exercise. J Rehabil Res 1994;31:50-61. BeDell KK, Scremin AME, Perell KL, Kunkel CF. Effects of functional electrical stimulation-induced lower extremity cycling on bone density of spinal cord injured patients. Am J Phys Med Rehabil 1996;75:29-34. Needham-Shropshire BM, Broton JG, Klose J, Lebwohl N, Guest RS, Jacobs PL. Evaluation of a training program for persons with SC1 paraplegia using the Parastep@ 1 ambulation system: part 3. Lack of effect on bone mineral density. Arch Phys Med Rehabil 1997;78:799-803. Bloomfield SA, Mysiw WJ, Jackson RD. Bone mass and endocrine adaptations to training in spinal cord injured individuals. Bone 1996;19:61-8. Mohr T, Podenphant J, Biering-Sorensen F, Galbo H, Thamsborg G, Kjaer M. Increased bone mineral density after prolonged electrically induced cycle training of paralyzed limbs in spinal cord injured man. Calcif Tissue Int 1997;61:22-5.

PAMIDRONATE

78. Changlai SP, Kao CH. Bone mineral density in patients with spinal cord injuries. Nucl Med Commun 1996;17:385-8. 79. Saltzstein RJ, Hardin S, Hastings J. Osteoporosis in spinal cord injury: using an index of mobility and its relationship to bone density. J Am Paraplegia Sot 1992;15:232-4. 80. Chappard D: Minaire P, Privat C, Berard E, Mendoza-Sarmiento J, Tournebise H, et al. Effects of tiludronate on bone loss in paraplegic patients. J Bone Miner Res 1995;10:112-8. 81. Orr-Walker B, Wattie DJ, Evans MC, Reid IR. Effects of prolonged bisphosphonate therapy and its discontinuation on bone mineral density in post-menopausal osteoporosis. Clin Endocrinol1997;46: 87-92. 82. Chapurlat RD, Duboeuf FP, Liens D, Meunier PJ. Dual energy

AFTER

251

SCI, Nance

x-ray absortiometry in patients with lower limb reflex sympathetic dystrophy syndrome. J Rheumatol 1996;23: 1557-9. 83. Roux C, Ravaud P, Cohen-Solal M: de Verneioul MC: Guillemant S, Cherruau B, et al. Biologic, histologic andhensitometric effects of oral risedronate on bone in patients with multiple myeioma. Bone 1994;15:41-9. Suppliers a. Osteomark kit, Ostex International, Inc., 2203 Airport Way South, Suite 400, Seattle, WA 98 134. b. Lunar DPX; Lunar Corporation, 313 West Beltline Highway, Madison, WI 53713. C. CLRANOVA Software, Anova version 1.1; Clear Lake Research. Inc., 2300 East 14th Street. Tulsa OK 74104.

Arch

Phys

Med

Rehabil

Vol 80, March

1999