Ambulatory function is an important outcome measure for

Transcription

1 Gait Speed in Relation to Categories of Functional Ambulation After Spinal Cord Injury Neurorehabilitation and Neural Repair Volume 23 Number 4 May The Author(s) / Hubertus J. A. van Hedel, PhD, PT, for the EMSCI Study Group Objective. The aim of the present study was to assess gait speeds that distinguished between levels of functional ambulation in subjects with a spinal cord injury. Methods. The data of 886 spinal cord injury subjects were derived from the European Multicenter Study for Human Spinal Cord Injury and analyzed at 1, 3, 6, and 12 months after injury. The indoor and outdoor mobility items from the Spinal Cord Independence Measure were combined into 5 clinically relevant categories: (1) wheelchair-dependent, (2) supervised walker with outdoor wheelchair dependency, (3) indoor walker with outdoor wheelchair dependency, (4) walker with aid, and (5) walker without aid. The preferred walking speed that distinguished between ambulation categories was derived from the 10-meter walking test and determined using receiver operating characteristic curves. Results. The walking speed correlated well (>0.84) with the ambulation categories. The average walking speed for each category was (1) 0.01 m/s, (2) 0.34 m/s, (3) 0.57 m/s, (4) 0.88 m/s, and (5) 1.46 m/s. The average (± SD) speed that distinguished between the categories was 0.09 ± 0.01 m/s (1 vs 2), 0.15 ± 0.08 m/s (2 vs 3), 0.44 ± 0.14 m/s (3 vs 4), and 0.70 ± 0.13 m/s (4 vs 5). The averaged sensitivity and specificity were above 0.98 and 0.94, respectively. Conclusion. In subjects with spinal cord injury, the preferred walking speed as assessed in the clinic can be used to estimate functional ambulation during daily life. The walking speed can distinguish between ambulation categories with high sensitivity and specificity. Keywords: 10-meter walking test; SCIM; Mobility; Daily life activities; Sensitivity and specificity Ambulatory function is an important outcome measure for present and future clinical trials in spinal cord injury (SCI). 1,2 To address changes in ambulatory function after SCI, several assessments have received increasing attention, including the 50-foot 3-5 and the 10-meter (10MWT) timed speed 6-8 and the 6-minute walking test (6MinWT). 3,6,7 Short-distance tests have several advantages over longdistance tests, although the results of short- and long-distance tests are comparable when applied between 1 and 6 months after SCI. 3,9 The test environment for short-distance tests of walking speed is easier to standardize and may be more reliable than long-distance testing as well as being considerably less time consuming. 9 One disadvantage of these tests might be that ceiling effects could potentially occur in SCI subjects with good ambulatory function. However, the most important limitation of these tests is whether the walking speed as assessed under strictly standardized conditions in the clinic can be used to estimate the level of functional ambulation in a continuously changing daily life environment. 10,11 To overcome this problem, additional assessments of measures reflecting activities of daily life (ADL) and independence, such as the Spinal Cord Independence Measure (SCIM) 12 are performed. However, as each additional test is a burden for both the therapist and the patient, it would be of value if a practical, valid, reliable, and responsive clinical walking test could provide an accurate estimation of the level of functional ambulation during daily life. Therefore, the aim of the present study was to determine the minimal walking speeds that characterize different levels of functional ambulation in subjects with SCI. We defined several patient ambulation categories using items from the revised SCIM. 12 For each category, we determined the minimal optimal (with respect to both sensitivity and specificity) walking speed required. We hypothesized that speed would differ between each successive ambulation category. Methods Data were derived from the European Multicenter Study of Human Spinal Cord Injury (EM-SCI; that prospectively assessed subjects after SCI. Patients from 18 European centers were assessed within 2 weeks and at 1, 3, 6, and 12 months after SCI. The assessments included the neurological status, which was determined in accordance with the protocol of the American Spinal Injury Association (ASIA), 13 ADL and independence, which were scored using the SCIM II 12 and walking capacity tests (the revised Walking Index for Spinal Cord Injury or WISCI II), 14 and timed walking tests. At the time of analysis, the EM-SCI included data from 1182 subjects (from 2001 to mid-2007). Included were subjects with traumatic SCI who had at least 2 measurements and who had From the Spinal Cord Injury Center, Balgrist University Hospital, Zurich, Switzerland. Address correspondence to Hubertus J. A. van Hedel, Spinal Cord Injury Center, Balgrist University Hospital, Forchstrasse 340, CH-8008 Zurich, Switzerland. hvanhedel@paralab.balgrist.ch. 343

2 344 Neurorehabilitation and Neural Repair completed a neurological assessment at 1 month after injury. The selected 886 subjects were rehabilitated in the following centers: Bad Wildungen (Germany), 33 subjects; Barcelona (Spain), 51; Bayreuth (Germany), 163; Bochum (Germany), 40; Frankfurt (Germany), 2; Halle (Germany), 47; Hamburg (Germany), 5; Heidelberg (Germany), 135; Karlsbad-Langensteinbach (Germany), 55; Murnau (Germany), 160; Nijmegen (Netherlands), 46; Garche (France), 47; Prague (Czech Republic), 22; Sion (Switzerland), 3; Tübingen (Germany), 7; Ulm (Germany), 20; and Zurich (Switzerland), 50 subjects. Detailed information about these subjects is provided in Table 1. Data from this cohort of subjects have been previously used to determine the validity and responsiveness of the SCIM II mobility items by comparing them to gait speed and the WISCI II. 15 SCIM II Mobility Items Table 1 Subject Demographics ASIA A ASIA B ASIA C ASIA D (n = 413) (n = 113) (n = 137) (n = 223) Age, y 39 ± ± ± ± 17 Gender, % female Body height, m 1.75 ± ± ± ± 0.09 Paraplegic, % LOS, wk a 23.4 ± ± ± ± 7.6 LEMS at 1 mo 1.4 ± ± ± ± 9.1 LEMS at 3 mo 1.8 ± ± ± ± 7.3 LEMS at 6 mo 2.1 ± ± ± ± 6.7 LEMS at 12 mo 3.0 ± ± ± ± 5.0 Abbreviations: ASIA, American Spinal Injury Association; LOS, length of stay; LEMS, lower extremity motor score. a The length of stay (LOS) of several subjects was missing, as many subjects had not completed the rehabilitation program at the time of the data analysis: 134 ASIA A subjects, 28 ASIA B and C subjects, and 44 ASIA D subjects. Two SCIM II mobility items were of interest, indoor mobility (<10 meters) and outdoor mobility (more than 100 meters). Mobility for moderate distances was not included in the analyses because it provides no additional information about walking ability compared to indoor mobility, whereas outdoor mobility does. 15 Indoor and outdoor mobility were similarly scored: (0) requires total assistance; (1) needs an electric wheelchair or partial assistance to operate manual wheelchair; (2) moves independently in a manual wheelchair; (3) requires supervision while walking (with or without devices); (4) walks with a walking frame or crutches (swing); (5) walks with crutches or 2 canes (reciprocal walking); (6) walks with 1 cane; (7) needs leg orthosis only; and (8) walks without aid. 12 We combined the items and scores into the following 5 functional ambulatory categories: (1) Subject is wheelchair-dependent (indoor and outdoor mobility score 2); (2) Subject is a supervised walker, ie, requires supervision to walk indoors and is wheelchair-dependent outdoors (indoor mobility = score 3; outdoor mobility score 2); (3) Subject can walk indoors, but is wheelchair-dependent outdoors (indoor mobility > score 3; outdoor mobility score 2); (4) Subject is an assisted walker, ie, requires a walking aid outdoors (indoor mobility > score 3; score 3 < outdoor mobility 7); and (5) Subject walks without assistance or aid both indoors and outdoors (indoor and outdoor mobility = score 8). We suggest that these 5 categories reflect the most important milestones concerning the recovery of ambulatory function that can be achieved during rehabilitation. The first category consists of subjects who are wheelchair-dependent at all times. Some subjects might perform some stepping, for example, during a therapy session, but are unable to transfer this ability into ADL. Second, supervised walkers require supervision to walk indoors for a short distance, but any further destination, especially outdoors, requires a wheelchair. These subjects could perform a 10MWT in the clinical environment, but in daily life, ambulation is limited to indoors and dependent on a second person. Third, subjects who can walk indoors might experience a large increase in quality of life, as they do not depend on personal assistance for indoor ambulation. Still, their walking function is limited, and for longer distances a wheelchair is required. Fourth, the mobility of assisted walkers during daily life is less influenced by obstacles that hinder subjects who are wheelchair-dependent. Still, these subjects require some sort of aid. Finally, those who can walk without aids are considered to be in the highest ambulatory category. Although these subjects might not have achieved normal ambulation, their daily life ambulation is minimally affected by this. Indeed, we assume that most SCI subjects can be categorized into these ambulation categories. Ten-Meter Walk Test The 10MWT was scored by trained physical therapists who measured the time (in seconds) needed to walk 10 meters at the subject s preferred speed. The test was performed on a flat, smooth, nonslippery surface without any disturbing factors. The tester walked beside the SCI subject for reasons of safety and measurement accuracy. A flying start was performed, ie, while the subject walked about 14 meters, the time was measured for walking the intermediate 10 meters (for details see also van Hedel et al 11 ). The time (seconds) was converted to walking speed (m/s). The speed was set to 0 m/s when the WISCI II 14 score was 0 or 1 (unable to walk 10 meters). Statistics Differences in walking speed between ambulatory categories were calculated with the nonparametric Kruskal-Wallis test. The factor group had 5 levels (ie, the 5 ambulatory categories as previously defined). Multiple pair-wise comparisons were performed with the Mann-Whitney U test (alpha was corrected for 4 comparisons and set at ). The relationship between the 5 ambulation categories and the preferred walking speed derived from the 10MWT was

3 van Hedel / Gait Speed and Ambulation Categories After SCI 345 Figure 1 Preferred Walking Speed per Ambulatory Group Note: Scatter-plots show the distribution of walking speeds per ambulatory category for 1, 3, 6, and 12 months after injury. The dotted lines indicate the cut-off points. n.s. indicates not significant. quantified using Spearman s correlation coefficient (ρ) and linear regression analysis. To determine the preferred walking speed that could distinguish between a certain level of functional ambulation (and higher) versus lower ambulation levels, receiver-operating characteristic (ROC) curves were used to determine sensitivity and specificity. The sensitivity is the true positive rate, ie, the probability of a positive test result in subjects with the condition. Specificity is the true negative rate, ie, the probability of a negative test result in subjects without the condition. The optimal value of the walking speed required to be at least in the ambulatory category of interest was determined by taking the maximum value of the sum of the sensitivity and specificity values for each speed. For each analysis, the data of all subjects were included. This means that the specific walking speed found using the ROC curves differentiates between SCI subjects walking poorer than the ambulatory category of interest and those who belong to the category or higher (dichotomized). The closer the area under the ROC curve approaches the 1 value, the better the accuracy. To verify the results, the analysis was repeated for the 1, 3, 6, and 12 months after injury assessments. Results The 5 functional ambulation categories covered more than 99% of the SCI subjects who were scored with the 18 original SCIM II mobility items. Only a small number of subjects were not included: at 1 month after SCI, 9 out of 841 subjects; at 3 months, 9 out of 753 subjects; at 6 months, 2 out of 591 subjects; and at 12 months, 3 out of 420 subjects. Mean Walking Speeds The distribution of the observations is shown in Figure 1. Each dot represents an SCI subject. In general, subjects in a higher category walk at higher speeds. For each time point, the walking speed differed between the ambulatory categories (for all, P <.001). The mean walking speeds (with 95% confidence interval [CI] boundaries for the mean) for each ambulation category and calculated for each time point are presented in Table 2. The overall mean walking speed (±SD) was calculated over the 4 time points and amounted to 0.01 ± 0.01 m/s (wheelchair-dependent), 0.34 ± 0.10 m/s (supervised walkers), 0.57 ± 0.17 m/s (indoor

4 346 Neurorehabilitation and Neural Repair Table 2 Preferred Walking Speeds Relevant to Ambulation Categories a Time Number of Mean Walking Speed, Minimal Required Point, mo Subjects b m/s (95% CI for Mean) Speed, m/s Sensitivity Specificity AUC (95% CI) Wheelchair-dependant ( ) ( ) ( ) ( ) Mean ± SD 470 ± ± Supervised walker ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Mean ± SD 13 ± ± ± ± ± ± 0.02 Walker indoor, ( ) ( ) wheelchair outdoor ( ) ( ) ( ) ( ) ( ) ( ) Mean ± SD 25 ± ± ± ± ± ± 0.01 Walker with aid ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Mean ± SD 41 ± ± ± ± ± ± 0.01 Walker without aid ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Mean ± SD 62 ± ± ± ± ± ± 0.01 Abbreviations: AUC, area under the curve; SD, standard deviation; CI, confidence interval. a The total number of subjects was 792 at 1 month, 714 at 3 months, 540 at 6 months, and 393 at 12 months after spinal cord injury. b Only the number of subjects of each specific category is presented. walker, wheelchair-dependent outdoors), 0.88 ± 0.04 m/s (assisted walker), and 1.46 ± 0.04 m/s (walker without any aid). In general, the preferred walking speed was higher in supervised walkers compared to wheelchair-dependent subjects, in assisted walkers compared to those who can walk indoors but require a wheelchair outdoors, and walkers who need no aid compared to assisted walkers (see Figure 1). A difference between supervised walkers and those who can walk indoors but require a wheelchair for outdoors was observed only once. Relationship Between Walking Speed and Ambulation Categories The relationship between the 5 ambulation categories and gait speed was good (Table 3). The correlation coefficients amounted to 0.84 and higher, and the explained variance of the linear regression model was above 76%, indicating that by knowing the variability in walking speed, more than 76% of the variability in the ambulation categories can be explained. Walking Speeds That Distinguish Between Ambulation Categories The walking speeds that distinguished between an ambulation category (and lower) versus the higher ambulation categories were determined using ROC curves (see also Figure 2 for the analysis at 6 months after SCI) and are presented in Table 2. When averaged for the 4 time points, the mean minimal walking speed to be considered a supervised walker or better was 0.09 ± 0.01 m/s. To be considered at least an indoor walker who is wheelchair-dependent outdoors, a minimal speed of 0.15 ± 0.08 m/s was required. Assisted walkers (and better) ambulated at least at 0.44 ± 0.14 m/s, which was 3 to 4 times faster than supervised walkers and indoor walkers with outdoor wheelchair dependency. Finally, subjects who walk without any aid were characterized by a minimal speed of 0.70 ± 0.13 m/s, which was almost 0.3 m/s faster when compared to assisted walkers. The average sensitivity was above 0.98 (excellent), whereas the averaged specificity exceeded 0.94 (excellent). The average area under the curve (AUC) was never less than 0.99, indicating good accuracy (Table 2). We determined speed levels that distinguished between categories of ambulation. However, when we categorized the SCI subjects according to the walking speed, the number of real supervised walkers that fell outside this category was 5 (1 month), 2 (3 months), and 0 (6 and 12 months). These numbers were 0 (1 and 3 months), 2 (6 months), and 1 (12 months) for those who walk indoors and are wheelchair-dependent outdoors, and 0 (1 and 3 months), 4 (6 months), and 6 (12 months) for assisted walkers. Finally, for those who walk without aid, these numbers amounted to 1 (1 month), 2 (3 months), and 0 (6 and 12 months).

5 van Hedel / Gait Speed and Ambulation Categories After SCI 347 Table 3 Relationship Between Ambulation Categories and Preferred Walking Speed Amb Cat = b 0 + Time, mo ρ (P Value) b 1 Speed R 2, % 95% CI b 1 P Value of b (<.001) speed < (<.001) speed < (<.001) speed < (<.001) speed <.001 Abbreviations: ρ, Spearman s correlation coefficient; Amb Cat, ambulation category; b0, constant; b1, regression coefficient; R2, explained variance; CI, confidence interval for the regression coefficient. Figure 2 Receiver Operating Characteristics (ROC) Curves Note: ROC curves show the relation between sensitivity and specificity of the minimal walking speed assessed at 6 months after injury required for each ambulatory category (and better). The arrow indicates the speed cut-off point (white dot). The diagonal represents the line of no discrimination. Points above this line indicate good classification results. Distinguishing Between Minor and Strong Dependence on Walking Aids We performed an additional analysis to determine differences between subjects who depended strongly on walking aids (those who walked with a walking frame, crutches, or 2 canes) versus subjects with minor dependency (only 1 cane or a leg orthosis). Again, the minimal speed required for those depending heavily on walking aids was 0.44 m/s (see Table 2). The minimal speed for walking depending only on a cane or an orthosis was 0.64 m/s (3 months) and 0.68 m/s (6 months). However, only a small number of subjects were included in

6 348 Neurorehabilitation and Neural Repair this ambulatory category (at 3 months, 8 subjects; at 6 months, 12; zero at the other time points). Discussion The aim of this study was to evaluate walking speeds that characterize functional ambulation categories in subjects after SCI. Preferred walking speed differed between consecutive ambulatory groups, except between supervised walkers and indoor walkers with outdoor wheelchair dependency. For both groups, the minimal walking speed that separated them from the lower ambulation categories was about 0.10 m/s to 0.15 m/s. Spinal cord injury subjects who walk at least 0.45 m/s were categorized as assisted walkers or better. Finally, a minimal walking speed of 0.70 m/s separated the walkers who needed aids from those who did not. An increase of about 0.3 m/s is needed to change from indoor walker with outdoor wheelchair dependency to assisted walker. A similar increase is required to change from assisted walker to outdoor ambulator without aid. In general, a strong relationship between walking speed and ambulation categories was found. This enables therapists who assess the preferred walking speed in the clinical environment to estimate the level of ambulatory function in the daily life environment. The supplemental analysis showed that the minimal walking speed required for subjects depending on a cane or an orthosis is about 0.65 m/s. Although this particular speed plausibly fills the gap between heavily and mildly aid-dependent subjects, the number of subjects was very small, which might have negatively influenced the accuracy of this result. To our knowledge this is the first study contrasting gait speed with a SCI disability scale in order to determine relevant walking speeds for functional ambulation categories in subjects with SCI. The ambulation categories were comprised of SCIM II mobility items that assess ambulatory function during daily life. Walking speed is considered a surrogate for the overall quality of gait (and motor function). 16 However, there is no general consensus concerning the relevance of a certain (increment in) gait speed, as assessed in the clinical environment for daily life ambulation in SCI subjects. In a previous study, SCI subjects were grouped as functional or nonfunctional walkers, depending on whether their walking speed was sufficient to safely cross a street (0.6 m/s). 17 Apparently, only SCI subjects who walk outdoors requiring little to no walking aid are able to do this. One study found that the speed required to cross a street safely was 1.22 m/s 18 and, consequently, it was concluded in another study that only a small proportion of the incomplete SCI subjects could cross safely. 19 These studies assume that the walking speed as determined in the clinical environment can be directly translated in the daily life environment on a 1:1 scale. However, more likely, the actual walking speed during ADL depends on many factors such as the motivation of the subject as well as environmental and psychological factors. In SCI subjects, information is lacking as to whether the walking capacity as tested in the clinic can be used to estimate the level of functional ambulation during daily life. For example, one study divided subjects after stroke into 3 functional ambulatory categories based on walking speed: household ambulation (<0.4 m/s), limited community ambulation ( m/s), and full community ambulation (>0.8 m/s). 20 Subjects who graduated from one ambulatory group to the next showed improvement in quality of life and participation. 21 It is notable that the limited community ambulation speed fits nicely with the speed found for walkers who depend on aids and that the full community ambulation threshold speed is almost achieved by the speed required for walking without aid, as found in this study. In elderly subjects, a walking speed above 1.0 m/s was associated with an independent lifestyle, 22 while a substantially relevant change in gait speed was 0.10 m/s. 23 We consider the latter value to be rather low in comparison to the findings of the present study in which the minimal increase in speed needed to graduate from outdoor wheelchair dependency to assisted walking is about 0.3 m/s. The differences in the values reported in the previous studies when compared to those in the present one are potentially due to the patient population, the outcome measures, and the statistical analyses performed. Methodological Considerations The purpose of this study was to assess if levels of walking speed can distinguish between clinically relevant ambulation categories. We focused only on walking speed to determine the level of functional ambulation. Of course, many factors can influence locomotion after SCI, both more general factors, such as the age of the subject, 24 spasticity, and impairments in strength and balance. 25 Walking can be cognitively more challenging for SCI subjects compared to healthy subjects. 26 This might minimally affect the performance of a 10MWT, where the subject can fully concentrate on walking itself without consideration of variables such as irregular surfaces, obstacles, a dark environment, or other disturbing factors. Secure ambulation in a challenging environment could be severely compromised by an attentional deficit. It has therefore been questioned whether walking speed as assessed in the clinic can be used to declare the level of ambulation during daily life. Another methodological consideration is if the subjects in this study are truly representative of SCI populations. We believe that the subjects included in this study fully reflect the European population of acute traumatic SCI subjects, as the data were gathered from 18 rehabilitation centers located throughout 5 European countries. Indeed, a large proportion of subjects had no walking ability, and consequently these subjects were categorized in the lowest ambulation category (wheelchair-dependent) with a walking speed of 0 m/s. This might have contributed to the high sensitivity and specificity levels found in this study. It is also noteworthy that the minimal speeds that distinguished between ambulation categories sometimes differed

7 van Hedel / Gait Speed and Ambulation Categories After SCI 349 between different time points after SCI (see also Table 2). We assume that these differences could partially be explained by the different number of subjects included in each analysis. In addition, the minimal cut-off speed was determined using the maximum sum of the sensitivity and specificity values. At some time points, the relative contribution of the sensitivity and specificity was different, which might have influenced the speed cut-off value found. In addition, we found no difference between supervised walkers and those who walk indoors but need a wheelchair for outdoor mobility. However, the number of SCI subjects in these ambulation categories was relatively small. Therefore, we might consider combining these groups into a single milestone, both for ambulatory rehabilitation and future studies. Finally, the new version of the SCIM II the SCIM III contains similar indoor and outdoor mobility items as version II. 27 Both measures have been applied successfully in European, Middle Eastern, as well as North American centers to document changes in the rehabilitation outcome or differences between subgroups of subjects. 12,27-29 We are therefore confident that these results would also apply to ambulation categories based upon SCIM III items. Conclusion Previously it was shown that a change in the preferred walking speed can be a valid and sensitive indicator of ambulatory function after SCI. 6,7 The present results show that specific speed levels as assessed in the clinical environment correspond to functional ambulation categories and can distinguish between these categories with high sensitivity and specificity. However, additional longitudinal studies are needed to further validate the finding of what change in gait speed is required to change from one functional ambulation category to another in subjects with SCI. Finally, by knowing the walking speed of a patient with an SCI, the level of community ambulation can be estimated well, without applying a battery of tests. Acknowledgments This work was sponsored by the International Spinal Research Trust (Clinical Initiative; CLI006) and the Interna-tional Research Institute for Paraplegia (P66). The author thanks Markus Wirz and Professor Volker Dietz for the helpful discussions and Danie Meyer and Rachel Jurd for help with the English language. Furthermore, we are grateful to the patients and the cooperating centers of the European Multi-center Study on Human Spinal Cord Injury (EM-SCI), especially to those therapists who recruited and tested the patients: Dr T. Meiners, Werner Wicker Klinik Bad Wildungen, Bad Wildungen (Germany); Dr J. Benito, Institut Guttmann Hospital de Neurorehabilitació, Barcelona (Spain); Professor R. Abel, Krankenhaus Hohe Warte Bayreuth, Bayreuth (Germany); Dr R. Meindl, BG Kliniken Bergmannsheil Universitätsklinik, Bochum (Germany); Dr O. Marcus, Berufsgenossenschaftliche Unfallklinik, Frankfurt am Main (Germany); Dr K. Röhl, Berufsgenossenschaftliche Kliniken Bergmannstrost, Halle (Germany); Dr R. Thietje, BG Unfallkrankenhaus, Hamburg (Germany); Professor H. J. Gerner, Orthopädische Univer-sitätsklinik Heidelberg, Heidelberg (Germany); Professor J. Harms, SRH Klinikum Karlsbad- Langensteinbach, Karlsbad-Langensteinbach (Germany); Dr M. Potulski and D. Maier, BG Unfallklinik Murnau, Murnau (Germany); Professor J. Duysens, Rehabilitation Centre Nijmegen, Nijmegen (Netherlands); Professor B. Bussel, Hopital Raymond- Poincarè, Garche (France); Dr J.Kriz, Motol Hospital, Prague (Czech Republic); Dr A. Al-Khodairy, Clinique Romande de Réadaptation, Sion (Switzerland); Dr H. P. Kaps, BG-Unfallklinik Tübingen, Tübingen (Germany); and Dr Y.-B. Kalke, Rehabilitationskrankenhaus Ulm, Ulm (Germany). References 1. Curt A, Schwab ME, Dietz V. Providing the clinical basis for new interventional therapies: refined diagnosis and assessment of recovery after spinal cord injury. Spinal Cord. 2004;42: Ditunno JF Jr, Burns AS, Marino RJ. Neurological and functional capacity outcome measures: essential to spinal cord injury clinical trials. J Rehabil Res Dev. 2005;42: Barbeau H, Elashoff R, Deforge D, Ditunno J, Saulino M, Dobkin BH. Comparison of speeds used for the 15.2-meter and 6-minute walks over the year after an incomplete spinal cord injury: the SCILT Trial. Neurorehabil Neural Repair. 2007;21: Dobkin B, Apple D, Barbeau H, et al. Weight-supported treadmill vs overground training for walking after acute incomplete SCI. Neurology. 2006;66: Dobkin B, Barbeau H, Deforge D, et al. The evolution of walking-related outcomes over the first 12 weeks of rehabilitation for incomplete traumatic spinal cord injury: the multicenter randomized Spinal Cord Injury Locomotor Trial. Neurorehabil Neural Repair. 2007;21: van Hedel HJ, Wirz M, Curt A. Improving walking assessment in subjects with an incomplete spinal cord injury: responsiveness. Spinal Cord. 2006;44: van Hedel HJ, Wirz M, Dietz V. Assessing walking ability in subjects with spinal cord injury: validity and reliability of 3 walking tests. Arch Phys Med Rehabil. 2005;86: Wirz M, van Hedel HJ, Rupp R, Curt A, Dietz V. Muscle force and gait performance: relationships after spinal cord injury. Arch Phys Med Rehabil. 2006;87: van Hedel HJ, Dietz V, Curt A. Assessment of walking speed and distance in subjects with an incomplete spinal cord injury. Neurorehabil Neural Repair. 2007;21: Lam T, Noonan VK, Eng JJ. A systematic review of functional ambulation outcome measures in spinal cord injury. Spinal Cord. 2008;46: van Hedel HJ, Wirz M, Dietz V. Standardized assessment of walking capacity after spinal cord injury: the European network approach. Neurol Res. 2008;30: Catz A, Itzkovich M, Steinberg F, et al. The Catz-Itzkovich SCIM: a revised version of the Spinal Cord Independence Measure. Disabil Rehabil. 2001;23: ASIA. International Standards for Neurological Classification of Spinal Cord Injury, revised Chicago, IL: American Spinal Injury Association; Ditunno JF, Barbeau H, Dobkin BH, et al. Validity of the walking scale for spinal cord injury and other domains of function in a multicenter clinical trial. Neurorehabil Neural Repair. 2007;21:

8 350 Neurorehabilitation and Neural Repair 15. van Hedel HJ, Dietz V. Walking during daily life can be validly and responsively assessed in subjects with a spinal cord injury. Neurorehabil Neural Repair. In press. 16. Dobkin BH. Short-distance walking speed and timed walking distance: redundant measures for clinical trials? Neurology. 2006;66: Zörner B, Dietz V, Curt A. Clinical algorithm for predicting walking capacity based on the ASIA motor score in acute SCI. Paper presented at: Combined ASIA/ISCOS Meeting; 2006; Boston, MA. 18. Hoxie RE, Rubenstein LZ. Are older pedestrians allowed enough time to cross intersections safely? J Am Geriatr Soc. 1994;42: Lapointe R, Lajoie Y, Serresse O, Barbeau H. Functional community ambulation requirements in incomplete spinal cord injured subjects. Spinal Cord. 2001;39: Perry J, Garrett M, Gronley JK, Mulroy SJ. Classification of walking handicap in subjects in the stroke population. Stroke. 1995;26: Schmid A, Duncan PW, Studenski S, Lai SM, Richards L, Perera S, Wu SS. Improvements in speed-based gait classifications are meaningful. Stroke. 2007;38: Cunningham DA, Paterson DH, Himann JE, Rechnitzer PA. Determinants of independence in the elderly. Can J Appl Physiol. 1993;18: Perera S, Mody SH, Woodman RC, Studenski SA. Meaningful change and responsiveness in common physical performance measures in older adults. J Am Geriatr Soc. 2006;54: Bohannon RW. Comfortable and maximum walking speed of adults aged years: reference values and determinants. Age Ageing. 1997;26: Scivoletto G, Romanelli A, Mariotti A, et al. Clinical factors that affect walking level and performance in chronic spinal cord lesion patients. Spine. 2008;33: Lajoie Y, Barbeau H, Hamelin M. Attentional requirements of walking in spinal cord injured patients compared to normal subjects. Spinal Cord. 1999;37: Catz A, Itzkovich M, Tesio L, et al. A multicenter international study on the Spinal Cord Independence Measure, version III: Rasch psychometric validation. Spinal Cord. 2007;45: van Hedel HJ, Curt A. Fighting for each segment: estimating the clinical value of cervical and thoracic segments in SCI. J Neurotrauma. 2006;23: Wirth B, van Hedel HJ, Kometer B, Dietz V, Curt A. Changes in activity after a complete spinal cord injury as measured by the Spinal Cord Independence Measure II (SCIM II). Neurorehabil Neural Repair. 2008;22: For reprints and permission queries, please visit SAGE s Web site at