Inflammatory myelopathies and traumatic spinal cord lesions: comparison of functional and neurological outcomes.

The outcome of nontraumatic and traumatic spinal cord lesions is still a matter of debate, probably because of the different features of the 2 populations. Patients with traumatic spinal cord lesions usually are younger and often have complete lesions and a shorter interval from onset of lesion to rehabilitation admission, whereas patients with nontraumatic spinal cord lesions usually are older and often have incomplete lesions and a longer interval from onset of lesion to rehabilitation admission. (1) Due to these differences, it is difficult to compare the 2 groups of patients. In a recent retrospective study involving 1,085 patients with nontraumatic spinal cord injury (SCI) and 250 patients with traumatic SCI, Catz et al (2) found a better prognosis for recovery in the first group. McKinley et al, (1) however, found better functional outcome in patients with traumatic SCI who were younger and more severely injured. Scivoletto et al, (3) in 2003, reported that neurological and functional outcomes were influenced more by age than by etiology. Ditunno (4) noted the low attention paid to nontraumatic spinal cord lesions in the literature. Although the incidence of nontraumatic SCI is considered to be between 25% and 80% of all cases of SCI (1,5-10) and although this percentage will probably increase in an aging society, most of the information available on recovery and outcomes after spinal cord lesions is based on research on traumatic spinal cord lesions.

Ditunno (4) cautioned researchers not to generalize about nontraumatic spinal cord lesions because this area is very diverse and includes many diseases with different prognoses. Studies on the outcomes of the various nontraumatic etiologies are even fewer and only cover some specific pathologies (eg, degenerative spine disease with cord involvement, vascular and neoplastic myelopathies). (11-13) No data are available on inflammatory diseases of the spinal cord.

The aim of the present work was to evaluate the functional and neurological outcomes of patients with inflammatory spinal cord lesions (ISCLs) and compare the outcomes with those of patients with traumatic SCI.

Subjects and Method

We examined the charts of 457 patients admitted to our spinal unit between 1996 and 2006 for a first rehabilitation treatment and selected patients with traumatic SCI or with ISCLs (bacterial, viral, post-infective, and post-vaccine myelitis). The latter group was as stable as the traumatic one because the few cases that showed disease progression were excluded from the study.

We recorded the following data: onset of lesion to admission time (LTA), injury variables (etiology and medical complications), length of stay as inpatients (LOS), and destination at discharge. The American Spinal Injury Association (ASIA) standards (14) were adopted to assess subjects' neurological status. We used the ASIA Impairment Scale to evaluate sensory and motor function and neurological level. However, because it has been shown that motor evaluation is the best predictor of impairment in patients with SCI, (15) we excluded data on sensory function and used only data on motor function in the data analysis. Lesions were considered incomplete whenever subjects showed signs of sacral sparing (conservation of motor or sensory function in the sacral segments). (16) Regarding neurological level, subjects were classified as having cervical, thoracic, or lumbar lesions. Neurological recovery was considered present if subjects improved by at least one ASIA Impairment Scale grade. Functional status at admission and at discharge was assessed with the Barthel Index (BI) (a 0-100 scale, with lower scores denoting less independence in activities of daily living), (17) the Rivermead Mobility Index (RMI) (a 0-15 scale designed to assess mobility), (18) and the Walking Index for Spinal Cord Injury (WISCI) (a 0-20 scale designed to evaluate walking based on the need for physical assistance, braces, and assistive devices). (19-21) Barthel Index items also were recorded to assess different areas of activities of daily living. Barthel Index and RMI scores were copied directly from the scales in the charts. As in the previous study by Scivoletto et al, (3) WISCI scores were taken retrospectively from medical chart records. (3) Only one researcher, who had experience with the scale (GS), assigned the WISCI scores based on the description of walking derived from the records.

ASIA motor score, BI, RMI, and WISCI score changes were calculated based on the difference between scores at admission and at discharge. ASIA motor function, BI, RMI, and WISCI efficiency scores also were calculated. Efficiency score refers to the difference in scoring at admission and at discharge related to the duration of the treatment (score changes for each scale divided by duration of rehabilitation stay). The efficiency score provides a basis for measuring the success of rehabilitation in terms of both individual patient performance and rehabilitation center performance= and has been used as an outcome measure in patients with SCI. (23,24)

Bladder voiding modalities and autonomy in bowel management were assessed according to the Gruppo Italiano Studio Epidemiologico Mielolesioni (GISEM) study. (10) Finally, we evaluated subjects' destination at discharge.

Matching Procedure

To correct for the concurrent effects of the different features of the 2 groups on neurological recovery and rehabilitation outcomes, we applied a matching procedure with 4 matching variables:

* Level of lesion. As in the study by Catz et al, (2) we divided the subjects into 3 groups according to level of lesion: cervical, thoracic, and lumbar. Although there might have been functional outcome differences within these subdivisions, the small number of subjects did not allow for a better definition of the lesion level effect.

* ASIA Impairment Scale classification. ASIA Impairment Scale classification is considered the major determinant of both functional and neurological outcomes. (25) We, therefore, grouped subjects according to each ASIA Impairment Scale level.

* Age and LTA. Based on previous experience, we chose a cutoff age of 50 years (3) and a cutoff interval of 40 days. (26,27)

* Subjects" sex. Scivoletto et al (28) recently demonstrated that the sex of people with SCI does not affect their outcomes. Although this issue is a matter of debate, (29,30) categorization by sex was not applied to increase the numbers of subjects in the matching cohorts.

Each subject was identified by an injury type, age, or LTA group and categorized by etiology. Subjects were selected from each etiology group to create matched dyads based on their classification. In this way, 76 subjects (38 dyads) were selected (Tab. 1).

Data Analysis

Descriptive values (mean [+ or -] SD) were calculated for all continuous clinical data. The data of the entire group of 248 subjects were analyzed using the Student t test for independent samples and the chi-square test.

Data from the dyads were analyzed by means of a paired t test, and the McNemar chi-square test was applied to assess contingency differences. [3] Data from the dyads are presented separately for each ASIA Impairment Scale classification level. Differences were taken as significant at P<.05.

Results

The entire sample comprised 181 subjects with traumatic SCI and 67 subjects with inflammatory myelopathies. Subjects in the SCI group differed significantly from those in the ISCL group in several respects. Subjects in the ISCL group were more often female (P=.003), were significantly older (P<.001), and had longer LTAs (P<.001) and LOSs (P<.001) compared with subjects in the SCI group (Tabs. 2 and 3, Figs. 1 and 2). They also had a higher frequency of incomplete lesions (P=.001) and a lower frequency of cervical lesions (P=.001) (Tab. 3). As regards ASIA motor scores, RMI, BI, and WISCI scores, the ISCL group showed significantly better functional status at admission compared with the SCI group (Tab. 2, Fig. 2). At discharge, the subjects in the ISCL group had better outcomes than the

subjects in the SCI group, but the differences were not significant (Tab. 2, Fig. 2).

[FIGURE 1 OMITTED]

In the matching cohorts, the SCI group had lower LTAs and higher LOSs compared with the ISCL group, but these differences were not significant. The ISCL group had a slightly higher frequency of neurological improvement compared with the SCI group (10/38 versus 9/38), but this difference did not reach statistical significance (P=.8) (Tabs. 4 and 5). With regard to the evaluation scales, at admission, the SCI group had significantly lower BI scores (P=.04) and nonsignificantly lower RMI and WISCI scores. Barthel Index subscale item scores were available only for 13 dyads. The SCI group had lower values for all BI subscale items, although none of these differences were statistically significant. The ASIA Impairment Scale motor scores at admission were perfectly comparable between groups (Tab. 4). At admission, the SCI group showed a higher (but not significant) frequency of complications (Tab. 5). At discharge, BI scores were comparable between groups, although the increase in BI scores was significantly higher in the SCI group. The RMI and WISCI scores at discharge were slightly, but not significantly, lower in the SCI group. The increase in RMI scores was comparable between the 2 groups, whereas the increase in WISCI scores was higher in the ISCL group. The ASIA Impairment Scale motor scores at discharge and the increases in motor scores were comparable between the 2 groups (Tab. 4). The efficiency scores for the 4 scales also were comparable between groups. Bladder-voiding modalities, bowel management independence, and discharge disposition also were comparable between groups (Tab. 5). When each ASIA Impairment Scale classification level was examined separately, subjects who were classified as ASIA A and B did not show any significant difference at both admission and discharge. Subjects in the SCI group who were classified as ASIA C showed significantly lower RMI scores at admission compared with subjects in the ISCL group (1.25 [+ or -] 1.39 versus 0.00, P=.04). Subjects classified as ASIA D did not show significant differences (Supplemental Tabs. 1, 2, 3, 4, 5, and 6; available online only at: www. ptjournal.org).

[FIGURE 2 OMITTED]

Discussion

To our knowledge, this is the first study that has compared the outcomes of inflammatory myelopathies and traumatic SCIs. The few articles that have dealt with nontraumatic SCIs usually have grouped the various etiologies together. As already stated, different etiologies may have different outcomes; thus, generalizations can lead to biases. The demographic characteristics of our subjects are in line with what has already been reported for people with traumatic and nontraumatic spinal cord pathology. (31,32)

In our study, both groups of subjects showed significant differences with regard to several prognostic factors: age, sex, LTA, and completeness and level of the lesion. (3,26-29) Incomplete lesions were more frequent in the ISCL group, which also had fewer cervical lesions. Subjects in the SCI group were younger and had shorter LTAs. Therefore, the 2 groups were difficult to compare, and the outcomes of the total subject sample could have been biased by these differences. Because of the low frequencies of complete lesions and cervical lesions in the ISCL group, these subjects had a better functional and neurological status at admission. At discharge, the scores of the 3 evaluation scales (BI, RMI, and WlSCI) were higher (although not significantly) for the ISCL group than for the SCI group. Thus, as in previous research, (3) we decided to use a matching cohorts procedure to correct for the concurrent effects of the above-mentioned prognostic factors.

With regard to neurological outcome, both groups showed comparable neurological amelioration: 26% of the subjects in the ISCL group and 24% of the subjects in the SCI group improved their ASIA Impairment Scale level between admission and discharge. Motor scores at admission and discharge and motor score increase and efficiency also were comparable. Until now, only one study (2) has examined the neurological outcome of patients with nontraumatic spinal cord lesions of different etiologies. In this investigation with a large sample, Catz et al found that the odds of any neurological recovery and of functional neurological recovery were significantly higher in patients with myelitis. In the past, the importance of adopting procedures to correct for the covariant effects of prognostic factors while examining different populations of patients (3,4) was emphasized. The study by Catz et al was probably a case in which such procedures were needed. The differences between the findings of Catz and colleagues' study and those of the present study can be explained by the methodological differences and, in particular, by the matching cohorts procedure that we used, which allowed us to have 2 groups of highly comparable subjects with regard to level and severity of lesion, age, and LTA.

With regard to functional status, the SCI group had lower BI scores (ie, lower independence in activities of daily living) at admission compared with the ISCL group. This difference was not related to a specific area of activities of daily living, as subjects showed a trend to lower scores for all BI items. At discharge, BI scores were comparable between groups. Thus, it could be argued that the subjects in the SCI group had greater functional improvement, as rated with the BI, than the subjects in the ISCL group. This may represent a possible contradiction, considering the lack of effects of SCI etiology on the other indexes analyzed in the present study. Nevertheless, it should be considered that time-limited nonneurological trauma-associated factors (eg, the presence of associated lesions, the need to wear an orthotic device, the sequelae of major surgery) might limit functional status at admission but not at discharge without affecting a person's neurological status. Similarly, these factors could slow down the timing of rehabilitation, as confirmed by the trend toward longer LOS in the SCI group.

Another possible explanation of the difference in BI scores at admission between the SCI and ISCL groups may be related to LTA differences. Because the subjects in the SCI group had longer LTAs compared with subjects in the ISCL group, they may have undergone rehabilitation treatment that, although a specific, produced an amelioration of their functional status. It could be suggested that differences in spinal shock severity may have contributed to the differences in "functional status" at admission. However, the effects of etiology on functional status due to differences in the spinal pathological processes (in particular, spinal shock) would be mirrored by differences in the neurological indexes. This was not the case in the present study. At admission, neurological status, as evaluated using impairment severity and motor scores, was highly comparable in the 2 groups. Furthermore, spinal shock is commonly said to resolve within 40 days, although recent research highlighted that some reflexes reappear a few hours or days after onset of the lesion. (33) In the present study, both in the entire sample and in the matched cohorts, the average LTA was about 50 days. Therefore, it is highly probable that most of the subjects were already out of the shock phase. Based on these considerations, we do not believe that the differences in functional status at admission could be attributed to a more severe spinal shock in patients with traumatic lesions. Thus, BI differences at admission can reasonably be ascribed to nonneurological trauma-associated factors.

Mobility (RMD and walking capacity (WISCI) at admission and at discharge were slightly lower in the SCI group than in the ISCL group, although the increases on the scales and their efficiency were comparable between groups. Both groups had the same outcome with regard to bladder and bowel management. As mentioned above, no data are available on the rehabilitation outcome of patients with inflanamatory spinal cord lesions. The only comparison we could make was with the studies of McKinley and colleagues. (l,34) They compared patients with traumatic SCI and patients with nontraumatic spinal cord lesions using a matching cohorts procedure and found better outcomes with regard to independence in activities of daily living for the patients with traumatic lesions. There are 2 possible explanations for the differences between the findings of our study and those of McKinley and colleagues. First, we selected inflammatory diseases that could show different outcomes than the other nontraumatic etiologies. Second, we used different statistical methods. In one study by McKinley et al, (1) the subjects were divided into only tetraplegic incomplete and paraplegic incomplete and complete groups, whereas we divided our subjects into 3 lesion levels and classified them according to each ASIA Impairment Scale classification level. In the second study, (34) the authors used a matching comparison similar to the matching procedure in our study, but they did not take into account the LTA.

Finally, with regard to discharge disposition, the 2 matched cohorts showed similar rates of returning home after discharge from rehabilitation (about 80%). As already underscored in other work, (35-37) discharge disposition and, most of all, the risk of being institutionalized are influenced by factors other than etiology of the lesion. In patients with SCI and in patients with other neurological diseases, discharge dispositions depend mostly on the level of independence in activities of dally living (with a BI score of 60 considered the point at which patients move from dependence to assisted independence and also considered the cutoff score for home discharge) (35,36) and on age. (37)

There are some limitations in the present study. One limitation is the possibility that the disease may worsen over time (eg, if myelitis represents the beginning of multiple sclerosis). Although patients who showed a worsening of their neurological status during the rehabilitation stay were excluded from the study, a subsequent deterioration cannot be excluded, and a follow-up study probably is needed. Another limitation is that, according to the GISEM study requirements, (10) in our study sample we grouped together all types of myelitis (bacterial, viral, post-infective, and post-vaccine). As indicated by Ditunno, (4) different etiologies could have an effect on outcomes.

To analyze the significance of the present study, it should be noted that the standard deviations in the 2 groups were very large in relation to the means. Thus, it is possible that in a larger sample some differences would emerge between the 2 groups. Accordingly, we performed a power analysis after the study (with BI score change as the primary outcome), which revealed that the power of the present matched cohort study was 57% and that, to reach a power of 80%, 58 pairs were required. Furthermore, the number of subjects for each ASIA Impairment Scale classification level was quite low, obliging us to group together ASIA A and B. As suggested by Ditunno, (4) multicenter studies that involve a larger number of patients are needed.

A third limitation is that, because of the number of subjects, it was impossible to analyze the effect of neurological factors such as pain and spasticity that are known to play a role in patients' function but are not included in the ASIA examination. Again, multicenter studies involving larger numbers of patients would help to clarify this issue.

Conclusions

People with traumatic SCI and those with inflammatory myelopathies are very different with regard to several clinical features that can confound the evaluation of outcomes. When the confounding effects of age and of lesion severity and level are eliminated, at admission patients with traumatic SCI have a greater physical dependence on others for assistance with activities of daily living compared with people with inflammatory spinal cord lesions. As this difference is not mirrored by a difference in neurological status, it could be argued that it is related to nonneurological factors (eg, the presence of associated lesions, major surgery, the need to wear an orthotic device). At discharge, the 2 patient populations achieve comparable neurological recovery and functional status.

Three major conclusions can be drawn from the present study. First, rehabilitation treatment seems to have the same positive effect on recovery in both patients with traumatic SCI and those with nontraumatic spinal cord lesions. Second, if studies with larger numbers of subjects confirm that the outcome of patients with SCI is determined more by factors such as lesion level and severity and age than by etiology, then both patient populations could be used for clinical studies aimed at evaluating the efficacy of rehabilitation. Third, there is a need for international multicenter studies in order to investigate a larger number of patients with nontraumatic SCI and, thus, to make possible an in-depth evaluation of the outcomes of this patient population.

This article was received February 10, 2007, and was accepted November 15, 2007.

DOI: 10.2522/ptj.20070049

References

(1) McKinley WO, Seel RT, Gadi RK, Tewksbury MA. Nontraumatic vs traumatic spinal cord injury. Am J Phys Med Rehabil. 2001;80:693-699.

(2) Catz A, Goldin D, Fishel B, et al. Recovery of neurologic function following nontraumatic spinal cord lesions in Israel. Spine. 2004;29:2278-2282.

(3) Scivoletto G, Morganti B, Ditunno P, et al. Effects of age on spinal cord lesion patients rehabilitation. Spinal Cord. 2003; 41:457-464.

(4) Ditunno JF. Point of view [editorial]. Spine. 2004;29:2283.

(5) Gnttman L. Spinal Cord Injuries: Comprehensive Management and Research. Oxford, United Kingdom: Blackwell; 1973.

(6) Murt-ay PK, Kusier MF. Epidemiology of nontraumatic and traumatic spinal cord injury. Arch Phys Med Rehabil. 1994;65:634.

(7) Buchan AC, Fulford GE, Jellineck E, et al. A preliminary survey of the incidence and etiology of spinal paralysis. Paraplegia. 1972;10:23-28.

(8) Celani MG, Spizzichino L, Ricci S, et al. Spinal cord injnry in Italy: a multicenter retrospective study. Arch Phys Med Rehabit. 2001;82:589-596.

(9) Citterio A, Franceschini M, Spizzichino L, et al. Nontraumatic spinal cord injury: an Italian survey. Arch Phys Med Rehabil. 2004;85:1483-1487.

(10) Pagliacci MC, Celani MG, Spizzichino L, et al; for the Gntppo Italiano Studio Epidemiologico Mielolesioni (GISEM) Group. Spinal cord lesion management in Italy: a 2-year survey. Spinal Cord. 2003;41:620-628.

(11) Salvador de la Barrera S, Barca-Buyo A, Montoto-Marques A, et al. Spinal cord infarction: prognosis and recovery in a series of 36 patients. Spinal Cord. 2001;39: 520-525.

(12) McKinley WO, Huang ME, Brunsvold KT. Neoplastic versus traumatic spinal cord injury: an outcome comparison after inpatient rehabilitation. Arch Phys Meal Rehabil. 1999;80:1253-1257.

(13) Ronen J, Goldin D, Itzokivich M, Bluvshtein V, et al. Outcomes in patients admitted for rehabilitation with spinal neurological lesions following intervertebral disc herniation. Spinal Cord. 2004;42:621-626.

(14) International Standards for Neurological Classifications of Spinal Cord Injury (Revised). Chicago, Ill: American Spinal Injury Association; 2000:1-23.

(15) Marina RJ, Rider-Forster D, Maissel G, Ditunno JH. Superiority of motor level over single neurological level in categorizing tetraplegia. Paraplegia. 1995;33:510-513.

(16) Waters RL, Adkins RH, Yakura JS. Definition of complete spinal cord injury. Paraplegia. 1991;29:573-581.

(17) Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index. Md State Med J. 1965;14:61-65.

(18) Collen FM, Wade DT, Robb GF, Bradshaw CM. The Rivermead Mobility Index: a further development of the Rivermead Motor Assessment. Int Disabil Stud. 1991;13: 50-54.

(19) Ditunno JF, Ditunno PL, Graziani V, et al. Walking index for spinal cord injury (WISCI): an international multicenter validity and reliability study. Spinal Cord. 2000;38:234-243.

(20) Ditunno PL, Ditunno J Jr [listed as "Dittuno" in MEDLINE]. Walking Index for Spinal Cord Injury (WISCI II): scale revision. Spinal Cord. 2001;39:654-656.

(21) Morganti B, Scivoletto G, Ditunno P, et al. Walking Index for Spinal Cord Injury (WISCI): criterion validation. Spinal Cord. 2005:43:27-33.

(22) Vanclay F. Functional outcome measures in stroke rehabilitation. Stroke. 1991;22: 105-108.

(23) Ottenbacher KJ, Smith PM, Illig SB, et al. Trends in length of stay, living setting, functional outcome, and mortality following medical rehabilitation. JAMA. 2004: 292:1687-1695.

(24) Ronen J, Itzkovich M, Bluvshtein V, et al. Length of stay in hospital following spinal cord lesions in Israel. Spinal Cord. 2004; 42:353-358.

(25) Coleman WP, Geisler FH. Injury severity as primary predictor of outcome in acute spinal cord injury: retrospective results from at large multicenter clinical trial. Spine J. 2004;4:373-378.

(26) Sumida M, Fujimoto M, Tokuhiro A, et al. Early rehabilitation effect for traumatic spinal cord injury. Arch Phys Med Rehabil. 2001;82:391-395.

(27) Scivoletto G, Morganti B, Molinari M. Early vcrsns delayed inpatient spinal cord injury (SCI) rehabilitation: an Italian study. Arch Phys Med Rehabil. 2005;86:512-516.

(28) Scivoletto G, Morganti B, Molinari M. Sexrelated differences of rehabilitation outcomes of spinal cord lesion patients. Clin Rehabil. 2004;18:709-713.

(29) Greenwald BD, Seel RT, Cifu DX, Shah AN. Gender-related differences in acute rehabilitation length of stay, charges and functional outcomes for a matched sample with spinal cord injury: a multicenter investigation. Arch Phys Med Rehabil. 2001; 82:1181-1187.

(30) Sipski ML, Jackson AB, Gomez-Marin O, et al. Effects of gender on neurologic and functional recovery after spinal cord injury. Arch Phys Med Rehabil. 2004:85: 1826-1836.

(31) Selhon LHS, Fehlings MG. Epidemiology, demographics and pathophysiology of acute spinal cord injury. Spine. 2001;24(suppl):S2-S-12.

(32) Van Asbeck FWA, Post MWM, Pangalila RF. An epidemiological description of spinal cord injuries in the Netherlands in 1994. Spinal Cord. 2000;38:420-424.

(33) Ditunno JF, Little JW, Tessler A, Burns AS. Spinal shock revisited: a four-phase model. Spinal Cord. 2004;42:383-395.

(34) McKinley WO, Seel RT, Hardman JT. Nontraumatic spinal cord injury: incidence, epidemiology, and functional outcome. Arch Phys Med Rehabil. 1999;80:619-623.

(35) Granger CV, Dewis LS, Peters NC, et al. Stroke rehabilitation: analysis of repeated Barthel Index measure. Arch Phys Med Rehabil. 1979;60:14-17.

(36) DeJong G. Branch LG, Corcoran PJ. Independent living outcomes in spinal cord injury: mnhivariate analyses. Arch Phys Med Rehabil. 1984;65:66-73.

(37) Anzai K, Young J, McCallum J, et al. Factors influencing discharge location following high lesion spinal cord injury rehabilitation in British Columbia, Canada. Spinal Cord. 2006;44:11-18.

G Scivoletto, MD, is Assistant Physician, Spinal Cord Unit, IRCCS S. Lucia Foundation, Rome, Italy. Dr Scivoletto's institutional address: Spinal Cord Unit, IRCCS Fondazione S. Lucia, Via Ardeatina 306, 00179 Rome, Italy. Address all correspondence to Dr Scivoletto at: [e-mail].

E Cosentino, MD, is Assistant Physician, Spinal Cord Unit, IRCCS S. Lucia Foundation, Rome, Italy.

A Mammone, PhD, is Consultant for Statistical Analysis, Department of Statistics, Probability and Applied Statistics, Sapienza University of Rome, Rome, Italy.

M Molinari, MD, PhD, is Chief Physician, Spinal Cord Unit, IRCCS S. Lucia Foundation, Rome, Italy.

[Scivoletto G, Cosentino E, Mammone A, Molinari M. Inflammatory myelopathies and traumatic spinal cord lesions: comparison of functional and neurological outcomes. Phys Ther. 2008;88:471484.]

Dr Scivoletto provided idea/research design. Dr Scivoletto, Dr Cosentino, and Dr Molinari provided writing. Dr Scivoletto and Dr Cosentino provided data collection. Dr Cosentino provided subjects. Dr Mammone provided data analysis. Dr Molinari provided project management and consultation.

The professional English-language editing of Claire Montagna is gratefully acknowledged.

Table 1. Composition of Dyads

Level of Lesion (a)   Age <50 y         Age >50 y       Total

                      LTA (b)   LTA     LTA     LTA
                      <40 d     >40 d   <40 d   >40 d

Cervical A            0         0       0        2       2
Cervical B            0         0       0        0       0
Cervical C            0         0       1        0       1
Cervical D            1         0       0        3       4
Thoracic A            0         2       2        7      11
Thoracic B            0         1       1        0       2
Thoracic C            1         2       2        1       6
Thoracic D            1         0       0        1       2
Lumbar A              0         0       0        0       0
Lumbar B              1         0       1        0       2
Lumbar C              0         1       0        1       2
Lumbar D              2         3       0        1       6
Total                 6         9       7       16      38

(a) Level of lesion based on American Spinal Injury
Association (ASIA) standards. (14)

(b) LTA-interval from onset of lesion to admission for
rehabilitation treatment.

Table 2.
Characteristics of Subjects With Traumatic Spinal Cord
Injury (SCI Group) and Subjects With Inflammatory Spinal
Cord Lesions (ISCL Group) According to Lesion Etiology:
Continuous Variables (a)

Variable             n     Mean     SD      t       df    P

Age (y)

  ISCL group          67    57.96   15.32
  SCI group          179    38.76   17.44    7.94   244   .000 *
  Total              246    43.99   18.91

LTA (d)

  ISCL group          55    75.04   48.87
  SCI group          157    50.03   39.82    3.77   210   .000 *
  Total              212    56.51   43.64

LOS (d)

  ISCL group          65    97.23   69.54
  SCI group          176   132.88   78.32   -3.23   239   .001 *
  Total              241   123.26   77.55

BI at admission

  ISCL group          62    33.69   22.03
  SCI group          178    19.19   18.99    4.96   238   .000 *
  Total              240    22.94   20.77

BI at discharge

  ISCL group          62    67.10   28.36
  SCI group          178    64.74   30.13    0.54   238   .590
  Total              240    65.35   29.64

BI score change

  ISCL group          62    33.40   21.79
  SCI group          178    45.49   26.30   -3.25   238   .001 *
  Total              240    42.37   25.72

BI efficiency

  ISCL group          59     0.42    0.37
  SCI group          173     0.46    0.49   -0.70   230   .484
  Total              232     0.45    0.46

RMI at admission

  ISCL group          57     2.12    2.77
  SCI group          173     0.95    2.41    3.06   228   .003 *
  Total              230     1.24    2.55

RMI at discharge

  ISCL group          57     6.44    4.73
  SCI group          172     5.58    4.56    1.22   227   .225
  Total              229     5.79    4.61

RMI score change

  ISCL group          57     4.32    3.66
  SCI group          172     4.62    3.98   -0.51   227   .608
  Total              229     4.55    3.90

RMI efficiency

  ISCL group          55     0.05    0.06
  SCI group          167     0.05    0.07    0.36   220   .720
  Total              222     0.05    0.06

WISCI at admission

  ISCL group          57     2.44    5.85
  SCI group          177     1.12    4.11    1.89   232   .060
  Total              234     1.44    4.62

WISCI at discharge

  ISCL group          57     8.79    8.73
  SCI group          177     6.77    8.34    1.57   232   .118
  Total              234     7.26    8.46

WISCI score change

  ISCL group          57     6.35    7.55
  SCI group          177     5.66    7.57    0.60   232   .547
  Total              234     5.82    7.56

WISCI efficiency

  ISCL group          54     0.07    0.10
  SCI group          173     0.06    0.11    0.51   225   .609
  Total              227     0.07    0.11

MS at admission

  ISCL group          34    60.47   17.29
  SCI group          127    50.61   18.54    2.79   159   .065
  Total              161    52.70   18.68

MS at discharge

  ISCL group          34    66.85   21.35
  SCI group          127    58.62   23.38    1.86   159   .065
  Total              161    60.36   23.15

MS score change

  ISCL group          34     6.38    6.99
  SCI group          127     8.01   10.16   -0.88   159   .381
  Total              161     7.66    9.58

MS efficiency

  ISCL group          33     0.07    0.10
  SCI group          122     0.09    0.16   -0.90   153   .369
  Total              155     0.09    0.15

(a) LTA=interval from onset of lesion to admission for
rehabilitation treatment, LOS=length of stay as inpatients,
BI=Barthel Index, RMI=Rivermead Mobility Index, WISCI=Walking
Index for Spinal Cord Injury, MS=American Spinal Injury
Association (ASIA) motor scores. Asterisk indicates
significant results.

Table 3.
Characteristics of Subjects With Traumatic Spinal Cord
Injury (SCI Group) and Subjects With Inflammatory Spinal
Cord Lesions (ISCL Group) According to Lesion Etiology:
Dichotomous Variables (a)

Variable           ISCL    SCI     Total   [chi      df   P
                   Group   Group           square]

Sex

  Female           27       37      64
  Male             40      144     184      9.06     1     .003 *
  Total            67      181     248

Lesion level

  Cervical          8       59      67
  Thoracic         47       70     117     21.7      2     .000 *
  Lumbar           10       51      61
  Total            65      180     245

Complications
at admission

  Absence          50      110     160
  Presence         17       71      88      3.52     1     .052
  Total            67      181     248

ASIA Impairment
Scale level at
admission

  A                16       89     105
  B                 4       14      18
  C                25       50      75     16.43     3    0.01 *
  D                22       28      50
  Total            67      181     248

ASIA Impairment
Scale level at
discharge

  A                13       83      96
  B                 2        5       7
  C                13       27      40     16.83     4     .002 *
  D                37       61      98
  E                 0        4       4
  Total            65      180     245

ASIA Impairment
Scale
improvement

  No               47      130     177
  Yes              18       50      68      0.000    1    1.000
  Total            65      180     245

Good bladder
control

  No               35      126     161
  Yes              25       53      78      2.45     1     .111
  Total            60      179     239

Bowel management
independence

  No               20       50      70
  Yes              44      129     173      0.117    1     .632
  Total            64      179     243

Destination at
discharge

  Others           12       39      51
  Home             53      137     190      0.199    1     .597
  Total            65      176     241

(a) ASIA=American Spinal Injury Association. Asterisk
indicates significant results.

Table 4.
Characteristics of Dyads of Subjects With Traumatic Spinal
Cord Injury (SCI Group) and Subjects With Inflammatory
Spinal Cord Lesions (ISCL Group) According to Lesion
Etiology: Continuous Variables (a)

Variable              n    Mean     SD      t       df   P

Age (y)

  ISCL group          38    51.44   16.33
  SCI group           38    50.76   17.94    0.40   37   .694
  Paired difference   38     0.68   10.61

LTA (d)

  ISCL group          27    71.63   55.67
  SCI group           27    63.15   46.07    0.87   26   .393
  Paired difference   27     8.48   50.74

LOS (d)

  ISCL group          35    99.74   71.06
  SCI group           35   115.29   60.89   -1.1    34   .279
  Paired difference   35   -15.54   83.64

BI at admission

  ISCL group          34    34.09   23.11
  SCI group           34    27.32   22.99    2.16   33   .038 *
  Paired difference   34     6.76   18.24

BI at discharge

  ISCL group          34    65.76   28.82
  SCI group           34    69.18   27.73   -0.89   33   .380
  Paired difference   34    -3.41   22.35

BI score change

  ISCL group          34    31.68   19.78
  SCI group           34    41.56   25.27   -2.17   33   .038 *
  Paired difference   34    -9.88   26.59

BI efficiency

  ISCL group          31     0.419   0.35
  SCI group           31     0.574   0.81   -1.00   30   .324
  Paired difference   31    -0.155   0.86

RMI at admission

  ISCL group          33     2.00    2.36
  SCI group           33     1.18    2.69    1.89   32   .067
  Paired difference   33     0.82    2.48

RMI at discharge

  ISCL group          33     6.36    4.96
  SCI group           33     5.82    4.07    0.76   32   .454
  Paired difference   33     0.55    4.13

RMI score change

  ISCL group          33     4.36    3.67
  SCI group           33     4.64    3.05   -0.48   32   .637
  Paired difference   33    -0.27    3.29

RMI efficiency

  ISCL group          30     0.059   0.05
  SCI group           30     0.069   0.09   -0.65   29   .522
  Paired difference   30    -0.01    0.09

WISCI at admission

  ISCL group          31     2.35    6.04
  SCI group           31     1.42    4.57    0.79   30   .433
  Paired difference   31    -0.94    6.55

WISCI at discharge

  ISCL group          31     8.71    9.05
  SCI group           31     7.23    8.40    1.20   30   .239
  Paired difference   31     1.48    6.87

WISCI score change

  ISCL group          31     6.35    8.11
  SCI group           31     5.81    7.35    0.38   30   .709
  Paired difference   31     0.55    8.09

WISCI efficiency

  ISCL group          28     0.087   0.12
  SCI group           28     0.095   0.19   -0.21   27   .838
  Paired difference   28    -0.008   0.18

MS at admission

  ISCL group          16    58.38   13.43
  SCI group           16    56.94   10.94    1.08   15   .297
  Paired difference   16     1.44    5.32

MS at discharge

  ISCL group          16    62.88   18.07
  SCI group           16    62.38   15.84    0.28   15   .783
  Paired difference   16     0.50    7.14

MS score change

  ISCL group          16     4.50    7.17
  SCI group           16     5.44    7.30   -0.43   15   .670
  Paired difference   16    -0.94    8.62

MS efficiency

  ISCL group          14     0.057   0.08
  SCI group           14     0.132   0.34   -0.95   13   .358
  Paired difference   14    -0.085   0.33

(a) LTA=interval from onset of lesion to admission for
rehabilitation treatment, LOS=length of stay as inpatients,
BI=Barthel Index, RMI=Rivermead Mobility Index,
WISCI=Walking Index for Spinal Cord Injury, MS=American
Spinal Injury Association (ASIA) motor scores. Asterisk
indicates significant results.

Table 5.
Characteristics of Dyads of Subjects With Traumatic Spinal
Cord Injury (SCI Group) and Subjects With Inflammatory
Spinal Cord Lesions (ISCL Group) According to Lesion
Etiology: Dichotomous Variables (a)

             Variable                SCI Group

             ASIA Impairment Scale   A          B        C
             level at admission

               A                     12          1
               B                                 4
ISCL Group     C                                          9
               D                                          1
               Total                 12          5       10

             ASIA Impairment Scale   A          B        C
             level at discharge

               A                      9          1        1
               B                      1          1
ISCL Group     C                      1                   2
               D                      1                   3
               Total                 12          2        6

             ASIA Impairment Scale   No         Yes      Total
             level improvement

               No                    23          5       28
ISCL Group     Yes                    6          4       10
               Total                 29          9       38

             Lesion level            Cervical   Lumbar   Thoracic

               Cervical               7
ISCL Group     Lumbar                           10
               Thoracic                                  21
               Total                  7         10       21

             Sex                     Female     Male     Total

               Female                 6          9       15
ISCL Group     Male                   4         19       23
               Total                 10         28       38

             Complications at        No         Yes      Total
             admission

               No                    16         10       15
ISCL Group     Yes                    4          8       12
               Total                 20         18       38

             Good bladder control    No         Yes      Total

               No                    14          6       20
ISCL Group     Yes                    6          8       14
               Total                 20         14       34

             Bowel management        No         Yes      Total
             independence

               No                     5          8       13
ISCL Group     Yes                    5         19       24
               Total                 10         27       37

             Destination at          Home       Others   Total
             discharge

               Home                  24          6       30
ISCL Group     Others                 6          2        8
               Total                 30          8       38

Variable                SCI Group       P

ASIA Impairment Scale   D       Total
level at admission

  A                             13
  B                              4
  C                              9      1.000
  D                     11      12
  Total                 11      38

ASIA Impairment Scale   D       Total
level at discharge

  A                      1      12
  B                              2
  C                      2       5      1.000
  D                     15      19
  Total                 18      38

ASIA Impairment Scale
level improvement

  No
  Yes                                   1.000
  Total

Lesion level            Total

  Cervical               7
  Lumbar                10              1.000
  Thoracic              21
  Total                 38

Sex

  Female
  Male                                   .267
  Total

Complications at
admission

  No
  Yes                                    .180
  Total

Good bladder control

  No
  Yes                                   1.000
  Total

Bowel management
independence

  No
  Yes                                    .581
  Total

Destination at
discharge

  Home
  Others                                1.000
  Total

(a) ASIA=American Spinal Injury Association.