SURGICAL SPINE INTERVENTION after traumatic spinal

Transcription

1 1818 Outcomes of Early Surgical Management Versus Late or No Surgical Intervention After Acute Spinal Cord Injury William McKinley, MD, Michelle A. Meade, PhD, Steven Kirshblum, MD, Barbara Barnard, DO From the Department of Physical Medicine and ilitation, Virginia Commonwealth University, Richmond, VA (McKinley, Meade, Barnard); and Department of Physical Medicine and ilitation, University of Medicine and Dentistry of New Jersey, Newark, NJ (Kirshblum). Supported by the National Institute on Disability and ilitation Research, Office of Special Education and ilitation Services, US Department of Education (grant no. H133N ). No commercial party having a direct interest in the results of the research supporting this article has or will confer a benefit on the author(s) or on any organization with which the author(s) is/are associated. Reprint requests to William McKinley, MD, PO Box , Richmond, VA 23298, wmckinle@hsc.vcu.edu /04/ $30.00/0 doi: /j.apmr ABSTRACT. McKinley W, Meade MA, Kirshblum S, Barnard B. Outcomes of early surgical management versus late or no surgical intervention after acute spinal cord injury. Arch Phys Med il 2004;85: Objective: To compare neurologic, medical, and functional outcomes of patients with acute spinal cord injury (SCI) undergoing early ( 24h and 24 72h) and late ( 72h) surgical spine intervention versus those treated nonsurgically. Design: Retrospective case series comparing outcomes by surgical and nonsurgical groups during acute care, rehabilitation, and at 1-year follow-up. Setting: Multicenter National Spinal Cord Injury Database. Participants: Consecutive patients with acute, nonpenetrating, traumatic SCI from 1995 to 2000, admitted in the first 24 hours after injury. Surgical spinal intervention was likely secondary to nature of injury and the need for spinal stabilization. Interventions: Not applicable. Main Outcome Measures: Changes in neurologic outcomes (motor and sensory levels, motor index score, American Spinal Injury Association [ASIA] Impairment Scale [AIS]), medical complications (pneumonia and atelectasis, deep vein thrombosis and pulmonary embolism, pressure ulcers, autonomic dysreflexia, rehospitalization), and functional outcomes (acute and rehabilitation length of stay [LOS], hospital charges, FIM instrument score, FIM motor efficiency scores). Results: Subjects in the early surgery group were more likely (P.05) to be women, have paraplegia, and have SCI caused by motor vehicle collisions. The nonsurgical group was more likely (P.05) to have an older mean age and more incomplete injuries. ASIA motor index improvements (from admission to 1-y follow-up) were more likely (P.05) in the nonsurgical groups, as compared with the surgical groups. Those with late surgery had significantly (P.05) increased acute care and total LOS and hospital charges along with higher incidence of pneumonia and atelectasis. No differences between groups were found for changes in neurologic levels, AIS grade, or FIM motor efficiency. Conclusions: ASIA motor index improvements were noted in the nonsurgery group, though likely related to increased incompleteness of injuries within this group. Early versus late spinal surgery was associated with shorter LOS and reduced pulmonary complications, however, no differences in neurologic or functional improvements were noted between early or late surgical groups. Key Words: Neurologic disorders; Recovery of function; ilitation; Spinal cord injuries; Surgery by the American Congress of ilitation Medicine and the American Academy of Physical Medicine and ilitation SURGICAL SPINE INTERVENTION after traumatic spinal cord injury (SCI) typically involves vertebral stabilization and/or decompression of neural tissue. 1-5 Although the past few decades have seen much advancement in spinal surgical techniques and an increase in spinal surgeries after SCI, controversy still exists regarding the potential neurologic benefits and timing of the surgery Traumatic SCI affects over 200,000 people in the United States, with nearly 10,000 new injuries annually People are typically of young age (average age, 32y), with life expectancies only slightly reduced when compared with the non-sci population; thus, the medical and functional care decisions can lead to tremendous annual and lifetime costs Long-term functional outcomes after SCI result from a combination of acute neurologic recovery in addition to medical intervention, rehabilitation, and community reintegration. Early interventions have been a focus of treatment for enhancing neurologic potential after acute SCI. In the early 1990s, the National Acute Spinal Cord Injury Study (NASCIS- II) concluded that the administration of methylprednisolone within the first 8 hours (30mg/kg bolus and 5.4mg/kg hourly for 23h) after traumatic nonpenetrating SCI may enhance the potential for neurologic improvement, 26 thus indicating the potential of early intervention on secondary SCI. The NASCIS- III study further revealed significant improvement in motor scores on those with methylprednisolone and tirilazad started 3 to 8 hours after injury. 27 Because these findings have not been consistently replicated, recommendations have not been universally adopted Criticism of the NASCIS studies has been expressed that their data were noncompelling, that use of the data remains experimental, and that clinicians should adhere to strict inclusion criteria when considering its use. Research using animal models has provided evidence that early decompressive surgery could lead to improved neurologic recovery after SCI, 11,12 further suggesting the importance of early intervention. The number of human spinal surgical studies using prospective, randomized, or controlled methodologies is limited. The current literature available has shown varying results, some indicating the potential neurologic benefit associated with early spinal decompression 13,14 and others not. 3,5,15-18 Although uncertainties remain as to the potential neurologic benefits or optimal timing of surgical intervention after SCI, increased medical complications have not been reported. 2,18 Schlegel et al 28 reported fewer medical complications in patients who underwent decompressive surgery less than 72 hours after injury compared with those who had surgical treatment after 72 hours. Nonetheless, spinal surgical procedures carry risks such as neurologic deterioration, bleeding, infection, and

2 SURGICAL MANAGEMENT AFTER SPINAL CORD INJURY, McKinley 1819 dysphagia. Additionally, several studies have noted the beneficial effects of surgical intervention for earlier patient mobilization, which contributes to earlier transition to rehabilitation, shorter hospital length of stay (LOS), and decreased hospital costs. 15,24,25,29 Decreased LOS is associated with decreased medical morbidity. The objective of this study was to compare patient outcomes in a multicenter database of acute SCI treated with early ( 24h and 24 72h) versus later ( 72h) surgical spine intervention and nonsurgical management. It is important to note that patients likely underwent surgical intervention secondary to spinal instability, whereas the nonsurgical group likely had spinal stability. This information will enhance the understanding of the outcomes of spinal surgical intervention on potential neurologic improvement, medical complications, acute and rehabilitation hospital course, and costs of care. METHODS Participants The sample consisted of 779 patients with nonpenetrating, traumatic SCI who completed inpatient rehabilitation programs at 1 of 18 Model Spinal Cord Injury Systems (MSCIS) between 1995 and People were selected based on injury type and completeness of data. Patients with penetrating injuries (ie, gunshot wounds) were excluded from analysis because of the potential for spinal cord transection and subsequent diminished potential for neurologic recovery. The centers were participants in the MSCIS program funded by the US Department of Education s National Institute on Disability and ilitation Research. Participants were predominantly men (78.8%) and white (76.7%). The percentages of African American, Asian, and other ethnic groups represented were 18.1%, 2.9%, and 2.3%, respectively. Patients mean age at the time of injury standard deviation (SD) was years (median, 35.0y). Motor vehicle collisions (MVCs) were the primary causal factor (52.9%), followed by falls (28.2%) and sports accidents (including diving) (9.1%). Injuries fell into the neurologic impairment classifications of incomplete tetraplegia (32.9%), complete paraplegia (27.2%), complete tetraplegia (22.1%), and incomplete paraplegia (17.8%) at initial MSCIS admission. Procedure Patient information was collected according to standard MS- CIS procedures at 4 time intervals: (1) admission to acute care, (2) admission to inpatient rehabilitation, (3) discharge from inpatient rehabilitation, and (4) 1 year postinjury (follow-up). For this analysis, patients were grouped according to whether and when they received spinal surgery. Because surgery data had been recorded by date (and not time of day), we equated these with hours after injury. Spinal surgery on day of admission would equate to less than 24 hours. Surgery on day 2 would equate to surgery between 1 hour and 48 hours. Surgery on day 3 would equate to surgery between 24 to 72 hours. Those who received spinal surgery on day 1, 2, or 3 were placed in the early surgery group. Those whose spinal surgery did not occur until after 72 hours were placed in the late surgery group. Those who did not receive spinal surgery were placed in the no surgery group. Surgeries included laminectomies, spinal decompressions, spinal fusions, and internal fixations; however, type of surgery was not used to further classify the groups. Measures The following information was collected during acute care, inpatient rehabilitation, and at year-1 follow-up. American Spinal Injury Association motor index total score. The American Spinal Injury Association (ASIA) motor index total score was calculated. 30 Scores ranging from 0 to 100 were assigned at the time of acute care admission, at Characteristics Table 1: Demographic and Injury Characteristics Overall (N 779) Nonsurgical Early Surgery Late Surgery 2 and P Mean age SD (y) F 12.79, P.001 Gender, % (n 779) Male 78.8 (614) 84.7 (149) 74.3 (228) 80.1 (237) , P.021 Female 21.2 (165) 15.3 (27) 25.7 (79) 19.9 (59) Ethnicity, % (n 751) White 76.7 (516) 70.8 (121) 82.7 (249) 73.8 (206) , P.082 African American 18.1 (136) 23.4 (40) 14.3 (43) 19 (53) Native American/Aleutian 0.8 (6) 0.6 (1) 0.7 (2) 1.1 (3) Asian 2.9 (22) 2.9 (5) 1.7 (5) 4.3 (12) Other 1.5 (11) 2.3 (4) 0.7 (2) 1.8 (5) Neurologic impairment , P.004 (admission), % (n 779) Paraplegia, incomplete 17.8 (139) 19.3 (34) 21.2 (65) 13.5 (40) Paraplegia, complete 27.2 (212) 19.3 (34) 30.3 (93) 28.7 (85) Tetraplegia, incomplete 32.9 (256) 41.5 (73) 26.7 (82) 34.1 (101) Tetraplegia, complete 22.1 (172) 19.9 (35) 21.8 (67) 23.6 (70) Etiology, % (n 779) MVC 52.9 (412) 50.6 (89) 57.0 (175) 50.0 (148) , P.001 Falls 28.2 (220) 29.5 (52) 24.4 (75) 31.4 (93) Sports 9.1 (71) 4.0 (7) 9.4 (29) 11.8 (35) Med/surgical complication 2.8 (22) 9.7 (17) 1.3 (4) 0.3 (1) Other violence 1.0 (8) 1.1 (2) 1.0 (3) 1.0 (3) Other 5.9 (46) 5.1 (9) 6.9 (21) 5.4 (16)

3 1820 SURGICAL MANAGEMENT AFTER SPINAL CORD INJURY, McKinley Table 2: FIM Scores, LOS, and Charges Variables Time Period Overall (N 779) No Surgery Early Surgery Late Surgery ANOVA, 2 and P FIM motor change ilitation (n 167) Discharge to follow-up (n 145) Admit to follow-up (n 147) FIM motor ilitation efficiency (n 167) LOS (d) Acute care ilitation Total Charges ($) Acute care 113, ,859 98,353 12,3621 (n 591) (n 130) ilitation 110, , , ,548 (n 637) (n 140) Total hospital 218, , , ,477 (n 605) (n 131) (n 289) (n 261) (n 262) (n 289) ,001 79,746 (n 243) 97,821 93,276 (n 254) 193, ,835 (n 242) (n 287) (n 248) (n 250) (n 287) , ,510 (n 218) 122, ,220 (n 243) 250, ,398 (n 232) F 0.48, P.616 F 0.01, P.988 F 0.21, P , P , P , P.307 F 3.80, P , P , P , P.001 NOTE. Values are mean SD or as otherwise indicated. admission, and at discharge from inpatient rehabilitation, and at 1-year follow-up. ASIA motor index change scores were calculated between initial MSCIS admission and rehabilitation discharge, rehabilitation discharge and 1-year follow- up, and initial admission and 1-year follow-up. ASIA motor index efficiency scores were calculated by dividing the ASIA motor index change scores by the number of days. ASIA Impairment Scale. The ASIA Impairment Scale (AIS) was assessed at MSCIS admission (for patients admitted on day of injury), discharge from inpatient rehabilitation, and first anniversary of injury. For the purpose of analysis, AIS injury level categories were recoded into numeric format so that change scores could be computed (eg, grade A [complete] into 0; grade B [sensory only] into 1; grade C [motor, not functional] into 2; grade D [motor, functional] into 3; grade E [normal] into 4). Therefore, a change score of 1 could represent the change from an AIS grade A to an AIS grade B or the change from an AIS grade D to normal functioning (grade E). Neurologic, motor, and sensory levels. Neurologic as well as motor and sensory levels of injury were assessed at the 4 defined time periods. The change scores for motor and sensory levels of injury (for each side) were calculated between initial admission and rehabilitation discharge and between initial admission and 1-year follow-up. FIM instrument motor scores. FIM instrument motor scores range from 13 to 91, with higher scores reflecting greater levels of independence. 31 FIM motor change scores were calculated based on the difference between rehabilitation admission and discharge and between initial admit and 1-year follow-up. FIM motor efficiency scores were calculated by dividing change scores by the respective duration of stay in rehabilitation. Length of stay. We determine length of stay (LOS), number of hospitalized days in acute care, inpatient rehabilitation, and total LOS (acute care plus rehabilitation LOS). If patients spent time in a subacute setting prior to attending inpatient rehabilitation, that time was not counted as either inpatient or acute (and likely no information was provided on that person). Charges. Estimated or actual hospital charges for each patient were assessed during acute care and rehabilitation. Charges were adjusted for inflation (to 2002 US dollars based on discharge year) to eliminate the possibility that differences in dollars may be reflected by date of admission across the 3 groups. Medical complications. Secondary medical complications common to SCI, 32 and those collected on the database including autonomic dysreflexia, pressure ulcers, deep vein thrombosis (DVT), pulmonary embolism (PE), and pneumonia were assessed during acute care, rehabilitation, total hospitalization (acute plus rehabilitation), and for the period between discharge and 1-year follow-up. Although date of death was examined, only 1 person (in the no-surgery group) in this data set died before the first anniversary of injury. Therefore, no analyses were conducted on this information. Rehospitalizations. The number of times that patients were admitted to a hospital in the first year after injury as well as the number of days they spent in the hospital was recorded. Comparison of Early Surgery Group Comparisons were also made between day 1 (surgery 24h), day 2 (surgery 48h), and day 3 (surgery 24 72h) surgical interventions within the early surgery group for changes in neurologic (motor and sensory) levels, ASIA motor index, and AIS grade. This comparison was designed to take a more in-depth look at how timing of surgery impacts outcomes. Data Analysis Statistics, including proportions, means, and standard deviation, were compiled for all demographic and outcome measures. One-way analysis of variance (ANOVA) procedures were used to examine the effects of surgery or con-

4 SURGICAL MANAGEMENT AFTER SPINAL CORD INJURY, McKinley 1821 Table 3: Changes in Neurologic Level and ASIA Motor Index Scores Variables Time Period Overall (N 779) No Surgery Early Surgery Late Surgery ANOVA, 2 and P ASIA motor index Acute care admit to rehab (n 654) (n 148) admit to DC (n 156) (n 95) Admit to follow-up (n 92) ASIA motor index Acute care efficiency (n 144) (n 156) Neurologic level Admit to rehab DC (n 166) (n 114) Admit to follow-up (n 110) Motor level Admit to rehab DC (n 167) (n 106) Admit to follow-up (n 103) Sensory level Admit to rehab DC (n 166) (n 105) Admit to follow-up (n 101) AIS grade Admit to rehab DC (n 170) (n 114) Admit to follow-up (n 111) (n 256) (n 269) (n 164) (n 154) (n 254) (n 269) (n 289) (n 173) (n 290) (n 168) (n 166) (n 288) (n 168) (n 163) (n 297) (n 179) (n 181) (n 250) (n 268) (n 153) (n 140) (n 236) (n 268) (n 285) (n 162) (n 161) (n 286) (n 145) (n 145) (n 282) (n 146) (n 145) (n 286) (n 163) (n 162) F 115, P.860 F 0.87, P , P.215 F 4.05, P , P , P.304 F 1.70, P , P , P.056 F 0.78, P , P , P.098 F 1.26, P , P , P , P.366 F 0.36, P , P.764 NOTE. Values are mean SD or as otherwise indicated. Abbreviations: DC, discharge;, rehabilitation. ventional treatment on demographic, medical, and functional outcomes. Those variables that violated assumptions of homoscedastisity were analyzed using nonparametric statistics (Kruskal-Wallis 2 ). Chi-square nonparametric statistics were also used to examine group differences in nominal and ordinal level outcomes. Because of the number of variables under examination, a Bonferroni adjustment was used for post hoc between-group comparisons. Consistent with traditional statistical procedures, an level of P less than or equal to.05 was deemed acceptable. RESULTS Demographics and Injury Characteristics Significant differences in group composition were found for age, gender, neurologic impairment, and traumatic etiology (table 1). In this study, participants were not stratified into groups based on injury types; groups were natural outgrowths of the types of interventions that were performed, which was based on the injury and the clinical assessment of treating physicians. Subjects in the nonsurgical group were significantly older than subjects who received either early or late surgery. Women were significantly more likely to be represented in the early spine surgery group than in either of the other 2 groups. Further evaluation revealed no gender-related differences in mean age or neurologic impairment; however, significant (P.001) differences existed between genders with regard to etiology of injury (MVC seen in women 66% vs 50% in men). With regard to differences in neurologic impairment at system admission, the early surgery group had the highest percentage of subjects with incomplete paraplegia and the lowest percentage of subjects with incomplete tetraplegia. Finally, the early spine surgery group was composed of a higher proportion of people with injuries resulting from MVCs. No between-group differences were found for ethnicity. Functional Outcome, LOS, and Charges FIM instrument. There were no significant FIM motor changes between groups for any time period. No significant

5 1822 SURGICAL MANAGEMENT AFTER SPINAL CORD INJURY, McKinley Table 4: Comparison With Early Surgery Group on Changes in Neurologic Level and ASIA Motor Index Scores Variables Time Period Surgery on Day of Injury (n 73) Surgery on Day 1 After Injury (n 143) Surgery on Day 2 After Injury (n 91) 2 and P ASIA motor index Acute care (n 59) (n 63) (n 36) Admit to follow-up (n 33) ASIA motor index Acute care efficiency (n 59) (n 63) Neurologic level Admit to rehab DC (n 69) (n 36) Admit to follow-up (n 34) Sensory level Admit to rehab DC (n 67) (n 34) Admit to follow-up (n 32) Motor level Admit to rehab DC (n 66) (n 35) Admit to follow-up (n 33) AIS Admit to rehab DC (n 70) (n 36) Admit to follow-up (n 34) (n 121) (n 123) (n 76) (n 75) (n 121) (n 123) (n 134) (n 86) (n 85) (n 136) (n 83) (n 80) (n 82) (n 82) (n 82) (n 141) (n 88) (n 90) (n 76) (n 83) (n 52) (n 46) (n 74) (n 83) (n 86) (n 54) (n 54) (n 85) (n 51) (n 51) (n 88) (n 51) (n 51) (n 86) (n 55) (n 57) , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P.037 NOTE. Values are mean SD or as otherwise indicated. differences were found between groups for FIM motor efficiency (table 2). Length of stay. Significant (P.05) differences were found between groups for acute and total LOS and but not for LOS in inpatient rehabilitation. Patients receiving no spinal surgeries or early spine surgery had significantly shorter acute care and total LOS than those with late surgery. Patient charges. Significant differences (P.001) were found between groups for acute care hospital charges and for total system charges, with the late surgery group having the highest charges. Neurologic Levels and ASIA Changes ASIA motor index changes. Differences were found between groups when the period between admission and 1-year anniversary was examined (F 4.05, P.018); subjects in the nonsurgery group had significantly higher change scores than the other 2 groups (table 3). Significant differences between groups were found for acute care ASIA motor index efficiency, with subjects in the nonsurgery group having the highest efficiency scores and subjects in the no-surgery group having the highest mean scores. No between-group differences were found for ASIA motor index efficiency in rehabilitation. Neurologic motor and sensory changes. Level of neurologic preservation changed an average of levels from admission to discharge and levels from discharge to 1-year follow-up. The mean motor level change between admission to and discharge from rehabilitation was ; mean motor level change from discharge to 1-year follow-up was The mean sensory level change from admission to discharge from rehabilitation was ; mean sensory level change from discharge to 1-year follow-up was No significant differences between groups were found for changes in neurologic, motor, or sensory levels or AIS grade at any of the time periods. Comparison of Day 1, 2, and 3 Admissions To determine whether there were any differences with regard to neurologic outcomes for those in the early surgery group, this group was subdivided based on day of surgery (table 4). Additional analyses were then conducted to compare these 3 subgroups. Of the 307 subjects in the early surgery group, 73

6 SURGICAL MANAGEMENT AFTER SPINAL CORD INJURY, McKinley 1823 Table 5: Medical Complications Complications Time Period Overall No Surgery Early Surgery Late Surgery 2 and P Pneumonia and atelectasis Acute care 38.1% 32% 11.0% 10.9% (n 774) Total hospitalization 42.4% 35.4% 3.8% 4.3% (n 738) (n 162) DVT Acute 4.9% 3.4% (n 775) (n 174) 7.1% 8.0% Total hospitalization 11.3% 10.9% (n 777) 2.4% 1.2% (n 735) (n 161) PE Acute care 1.7% 0.6% (n 777) 1.2% 1.7% (n 775) Total hospitalization 2.7% 2.3% (n 777) 0.8% 0.6% (n 736) (n 161) Autonomic Acute care 1.5% 2.3% dysreflexia (n 775) (n 174) 8.5% 8.6% (n 774) Total hospitalization 9.7% 9.7% 10.5% 5.6% (n 736) (n 162) Pressure ulcers Acute 19.0% 19.9% (n 779) 20.5% 16% Total hospitalization 34.7% 32% (n 779) 16.6% 17.5% (n 473) (n 114) No. of days rehospitalized (n 166) No. of rehospitalizations (n 165) 34.6% 9.8% 39.9% 2.4% (n 297) 5.2% (n 305) 7.8% 12.4% 2.7% (n 297) 2.0% 0.7% 2.6% 0.3% (n 297) 1.0% 6.2% 7.2% 10.1% 18.2% (N 307) 21.9% 35.6% 13.9% (n 187) (n 293) (n 293) 45.4% 12.3% (n 293) 49.2% (n 195) 5.0% (n 279) 5.4% 5.8% 10.5% 2.9% (n 277) 2.0% 1.4% (n 294) 3.0% (n 196) 1.4% (n 178) 1.7% 10.9% (n 293) 12.2% 13.7% (n 278) 19.3% 21.7% (n 195) 34.9% 13.4% (n 172) (n 276) (n 280) , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P , P.129 NOTE. Values are mean SD or as otherwise indicated. (23.8%) received surgery on the day of injury, 143 (46.6%) received surgery on the first day after injury, and 91 (29.6%) received surgery on the second day after injury. There were no between-group differences in neurologic, motor, or sensory levels or ASIA motor index. Only changes in ASIA impairment from admission to follow-up differed between groups, with subjects having surgery on the day of injury showing the most change and those whose surgery was on the second day after injury showing the least. Medical Complications In examining differences between no surgery, early spine surgery, and late spine surgery groups, significant (P.05) differences were found for 2 types of medical complications: the occurrence of pneumonia and atelectasis in acute care and total hospitalization (with a higher percentage in the late surgery group); and the occurrence of autonomic dysreflexia in the late surgery group, the year after discharge (table 5). No significant between-group differences were noted for DVT and PE or pressure ulcers. No significant between-group differences were found for number of rehospitalizations or total rehospitalization days. DISCUSSION Spinal surgery after SCI is not uncommon and is usually performed secondary to vertebral instability or in the presence of neurologic decline. In this study, which reviewed subjects with nonpenetrating traumatic SCI, 77% (603/779 subjects)

7 1824 SURGICAL MANAGEMENT AFTER SPINAL CORD INJURY, McKinley underwent either early or late spinal surgical intervention. The controversy remains whether spinal surgery after SCI leads to additional benefits in the areas of neurologic and functional outcome. Fixation for stabilization is important if one has an unstable spine. However, it has not been demonstrated whether immediate decompression is better than nonsurgical intervention (and immobilization with an orthosis) to provide adequate stabilization and improve outcomes. Additionally, risks and benefits of potential perioperative medical complications must be taken into account. This large multicenter database allows for further study of these issues. Demographics and Injury Characteristics Although some demographic differences were noted in our study, they are likely related to the etiology of injury and clinical presentation. Patients with MVC etiologies were more likely noted to undergo early surgical intervention. This may reflect the mechanism and increased force of injury associated with MVC-induced SCI, thus increasing the likelihood of unstable vertebral fractures. Subjects with SCI secondary to falls, sports, or medical and surgical complications may be less likely to have vertebral fractures or vertebral instability. Our study noted that women were more likely to undergo early surgery. This may represent the fact that women are more likely to be in the MVC etiologic group. Incomplete injuries were more commonly noted in the no-surgery group, again a reflection of the higher incidence of injuries less often associated with spinal instability (ie, medical complication and falls). Neurologic and Functional Outcomes ASIA motor index improvements were noted in the nosurgery group; however, this group was more likely to have subjects with incomplete injuries. No significant differences in changes of neurologic status were noted between early and late surgical and nonsurgical groups, either for neurologic level of injury or ASIA motor changes. No significant differences were found for changes in FIM motor scores or FIM motor efficiency between these groups. Thus, this study did not reveal neurologic benefits associated with early or late vertebral surgical intervention as compared with those who did not undergo surgery. Additionally, further comparisons of day 1 ( 24h), day 2 (1 48h), and day 3 (24 72h) did not reveal significant differences. Acute care and total LOS were longer in the late surgery group. Earlier surgery may promote earlier patient mobilization, transition to rehabilitation, and, subsequently, earlier discharge to community. Patients within the late surgical group also had higher acute care and total hospitalization charges. This was likely related to increased LOS within this group combined with surgical and postoperative charges. Medical Complications This study noted an increased acute care incidence of pulmonary complications (pneumonia, atelectasis) within the late surgery groups and for autonomic dysreflexia within both surgical groups. Despite similar acute care LOS for the early surgical and nonsurgical groups, these findings may be related to the inherent risk of surgical intervention and immediate postoperative immobility, thereby increasing the potential for pulmonary atelectasis and pneumonia. The longer acute care and total hospitalization LOS were likely contributing factors. Our data did not address the timing of medical complications and, as such, cannot differentiate between complications that may have occurred prior to surgical interventions (and thus delayed surgery) or as a sequela of surgical intervention. Additionally, information pertaining to the use of methylprednisolone or DVT prophylaxis, within groups, was not available. Future prospective studies are recommended to address these issues. This was not a randomized study. Therefore, observed differences in outcomes may be due to differences in characteristics such as neurologic impairment of the patients in each group, where significant differences were found (see table 1). Adjustment for these differences in the analysis (eg, by multiple linear or logistic regression analyses as appropriate) would eliminate this alternative explanation but would still not account for any possible differences across groups in unmeasured characteristics that may relate to the decision to operate and the timing of the surgery. CONCLUSIONS Early spinal surgical intervention ( 72h after injury) was associated with earlier transition from the acute care hospital to rehabilitation and decreased the overall hospital LOS, relative to surgery after 72 hours. Costs were higher in the late surgery groups. No significant differences in neurologic or functional changes were noted between surgical groups. Pulmonary complications such as pneumonia and atelectasis appear to have occurred more frequently in those with late surgical intervention. References 1. Donovan W. Operative and nonoperative management of spinal cord injury. A review. Paraplegia 1994;32: Marshall LF, Knowlton S, Garfin SR, et al. Deterioration following spinal cord injury. A multicenter study. J Neurosurg 1987;66: Waters R, Adkins R, Yakura J, Sie I. Effect of surgery on motor recovery following traumatic spinal cord injury. Spinal Cord 1996;34: Chapman JR, Grant GA, Newell DW. Emergent treatment of spinal cord injury. Spine State Art Rev 1999;13: Fehlings MG, Tator CH. An evidence-based review of decompressive surgery in acute spinal cord injury: rationale, indications, and timing based on experimental and clinical studies. J Neurosurg 1999;91: Amar A, Levy M. Surgical controversies in the management of spinal cord injury. J Am Coll Surg 1999;188: Short DJ, El Masry WS, Jones PW. High dose methylprednisolone in the management of acute spinal cord injury; a systematic review from a clinical perspective. Spinal Cord 2000;38: Pointillart V, Petitjean ME, Wiart L, et al. Pharmacological therapy of spinal cord injury during the acute phase. Spinal Cord 2000;38: Nesathurai S. Steroids and spinal cord injury: revisiting the NAS- CIS 2 and NASCIS 3 trials. J Trauma 1998;45: Hurlbert RJ. Methylprednisolone for acute spinal cord injury: an appropriate standard of care. J Neurosurg 2000;93: Delamarter RB, Sherman J, Carr JB. Pathophysiology of spinal cord injury: recovery after immediate and delayed decompression. J Bone Joint Surg Am 1995;77: Dimar JR II, Glassman SD, Raque GH. The influence of spinal canal narrowing and timing of decompression on neurologic recovery after spinal cord contusion in a rat model. Spine 1999;24: Gaebler C, Maier R, Kutscha-Lissberg F. Results of spinal cord decompression and thoracolumbar pedicle stabilization in relation to the time of operation. Spinal Cord 1999;37: Cotler JM, Herbison GJ, Nasuti JF. Closed reduction of traumatic cervical spine dislocation using traction weights up to 140 pounds. Spine 1993;18:

8 SURGICAL MANAGEMENT AFTER SPINAL CORD INJURY, McKinley Asazuma T, Satomi K, Suzuki N, Fujimura Y, Hirabayashi K. Management of patients with an incomplete cervical spinal cord injury. Spinal Cord 1996;34: Mirza SK, Krengel WF III, Chapman JR, et al. Early versus delayed surgery for acute cervical spinal cord injury. Clin Orthop 1999;Feb(359): Tator C, Fehlings M, Thorpe K, Taylor W. Current use and timing of spinal surgery for management of acute spinal surgery for management of acute spinal cord injury in North America: results of a retrospective multicenter study. J Neurosurg 1999;91(Suppl 1): Vaccaro A, Daugherty R, Sheehan T, et al. Neurologic outcome of early versus late surgery for cervical spinal cord injury. Spine 1997;22: Facts and figures: National Spinal Cord Injury Statistical Center. J Spinal Cord Med 2002;25: Fine PR, DeVivo MJ, McEachran AB. Re: incidence of acute traumatic hospitalized spinal cord injury in the United States, Am J Epidemiol 1982;115: Kraus JF, Franti CE, Riggins RS, Richards D, Borhani NO. Incidence of traumatic spinal cord lesions. J Chronic Dis 1975;28: Go BK, DeVivo MJ, Richards JS. The epidemiology of spinal cord injury. In: Stover SL, DeLisa JA, Whiteneck GG, editors. Spinal cord injury: clinical outcomes from the Model Systems. Gaithersburg: Aspen; p DeVivo MJ, Whiteneck GG, Charles ED Jr. The economic impact of spinal cord injury. In: Stover SL, DeLisa JA, Whiteneck GG, editors. Spinal cord injury: clinical outcomes from the Model Systems. Gaithersburg: Aspen; p DeVivo M, Rutt R, Black K, Stover S. Trends in spinal cord demographics and treatment outcomes between 1973 and Arch Phys Med il 1992;73: Berkowitz M, O Leary PK, Kruse DL, Harvey C. Spinal cord injury: an analysis of medical and social costs. New York: Demos; Bracken MB, Shepard MJ, Collins WF, et al. A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal cord injury. Result of the Second National Acute Spine Cord Injury Study [see comments]. N Engl J Med 1990; 322: Bracken MB, Shepard MJ, Holford TR, et al. Administration of methylprednisolone for 24 or 48 hours or tirilazed mesylate for 48 hours in the treatment of acute spinal cord injury. Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial. National Acute Spinal Cord Injury Study. JAMA 1997;277: Schlegel J, Bayley J, Yuan H. Timing of surgical decompression and fixation of acute spinal fractures. J Orthop Trauma 1996;10: Campagnolo D, Esquieres B, Kopacz K. Effect of timing of stabilization on length of stay and medical complications following spinal cord injury. J Spinal Cord Med 1997;20: American Spinal Injury Association, International Medical Society of Paraplegia. International standards for neurological and functional classification of spinal cord injury patients (revised). Chicago: ASIA; Hamilton B, Granger C, Sherwin F, Zielezny M, Tashman JA. Uniform National Data System for Medical ilitation. In: Fuhrer MJ, editor. ilitation outcomes: analysis and measurement. Baltimore: Brooks; p McKinley WO, Jackson AB, Cardenas DD, DeVivo MJ. Longterm medical complications after traumatic spinal cord injury: a regional model systems analysis. Arch Phys Med il 1999; 80: