Effects of Body Weight-Support Treadmill Training in Incomplete Spinal Cord Injury: A Critical Review

Problem statement: The present study reviewed the efficacy of body-we ight support treadmill training in patients with incomplete spin al cord injury. Approach: We developed a computer-supported search strategy for finding stud ies in the main data bases: Pubmed/Medline, ISI Web of Knowledge and Scielo. We also developed a ma nual search withing all electronically references found. The search terms gait, locomotor training, spinal cord injury, SCI, body-weight support treadmill training were used. Only papers p ublished in English and conducted from 1991 up to 2012 were preferentially reviewed. The inclusion cr iteria were: (a) studies using body-weight support treadmill training in patients with iSCI; (b) studi es with iSCI patients classified as ASIA C or D. Th e exclusion criteria were: (a) studies comparing the BWSTT with manual and robotic techniques and associated BWSTT with the use of drugs. Results: After a careful electronic search in the data base s, forty seven articles were found. Studies with patie nts of ASIA A and/or B classification, studies comparing the BWSTT with manual and robotic techniq ues and studies associating BWSTT with the use of drugs were excluded of the study. Thus, only 21 articles were selected. Conclusion: iSCI is a severe neurological condition that causes serious c mpromises, such as gait disability. The wide use o f BWSTT seems to be an effective, safe and reliable m ethod for functional rehabilitation of gait. Even though, there is no sufficient scientific evidence to confirm that the BWSTT is superior to the other rehabilitation techniques.


INTRODUCTION
Approximately 259,000 people with a Spinal Cord Injury (SCI) live in the United States. SCI is classified according two criteria: the neurological level and if the injury is complete or incomplete. Of those SCI patients, approximately 30.1% have incomplete tetraplegia. This level of lesion is classified "D" on the American Spinal Injury Association Impairment Scale (AIS), due to the preservation of some sensorimotor functions below the level of injury, often at the level of cervical backbone (NSCISC, 2009).
The ability of muscle to function against gravity is preserved in more than half of all muscle groups below the neurological level of injury. However gait patterns in this population often reflect incoordination of muscle action, leading to inefficient gait, with high-level energy consumption (Cunha-Filho et al., 2003). Although able to ambulate, these people are often unable to isolate specific muscle patterns to ambulate at functional speeds and without significant gait deviations. Gait impairments in people with incomplete SCI (iSCI) are not only induced by alterations in motor control, but also by muscle weakness, decreased range of motion and decreased muscle extensibility (Gracies, 2005;Biering-Sorensen et al., 2009), improving quality of life through functional independence, improving self esteem and social inclusion of these patients .
In addition to its severity and irreversibility, the iSCI requires long and costly rehabilitation programs that in general lead to functional recovery. In front of the difficulties of the traditional gait training, Body Weight Support Treadmill Training (BWSTT) was created (Cunha et al., 2002). It allows for the repetitive practice of walking/stepping in a controlled environment, which the patient's weight is partially unloaded and sensory input is provided that facilitates normal walking parameters (Cunha et al., 2002;Visitin et al., 1998;Wilson et al., 2001;Field-Fote, 2000) Research has shown improvements in functional walking ability in both acute and chronic ISCI populations after locomotor training with BWSTT (Hicks et al., 2005;Wernig et al., 1998;Wirz et al., 2005).
The theoretical basis for the emergence of BWSTT results from studies in cats with spinal cord injuries that have passed through gait training (Threlkeld et al., 2003). From these studies it was found a central pattern generator that was responsible for generating the cyclical pattern of movement in the central nervous system of these animals, even with little training after SCI (Duysens and Crommert, 1998;Schindl et al., 2000;Miyai et al., 2000). Historically, BWSTT is linked to neurological impairment, which was firstly used in stroke patients (Hesse et al., 1999;Barbeau and Visintin, 2003) and SCI (Visintin and Barbeau, 1989;Gardner et al., 1998), but now it has been used to treat various neurological and orthopedic diseases. Thus, this study aims to investigate the efficacy of BWSTT in iSCI rehabilitation.

MATERIALS AND METHODS
The present paper reviewed the efficacy of bodyweight support treadmill training in patients with iSCI. We developed a computer-supported search strategy for finding studies in the main data bases: Pubmed/Medline, ISI Web of Knowledge and Scielo. We also developed a manual search withing all electronically references found. The search terms gait, locomotor training, spinal cord injury, SCI, bodyweight support treadmill training were used. Only papers published in English and conducted from 1991 up to 2009 were preferentially reviewed. The inclusion criteria were: (a) studies using BWSTT in patients with iSCI; (b) studies with iSCI patients classified as ASIA C or D; (c) studies using BWSTT in combination with electrical stimulation in patients with iSCI. The exclusion criteria were: (a) studies comparing the BWSTT with manual and robotic techniques and associating BWSTT with the use of drugs.

RESULTS
After a careful electronic search in the data bases, forty seven articles were found. Studies with patients of ASIA A and/or B (Maynard et al., 1997) classification, studies comparing the BWSTT with manual and robotic techniques and studies associating BWSTT with the use of drugs were excluded of the study. Thus, 14 studies were selected and the main were summarized in Table 1 with their respective methods and outcomes.
Three out 21 studies (Wernig et al., 1995;Field-Fote et al., 2005) were randomized clinical trials and/or experimental studies of more than 20 patients. The rest were case reports, case-control pilot study and /or experimental studies of less than 20 patients. Despite the positive outcomes found in favor of BWSTT, reliable scientific evidence is needed in order to confirm these findings.
BWSTT provides specific activation of the neuromuscular system below the level of injury in order to "retraining" the CNS and restore motor skills. Researchers examined the role of the spinal cord to control the walk and found that cats with complete transection of thoracic spinal cord could restore locomotion, after intense practice of walking exercises, in order to improving normal gait parameters, such as, walking speed and appropriate kinematic (Behrman and Harkema, 2007).
Within this context, BWSTT is tough to be superior compared to conventional gait training in neurological disorders, such as stroke (Cunha et al., 2002;Daly et al., 2004;Visintin and Barbeau, 1998), Cerebral Palsy (CP) (Schindl et al., 2000) and Parkinson's Disease (PD) (Miyai et al., 2000). In stroke patients, a randomized clinical trial was conducted comparing the conventional physical therapy intervention (4 daily sessions, 3 h each) with BWSTT (4 daily sessions, 20 min. each) in 15 stroke patients in acute phase (< 6 weeks). Investigators verified the functional gait, the speed obtained in 5 meters, the distance traveled in 5 min and the energy expenditure (oxygen consumption -VO 2 max.).  Wernig and Müller (1992) The time of treatment was 1½ to The distance increased after training. (5-20 years after injury) 7 months, therapy session ranged The important phenomenon observed from 30-60 min. The use of was that patients with absent BWSTT started with 40%.
voluntary activity in the limb at rest, was able to ambulate 100-200 meters (locomotor automatism). Wernig et al. (1995) 89 iSCI (male) The treatment with BWSTT had Patients acquired independent gait, 44 in chronic phase and 45 an average of 10.5 weeks (3-20 weeks), speed and performance of the gait. (in acute phase) 64 controls and patients who ambulate, improved Although there has been little therapy session 30 min. (5 times/week). improvement in voluntary muscle activity. Gardner et al. (1998) 1 iSCI (male) Patient injured between C5-C6. Improved performance in Training in the BWSTT with 40% the race and in the weight bearing, for 6 weeks cadence of the gait. (3 times/week) 16 sessions (20 min each). Wernig et al. (1998) 76 iSCI (male) Patients trained in the BWSTT an Improved function of moving from (35 in chronic phase and average of 12 weeks (sessions wheelchair to walking. 41 in acute phase) with 30 min.(5 times/week) Ambulation patients, improved gait and patients in acute phase, speed and endurance and keep trained in BWSTT, an average of the winnings for an 10 weeks (sessions with 30 extended period after training. min, 5 times/week). Nymark et al. (1998) 5 iSCI (male) Patients injury among C2-T10. Improved range of motion (sub acute phase, 2 All participants trained in the and kinematic measures months after injury) BWSTT for 36 sessions temporal and spatial) of the gait. (1 h each) along 3 months.
(More weight bearing in lower limbs. Behrman and Harkema 3 iSCI (male) Patients ASIA C and D training Case 1-Improved walking speed (2000) in the BWSTT. and independence at home, Case 1-Men, 20 years, just starting to use a cane for mobility. injury in T5, ASIA C, one After the whole training, month after the event, with key progressed to Asia D and Key muscles motor score 2/50 in lower limbs. muscles improved to 38/50. Case 2-Men 43 years, injury Case 2-After training, in C6, 8 months after trauma, remained ASIA D, key muscles ASIA D, key muscles score 32/50. score improved to 34/50. Case 3-Men, 45 years, injury in After complete training became T-9, ASIA D, lower limbs full time ambulant with crutches. key muscles score 46/50. Improvement in walking speed, balance, fall risk and quality of life. Case 3-He did not have any change in classification and motor score of ASIA, improved in climbing stairs and walking speed. Protas et al. (2001) 3 iSCI (male) Patients iSCI injured in thoracic All three patients improved level, 2 cases ASIA D and 1 case in walking speed and efficiency, ASIA C. Training in the BWSTT with reduced O2 consumption in 68%.

% weight bearing, sessions
The muscle function did not change. of 20 min, 5 times/ week during 3 months. Field-Fote (2001) 19 iSCI (male) (<1 year after injury) Assess Individual with some ability to the use of BWSTT associated ambulate received benefits from with electrical stimulation in peroneal BWSTT associated with electrical nerve. The treatment lasted 3 stimulation, improved gait speed. months (3 times/week) with sessions of 1.5 h. Field-Fote and 14 iSCI (male) Patients SCI ASIA C, injury below Patients walked at a speed Tepavac (2002) 3 controls (male) T-10, average of 70 months 84% higher than the initial after injury. 12 weeks of and improved coordination. treatment (3 times/week) total of 36 sessions. Postans et al. (2004) 12 iSCI (male) Patients ASIA C e D. The group that received BWSTT (acute phase) The treatment lasted 4 weeks obtained better results on No change in body fat. Phillips et al. (2004) 9 iSCI (male) Patients ASIA C and developed Improved blood glucose regulation the treatment using BWSTT for (increased glucose intolerance) 6 months (3times/week) and increased insulin sensitivity. 68 total sessions. Started with 65±3% of body weight support. Behrman et al. (2005) 1 iSCI (male) Patient, injury in C6-7, ASIA D, Evolved from home ambulant ambulation for three years with a rolling with orthesis to community ambulation walker and right AFO brace.
using crutches, improve gait speed, 45 sessions of locomotor training becoming more symmetrical, with the were performned, 5 balance of the upper limbs, sessions per week, 30 min. coordination and appropriate steps. BWSTT training, 20 min.
The number of steps measured Walk on the ground and for 24 h also increased. community ambulation. Field-Fote et al. (2005) 27 iSCI (male) Patients with injury above T-10, An improvement in walking speed, (<1 year after injury) received treatment with BWSTT stride length and during 12 weeks (5 times/ symmetry was observed. week), 60 min/ session. Effing et al. (2006) 3 iSCI (male) Patients, 2 ASIA C (>4 years injury) and 1 ASIA D performed BWSTT A discreet improvement was for 12 weeks (5times/ observed in quality of life. week) 30 min each session. Lucareli et al. (2008) 12 iSCI (male) 12 patients iSCI, ASIA C and D Increased speed distance, cadence, (<1 year after injury) accomplished 30 sessions stride length, total cycle time 12 controls (male) (2 times/week) during 4 months and decreased support time on the gait. with BWSTT and 12 patients The BWST was more effective received physical therapy in than conventional physiotherapy the control group with on gait kinematic parameters. the same procedure. Lucareli et al. (2008) 12 iSCI (male) Assessed the use of BWSTT (>1 year after injury) in patients ASIA C e D, ambulant. There was improvement in Treatment lasted 4 months the temporo-spatial parameters 30 sessions, 30 min each), of gait, three showed no (2 times/week. improvement in quality of life. Behrman et al. (2008) 1 iSCI (4 ½ years) Training in BWSTT, ASIA C, Showed after the first month of (male) injury in C-7, 3 months after treatment: voluntary steps, injury, wheelchair dependent, progressing with the ability to walk LEMS 4/50. 16 months of treatment, with a rolling walker. At the end of 76 total sessions, 20-30 min of BWST.
locomotor training, walking independently, with significant There was no change in LEMS score. Musselman et al. (2009)

iSCI (male)
Assessed 4 SCI patients ASIA C. All There was no improvement in patients started BWSTT treatment.
static and/or dynamic balance. After, 2 of them receive skills training The walking speed improved (walking on different surfaces, gait during treatment with motor training with obstacles, walking long skills, higher than BWSTT. distances and others) and the rest continued with BWSTT. In the third phase of treatment, reversed the order of treatment. Each treatment phase lasted 3 months, 1 h each session (5 times/week). The sample size probably did not allow a statistically significant difference, however the investigators were able to observe large differences in distance traveled and energy expenditure of the group that received BWSTT compared to conventional physical therapy group (Cunha et al., 2002). In another study, 43 poststroke patients received BWST treatment and 36 patients were treated with conventional gait training. Both groups were treated for 6 weeks (2 sessions per week, 20 min each). Investigators verified the balance, motor recovery, walking speed and strength. There were statistical differences for all parameters, with better results for BWSTT group. They conclude that increasing gradually the body weight is more advantageous and contributes to better balance and gait capacity when compared to conventional training on the treadmill (Visintin and Barbeau, 1998). Daly et al. (2004), studied 16 post-stroke patients over 12 weeks, 4 times a week, 90 min each session. They compared a group of 9 patients who were treated with BWSTT in combination to Functional Neuromuscular Stimulation (FNS) with Intramuscular (IM) electrodes (FNS-IM) to a group of 7 patients who received only BWSTT. Outcome measures were kinematics of gait swing phase. There were no statistical significant differences between groups, with no significant pre-/post-treatment gains in peak swing hip flexion. BWSTT had no significant gains in other gait components at posttreatment or at follow-up. BWSTT+FNS-IM group had significant gains in peak swing knee flexion and midswing ankle dorsiflexion that were maintained for 6 months.
In addition, in patients with CP, BWSTT also aided in gait and motor skills, posture, balance and improved scores on the Gross Motor Function Measure (GMFM) scale, in 6 out 10 patients that cannot walk, demonstrating the effectiveness of BWSTT also in patients depending on the gait (Schindl et al., 2000).
In PD patients, BWSTT was superior compared to conventional physical therapy treatment. Ten PD patients were investigated and statistical difference was observed between conventional physical therapy and BWSTT groups. An improvement in motor performance and mobility of gait were found, leading to an enhancement of activities of daily life (Miyai et al., 2000).
With respect to iSCI, several studies showed significant clinical effects (Field-Fote et al., 2005;Gardner et al., 1998;Field-Fote and Tepavac, 2002;Wernig and Wernig, 1992;Wernig et al., 1995;Nymark et al., 1998;Behrman and Harkema, 2000;Protas et al., 2001;Field-Fote, 2001;Postans et al., 2004;Field-Fote et al., 2005;Lucareli et al., 2008;Behrman et al., 2008;Musselman et al., 2009;Phillips et al., 2004;Behrman et al., 2005). For instance, the study of Field-Fote et al. (2005) assessed gait of 27 iSCI patients (<1 year after injury, above T-10). After 12 weeks (5 sessions/week) treatment with BWSTT there was improvement in gait speed and step length and symmetry (Field-Fote et al., 2005). Another study examined 12 iSCI patients (ASIA C and D) in acute phase. Each patient was his own control in second period. In the first period, they received conventional physical therapy and BWSTT, beginning with a withdrawal of 40% of body weight associated with FNS. The treatment lasted four weeks (5 sessions/week) with 25 min per session for both treatment periods. The findings of first period were superior compared to second period on performance and walking speed. However, spasticity and muscle strength were not significant different between periods of treatment (Postans et al., 2004). Stewart et al. (2004) showed that after 6 months of treatment (3 sessions/week), iSCI patients (ASIA C) showed improvement in lipid profile with decreased LDL, increased muscle fiber area of type I and IIa, but no changes in body fat mass. Similar results were found by Phillips et al. (2004). They examined iSCI patients (ASIA C) and treated them with BWSTT for 6 months (3 times/week) with a total of 68 sessions, observing improvement in regulation of blood glucose (glucose tolerance) and increased insulin sensitivity.

CONCLUSION
iSCI is a severe neurological condition that causes serious compromises, such as gait disability. Most of studies of BWSTT showed better physiological and psychological effects than studies using conventional gait training. BWSTT appears to be more effective when combined with electrical stimulation. In contrast, no benefits were found for BWSTT in the improvement of muscular strength and bone density, spasticity and body fat. Moreover, BWSTT showed reduction in energy expenditure measured by oxygen consumption (VO 2 max) and decrease in heart rate after training. These findings indicate decrease in chances of a traumatic event such as cardiac arrest or shortness of breath during training and decrease in fatigue during and after training. However, BWSTT does not seem to prevent loss of bone density in both acute and chronic phases of iSCI. Even though, BWSTT seems to be effective, safe and reliable method for functional rehabilitation of gait, there is still no sufficient scientific evidence to confirm it. Therefore, more controlled studies are required to better understand the efficacy of BWSTT in iSCI.