739
Journal of Back and Musculoskeletal Rehabilitation 28 (2015) 739–747 DOI 10.3233/BMR-140577 IOS Press
Exercise and gait training in persons with paraplegia and its effect on muscle properties Rohit Prakash Bhidea,∗, Cassandra Solomonsc, Suresh Devsahayamc and George Tharionb a
Clinical Fellow, Division of Physiatry, Department of Medicine, University of Toronto, Toronto, ON, Canada Department of Physical Medicine and Rehabilitation, Christian Medical College, Vellore, Tamil Nadu, India c Department of Bioengineering, Christian Medical College, Vellore, Tamil Nadu, India b
Abstract. BACKGROUND: Upper extremity strengthening and gait training with orthoses form a major part of inpatient rehabilitation of paraplegic patients in developing countries. This helps to overcome architectural barriers and limited wheelchair accessible environment in the community. OBJECTIVES: To evaluate the changes in physiological properties of the Triceps Brachii muscle following exercise training in individuals with paraplegia. The authors also explored the correlation between muscle property changes and gait parameters using orthoses in paraplegic persons. METHODS: Twelve subjects with complete paraplegia and neurological level of injury (NLI) from T9 to L1, underwent exercise training for a mean 64.1 ± 4.1 days. Triceps brachii was chosen as the sample muscle. Variables like arm circumference, time to fatigue and mean power frequency (MF) (surface EMG parameter), were recorded at the beginning and the end of training, during a sub-maximal isometric elbow extension. Non-parametric tests were used to assess statistical significance between the two recordings. Additionally, gait parameters like walking speed and distance (with the help of orthoses) were obtained and compared with the above variables, to determine impact of upper extremity strengthening on gait improvements in such patients. RESULTS: Statistically significant changes were noted in bilateral arm circumferences (p = 0.003 bilaterally) and MF drop, expressed as percentage (right p = 0.04, left p = 0.01), indicative of better muscle resilience and adaptation. Significant positive correlation was observed between ‘time to fatigue’ and the orthoses-aided total walking distance (right ρ = 0.65, left ρ = 0.69). CONCLUSIONS: Exercise training induces noticeable changes in the muscles of upper extremities favoring better muscle adaptation. Furthermore, positive correlation between ‘time to fatigue’ and (orthotic) aided walking distance highlights the positive impact of strengthening program on gait parameters in paraplegic patients. These findings are important and relevant in developing countries with environmental barriers. Upper extremity strengthening should be included in the rehabilitation of paraplegic patients who are being trained for ambulation with orthoses. Keywords: Exercise, paraplegia, muscle fatigue, surface electromyography, triceps
1. Background Spinal cord injury (SCI) has far reaching consequences on an individual’s life. In a developing country, wheelchairs are often not the best mode ∗ Corresponding author: Rohit Prakash Bhide, Clinical Fellow, Division of Physiatry, Department of Medicine, University of Toronto, UHN – Toronto Rehabilitation Institute, 550 University Avenue, Toronto, M5G 2A2 Ontario, Canada. Tel.: +1 416 6252843; E-mail: [email protected], [email protected].
of transportation. Patients with paraplegia encounter significant architectural and attitudinal barriers compared to their western counterparts. Ambulation training in paraplegic patients, with the help of orthoses like Knee-Ankle-Foot orthoses (KAFOs) and forearm crutches, can potentially overcome this issue. In paraplegics with caudal neurological levels, majority of the rehabilitation training focuses on upper extremity strengthening and ambulation training, in order to condition the upper extremity for activities which were earlier performed by the lower limbs [1]. Dur-
c 2015 – IOS Press and the authors. All rights reserved ISSN 1053-8127/15/$35.00
740
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
ing rehabilitation, efforts are taken to meet the new demands through exercises and endurance training program. Physical fitness directly correlates with functional capability and highlights the importance of systematic rehabilitation and conditioning exercise programs. Dallmeijer et al. have reported notable improvements in physical capacity and abilities to perform daily living activities during and after rehabilitation in persons with SCI [2,3]. On a physiological level, skeletal muscle is known for its adaptability and plasticity, as commonly seen in the results of athletic training. Fitts and Gollnick have documented changes occurring in several aspects of skeletal muscle such as micro-architecture, histochemical and fiber characteristics after long term focused endurance training [4–6]. Schantz reported better muscle adaptations in upper extremities of individuals with spinal cord injury and successively higher proportions of type I (non-fatigable) fibers following endurance training [7]. The gold standard for documenting these changes is muscle biopsy [8], which, being an invasive procedure, has its own disadvantages [9]. Any damage or complication in the upper extremities of paraplegic patients, with already non-functioning lower extremities, can greatly hamper the adapted lifestyle. There is evidence regarding parameters of surface electromyography (SEMG) being used non-invasively to study manifestations of fatigue [10–12], with negligible discomfort to the individual. Spectral analysis studies to evaluate metabolic muscle fatigue have shown that muscles with type II fiber predominance have faster rates of EMG ‘Mean Power Frequency’ (MF) reduction with time [13]. This study was undertaken to document non-invasively the changes occurring in upper extremity (triceps brachii muscle) of paraplegic subjects following exercise training. The second objective was to explore the impact of these changes in muscle properties with gait parameters like walking distance and speed, using orthoses. Triceps brachii was chosen as the sample muscle, as it is involved in transfers, push-ups and gait training in persons with SCI.
2.1. Study subjects A sample size of convenience of 12 paraplegic subjects was taken satisfying the inclusion criteria: age 18–60 years, neurological level of injury (NLI) T9 and below with American Spinal Injury Association (ASIA) impairment scale (AIS) A or B (motor complete paraplegia). NLI T9 and below was chosen after checking Beevor’s sign for abdominal muscle tone and evidence of fair truncal control to aid standing balance and stability. None of the subjects had received formal rehabilitation and exercise training prior to the admission. Proposed goal of ambulation with bilateral KAFOs and forearm crutches were set for all subjects. Patients with AIS C or D, NLI above T9 and patients with upper extremity issues like shoulder injury, fractures and neuropathies were excluded. 2.2. Exercise and rehabilitation training All subjects underwent similar structured rehabilitation training which included Physical and Occupational therapy to maximize mobility and activities of daily living. It consisted of daily six hours of therapy (three 2-hour therapy sessions), five days a week. During the sessions, patients were supervised by the therapist who ensured maximum utilization of the therapy time. Upper extremity strengthening program consisted of endurance and weight training for shoulder abductors, depressors and elbow flexors/extensors, wrist extensors and hand grip strengthening. Gradual progression in terms of repetitions and number of sets performed for each exercise was monitored during the entire training. Simultaneously, gait training was initiated which involved orientation to vertical with the help of tilt table and Standing frame. Following this, patients worked on their standing balance with KAFOs in parallel bars followed by graduated gait training within the parallel bars, progressing to walker and then forearm crutches. Once a subject started walking with KAFOs and forearm crutches, they were encouraged to practice during evening hours and on weekends. 2.3. The experiment set-up
2. Methods The study was conducted in a tertiary care rehabilitation institute of a University teaching hospital following approval from the Institutional Review Board and Ethics Committee. Informed consent was obtained from all participants.
Each patient was subjected to the experimental measurements in both arms at two points – initial and final assessment. Three variables were obtained during each recording: arm circumference, time to fatigue and changes in Surface electromyography (SEMG) parameter – Mean Power Frequency (MF).
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
741
Fig. 1. The fabricated apparatus with force transducer (load cell) fixed at the end of the forearm beam. The proximal end of the load cell is fixed to the forearm beam. On application of extension force at the distal end of the load cell, corresponding changes are seen on the monitor.
Arm circumference measurements were taken from both arms at the beginning of each recording by placing the upper limb besides the trunk with elbow flexed at 90 degrees. Arm circumference was measured from a point midway between the acromion and olecranon process. These measurements did not control for body fat, which we later noted to be one of the limitation. This was followed by surface EMG recordings from each extremity. A structured frame was fabricated to stabilize the upper extremity, which consisted of two longitudinal bars hinged at 90 degrees, with a trough attached on the reference beam to match the arm contour (Fig. 1). Force transducer (load cell) was fixed at the distal end of wooden bar to measure the extension force. The forearm when placed in semi-prone position and strapped with Velcro against the force transducer, fixed the upper limb in a rigid position with elbow and shoulder joint stabilized and maintaining constant muscle length of the triceps brachii. Standard protocol was followed which began with marking of points as per SENIAM (Surface Electromyography for the Non-Invasive Assessment of Muscles) recommendations [14]. Skin preparation was done by shaving and surgical alcohol application. A pair of bipolar surface electrodes (circular silver discs, diameter 0.01 metre, 0.015 metre inter-electrode spacing) with conductive gel were placed on the lateral
head of the triceps brachii muscle along the direction of muscle fibers. A single carbon gel adhesive electrode, used as a ground, was placed off the muscle. Lateral head of Triceps brachii was specifically selected as it is superficial, has one-joint action, and most importantly, shoulder position and forearm rotation does not affect its action and contractility [15]. Custom-made ‘CMC-Daq’ data acquisition hardware and software, compatible with Windows XP, was used for data recordings and collection. CMC-Daq is a four channel data acquisition system, with 10/12 bit data acquisition. The force transducer (load cell) from the wooden frame was connected to first channel of CMC-Daq, providing visual biofeedback during elbow extension. SEMG electrodes from the lateral head of triceps brachii, were connected to second channel. All data readings were recorded at a sampling rate of 2000 Hz and later processed using CMC-Daq software on a personal computer. 2.4. The procedure During the procedure, the subjects lay supine with electrodes placed and arm fixed in the frame as described earlier. They were then asked to perform maximal elbow extension isometrically against the force transducer lasting 30 seconds, displayed graphically
742
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
Fig. 2. Changes seen on the monitor on application of extension force. The rise in the ‘force line’ is seen by the subject on the screen (Channel 1) and acts as visual biofeedback. During the same recording, the surface EMG from the lateral head of Triceps Brachii is recorded via the second channel.
on the computer screen (Fig. 2). This was done three times, at an interval of 5 minutes. Since the fabricated frame fixes the point of force measurement, torque is proportional to the measured force. Force produced during these contractions was termed as Maximal Voluntary Contraction (MVC) and averaged from three contractions. A point corresponding to the 50% MVC was determined for each subject. After 15 minutes of rest, the subjects were asked to perform a sustained fatiguing isometric voluntary contraction at 50% MVC. Continuous visual biofeedback from the computer screen helped to maintain the level of force. Efforts were taken to reduce central fatigue by constant verbal encouragement throughout the contraction [16]. Recordings from both upper extremities were taken separately. The same procedure with all the steps was repeated on completion of training. Spectral analysis was done of the collected data using averaging of periodograms and fast Fourier transform to calculate MF, with a frequency band of 20– 500 Hz (Fig. 3). Decrease in MF during the fatigu-
ing isometric contraction was expressed as percentage reduction compared to baseline, rather than absolute difference, to eliminate factors like electrode orientation, spacing and to permit comparison between the two recordings [17]. ‘Time to fatigue’ was calculated from the same recording, as the duration for which the sustained sub-maximal contraction lasted and was expressed in seconds. 2.5. Gait parameters Functional walkers were those who could walk at least 250 meters at a stretch and were able to negotiate architectural barriers [18,19]. Gait parameters like walking distance and speed were recorded, on a marked path, with the help of a stopwatch. This recording was done only at the end of rehab training as all patients were either wheelchair or bed-bound at the beginning of rehab training. Statistical analyses were done using SPSS 16.0. Non-parametric tests were used in view of the small sample size.
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
743
Fig. 3. The Mean power frequency (MF) calculated from the surface EMG spectral analysis at one point in time during the isometric contraction. Table 1 Demographic variables of the study participants Demographic variables Age distribution (mean-29.66 yrs) 15–20 20–30 30–40 40-above Level of injury T9-T10 T11-T12 L1-L2 Duration of stay in rehab (average – 62.58 days, 9 weeks) 40–50 50–60 60–70 > 70
Number of subjects (n = 12) 1 6 5 0 5 6 1
2 3 5 2
3. Results All the 12 recruited patients were in the middle age group with a median age of 28.5 years (range19–40 years). 90% of patients had NLI between T9T12. Average duration prior to inpatient admission was 6.2 months (range-0.5–14 months). The mean duration of exercise and rehab training was 64.1 ± 4.1 days (range-38–92 days) (Tables 1 and 2). 3.1. Changes in muscle properties Wilcoxon Signed ranks test was used to calculate the statistical significance of changes in each variable.
At the end of exercise training, arm circumferences of both upper limbs showed significant increase; right 0.021 ± 0.011 meters and left 0.022 ± 0.011 meters, p =0.003 bilaterally. ‘Time to fatigue’ improved during training with a mean increase of 18.1 ± 37.9 seconds for the right side (p = 0.170) and 29 ± 43.7 seconds for the left (p = 0.05). MF calculated from spectral analysis of SEMG decreased linearly over the course of sustained contraction. This drop in MF when calculated during the final reading was significantly lesser compared to that in the initial reading. A difference of 7.8 ± 10.5 %, p = 0.04 for right side and 10.82 ± 1.14 %, p = 0.01 for left side was seen, implying better metabolic resilience (Table 3). 3.2. Gait parameters Ambulation parameters measured at the end of exercise training could be directly related to the overall improvement due to exercise training. Correlation analyses for the improvement in above variables with final ambulation outcomes such as walking speed and distance was done using Spearman’s ρ (rho). Ambulation parameters included recording of walking distance and speed (using orthoses). Average walking distance was 362.5 ± 82.9 meters at a speed of 0.205 ± 0.046 meters/second. Correlation analysis yielded positive correlation between improvement in ‘time to fatigue’ in both arms following training with walking distance (right, p =
19 24 26 33 29 37 37 38 21 24 40 28
L1-AIS A T12-AIS B T10-AIS A T11-AIS A T12-AIS A T11-AIS A T10-AIS A T10-AIS A T9-AIS A T9-AIS A T11-AIS A T12-AIS A
Age Neu. Lev. (yrs) of injury
56 82 68 92 45 54 61 53 70 61 65 44
Total rehab stay (days)
Initial 0.225 0.21 0.3 0.25 0.3 0.26 0.27 0.275 0.23 0.25 0.28 0.225
Arm circumference (meters) Right Left Final Difference Initial Final Difference 0.24 0.015 0.225 0.25 0.025 0.235 0.025 0.21 0.23 0.02 0.31 0.01 0.29 0.31 0.02 0.28 0.03 0.25 0.27 0.02 0.3 0 0.305 0.31 0.005 0.27 0.01 0.27 0.27 0 0.3 0.03 0.265 0.3 0.035 0.3 0.025 0.265 0.3 0.035 0.27 0.04 0.23 0.27 0.04 0.275 0.025 0.25 0.275 0.025 0.3 0.02 0.28 0.3 0.02 0.245 0.02 0.22 0.245 0.025 Initial 65 51.5 92.5 85 79 117 76 145.5 98 62 75 80
Time to fatigue (Seconds) Right Left Final Difference Initial Final Difference 56.5 −8.5 60 40 −20 66 14.5 57.7 101.5 43.7 188 95.5 92 151 59 90 5 93.2 101 7.7 120 41 61 156 95 115 −2 106 93.5 −12.5 132 56 69 153 84 111 −34.5 150.5 111 −39.5 119 21 98 121 23 47.7 −14.2 33 77.5 44.5 134 59 44 114.5 70.5 65.5 −14.5 111 104 −7
Table 2 Details of the recordings of all the patients Mean frequency percentage decrease (%) Walking Walking Right Left distance speed Initial Final Difference Initial Final Difference (Meters) (meters/sec) 30.4 16.6 13.8 21.8 11.5 10.3 330 0.23 38.8 13.9 24.9 24.4 15.6 8.8 435 0.25 25.0 15.8 9.2 29.9 20.6 9.3 455 0.21 41.2 13.7 27.5 37.5 15.9 21.6 390 0.21 25.2 25.3 −0.1 14.6 16.2 −1.6 455 0.25 21.1 30.7 −9.6 20.5 26.7 −6.2 295 0.2 17.3 9.2 8.1 26.2 12.1 14.1 400 0.25 24.1 23.8 0.3 42.5 25.7 16.8 250 0.13 36.9 29.4 7.5 42.9 25.8 17.1 290 0.23 8.6 9.0 −0.4 9.5 17.9 −8.3 390 0.1 39.2 36.2 3.0 37.5 19.6 17.9 405 0.2 30.5 21.0 9.5 48.8 18.5 30.3 255 0.2
744 R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
0.65 and left, p = 0.69). No other variable had a significant positive or negative impact on walking distance and speed (Table 4).
4. Discussion The ability to walk, albeit with calipers, is one of the important goals for rehabilitation for lower neurological level paraplegic patients. In our institute, persons with complete paraplegia and NLI T9 and below are encouraged and trained to walk upto 500 meters using orthoses. It is essential to identify factors which improve these gait parameters. The three variables chosen helped in assessing the muscle physiological properties non-invasively. In a similar study by Umezu et al., MF and ‘time to fatigue’ were compared between marathon wheelchair athletes and recreational wheelchair paraplegic athletes [10]. They reported longer ‘time to fatigue’ and significantly less decline of MF in marathon wheelchair athletes when compared to controls, suggestive of better endurance. Similarly, in this study, the above variables were used in a single set of subjects, but at different time intervals. In addition, arm circumferences were also measured to quantify the actual changes happening over the training duration. Arm circumference measured during each recording highlighted the hypertrophic changes occurring in the muscles due to exercise training. Exercise and strength training are known to cause muscle hypertrophy, but it also includes changes in the other structures such as capillaries, micro-architecture and fat distribution. It helped in supporting the fact that there were positive changes seen due to exercise training. Significant increase (p = 0.003) seen in both arm circumferences showed the impact of exercise therapy in improving upper extremity muscle mass. The authors, however, did not take into account body mass and fat distribution. The rationale behind this was the focus on changes in arm circumference, rather than actual arm circumference measurement. ‘Time to fatigue’ helped in assessing the improvement in the metabolic resilience of the muscle fibers. The ability of muscle to sustain a contraction is determined by muscle fiber type distribution. Accumulation of the metabolic byproducts in a muscle, when subjected to a prolonged contraction, decreases the conduction velocity of muscle fibers. Mortimer documented that the decrease in conduction velocity was less pronounced in muscles with lesser fast-twitch
745
fibers probably due to fewer acidic byproducts during muscular contractions [20]. A muscle with more type I fibers displays longer time to fatigue compared to a muscle with predominant type II fibers. ‘Time to fatigue’ recorded from the sustained sub-maximal isometric contraction, showed improvement with training, more in the non-dominant upper extremity, although these changes were not statistically significant. Four patients showed a pronounced improvement in ‘time to fatigue’ during the training period while only one patient had an appreciable worsening. A difference of 18.1 ± 37.9 seconds for the right and 29.1 ± 43.7 seconds for the left upper extremity was noted between the final and initial readings, suggestive of more improvement in the non-dominant upper extremity. Umezu et al. had noted a difference of 45 seconds between elite marathon and recreational wheelchair paraplegic athletes. Although the difference in this study was not the same as in Umezu’s study, it showed a positive trend. Perhaps, if the training was continued for an extended period of time, further improvement in ‘time to fatigue’ could have been achieved. MF calculated from SEMG, decreases during a sustained muscle contraction and has been attributed to metabolic fatigue and lactic acid accumulation in the muscle. EMG spectral analysis is useful to evaluate muscle endurance of upper limbs [13]. It can therefore be used as a non-invasive substitute to histochemical methods. Recordings taken at two different time intervals, showed a smaller amount of frequency reduction during the final recording when compared to initial recording, indicating better muscle adaptation following exercise therapy (right 7.8 ± 10.5 %, p = 0.04 and left 10.82 ± 1.14%, p = 0.01). In the study by Umezu, MF reduction in elite marathon wheelchair athletes was 8.9%. Results from both the studies are comparable and suggest that frequency reduction is smaller in a conditioned and trained muscle, when compared to a non-conditioned muscle. It is interesting to note that all parameters showed better response in the non-dominant extremity (left) compared to dominant one (right). On further analysis, it was noticed that all participants in the study were right hand dominant. Calmels et al. reported that patients with SCI have more symmetrical bilateral elbow flexor and extensor muscle strength and mass than nonimpaired subjects, suggesting attenuation of extremity dominance due to long term wheelchair mobility demands [21]. It is worth noting that the average duration of paraplegia was 12.5 years in their study. Readings taken after such a long duration of paraplegia revealed
746
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI Table 3 The Mean, Standard deviation, 50th percentile and the p values for all the variables Variable Arm circumference (cms)
Side Right Left
Time to fatigue (Seconds)
Right Left
Mean frequency percentage decrease
Right Left
∗ Statistically
Reading no. Initial Final Initial Final
Number (n) 12 12 12 12
Mean ± standard deviation 0.256 ± 0.029 0.277 ± 0.026 0.255 ± 0.029 0.277 ± 0.029
Initial Final Initial Final
12 12 12 12
85.5 ± 25.5 103.7 ± 40.3 81.2 ± 33.1 110.3 ± 33.3
Initial Final Initial Final
12 12 12 12
28.2 ± 9.8 20.4 ± 8.8 29.7 ± 12.2 18.8 ± 5.1
P value 0.003∗ 0.003∗ 0.17 0.05 0.04∗ 0.01∗
significant, p < 0.05.
Table 4 Correlation statistics between the measured variables of both arms and ambulation parameters (walking distance and walking speed) Variables Right Arm circumference difference Left Arm circumference difference Time to fatigue – Right Time to fatigue – Left MF percentage difference – Right MF percentage difference – Left
Walking distance Correlation coefficient −0.33 −0.41 0.65 0.69 0.12 −0.52
P value 0.29 0.17 0.02∗ 0.01∗ 0.71 0.08
Walking speed Correlation coefficient 0.04 −0.06 0.53 0.42 0.42 −0.06
P value 0.89 0.85 0.07 0.16 0.16 0.85
Nonparametric Spearman’s rho (ρ) correlation coefficient used and statistical significance set at p < 0.05. ∗ Correlation is significant at 0.05 (2-tailed).
symmetrical bilateral extremity strength. We propose that in order to achieve this, the non-dominant upper extremity will display greater improvement during the initial training period as the dominant one is already better conditioned. The findings from our study corroborate this point as better response was seen in the non-dominant upper limb in all the three variables. In terms of whether the changes in above variables actually helped the patients to improve daily activities and ambulation, we compared these variables with gait parameters, such as the walking distance and walking speed. Community ambulation after SCI requires independence in transfers, ability to stand from sitting position and ambulate reasonable distances with/without braces and assistive devices. Ambulating with the help of KAFOs and forearm crutches is physically taxing. Ambulation parameters assessed at the end of training helped in quantifying the degree of independence and gains as a result of exercise training. Correlation analysis between above variables and gait parameters showed significant positive correlation between improvement in ‘time to fatigue’ and walking distance. It is worth noting that although the change in time to fatigue following exercise therapy was not statistically significant, it was the only parameter which
had positive impact on walking distance. Individuals with marked improvement in time to fatigue were able to ambulate longer distances compared to others, as seen in Table 2. Lack of any other correlation between measured variables and ambulation parameters could be due to other factors involved during walking like body mass index, balance and hip hiking technique. Further studies are warranted taking into account these factors. This study being a pilot study, can form a basis for future research to observe changes in the above properties over a longer period of time in a large sample size. 4.1. Limitations There were a few limitations in the current study. Firstly, the average duration between spinal cord injury and initiation of rehabilitation was approximately six months. We would have preferred to recruit patients earlier in the post-injury period. However, most of the patients managed in our centre come from all parts of the country and usually have exhausted other options of rehab/treatment before considering inpatient rehabilitation. Secondly, for the arm circumference measurement, Body Mass Index was not calculated, as the
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
focus was on changes in circumference rather than the actual circumference measurement. Thirdly, smaller sample size could have resulted in less statistical power and the results could have been different had there been more number of patients evaluated.
5. Conclusion A simple non-invasive method to measure the changes in properties of skeletal muscle following exercise is discussed. This study highlights three points; firstly, focused and planned endurance training can induce changes in the physiological properties of the muscle. Salient changes in muscle properties like increase in time to fatigue, arm circumference and reduction in drop of MF, were suggestive of better muscle endurance and adaptation. Secondly, changes occurring are more in the non-dominant upper extremity compared to dominant one. Thirdly, improvement in time to fatigue has a positive impact on walking distance in a paraplegic person.
Acknowledgements This study was conducted using the Institutional Fluid Research grant. The authors would like to acknowledge Dr. Elizabeth Tharion, Department of Physiology and Dr. Rajdeep Ojha, Department of Bioengineering, Christian Medical College, Vellore.
References [1]
[2]
[3]
[4]
Nyland J, Quigley P, Huang C, Lloyd J, Harrow J, Nelson A. Preserving transfer independence among individuals with spinal cord injury. Spinal Cord. 2000 Nov; 38(11): 649–57. Dallmeijer AJ, van der Woude LH, Hollander PA, Angenot EL. Physical performance in persons with spinal cord injuries after discharge from rehabilitation. Med Sci Sports Exerc. 1999 Aug; 31(8): 1111–7. Dallmeijer AJ, van der Woude LH, Hollander AP, van As HH. Physical performance during rehabilitation in persons with spinal cord injuries. Med Sci Sports Exerc. 1999 Sep; 31(9): 1330–5. Fitts RH, Widrick JJ. Muscle mechanics: adaptations with
747
exercise-training. Exerc Sport Sci Rev. 1996; 24: 427–73. Fitts RH. Effects of regular exercise training on skeletal muscle contractile function. Am J Phys Med Rehabil. 2003 Apr; 82(4): 320–31. [6] Gollnick PD, Armstrong RB, Saltin B, Saubert CW, Sembrowich WL, Shepherd RE. Effect of training on enzyme activity and fiber composition of human skeletal muscle. J Appl Physiol. 1973 Jan; 34(1): 107–11. [7] Schantz P, Sjöberg B, Widebeck AM, Ekblom B. Skeletal muscle of trained and untrained paraplegics and tetraplegics. Acta Physiol Scand. 1997 Sep; 161(1): 31–9. [8] Behan WMH, Cossar DW, Madden HA, McKay IC. Validation of a simple, rapid, and economical technique for distinguishing type 1 and 2 fibres in fixed and frozen skeletal muscle. J Clin Pathol. 2002 May; 55(5): 375–80. [9] Kirby RL, Bonen A, Belcastro AN, Campbell CJ. Needle muscle biopsy: techniques to increase sample sizes, and complications. Arch Phys Med Rehabil. 1982 Jun; 63(6): 264–8. [10] Umezu Y, Shiba N, Tajima F, Mizushima T, Okawa H, Ogata H, et al. Muscle endurance and power spectrum of the triceps brachii in wheelchair marathon racers with paraplegia. Spinal Cord. 2003 Sep; 41(9): 511–5. [11] Bigland-Ritchie B, Donovan EF, Roussos CS. Conduction velocity and EMG power spectrum changes in fatigue of sustained maximal efforts. J Appl Physiol. 1981 Nov; 51(5): 1300–5. [12] Merletti R, Rainoldi A, Farina D. surface electromyography for noninvasive characterization of muscle. Exerc Sport Sci Rev. 2001; 29(1): 20–5. [13] Komi PV, Tesch P. EMG frequency spectrum, muscle structure, and fatigue during dynamic contractions in man. Eur J Appl Physiol Occup Physiol. 1979 Sep; 42(1): 41–50. [14] Welcome to SENIAM [Homepage on the Internet]. Enschede, the Netherlands; 2014 [cited 2014 Sept 20]. Available from: http://www.seniam.org/. [15] Levangie PK, Norkin CC. Joint structure and function: A comprehensive analysis. F.A. Davis Co.; 2005, p. 588. [16] McNair PJ, Depledge J, Brettkelly M, Stanley SN. Verbal encouragement: Effects on maximum effort voluntary muscle action. Br J Sports Med. 1996 Sep; 30(3): 243–5. [17] Tharion E. A study of fatiguing isometric contractions of the human first dorsal interosseous muscle. Indian J Physiol Pharmacol. 2006 Sep; 50(3): 319–21. [18] Rosman N, Spira E. Paraplegic use of walking braces: a survey. Arch Phys Med Rehabil. 1974 Jul; 55(7): 310–4. [19] Gopalan L, Tharion G, Prasad R, Nagarajan G, Bhattacharji S. A Follow up study to evaluate long term mobility in persons with paraplegia. Asia Pacific Disab Rehab J. 1997; 8(2): 47– 54. [20] Mortimer JT, Magnusson R, Petersén I. Conduction velocity in ischemic muscle: effect on EMG frequency spectrum. Am J Physiol. 1970 Nov; 219(5): 1324–9. [21] Calmels P, Berthouze S, Barral FG, Domenach M, Minaire P. A comparative study of the muscle strength and mass of the arm flexors and extensors in paraplegic and in non paraplegic basketball players. Paraplegia. 1992 Jul; 30(7): 509–16. [5]
Journal of Back and Musculoskeletal Rehabilitation 28 (2015) 739–747 DOI 10.3233/BMR-140577 IOS Press
Exercise and gait training in persons with paraplegia and its effect on muscle properties Rohit Prakash Bhidea,∗, Cassandra Solomonsc, Suresh Devsahayamc and George Tharionb a
Clinical Fellow, Division of Physiatry, Department of Medicine, University of Toronto, Toronto, ON, Canada Department of Physical Medicine and Rehabilitation, Christian Medical College, Vellore, Tamil Nadu, India c Department of Bioengineering, Christian Medical College, Vellore, Tamil Nadu, India b
Abstract. BACKGROUND: Upper extremity strengthening and gait training with orthoses form a major part of inpatient rehabilitation of paraplegic patients in developing countries. This helps to overcome architectural barriers and limited wheelchair accessible environment in the community. OBJECTIVES: To evaluate the changes in physiological properties of the Triceps Brachii muscle following exercise training in individuals with paraplegia. The authors also explored the correlation between muscle property changes and gait parameters using orthoses in paraplegic persons. METHODS: Twelve subjects with complete paraplegia and neurological level of injury (NLI) from T9 to L1, underwent exercise training for a mean 64.1 ± 4.1 days. Triceps brachii was chosen as the sample muscle. Variables like arm circumference, time to fatigue and mean power frequency (MF) (surface EMG parameter), were recorded at the beginning and the end of training, during a sub-maximal isometric elbow extension. Non-parametric tests were used to assess statistical significance between the two recordings. Additionally, gait parameters like walking speed and distance (with the help of orthoses) were obtained and compared with the above variables, to determine impact of upper extremity strengthening on gait improvements in such patients. RESULTS: Statistically significant changes were noted in bilateral arm circumferences (p = 0.003 bilaterally) and MF drop, expressed as percentage (right p = 0.04, left p = 0.01), indicative of better muscle resilience and adaptation. Significant positive correlation was observed between ‘time to fatigue’ and the orthoses-aided total walking distance (right ρ = 0.65, left ρ = 0.69). CONCLUSIONS: Exercise training induces noticeable changes in the muscles of upper extremities favoring better muscle adaptation. Furthermore, positive correlation between ‘time to fatigue’ and (orthotic) aided walking distance highlights the positive impact of strengthening program on gait parameters in paraplegic patients. These findings are important and relevant in developing countries with environmental barriers. Upper extremity strengthening should be included in the rehabilitation of paraplegic patients who are being trained for ambulation with orthoses. Keywords: Exercise, paraplegia, muscle fatigue, surface electromyography, triceps
1. Background Spinal cord injury (SCI) has far reaching consequences on an individual’s life. In a developing country, wheelchairs are often not the best mode ∗ Corresponding author: Rohit Prakash Bhide, Clinical Fellow, Division of Physiatry, Department of Medicine, University of Toronto, UHN – Toronto Rehabilitation Institute, 550 University Avenue, Toronto, M5G 2A2 Ontario, Canada. Tel.: +1 416 6252843; E-mail: [email protected], [email protected].
of transportation. Patients with paraplegia encounter significant architectural and attitudinal barriers compared to their western counterparts. Ambulation training in paraplegic patients, with the help of orthoses like Knee-Ankle-Foot orthoses (KAFOs) and forearm crutches, can potentially overcome this issue. In paraplegics with caudal neurological levels, majority of the rehabilitation training focuses on upper extremity strengthening and ambulation training, in order to condition the upper extremity for activities which were earlier performed by the lower limbs [1]. Dur-
c 2015 – IOS Press and the authors. All rights reserved ISSN 1053-8127/15/$35.00
740
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
ing rehabilitation, efforts are taken to meet the new demands through exercises and endurance training program. Physical fitness directly correlates with functional capability and highlights the importance of systematic rehabilitation and conditioning exercise programs. Dallmeijer et al. have reported notable improvements in physical capacity and abilities to perform daily living activities during and after rehabilitation in persons with SCI [2,3]. On a physiological level, skeletal muscle is known for its adaptability and plasticity, as commonly seen in the results of athletic training. Fitts and Gollnick have documented changes occurring in several aspects of skeletal muscle such as micro-architecture, histochemical and fiber characteristics after long term focused endurance training [4–6]. Schantz reported better muscle adaptations in upper extremities of individuals with spinal cord injury and successively higher proportions of type I (non-fatigable) fibers following endurance training [7]. The gold standard for documenting these changes is muscle biopsy [8], which, being an invasive procedure, has its own disadvantages [9]. Any damage or complication in the upper extremities of paraplegic patients, with already non-functioning lower extremities, can greatly hamper the adapted lifestyle. There is evidence regarding parameters of surface electromyography (SEMG) being used non-invasively to study manifestations of fatigue [10–12], with negligible discomfort to the individual. Spectral analysis studies to evaluate metabolic muscle fatigue have shown that muscles with type II fiber predominance have faster rates of EMG ‘Mean Power Frequency’ (MF) reduction with time [13]. This study was undertaken to document non-invasively the changes occurring in upper extremity (triceps brachii muscle) of paraplegic subjects following exercise training. The second objective was to explore the impact of these changes in muscle properties with gait parameters like walking distance and speed, using orthoses. Triceps brachii was chosen as the sample muscle, as it is involved in transfers, push-ups and gait training in persons with SCI.
2.1. Study subjects A sample size of convenience of 12 paraplegic subjects was taken satisfying the inclusion criteria: age 18–60 years, neurological level of injury (NLI) T9 and below with American Spinal Injury Association (ASIA) impairment scale (AIS) A or B (motor complete paraplegia). NLI T9 and below was chosen after checking Beevor’s sign for abdominal muscle tone and evidence of fair truncal control to aid standing balance and stability. None of the subjects had received formal rehabilitation and exercise training prior to the admission. Proposed goal of ambulation with bilateral KAFOs and forearm crutches were set for all subjects. Patients with AIS C or D, NLI above T9 and patients with upper extremity issues like shoulder injury, fractures and neuropathies were excluded. 2.2. Exercise and rehabilitation training All subjects underwent similar structured rehabilitation training which included Physical and Occupational therapy to maximize mobility and activities of daily living. It consisted of daily six hours of therapy (three 2-hour therapy sessions), five days a week. During the sessions, patients were supervised by the therapist who ensured maximum utilization of the therapy time. Upper extremity strengthening program consisted of endurance and weight training for shoulder abductors, depressors and elbow flexors/extensors, wrist extensors and hand grip strengthening. Gradual progression in terms of repetitions and number of sets performed for each exercise was monitored during the entire training. Simultaneously, gait training was initiated which involved orientation to vertical with the help of tilt table and Standing frame. Following this, patients worked on their standing balance with KAFOs in parallel bars followed by graduated gait training within the parallel bars, progressing to walker and then forearm crutches. Once a subject started walking with KAFOs and forearm crutches, they were encouraged to practice during evening hours and on weekends. 2.3. The experiment set-up
2. Methods The study was conducted in a tertiary care rehabilitation institute of a University teaching hospital following approval from the Institutional Review Board and Ethics Committee. Informed consent was obtained from all participants.
Each patient was subjected to the experimental measurements in both arms at two points – initial and final assessment. Three variables were obtained during each recording: arm circumference, time to fatigue and changes in Surface electromyography (SEMG) parameter – Mean Power Frequency (MF).
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
741
Fig. 1. The fabricated apparatus with force transducer (load cell) fixed at the end of the forearm beam. The proximal end of the load cell is fixed to the forearm beam. On application of extension force at the distal end of the load cell, corresponding changes are seen on the monitor.
Arm circumference measurements were taken from both arms at the beginning of each recording by placing the upper limb besides the trunk with elbow flexed at 90 degrees. Arm circumference was measured from a point midway between the acromion and olecranon process. These measurements did not control for body fat, which we later noted to be one of the limitation. This was followed by surface EMG recordings from each extremity. A structured frame was fabricated to stabilize the upper extremity, which consisted of two longitudinal bars hinged at 90 degrees, with a trough attached on the reference beam to match the arm contour (Fig. 1). Force transducer (load cell) was fixed at the distal end of wooden bar to measure the extension force. The forearm when placed in semi-prone position and strapped with Velcro against the force transducer, fixed the upper limb in a rigid position with elbow and shoulder joint stabilized and maintaining constant muscle length of the triceps brachii. Standard protocol was followed which began with marking of points as per SENIAM (Surface Electromyography for the Non-Invasive Assessment of Muscles) recommendations [14]. Skin preparation was done by shaving and surgical alcohol application. A pair of bipolar surface electrodes (circular silver discs, diameter 0.01 metre, 0.015 metre inter-electrode spacing) with conductive gel were placed on the lateral
head of the triceps brachii muscle along the direction of muscle fibers. A single carbon gel adhesive electrode, used as a ground, was placed off the muscle. Lateral head of Triceps brachii was specifically selected as it is superficial, has one-joint action, and most importantly, shoulder position and forearm rotation does not affect its action and contractility [15]. Custom-made ‘CMC-Daq’ data acquisition hardware and software, compatible with Windows XP, was used for data recordings and collection. CMC-Daq is a four channel data acquisition system, with 10/12 bit data acquisition. The force transducer (load cell) from the wooden frame was connected to first channel of CMC-Daq, providing visual biofeedback during elbow extension. SEMG electrodes from the lateral head of triceps brachii, were connected to second channel. All data readings were recorded at a sampling rate of 2000 Hz and later processed using CMC-Daq software on a personal computer. 2.4. The procedure During the procedure, the subjects lay supine with electrodes placed and arm fixed in the frame as described earlier. They were then asked to perform maximal elbow extension isometrically against the force transducer lasting 30 seconds, displayed graphically
742
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
Fig. 2. Changes seen on the monitor on application of extension force. The rise in the ‘force line’ is seen by the subject on the screen (Channel 1) and acts as visual biofeedback. During the same recording, the surface EMG from the lateral head of Triceps Brachii is recorded via the second channel.
on the computer screen (Fig. 2). This was done three times, at an interval of 5 minutes. Since the fabricated frame fixes the point of force measurement, torque is proportional to the measured force. Force produced during these contractions was termed as Maximal Voluntary Contraction (MVC) and averaged from three contractions. A point corresponding to the 50% MVC was determined for each subject. After 15 minutes of rest, the subjects were asked to perform a sustained fatiguing isometric voluntary contraction at 50% MVC. Continuous visual biofeedback from the computer screen helped to maintain the level of force. Efforts were taken to reduce central fatigue by constant verbal encouragement throughout the contraction [16]. Recordings from both upper extremities were taken separately. The same procedure with all the steps was repeated on completion of training. Spectral analysis was done of the collected data using averaging of periodograms and fast Fourier transform to calculate MF, with a frequency band of 20– 500 Hz (Fig. 3). Decrease in MF during the fatigu-
ing isometric contraction was expressed as percentage reduction compared to baseline, rather than absolute difference, to eliminate factors like electrode orientation, spacing and to permit comparison between the two recordings [17]. ‘Time to fatigue’ was calculated from the same recording, as the duration for which the sustained sub-maximal contraction lasted and was expressed in seconds. 2.5. Gait parameters Functional walkers were those who could walk at least 250 meters at a stretch and were able to negotiate architectural barriers [18,19]. Gait parameters like walking distance and speed were recorded, on a marked path, with the help of a stopwatch. This recording was done only at the end of rehab training as all patients were either wheelchair or bed-bound at the beginning of rehab training. Statistical analyses were done using SPSS 16.0. Non-parametric tests were used in view of the small sample size.
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
743
Fig. 3. The Mean power frequency (MF) calculated from the surface EMG spectral analysis at one point in time during the isometric contraction. Table 1 Demographic variables of the study participants Demographic variables Age distribution (mean-29.66 yrs) 15–20 20–30 30–40 40-above Level of injury T9-T10 T11-T12 L1-L2 Duration of stay in rehab (average – 62.58 days, 9 weeks) 40–50 50–60 60–70 > 70
Number of subjects (n = 12) 1 6 5 0 5 6 1
2 3 5 2
3. Results All the 12 recruited patients were in the middle age group with a median age of 28.5 years (range19–40 years). 90% of patients had NLI between T9T12. Average duration prior to inpatient admission was 6.2 months (range-0.5–14 months). The mean duration of exercise and rehab training was 64.1 ± 4.1 days (range-38–92 days) (Tables 1 and 2). 3.1. Changes in muscle properties Wilcoxon Signed ranks test was used to calculate the statistical significance of changes in each variable.
At the end of exercise training, arm circumferences of both upper limbs showed significant increase; right 0.021 ± 0.011 meters and left 0.022 ± 0.011 meters, p =0.003 bilaterally. ‘Time to fatigue’ improved during training with a mean increase of 18.1 ± 37.9 seconds for the right side (p = 0.170) and 29 ± 43.7 seconds for the left (p = 0.05). MF calculated from spectral analysis of SEMG decreased linearly over the course of sustained contraction. This drop in MF when calculated during the final reading was significantly lesser compared to that in the initial reading. A difference of 7.8 ± 10.5 %, p = 0.04 for right side and 10.82 ± 1.14 %, p = 0.01 for left side was seen, implying better metabolic resilience (Table 3). 3.2. Gait parameters Ambulation parameters measured at the end of exercise training could be directly related to the overall improvement due to exercise training. Correlation analyses for the improvement in above variables with final ambulation outcomes such as walking speed and distance was done using Spearman’s ρ (rho). Ambulation parameters included recording of walking distance and speed (using orthoses). Average walking distance was 362.5 ± 82.9 meters at a speed of 0.205 ± 0.046 meters/second. Correlation analysis yielded positive correlation between improvement in ‘time to fatigue’ in both arms following training with walking distance (right, p =
19 24 26 33 29 37 37 38 21 24 40 28
L1-AIS A T12-AIS B T10-AIS A T11-AIS A T12-AIS A T11-AIS A T10-AIS A T10-AIS A T9-AIS A T9-AIS A T11-AIS A T12-AIS A
Age Neu. Lev. (yrs) of injury
56 82 68 92 45 54 61 53 70 61 65 44
Total rehab stay (days)
Initial 0.225 0.21 0.3 0.25 0.3 0.26 0.27 0.275 0.23 0.25 0.28 0.225
Arm circumference (meters) Right Left Final Difference Initial Final Difference 0.24 0.015 0.225 0.25 0.025 0.235 0.025 0.21 0.23 0.02 0.31 0.01 0.29 0.31 0.02 0.28 0.03 0.25 0.27 0.02 0.3 0 0.305 0.31 0.005 0.27 0.01 0.27 0.27 0 0.3 0.03 0.265 0.3 0.035 0.3 0.025 0.265 0.3 0.035 0.27 0.04 0.23 0.27 0.04 0.275 0.025 0.25 0.275 0.025 0.3 0.02 0.28 0.3 0.02 0.245 0.02 0.22 0.245 0.025 Initial 65 51.5 92.5 85 79 117 76 145.5 98 62 75 80
Time to fatigue (Seconds) Right Left Final Difference Initial Final Difference 56.5 −8.5 60 40 −20 66 14.5 57.7 101.5 43.7 188 95.5 92 151 59 90 5 93.2 101 7.7 120 41 61 156 95 115 −2 106 93.5 −12.5 132 56 69 153 84 111 −34.5 150.5 111 −39.5 119 21 98 121 23 47.7 −14.2 33 77.5 44.5 134 59 44 114.5 70.5 65.5 −14.5 111 104 −7
Table 2 Details of the recordings of all the patients Mean frequency percentage decrease (%) Walking Walking Right Left distance speed Initial Final Difference Initial Final Difference (Meters) (meters/sec) 30.4 16.6 13.8 21.8 11.5 10.3 330 0.23 38.8 13.9 24.9 24.4 15.6 8.8 435 0.25 25.0 15.8 9.2 29.9 20.6 9.3 455 0.21 41.2 13.7 27.5 37.5 15.9 21.6 390 0.21 25.2 25.3 −0.1 14.6 16.2 −1.6 455 0.25 21.1 30.7 −9.6 20.5 26.7 −6.2 295 0.2 17.3 9.2 8.1 26.2 12.1 14.1 400 0.25 24.1 23.8 0.3 42.5 25.7 16.8 250 0.13 36.9 29.4 7.5 42.9 25.8 17.1 290 0.23 8.6 9.0 −0.4 9.5 17.9 −8.3 390 0.1 39.2 36.2 3.0 37.5 19.6 17.9 405 0.2 30.5 21.0 9.5 48.8 18.5 30.3 255 0.2
744 R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
0.65 and left, p = 0.69). No other variable had a significant positive or negative impact on walking distance and speed (Table 4).
4. Discussion The ability to walk, albeit with calipers, is one of the important goals for rehabilitation for lower neurological level paraplegic patients. In our institute, persons with complete paraplegia and NLI T9 and below are encouraged and trained to walk upto 500 meters using orthoses. It is essential to identify factors which improve these gait parameters. The three variables chosen helped in assessing the muscle physiological properties non-invasively. In a similar study by Umezu et al., MF and ‘time to fatigue’ were compared between marathon wheelchair athletes and recreational wheelchair paraplegic athletes [10]. They reported longer ‘time to fatigue’ and significantly less decline of MF in marathon wheelchair athletes when compared to controls, suggestive of better endurance. Similarly, in this study, the above variables were used in a single set of subjects, but at different time intervals. In addition, arm circumferences were also measured to quantify the actual changes happening over the training duration. Arm circumference measured during each recording highlighted the hypertrophic changes occurring in the muscles due to exercise training. Exercise and strength training are known to cause muscle hypertrophy, but it also includes changes in the other structures such as capillaries, micro-architecture and fat distribution. It helped in supporting the fact that there were positive changes seen due to exercise training. Significant increase (p = 0.003) seen in both arm circumferences showed the impact of exercise therapy in improving upper extremity muscle mass. The authors, however, did not take into account body mass and fat distribution. The rationale behind this was the focus on changes in arm circumference, rather than actual arm circumference measurement. ‘Time to fatigue’ helped in assessing the improvement in the metabolic resilience of the muscle fibers. The ability of muscle to sustain a contraction is determined by muscle fiber type distribution. Accumulation of the metabolic byproducts in a muscle, when subjected to a prolonged contraction, decreases the conduction velocity of muscle fibers. Mortimer documented that the decrease in conduction velocity was less pronounced in muscles with lesser fast-twitch
745
fibers probably due to fewer acidic byproducts during muscular contractions [20]. A muscle with more type I fibers displays longer time to fatigue compared to a muscle with predominant type II fibers. ‘Time to fatigue’ recorded from the sustained sub-maximal isometric contraction, showed improvement with training, more in the non-dominant upper extremity, although these changes were not statistically significant. Four patients showed a pronounced improvement in ‘time to fatigue’ during the training period while only one patient had an appreciable worsening. A difference of 18.1 ± 37.9 seconds for the right and 29.1 ± 43.7 seconds for the left upper extremity was noted between the final and initial readings, suggestive of more improvement in the non-dominant upper extremity. Umezu et al. had noted a difference of 45 seconds between elite marathon and recreational wheelchair paraplegic athletes. Although the difference in this study was not the same as in Umezu’s study, it showed a positive trend. Perhaps, if the training was continued for an extended period of time, further improvement in ‘time to fatigue’ could have been achieved. MF calculated from SEMG, decreases during a sustained muscle contraction and has been attributed to metabolic fatigue and lactic acid accumulation in the muscle. EMG spectral analysis is useful to evaluate muscle endurance of upper limbs [13]. It can therefore be used as a non-invasive substitute to histochemical methods. Recordings taken at two different time intervals, showed a smaller amount of frequency reduction during the final recording when compared to initial recording, indicating better muscle adaptation following exercise therapy (right 7.8 ± 10.5 %, p = 0.04 and left 10.82 ± 1.14%, p = 0.01). In the study by Umezu, MF reduction in elite marathon wheelchair athletes was 8.9%. Results from both the studies are comparable and suggest that frequency reduction is smaller in a conditioned and trained muscle, when compared to a non-conditioned muscle. It is interesting to note that all parameters showed better response in the non-dominant extremity (left) compared to dominant one (right). On further analysis, it was noticed that all participants in the study were right hand dominant. Calmels et al. reported that patients with SCI have more symmetrical bilateral elbow flexor and extensor muscle strength and mass than nonimpaired subjects, suggesting attenuation of extremity dominance due to long term wheelchair mobility demands [21]. It is worth noting that the average duration of paraplegia was 12.5 years in their study. Readings taken after such a long duration of paraplegia revealed
746
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI Table 3 The Mean, Standard deviation, 50th percentile and the p values for all the variables Variable Arm circumference (cms)
Side Right Left
Time to fatigue (Seconds)
Right Left
Mean frequency percentage decrease
Right Left
∗ Statistically
Reading no. Initial Final Initial Final
Number (n) 12 12 12 12
Mean ± standard deviation 0.256 ± 0.029 0.277 ± 0.026 0.255 ± 0.029 0.277 ± 0.029
Initial Final Initial Final
12 12 12 12
85.5 ± 25.5 103.7 ± 40.3 81.2 ± 33.1 110.3 ± 33.3
Initial Final Initial Final
12 12 12 12
28.2 ± 9.8 20.4 ± 8.8 29.7 ± 12.2 18.8 ± 5.1
P value 0.003∗ 0.003∗ 0.17 0.05 0.04∗ 0.01∗
significant, p < 0.05.
Table 4 Correlation statistics between the measured variables of both arms and ambulation parameters (walking distance and walking speed) Variables Right Arm circumference difference Left Arm circumference difference Time to fatigue – Right Time to fatigue – Left MF percentage difference – Right MF percentage difference – Left
Walking distance Correlation coefficient −0.33 −0.41 0.65 0.69 0.12 −0.52
P value 0.29 0.17 0.02∗ 0.01∗ 0.71 0.08
Walking speed Correlation coefficient 0.04 −0.06 0.53 0.42 0.42 −0.06
P value 0.89 0.85 0.07 0.16 0.16 0.85
Nonparametric Spearman’s rho (ρ) correlation coefficient used and statistical significance set at p < 0.05. ∗ Correlation is significant at 0.05 (2-tailed).
symmetrical bilateral extremity strength. We propose that in order to achieve this, the non-dominant upper extremity will display greater improvement during the initial training period as the dominant one is already better conditioned. The findings from our study corroborate this point as better response was seen in the non-dominant upper limb in all the three variables. In terms of whether the changes in above variables actually helped the patients to improve daily activities and ambulation, we compared these variables with gait parameters, such as the walking distance and walking speed. Community ambulation after SCI requires independence in transfers, ability to stand from sitting position and ambulate reasonable distances with/without braces and assistive devices. Ambulating with the help of KAFOs and forearm crutches is physically taxing. Ambulation parameters assessed at the end of training helped in quantifying the degree of independence and gains as a result of exercise training. Correlation analysis between above variables and gait parameters showed significant positive correlation between improvement in ‘time to fatigue’ and walking distance. It is worth noting that although the change in time to fatigue following exercise therapy was not statistically significant, it was the only parameter which
had positive impact on walking distance. Individuals with marked improvement in time to fatigue were able to ambulate longer distances compared to others, as seen in Table 2. Lack of any other correlation between measured variables and ambulation parameters could be due to other factors involved during walking like body mass index, balance and hip hiking technique. Further studies are warranted taking into account these factors. This study being a pilot study, can form a basis for future research to observe changes in the above properties over a longer period of time in a large sample size. 4.1. Limitations There were a few limitations in the current study. Firstly, the average duration between spinal cord injury and initiation of rehabilitation was approximately six months. We would have preferred to recruit patients earlier in the post-injury period. However, most of the patients managed in our centre come from all parts of the country and usually have exhausted other options of rehab/treatment before considering inpatient rehabilitation. Secondly, for the arm circumference measurement, Body Mass Index was not calculated, as the
R.P. Bhide et al. / Effects of exercise and gait training on muscle properties in SCI
focus was on changes in circumference rather than the actual circumference measurement. Thirdly, smaller sample size could have resulted in less statistical power and the results could have been different had there been more number of patients evaluated.
5. Conclusion A simple non-invasive method to measure the changes in properties of skeletal muscle following exercise is discussed. This study highlights three points; firstly, focused and planned endurance training can induce changes in the physiological properties of the muscle. Salient changes in muscle properties like increase in time to fatigue, arm circumference and reduction in drop of MF, were suggestive of better muscle endurance and adaptation. Secondly, changes occurring are more in the non-dominant upper extremity compared to dominant one. Thirdly, improvement in time to fatigue has a positive impact on walking distance in a paraplegic person.
Acknowledgements This study was conducted using the Institutional Fluid Research grant. The authors would like to acknowledge Dr. Elizabeth Tharion, Department of Physiology and Dr. Rajdeep Ojha, Department of Bioengineering, Christian Medical College, Vellore.
References [1]
[2]
[3]
[4]
Nyland J, Quigley P, Huang C, Lloyd J, Harrow J, Nelson A. Preserving transfer independence among individuals with spinal cord injury. Spinal Cord. 2000 Nov; 38(11): 649–57. Dallmeijer AJ, van der Woude LH, Hollander PA, Angenot EL. Physical performance in persons with spinal cord injuries after discharge from rehabilitation. Med Sci Sports Exerc. 1999 Aug; 31(8): 1111–7. Dallmeijer AJ, van der Woude LH, Hollander AP, van As HH. Physical performance during rehabilitation in persons with spinal cord injuries. Med Sci Sports Exerc. 1999 Sep; 31(9): 1330–5. Fitts RH, Widrick JJ. Muscle mechanics: adaptations with
747
exercise-training. Exerc Sport Sci Rev. 1996; 24: 427–73. Fitts RH. Effects of regular exercise training on skeletal muscle contractile function. Am J Phys Med Rehabil. 2003 Apr; 82(4): 320–31. [6] Gollnick PD, Armstrong RB, Saltin B, Saubert CW, Sembrowich WL, Shepherd RE. Effect of training on enzyme activity and fiber composition of human skeletal muscle. J Appl Physiol. 1973 Jan; 34(1): 107–11. [7] Schantz P, Sjöberg B, Widebeck AM, Ekblom B. Skeletal muscle of trained and untrained paraplegics and tetraplegics. Acta Physiol Scand. 1997 Sep; 161(1): 31–9. [8] Behan WMH, Cossar DW, Madden HA, McKay IC. Validation of a simple, rapid, and economical technique for distinguishing type 1 and 2 fibres in fixed and frozen skeletal muscle. J Clin Pathol. 2002 May; 55(5): 375–80. [9] Kirby RL, Bonen A, Belcastro AN, Campbell CJ. Needle muscle biopsy: techniques to increase sample sizes, and complications. Arch Phys Med Rehabil. 1982 Jun; 63(6): 264–8. [10] Umezu Y, Shiba N, Tajima F, Mizushima T, Okawa H, Ogata H, et al. Muscle endurance and power spectrum of the triceps brachii in wheelchair marathon racers with paraplegia. Spinal Cord. 2003 Sep; 41(9): 511–5. [11] Bigland-Ritchie B, Donovan EF, Roussos CS. Conduction velocity and EMG power spectrum changes in fatigue of sustained maximal efforts. J Appl Physiol. 1981 Nov; 51(5): 1300–5. [12] Merletti R, Rainoldi A, Farina D. surface electromyography for noninvasive characterization of muscle. Exerc Sport Sci Rev. 2001; 29(1): 20–5. [13] Komi PV, Tesch P. EMG frequency spectrum, muscle structure, and fatigue during dynamic contractions in man. Eur J Appl Physiol Occup Physiol. 1979 Sep; 42(1): 41–50. [14] Welcome to SENIAM [Homepage on the Internet]. Enschede, the Netherlands; 2014 [cited 2014 Sept 20]. Available from: http://www.seniam.org/. [15] Levangie PK, Norkin CC. Joint structure and function: A comprehensive analysis. F.A. Davis Co.; 2005, p. 588. [16] McNair PJ, Depledge J, Brettkelly M, Stanley SN. Verbal encouragement: Effects on maximum effort voluntary muscle action. Br J Sports Med. 1996 Sep; 30(3): 243–5. [17] Tharion E. A study of fatiguing isometric contractions of the human first dorsal interosseous muscle. Indian J Physiol Pharmacol. 2006 Sep; 50(3): 319–21. [18] Rosman N, Spira E. Paraplegic use of walking braces: a survey. Arch Phys Med Rehabil. 1974 Jul; 55(7): 310–4. [19] Gopalan L, Tharion G, Prasad R, Nagarajan G, Bhattacharji S. A Follow up study to evaluate long term mobility in persons with paraplegia. Asia Pacific Disab Rehab J. 1997; 8(2): 47– 54. [20] Mortimer JT, Magnusson R, Petersén I. Conduction velocity in ischemic muscle: effect on EMG frequency spectrum. Am J Physiol. 1970 Nov; 219(5): 1324–9. [21] Calmels P, Berthouze S, Barral FG, Domenach M, Minaire P. A comparative study of the muscle strength and mass of the arm flexors and extensors in paraplegic and in non paraplegic basketball players. Paraplegia. 1992 Jul; 30(7): 509–16. [5]
