| | Incidence of Fractures in a Cohort of Veterans With Chronic Multiple Sclerosis or Traumatic Spinal Cord InjuryPresented to the American Spinal Association, September 2006, Las Vegas, NV, and the American Geriatrics Association, May 2006, Chicago, IL. Abstract Logan WC Jr, Sloane R, Lyles KW, Goldstein B, Hoenig HM. Incidence of fractures in a cohort of veterans with chronic multiple sclerosis or traumatic spinal cord injury. ObjectiveTo measure skeletal fractures in a cohort of veterans with spinal cord dysfunction (SCD) due to multiple sclerosis (MS) or trauma-related spinal cord injury (SCI). DesignRetrospective cohort analysis. ParticipantsStudy subjects were a subset of the 1996 Veterans Health Administration (VHA) National Spinal Cord Dysfunction Registry, from which 8150 patients were identified with either MS (n=1789) or SCI (n=6361). Inpatient and outpatient encounters for nonaxial fractures, based on International Classification of Diseases, Ninth Revision, Clinical Modification codes, were identified through VHA administrative databases between October 1996 and June 2005. VHA Beneficiary Identification Records Locator Subsystem death file identified time of death. InterventionsNot applicable. Main Outcome MeasuresData from the 1996 VHA National Spinal Cord Dysfunction Registry survey was used to determine duration of disease and motor impairment (4 categories of motor impairment based on self-report of the number of limbs involved and degree of motor loss). Proportional hazard modeling evaluated the time to first fracture and Poisson regression evaluated relative risk (RR) of fracture by cause of SCD and degree of motor impairment, adjusting for age, sex, race, and duration of SCD. ResultsSubjects were, on average, 52.5 years of age, acquired their SCD 22 years prior, and 386 of 8150 were deceased. During the study period, 4021 fracture encounters were identified representing 1738 unique fractures for 1085 of 7832 subjects, for a mean per-person fracture rate of 3.1 per 100 patient-years at risk. The RR of fracture differed according to cause of SCD and motor impairment. Fracture risk was increased by more than 2-fold in those with some motor impairment (RR=2.33, P<.001), by more than 80% with moderate motor impairment (RR=1.87, P<.001), and almost 70% for those with severe motor impairment (RR=1.67, P<.001), compared with those with little motor impairment. Trauma-related SCI increased the RR of fracture 80% (RR=1.82, P<.001) compared with MS. ConclusionsPersons with SCD have high rates of skeletal fractures. The highest fracture rates occurred in those with some to moderate motor impairment. There were significant differences in risk of fracture according to causal disease, controlling for motor impairment and duration. There appear to be unique contributors to risk of fracture beyond simply disuse. TRAUMATIC OR MEDICAL spinal cord dysfunction (SCD) can be devastating, resulting in severe disability. Subsequent to the injury or disease, persons with SCD may develop a variety of conditions that may primarily or secondarily further affect functional independence.1 Among these is osteoporosis leading to an increased rate of long-bone fractures with minimal trauma. Reports as early as 1941 note that persons with neurologic injury of any kind experienced increased bone loss and increased rates of fracture with minimal trauma when compared with the general population.2, 3, 4, 5, 6, 7, 8, 9 Trauma-related spinal cord injury (SCI) results in a rapid loss of bone density distal to the spinal level of injury,10, 11, 12 with measured rates of fractures ranging from 1.2 to 3.4 per 100 patient-years at risk.4, 5, 6, 13, 14, 15, 16 The dominant fracture location in all studies were the femoral shaft and proximal tibia/fibula—bones distal to the level of neurologic impairment.16, 17, 18 However, each of these prior studies was limited in length of follow-up and in representation of diverse levels of neurologic impairment and/or specification of the spinal level of injury. In persons with SCD due to multiple sclerosis (MS), osteoporosis and fractures have also been studied. Similar to trauma-related SCI, loss of bone density in MS occurs rapidly after diagnosis. The literature on fractures in MS primarily consists of case reports or case series with wide variation in the reported frequency of fractures.19, 20, 21 For example, Stenager and Jensen22 found that 5.7% of 299 persons with MS had a history of fractures; whereas Cosman et al23 found that 22% of their study subjects had experienced a prior fracture compared with 2% in controls. The most common locations of fractures in these studies were the hip and lower extremity. As with traumatic SCI, these studies were also limited in length of follow-up, number of persons included, and description of the degree of motor impairment. Depending on the level and completeness of spinal cord damage, SCD results in differing degrees of motor impairment. Osteoporosis in SCD is often attributed directly to disuse after injury. Osteoporosis seems to be more extensive and possibly more severe among those with more neurologic impairment. Szollar et al11 found that the extent of osteoporosis but not the severity was greater in tetraplegic men when compared with paraplegic men. Garland et al24 and Szollar11 have shown a linear relationship between higher level of spinal injury, more complete neurologic loss, and decreasing bone mineral density. However, when fractures were evaluated by level of spinal injury, both Wang25 and Ragnarsson4 and colleagues found that paraplegics were more likely to suffer an osteoporotic fracture than tetraplegics. The increased rate of fractures in paraplegics compared with tetraplegics, despite less neurologic impairment, may be due to their relatively greater activity level, exposing the patient to more potential opportunities to fracture. Comparing rates of fractures, according to degree of motor function, between persons with MS and SCI could shed light on this issue by allowing the opportunity to examine the impact of the neurologic disorder as opposed to the degree of motor impairment. The Veterans Health Administration (VHA) provides care for a large number of people with SCD. This study uses a subset of the VHA National Spinal Cord Dysfunction Registry to measure fractures in a large cohort of persons with either trauma-related SCI or MS and seeks to answer the following questions: What was the overall fracture rate in this cohort over a 9-year study period? Did the location of the fractures differ according to the cause of SCD? Did fracture rates differ according to degree of motor function? Methods  Study Design and Participants This is a retrospective cohort analysis. A subset of the VHA National Spinal Cord Dysfunction Registry developed in 1996 by Samsa et al26 was used for this study. Briefly, the VHA National Spinal Cord Dysfunction Registry was populated based on a national survey sent to veterans selected from the VHA population by diagnosis codes (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM]) that might indicate SCD and lists from the Paralyzed Veterans of American and VHA Spinal Cord Injury centers. A total of 18,038 persons returned surveys confirming that they had SCD, and verified their SCD diagnosis. Of that total, 8150 persons with affirmed SCD due to MS alone (n=1789) or to traumatic SCI alone (n=6361) were used for this study (ie, they did not have other conditions that might also affect the spinal cord as the primary cause of their SCD or comorbid with SCI or MS, such as tumor, spinal stenosis, or arthritis, as was reported by the persons not included in this analysis).27 Additional information was collected as a part of the national survey, including date when the SCD was diagnosed, information about physical function, and use of VHA and non-VHA services. The majority of the persons who used the VHA for care (81%) received 100% of their care for any illness within the VHA system, providing an opportunity to capture the vast majority of fracture-related encounters for these persons.28 We chose a study period beginning on October 1, 1996, because ICD-9-CM coding for inpatient and outpatient encounters was implemented in the VHA system on that date. Administrative data for encounters with primary diagnosis codes indicating fractures (defined below), including date of fracture-related visit or hospitalization, site of visit, and all additional diagnoses for the visits, was gathered for all inpatient and outpatient encounters recorded for the members of the study cohort beginning October 1, 1996, and ending June 8, 2005. Measures Dependent variables The primary outcome was nonaxial skeletal fracture as determined by either ICD-9-CM code or related Current Procedural Terminology code (ICD-9 codes 810−819 identified upper-extremity fractures, 820−829 identified lower-extremity fractures).29, 30 Axial fractures, including spine, skull, and face (identified by codes 800−804 and 805−809) were excluded due to potential use of these codes for persons with previous spinal injuries in recognition of the original spinal injury. After an initial review of all of the encounters, visits for the same diagnosis code that occurred within 120 days of one another were collapsed into single fractures—this was done by inspection of the data, clinical knowledge of fracture healing, and general clinical follow-up patterns. Encounters occurring within the 120-day period that were coded with a different 3-digit ICD-9 code (representing a different fracture location) were considered additional fractures. Independent variables We measured motor impairment by a categorical variable derived from data obtained as a part of the original VHA National Spinal Cord Dysfunction Registry Survey based on the self-reported number of affected limbs (0−4) and degree of motor impairment (full-, some-, or no-movement). As described elsewhere,27 the number of affected limbs and degree of motor impairment as collapsed into a 4-level categoric variable were as follows: little motor impairment included all who reported full movement, no matter how many limbs they reported affected by their disease; some motor impairment included persons reporting 0 to 2 limbs involved and also reporting either some or no movement; moderate motor impairment included persons with either 3 or 4 limbs involved and some movement as well as those with 3 limbs involved but no movement; and severe motor impairment identified persons reporting 4 limbs involved and no movement. In a separate study, this categoric measure of motor impairment has been shown to have good construct validity in relation to a self-reported impairment in activities of daily living.31 This measure of motor impairment is particularly useful for comparative study of persons with diverse causes of SCD (eg, trauma SCI, MS). Neither Frankel nor the American Spinal Injury Association (ASIA) grade, or other physician-reported neurologic levels of injury apply to persons with MS. The Kurtzke scale used in MS is not used after traumatic SCI. Thus, these commonly used scales could not be used in a study that measured fractures in both populations comparatively. Despite its advantages, the classification system used in this study is not equivalent to neurologic level, but rather it is descriptive of the functional impairment related to the underlying neurologic damage, which is measured more precisely by the other classification systems. For example, persons with complete tetraplegia at the C6 level may report that all 4 limbs were affected but that they have “some motor function” in the affected limbs due to preserved shoulder, elbow, and wrist function, whereas someone with complete C3 tetraplegia might report that 4 limbs were affected and that they had “no motor function” in the affected limbs. Other independent variables included in the multivariate models were cause of SCD (trauma SCI, MS), age, sex, race, and time since onset of SCD (duration of disease). Statistical Analysis We used descriptive statistics to describe fractures, stratified by both SCD cause and degree of motor impairment. Fracture location was determined by matching the 3-digit ICD-9 code to its respective definition. Fracture location was determined by grouping fractures by region: upper extremity; rib; hip, pelvis, and proximal femur; femoral shaft and distal femur; tibia and fibula; ankle, foot, and toes; and other or unclassifiable. Fracture locations were then expressed as a percentage of total fractures by cause of SCD. The Cochran-Mantel-Haenszel statistic was used to evaluate the overall difference in distribution of fracture location by cause of SCD. Per-person estimates of fracture rates were expressed as number of fracture episodes per 100 patient-years at risk over the study period. As a second step, multivariable Cox proportional hazards survival analysis was used to model the relative hazard of time to first fracture by both SCD cause and degree of motor impairment. Persons without a fracture by June 8, 2005, were right-censored, as were persons who died during the study period. In addition, Poisson regression was used to model fracture rate by SCD cause and degree of motor impairment. Multivariate analyses were adjusted for age on October 1, 1996, race, sex, and duration of SCD (based on self-reported date of onset). Unadjusted and adjusted logistic regression analyses were used to model any fracture versus no fracture over the time at risk. Results are reported with 95% confidence intervals (CIs) and P values where applicable. All statistical analyses were performed using SAS, version 9.2,a or Enterprise Guide, version 3.0.a This study received the approval of both the Duke University Medical Center institutional review board (IRB) and the Durham VA Medical Center IRB. Results Study cohort characteristics are shown in table 1. Of the original 8150 veterans in the original VHA National Spinal Cord Dysfunction Registry cohort, 7832 were alive on October 1, 1996 (MS=1700, trauma-related SCI=6132). The average age of study members on October 1, 1996, was 52.5±13.0 years, and they were predominantly men (95.9%) and of white race (77.8%). The mean reported duration of SCD in the study cohort was 22±13 years. Overall, 79.7% of the study cohort reported either some motor impairment (0 to 2 limbs involved and some or no movement) or moderate motor impairment (3 to 4 limbs involved and some movement or 3 limbs involved and no movement). Persons with MS most frequently reported moderate motor impairment (44.5%), whereas persons with trauma most frequently reported some motor impairment (54.5%). There were 1080 (13.7%) of 7832 veterans who experienced a total of 1738 fractures. The majority of the fractures occurred during a single episode (n=1293 fractures), with some persons having more than 1 fracture during the same fracture episode (ie, multiple bones were fractured simultaneously). There were 367 veterans who experienced 445 additional fractures subsequent to their first fractures. Persons with traumatic SCI experienced the majority of fractures (n=1519 fractures [87.4% of all fractures]). Figure 1 shows the distribution of fracture location by cause of SCD. The most frequent locations of fractures in persons with MS were the upper extremity and ankle, toe, or foot. When compared with the persons with trauma, the distribution of fracture locations in persons with MS was more evenly distributed. As expected, persons with trauma tended to fracture more frequently in either their distal femur or proximal tibia or fibula, accounting for more than 50% of their fractures. Overall, the distribution of fracture locations in persons with trauma was significantly different from the fracture locations in persons with MS (P=.001). Fracture rates differed significantly according to cause of SCD and amount of motor impairment. Figure 2 depicts this graphically. Persons with both MS and trauma-related SCI had higher mean fracture rates per 100 patient-years at risk if they had some motor impairment or moderate motor impairment than if they had little motor impairment or severe motor impairment. Those with severe motor impairment also had higher fracture rates than those with little motor impairment, but not to the same degree as those with some or moderate motor impairment. The highest fracture rate among the groups was seen in persons with trauma who had either some or moderate motor impairment—over 3.5 fractures per 100 patient-years at risk. The lowest fracture rate (≈1 per 100 patient-years at risk) was found in those with MS who had little motor impairment. The overall relationship of fracture rate to degree of motor impairment was nonlinear for both MS and trauma—higher in the some motor impairment and moderate motor impairment categories than in the little motor impairment or severe motor impairment categories. Results of adjusted Cox proportional hazard and Poisson regression models to test the strength of these relationships are shown in table 2 and figure 3 shows Kaplan-Meier curves for time to first fracture by degree of motor impairment and cause of SCD. The relative risk (RR) of fracture differed significantly according to cause of SCD and motor impairment, controlling for the presence of both in the same model. Compared with those with little impairment, some motor impairment increased the RR of fracture over 2-fold (RR=2.33, P<.001), moderate motor impairment increased the RR by 87% (RR=1.87, P<.001), and severe motor impairment increased the RR by 67% (RR=1.67, P<.001). Traumatic SCD increased the RR of fracture 82% (RR=1.82, P<.001). Survival analysis (time to first fracture) patterns were similar to the Poisson regression models (fig 3A). The hazard of fracture was significantly higher for those with some, moderate, or severe motor impairment or compared with those with little motor impairment, with the shortest time to fracture being seen in those with some motor impairment (hazard ratio [HR]=1.93, P<.001), and in those with traumatic SCD (HR=1.67, P<.001) compared with those with MS. Longer duration of SCD was a significant predictor of both time to first fracture and RR of fracture (P<.001), but the covariates age, race, and sex did not contribute significantly to either model. Thus, cause of SCD and motor impairment level appeared to be independent, additive risk factors for fracture. Logistic regression modeling with any fracture over the time at risk as the outcome showed results similar to both the Cox proportional hazards and the Poisson regression models—those with some motor impairment (odds ratio [OR]=1.93, P<.001) or moderate motor impairment (OR=1.65, P=.001) had almost 2-fold higher adjusted odds of having any fracture even in the presence, in the model, of cause of SCD. Cause of SCD, higher age (P<.001), and longer duration of SCD (P=.001) resulted in higher odds of any fracture in these logistic regression models, but race and sex did not make significant contributions to the models. Discussion This study took advantage of a unique opportunity to study fracture rate by degree of motor impairment across 2 diverse neurologic conditions. This study found a nonlinear relationship between degree of motor impairment and fractures. In addition, the study found differences in fracture rates between 2 neurologic conditions that both cause paralysis via damage to the spinal cord, controlling for degree of motor impairment. The classification system for motor impairment used in this study allows direct comparison between MS and trauma-related SCI, but it is not equivalent to standard classification systems for trauma-related SCI or MS. Therefore, comparison with other literature must be undertaken cautiously. Most previous studies of osteoporosis after trauma-related SCI showed a linear relationship between level of injury and both the degree and extent of bone loss—persons with lower, less complete cord injury seemed to develop osteoporosis to lesser extent and lower level of severity whereas persons with higher, more complete cord injury seemed to develop osteoporosis to a greater extent and severity.11, 18 Yet Ragnarsson4 and Wang25 and colleagues found that fractures were more common in persons with paraplegia than in persons with tetraplegia. Whitson et al32 in a study of fractures in persons after a stroke found that those with intermediate functional abilities (middle tertile score on the FIM instrument) suffered more fractures than those in the highest or lowest tertile of FIM scores. This study also found higher fracture rates among those with intermediate motor impairment, both for persons with trauma-related SCI and for those with MS. As a whole, the results of these studies suggest that increased fracture risk in persons with neurologic disease is associated with more than just the extent of bone loss—they suggest that risk of a fracture is related to having enough neurologic impairment to lose bone mass and enough motor function to engage in activities that might result in a fracture. As strikingly shown in figure 2, one of the most interesting findings in this study is the 2-fold higher skeletal fracture rate in those with trauma-related SCI compared with those with MS, irrespective of degree of motor impairment. This study cannot determine the reasons why there may be a difference in fractures for persons with MS versus trauma. However, review of the literature suggests several possible explanations. Recent work33, 34 indicates there may be a fundamental pathophysiologic difference between the osteoporosis that occurs after trauma-related SCI and osteoporosis that occurs under other circumstances, due to neurohormonal changes resulting from the complete loss of spinal innervation. The physiology underlying spinal cord damage with MS is very different from trauma-related SCI.35 Therefore, neurohormonal effects on bone metabolism may not be present to the same degree in persons with MS, particularly effects related to interruption of sympathetic pathways affecting bone intravenous shunts as have been described for trauma-related SCI.36 Perhaps bone loss in MS occurs both to a lesser degree and extent than after trauma-related SCI—even among persons with similar degrees of motor impairment. Patterns of bone loss have never been compared directly for these 2 causes of SCD. Another possible explanation may be differences in the patterns of functional impairment in the 2 patient populations, not fully controlled for by the measure of motor impairment used in this study, which in turn may relate to the differences in fracture location in the 2 populations. The limited prior data on site of fractures in SCD showed fractures most commonly occurring in the pelvis and lower extremities.17, 18 As expected, persons with trauma-related SCI indeed experienced the largest number of fractures in the distal femur and proximal tibia or fibula. For persons with MS, however, this study showed the most common fracture locations to be the upper extremity, rib, and ankle, foot, and toe. The disparity in location of fracture across the 2 groups may be due to differences in functional impairment found in trauma-related SCI versus MS. Hoenig et al27 found disease-specific differences in persons with trauma-related SCI compared with MS in the patterns of motor, visual, and cognitive impairment, and along with additional disease-specific differences in functional abilities. The dissimilar “fingerprints of disability” across the 2 conditions could explain some of the differences in fracture locations between the 2 groups. For example, the greater level of vision loss seen in the MS population does not preclude ambulation per se and therefore could lead to a greater tendency to fall from a standing position, whereas the more complete loss of lower-extremity function reported by persons with trauma-related SCI would tend to limit persons to wheelchair-based rather than standing mobility.27 Finally, there may be other unmeasured factors that differ according to the 2 conditions, for example, vitamin D exposure may be greater in the MS population due to being ambulatory for a longer period of time. Studies in able-bodied persons have outlined many contributing risk factors for fracture including age, sex, and race. In this study, those factors were not significant in the models (with the exception of age in the logistic regression model). Although adjusted models are presented, this study was not designed to evaluate the contributions of factors such as sex on fracture risk. Rather the goal was to examine fracture rates in a large cohort with either trauma or MS, and to test the contribution of causal disease and motor impairment to fracture risk, controlling for other relevant factors. For example, the preponderance of men in this veteran sample reduces the ability to examine the unique contribution of sex. Thus, negative results should not be interpreted to mean that age, sex, and race are not important contributors to fractures in persons with SCD. Study Limitations There are several other notable limitations to this study. First, it was done in a VHA population and may not be generalizable to nonveteran populations, particularly the MS population, which typically has a greater preponderance of women.37 Even though the MS group did have more women, there were still only 9% women. Second, this study used administrative data to identify fractures—relying on ICD-9 coding of visits by a broad variety of health professionals. The effect of coding variability was minimized both by grouping fractures based on the first 3 digits of the ICD-9 code and by collapsing adjacent visits for similar ICD-9-CM coded fractures into a single fracture event, the net effect of which could result in fracture undercounting. Another source of undercounting might be fracture care outside the VHA, even though these veterans reported receiving the majority of their health care in the VHA.28 Also, fractures of the axial skeleton were excluded. Although this eliminates possible duplicate coding for people whose original injury involved a spine fracture, it also eliminates the counting of vertebral compression fractures (which may be important in persons with MS treated with corticosteroids).38, 39 Thus, the estimated fracture rates in this study are conservative and may underestimate actual fracture rates to some degree. Use of ICD-9-CM coding limited the amount of detail that could be abstracted about the type of fracture. The study design is a retrospective, longitudinal cohort which limits definitive causal interpretation. Each of the analytic methods used to examine fracture rates have their respective methodologic limitations. What is notable is that the results were quite consistent across all 3 analytic approaches (logistic regression, Cox proportional hazard, Poisson regression). Finally, self-reported measures of motor function were used rather than traditional, observer-based measures. These self-reported measures allowed for capture of information from a large cohort and for comparison of the impact of motor impairment on fracture rates across persons with MS versus trauma-related SCI. They have been validated in this population31; nonetheless, they are self-reported and they do not have the same detailed specificity as the ASIA or Kurtzke measures of neurologic impairment. Use of a more detailed, precise measure might better delineate the relationships between specific aspects of motor impairment and both fracture rates and locations. Conclusions Future studies need to determine the physiologic reasons for the differences in fracture risk for MS versus trauma-related SCI. This in turn may enable the development of effective interventions to both prevent and to reverse osteoporosis due to SCD. The highest risk group in this study (those with trauma and some motor impairment) had a fracture rate of over 3.5 per 100 patient-years—a higher rate than those found in elderly white women with severe osteoporosis.40 The high rate of fractures underscores the importance of examining the impact of fractures in SCD persons on outcomes such as health care costs, morbidity, and quality of life. It might be all too easy to dismiss the importance of a fracture in someone with SCD on the basis of the person already being paralyzed. Just as the paralysis itself increases the risk of fracture, it is likely these persons have increased vulnerability to the impact of a fracture further reducing limited functional abilities, with consequent complications related to prolonged bedrest, prolonged hospitalization, and need for long-term nursing home placement. Clearly, preventive strategies need to be developed, both to reduce the severity of the osteoporosis after SCD and to reduce the risk of fractures among those with osteoporosis. Supplier Acknowledgment We thank Jennifer Hoff Lindquist with Health Services Research & Development, Durham VA Medical Center, for her assistance with identifying and linking VA administrative data files. References 1. 1McKinley WO, Jackson AB, Cardenas DD, DeVivo MJ. Long-term medical complications after traumatic spinal cord injury: a regional model systems analysis. Arch Phys Med Rehabil. 1999;80:1402–1410. Abstract |
Full-Text PDF (1195 KB)
|
CrossRef
2. 2Comarr AE, Hutchinson RH, Bors E. Extremity fractures of patients with spinal cord injuries. Am J Surg. 1962;103:732–739. MEDLINE |
CrossRef
3. 3Garland DE, Rhoades ME. Orthopedic management of brain-injured adults (Part II). Clin Orthop Relat Res. 1978;(131):111–122. 4. 4Ragnarsson KT, Sell GH. Lower extremity fractures after spinal cord injury: a retrospective study. Arch Phys Med Rehabil. 1981;62:418–423. MEDLINE 5. 5Freehafer AA, Hazel CM, Becker CL. Lower extremity fractures in patients with spinal cord injury. Paraplegia. 1981;19:367–372. MEDLINE 6. 6Ingram RR, Suman RK, Freeman PA. Lower limb fractures in the chronic spinal cord injured patient. Paraplegia. 1989;27:133–139. MEDLINE 7. 7Keating JF, Kerr M, Delargy M. Minimal trauma causing fractures in patients with spinal cord injury. Disabil Rehabil. 1992;14:108–109. MEDLINE |
CrossRef
8. 8Albright F, Burnett CH, Cope O, Parson W. Acute atrophy of bone (osteoporosis) simulating hyperparathyroidism. J Clin Endocrinol. 1941;1:711–716. 9. 9BeDell KK, Scremin AM, 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. MEDLINE |
CrossRef
10. 10Uebelhart D, Demiaux-Domenech B, Roth M, Chantraine A. Bone metabolism in spinal cord injured individuals and in others who have prolonged immobilisation (A review). Paraplegia. 1995;33:669–673. MEDLINE 11. 11Szollar SM, Martin EM, Sartoris DJ, Parthemore JG, Deftos LJ. Bone mineral density and indexes of bone metabolism in spinal cord injury. Am J Phys Med Rehabil. 1998;77:28–35. MEDLINE |
CrossRef
12. 12Dauty M, Perrouin VB, Maugars Y, Dubois C, Mathe JF. Supralesional and sublesional bone mineral density in spinal cord-injured patients. Bone. 2000;27:305–309. Abstract | Full Text |
Full-Text PDF (485 KB)
|
CrossRef
13. 13Zehnder Y, Luthi M, Michel D, et al. Long-term changes in bone metabolism, bone mineral density, quantitative ultrasound parameters, and fracture incidence after spinal cord injury: a cross-sectional observational study in 100 paraplegic men. Osteoporos Int. 2004;15:180–189.
CrossRef
14. 14Frisbie JH. Fractures after myelopathy: the risk quantified. J Spinal Cord Med. 1997;20:66–69. MEDLINE 15. 15Nottage WM. A review of long-bone fractures in patients with spinal cord injuries. Clin Orthop Relat Res. 1981;(155):65–70. 16. 16Lazo MG, Shirazi P, Sam M, Giobbie-Hurder A, Blacconiere MJ, Muppidi M. Osteoporosis and risk of fracture in men with spinal cord injury. Spinal Cord. 2001;39:208–214. MEDLINE |
CrossRef
17. 17Vestergaard P, Krogh K, Rejnmark L, Mosekilde L. Fracture rates and risk factors for fractures in patients with spinal cord injury. Spinal Cord. 1998;36:790–796. MEDLINE 18. 18Garland DE, Stewart CA, Adkins RH, et al. Osteoporosis after spinal cord injury. J Orthop Res. 1992;10:371–378. MEDLINE |
CrossRef
19. 19Cocksedge S, Freestone S, Martin JF. Unrecognised femoral fractures in patients with paraplegia due to multiple sclerosis. Br Med J (Clin Res Ed). 1984;289(6440):309. MEDLINE 20. 20Williams P, Frank A, Crawford CM, Alcorn T. Unrecognised femoral fractures in patients with paraplegia due to multiple sclerosis. Br Med J (Clin Res Ed). 1984;289(6443):501. MEDLINE 21. 21Kiessling WR, Maske B. [The problem of femoral neck fracture in multiple sclerosis]. Fortschr Neurol Psychiatr. 1989;57(3):107–109. MEDLINE |
CrossRef
22. 22Stenager E, Jensen K. Fractures in multiple sclerosis. Acta Neurol Belg. 1991;91:296–302. MEDLINE 23. 23Cosman F, Nieves J, Komar L, et al. Fracture history and bone loss in patients with MS. Neurology. 1998;51:1161–1165. MEDLINE 24. 24Garland DE, Adkins RH, Stewart CA. The natural history of bone loss in the lower extremity of complete spinal cord-injured males. Top Spinal Cord Inj Rehabil. 2005;11:48–60. 25. 25Wang CM, Chen Y, DeVivo MJ, Huang CT. Epidemiology of extraspinal fractures 22 associated with acute spinal cord injury. Spinal Cord. 2001;39:589–594. MEDLINE |
CrossRef
26. 26Samsa G, Hoenig H, Carswell J, et al. Developing a national registry of veterans with spinal cord dysfunction: experiences and implications. Spinal Cord. 1998;36:57–62. MEDLINE 27. 27Hoenig H, McIntyre L, Hoff J, Samsa G, Branch LG. Disability fingerprints: patterns of 27 in spinal cord injury and multiple sclerosis differ. J Gerontol A Biol Sci Med Sci. 1999;54:M613–M620. MEDLINE 28. 28Hoenig H, Hoff J, McIntyre L, Branch LG. The self-reported functional measure: predictive validity for health care utilization in multiple sclerosis and spinal cord injury. Arch Phys Med Rehabil. 2001;82:613–618. Abstract | Full Text |
Full-Text PDF (81 KB)
|
CrossRef
29. 29ICD-9-CM expert for hospitals: International Classification of Diseases 9th Revision Clinical Modification. 6th ed.. Salt Lake City: Ingenix; 2005;. 30. 30CPT 2005: Current procedural terminology. Chicago: AMA Pr; 2004;. 31. 31Hoenig H, Branch LG, McIntyre L, Hoff J, Horner RD. The validity in persons with spinal cord injury of a self-reported functional measure derived from the functional independence measure. Spine. 1999;24:539–543. MEDLINE |
CrossRef
32. 32Whitson HE, Pieper CF, Sanders L, Horner RD, Duncan PW, Lyles KW. Adding injury to insult: fracture risk after stroke in veterans. J Am Geriatr Soc. 2006;54:1082–1088. MEDLINE |
CrossRef
33. 33Maimoun L, Fattal C, Micallef JP, Peruchon E, Rabischong P. Bone loss in spinal cord-injured patients: from physiopathology to therapy. Spinal Cord. 2005;44:203–210. MEDLINE |
CrossRef
34. 34Maimoun L, Couret I, Mariano-Goulart D, et al. Changes in osteoprotegerin/RANKL system, bone mineral density, and bone biochemicals markers in patients with recent spinal cord injury. Calcif Tissue Int. 2005;76:404–411.
CrossRef
35. 35Kurtzke JF. Patterns of neurologic involvement in multiple sclerosis. Neurology. 1989;39:1235–1238. MEDLINE 36. 36Jiang SD, Dai LY, Jiang LS. Osteoporosis after spinal cord injury. Osteoporos Int. 2006;17:180–192.
CrossRef
37. 37Roth MP, Clayton J, Patois E, Alperovitch A. Gender distributions in parents and children concordant for multiple sclerosis. Neuroepidemiology. 1994;13:211–215. MEDLINE |
CrossRef
38. 38Kanis JA, Johansson H, Oden A, et al. A meta-analysis of prior corticosteroid use and fracture risk. J Bone Miner Res. 2004;19:893–899. MEDLINE |
CrossRef
39. 39Troiano RA, Jotkowitz A, Cook SD, Bansil S, Zito G. Rate and types of fractures in corticosteroid-treated multiple sclerosis patients. Neurology. 1992;42:1389–1391. MEDLINE 40. 40Barrett-Connor E, Siris ES, Wehren LE, et al. Osteoporosis and fracture risk in women of different ethnic groups. J Bone Miner Res. 2005;20:185–194. MEDLINE |
CrossRef
a Division of Geriatrics, Department of Medicine, Duke University Medical Center, Durham, NC b Geriatric Research Education and Clinical Center, Durham VA Medical Center, Durham, NC c Division of Geriatrics, Greenville Hospital System/University Medical Center, Greenville, SC d Center for the Study of Aging and Human Development, Duke University, Durham, NC e Endocrinology and Geriatrics, Duke University Medical Center, Durham, NC f Spinal Cord Injury and Disorders Strategic Healthcare Group, Department of Veterans Affairs, Puget Sound Health Care System, Seattle, WA g Duke University Medical Center, Durham, NC h Physical Medicine and Rehabilitation, Durham VA Medical Center, Durham, NC.  Reprint requests to William C. Logan, Jr, MD, Division of Geriatrics, Greenville Hospital System/University Medical Center, Center for Success in Aging, 255 Enterprise Blvd., Suite 101, Greenville, SC 29615
Supported by the Center For the Study of Aging and Human Development. A commercial party having a direct financial interest in the results of the research supporting this article has conferred or will confer a financial benefit upon the author or one or more of the authors. Lyles has received financial support from Novartis, the Alliance for Better Bone Health, and Amgen; he is a consultant to Novartis, Procter & Gamble, Merck, Amgen, GTx, and Bone Medical Ltd; he holds U.S. patent 20050272707 (methods for preventing or reducing secondary fractures after hip fracture); and has a provisional patent application (medications kits and formulations for preventing, treating, or reducing secondary fractures after previous fracture). PII: S0003-9993(07)01654-1 doi:10.1016/j.apmr.2007.08.144 © 2008 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT
01654-1/journalimage?img=PIIS0003999307016541.gr1.lrg.gif&fig=fig1&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)
01654-1/journalimage?img=PIIS0003999307016541.gr2.lrg.gif&fig=fig2&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)
01654-1/journalimage?img=PIIS0003999307016541.gr3.lrg.gif&fig=fig3&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)