Electrical stimulation: Can it increase muscle strength and reverse osteopenia in spinal cord injured individuals?

  • Marc Bélanger
    Affiliations
    Département de Kinanthropologie, Université du Québec à Montréal (Bélanger), and the Institut de réadaptation de Montréal (Leduc), Montreal, Quebec; and the Division of Neuroscience (Stein, Gordon) and the Rick Hansen Centre (Wheeler), University of Alberta, Edmonton, Alberta
    Search for articles by this author
  • Richard B. Stein
    Affiliations
    Département de Kinanthropologie, Université du Québec à Montréal (Bélanger), and the Institut de réadaptation de Montréal (Leduc), Montreal, Quebec; and the Division of Neuroscience (Stein, Gordon) and the Rick Hansen Centre (Wheeler), University of Alberta, Edmonton, Alberta
    Search for articles by this author
  • Garry D. Wheeler
    Affiliations
    Département de Kinanthropologie, Université du Québec à Montréal (Bélanger), and the Institut de réadaptation de Montréal (Leduc), Montreal, Quebec; and the Division of Neuroscience (Stein, Gordon) and the Rick Hansen Centre (Wheeler), University of Alberta, Edmonton, Alberta
    Search for articles by this author
  • Tessa Gordon
    Affiliations
    Département de Kinanthropologie, Université du Québec à Montréal (Bélanger), and the Institut de réadaptation de Montréal (Leduc), Montreal, Quebec; and the Division of Neuroscience (Stein, Gordon) and the Rick Hansen Centre (Wheeler), University of Alberta, Edmonton, Alberta
    Search for articles by this author
  • Bernard Leduc
    Affiliations
    Département de Kinanthropologie, Université du Québec à Montréal (Bélanger), and the Institut de réadaptation de Montréal (Leduc), Montreal, Quebec; and the Division of Neuroscience (Stein, Gordon) and the Rick Hansen Centre (Wheeler), University of Alberta, Edmonton, Alberta
    Search for articles by this author
  • Author Footnotes
    NO LABEL a. Hologic QDR-1000; Hologic, 590 Lincoln St, Waltham, MA 02451.
    NO LABEL b. Lunar DPX; Lunar Corp, 313 W Beltline Hwy, Madison, WI 53713.
    NO LABEL c. Cybex, Division of Lumex Inc, Ronkonkoma, NY 11799-0903.
    NO LABEL d. Chattanooga Corp, 101 Memorial Dr, Chattanooga, TN 37405.
    NO LABEL e. EMPI Inc, 599 Cardigan Rd, St. Paul, MN 55126-4099.
    NO LABEL f. Axon Instrument, 1101 Chess Dr, Foster City, CA 94404-1102.
    NO LABEL g. Quadstim, Biomotion, 1503, 10010-119 St, Edmonton, Alberta, Canada T5K 1Y8.
    NO LABEL h. Hydragym, Edmonton, Alberta, Canada.

      Abstract

      Bélanger M, Stein RB, Wheeler GD, Gordon T, Leduc B. Electrical stimulation: can it increase muscle strength and reverse osteopenia in spinal cord injured individuals? Arch Phys Med Rehabil 2000;81:1090-8. Objective: To study the extent to which atrophy of muscle and progressive weakening of the long bones after spinal cord injury (SCI) can be reversed by functional electrical stimulation (FES) and resistance training. Design: A within-subject, contralateral limb, and matching design. Setting: Research laboratories in university settings. Participants: Fourteen patients with SCI (C5 to T5) and 14 control subjects volunteered for this study. Interventions: The left quadriceps were stimulated to contract against an isokinetic load (resisted) while the right quadriceps contracted against gravity (unresisted) for 1 hour a day, 5 days a week, for 24 weeks. Main Outcome Measures: Bone mineral density (BMD) of the distal femur, proximal tibia, and mid-tibia obtained by dual energy x-ray absorptiometry, and torque (strength). Results: Initially, the BMD of SCI subjects was lower than that of controls. After training, the distal femur and proximal tibia had recovered nearly 30% of the bone lost, compared with the controls. There was no difference in the mid-tibia or between the sides at any level. There was a large strength gain, with the rate of increase being substantially greater on the resisted side. Conclusion: Osteopenia of the distal femur and proximal tibia and the loss of strength of the quadriceps can be partly reversed by regular FES-assisted training. © 2000 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation

      Keywords

      SPINAL CORD INJURY (SCI) commonly results in the loss of voluntary control of the limbs. Even when partial motor control remains, a frequent outcome is relative inactivity and disuse of the limbs. Gradually, the leg muscles atrophy,
      • Lieber RL
      Skeletal muscle adaptability. II. Muscle properties following spinal-cord injury.
      • Lieber RL
      • Friden JO
      • Hargens AR
      • Feringa ER
      Long-term effects of spinal cord transection on fast and slow rat skeletal muscle. II. Morphometric properties.
      • Lieber RL
      • Johansson CB
      • Vahlsing HL
      • Hargens AR
      • Feringa ER
      Long-term effects of spinal cord transection on fast and slow rat skeletal muscle. I. Contractile properties.
      • Gordon T
      Fatigue in adapted systems. Overuse and underuse paradigms.
      • Gordon T
      • Mao J
      Muscle atrophy and procedures for training after spinal cord injury.
      • Needham-Shropshire BM
      • Klose KJ
      • Tucker ME
      • Thomas CK
      Manual muscle test score and force comparisons after cervical spinal cord injury.
      • Gefen JY
      • Gelmann AS
      • Herbison GJ
      • Cohen ME
      • Schmidt RR
      Use of shoulder flexors to achieve isometric elbow extension in C6 tetraplegic patients during weight shift.
      • Cameron T
      • Calancie B
      Mechanical and fatigue properties of wrist flexor muscles during repetitive contractions after cervical spinal cord injury.
      and the long bones weaken, as a result of a rapid and severe loss of bone mineral density (BMD) in the paralyzed limbs (osteopenia).
      • Ragnarsson KT
      • Sell GH
      Lower extremity fractures after spinal cord injury: a retrospective study.
      • Rodgers MM
      • Glaser RM
      • Figoni SF
      • Hooker SP
      • Ezenwa BN
      • Collins SR
      • et al.
      Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training.
      • Garland DE
      • Stewart CA
      • Adkins RH
      • Hu SS
      • Rosen C
      • Liotta FJ
      • et al.
      Osteoporosis after spinal cord injury.
      • Keating JF
      • Kerr M
      • Delargy M
      Minimal trauma causing fractures in patients with spinal cord injury.
      • Saltzstein RJ
      • Hardin S
      • Hastings J
      Osteoporosis in spinal cord injury: using an index of mobility and its relationship to bone density.
      • Wilmet E
      • Ismail AA
      • Heilporn A
      • Welraeds D
      • Bergmann P
      Longitudinal study of the bone mineral content and of soft tissue composition after spinal cord section.
      • Chow YW
      • Inman C
      • Pollintine P
      • Sharp CA
      • Haddaway MJ
      • el-Masry W
      • et al.
      Ultrasound bone densitometry and dual energy X-ray absorptiometry in patients with spinal cord injury: a cross-sectional study.
      • Szollar SM
      • Martin EM
      • Parthemore JG
      • Sartoris DJ
      • Deftos LJ
      Demineralization in tetraplegic and paralegic man over time.
      Electrical stimulation can reverse the muscle weakness, although the extent of the reversal differs markedly in different studies. Most studies using electrical stimulation of muscle show little or no change in bone density.
      • Rodgers MM
      • Glaser RM
      • Figoni SF
      • Hooker SP
      • Ezenwa BN
      • Collins SR
      • et al.
      Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training.
      • Leeds EM
      • Klose KJ
      • Ganz W
      • Serafini A
      • Green BA
      Bone mineral density after bicycle ergometry training.
      • Hangartner TN
      • Rodgers MM
      • Glaser RM
      • Barre PS
      Tibial bone density loss in spinal cord injured patients: effects of FES exercise.
      • BeDell KK
      • Scremin AME
      • Perell KL
      • Kunkel CF
      Effects of functional electrical stimulation-induced lower extremity cycling on bone density of spinal cord–injured patients.
      • Mohr T
      • Podenphant J
      • Biering-Sorensen F
      • Galbo H
      • Thamsborg G
      • Kjaer M
      Increased bone mineral density after prolonged electrically induced cycle training of paralyzed limbs in spinal cord injured man.
      Because of bone weakness, persons with SCI have an increased risk for fractures as a result of mild trauma, such as when transferring to or from a wheelchair.
      • Ragnarsson KT
      • Sell GH
      Lower extremity fractures after spinal cord injury: a retrospective study.
      • Keating JF
      • Kerr M
      • Delargy M
      Minimal trauma causing fractures in patients with spinal cord injury.
      • Pilonchery G
      • Minaire P
      • Milan JJ
      • Revol A
      Urinary elimination of glycosaminoglycans during the immobilization osteoporosis of spinal cord injury patients.
      The incidence of long-bone fractures has been reported to be between 1% and 6%, and it is likely underestimated because numerous cases are either untreated or unreported at SCI centers.
      • Ragnarsson KT
      • Sell GH
      Lower extremity fractures after spinal cord injury: a retrospective study.
      • Nottage WM
      A review of long-bone fractures in patients with spinal cord injuries.
      Similarly, Nottage
      • Nottage WM
      A review of long-bone fractures in patients with spinal cord injuries.
      reported an incidence rate of 6.7%, with complication rates as high as 20% to 40% with open or closed treatment of extremity fractures. Strengthening muscle with electrical stimulation can further increase the risk, unless the bones are also strengthened. While it has been shown that mechanical stress is important in bone formation and remodeling,
      • Bohnor JA
      • Dilworth BB
      • Sullivan KM
      Exercise and osteoporosis: a critique of the literature.
      • Snow-Harter C
      • Marcus R
      Exercise, bone mineral density, and osteoporosis.
      no study has documented the level of physical training needed for a positive effect on BMD.
      • Schwarz P
      • Bulow JB
      • Kjaer M
      Bone metabolism and physical training] [Danish.
      Pettersson and colleagues
      • Pettersson U
      • Nordstrom P
      • Lorentzon R
      A comparison of bone mineral density and muscle strength in young male adults with different exercise level.
      reported that a group with a high level of physical activity (≈10 hours a week) showed higher BMD for total body and numerous sites compared with a reference group with a low level of physical activity (3 or fewer hours a week).
      Ayalon and associates
      • Ayalon J
      • Simkin A
      • Leichter I
      • Raifmann S
      Dynamic bone loading exercises for postmenopausal women: effect on the density of the distal radius.
      have also reported that in postmenopausal women exercise for 50 minutes three times per week for 5 months resulted in a 3.8% gain of bone density of the distal radius. Interestingly, de Bruin and colleagues
      • de Bruin ED
      • Frey-Rindova P
      • Herzog RE
      • Dietz V
      • Dambacher MA
      • Stussi E
      Changes of tibia bone properties after spinal cord injury: effects of early intervention.
      have reported no or insignificant loss of trabecular bone in subjects with SCI during the first 25 weeks after injury simply with weight bearing by standing and treadmill walking. It is not known whether this could be maintained over longer periods of time and, more important, whether atrophic changes that have occurred can be reversed substantially by electrical stimulation.
      The purpose of our investigation was twofold: (1) to determine if functional electrical stimulation (FES) training, which loads the muscles and bones, can slow or reverse osteopenia in patients with spinal cord lesion; and (2) to study whether the amount of loading (through progressive increases of isokinetic loading) has any effect on the strengthening of stimulated muscle. The reasons for wanting to increase the muscle strength were also twofold: (1) to increase the stress and loading on the bone (Doyle and colleagues
      • Doyle F
      • Brown J
      • Lachance C
      Relation between bone mass and muscle weight.
      have reported a strong relation between muscle weight and BMD); and (2) to increase the strength of the quadriceps muscles for FES-assisted standing and walking.

      Methods

       Subjects

      Subjects were recruited from regions near the research institutions in Edmonton and Montréal. All subjects with SCI were screened to exclude those who: (1) were on medications or had a medical condition that could interfere with bone formation, (2) had lesions of the quadriceps motoneurons that rendered stimulation ineffective, (3) could not tolerate the stimulation (ie, because of pain), or (4) had other medical problems (eg, previous cardiac problems, newly formed decubitus ulcers) that would put their health at risk by taking part in the study. Subjects were informed of the risks and signed a consent form before starting training, as had been approved by human ethics committees at the Universities of Québec and of Alberta.
      After the initial screening, the BMD of each subject was evaluated using dual energy x-ray absorptiometry (DEXA)
      a. Hologic QDR-1000; Hologic, 590 Lincoln St, Waltham, MA 02451.
      b. Lunar DPX; Lunar Corp, 313 W Beltline Hwy, Madison, WI 53713.
      Subjects were excluded if a fracture of either the femur or tibia was detected in this evaluation. The DEXA technique was used because of its availability, reproducibility, and good overall accuracy (5% to 8%) and precision (<2%) for a number of sites.
      • Hodgson SF
      • Johnston CC
      • Avioli LV
      • Heath III, HH
      • Khosla S
      • Kleerekoper M
      • et al.
      AACE Clinical practice guidelines for the prevention and treatment of postmenopausal osteoporosis.
      The accuracy and precision of the DEXA equipment that was used was well within these values. Characteristics of the SCI subjects who participated in this study are presented in table 1.
      Tabled 1Table 1: Description of subjects
      SubjectGenderAge (yrs)Mass (kg)Height (cm)Level of Lesion*Type of Lesion*Cause of LesionYears Postlesion
      1M32.973.3172.7C5CMVA7.5
      2M40.966.7182.9C6IMVA23
      3F28.750.7160T6CMVA4
      4M41.684120C6CMVA2.3
      5M39.765.3185.4T2CMVA14
      6F34.645.4162.6C6CGun shot17.8
      7M28.668.5168C6CMVA12.4
      8M35.469.3172.7C5IFall12.5
      9M36.473.3185.4C6CMVA18
      10F24.348.5167.6C6CMVA6.7
      11M25.757.8177.8C6CMVA6.5
      12M23.360177.8T3CMVA6.2
      13M28.082180C7CMVA1.2
      14M32.985179T5CMVA1.8
      SCI (mean ± SD)32.4 ± 5.966.4 ± 12.3170.9 ± 16.19.6 ± 6.6
      Control (mean ± SD)33.5 ± 6.375.4 ± 14.5174.5 ± 10.8
      * Level and completeness (C = complete, I = incomplete) of the lesion was based on the American Spinal Cord Injury Association (ASIA) scale. On the Frankel scale, all spinal cord injury (SCI) subjects were between A and C.
      Abbreviations: MVA, motor vehicle accident; SD, standard deviation.
      Fourteen age- and sex-matched control subjects with no known neurologic lesions were also tested to compare bone density.

       Initial evaluations

      Before starting the training program and on a weekly basis thereafter, the thigh circumference and quadriceps muscle strength and endurance for each limb were evaluated in the laboratory. The circumference was measured 15cm proximal to the superior patellar margin to test for gross changes in the bulk of the thighs. A skinfold measurement was also obtained from the 15-cm mark for the last 3 subjects (subjects 1, 8, and 11).
      Evoked muscle torques were measured over a range of 90° (ie, 90° of flexion to full extension) using a Cybex isodynamometer
      c. Cybex, Division of Lumex Inc, Ronkonkoma, NY 11799-0903.
      set at 30°/sec. The Cybex manual recommends this setting for knee torque measurements, but more important, the risk of fracture is less than with isometric testing. Because this isodynamometer did not automatically correct for gravity, this correction was performed offline on all the Cybex data. This was done by subtracting the torque generated by the weight of the limb and lever arm of the apparatus for the angles of knee extension where measurements were obtained.
      • Winter DA
      • Wells RP
      • Orr GW
      Errors in the use of isokinetic dynamometers.
      Strength and endurance were tested with a 40-Hz train lasting 2 seconds, repeated every 5 seconds for 4 minutes.
      • Stein RB
      • Gordon T
      • Jefferson J
      • Sharfenberger A
      • Yang JF
      • Totosy de Zepetnek J
      • et al.
      Optimal stimulation of paralyzed muscle after human spinal cord injury.
      The stimulus consisted of rectangular pulses of 0.5-msec duration and sufficient amplitude to produce a maximal twitch response. For the most part, the testing was done using the same multiweek
      d. Chattanooga Corp, 101 Memorial Dr, Chattanooga, TN 37405.
      e. EMPI Inc, 599 Cardigan Rd, St. Paul, MN 55126-4099.
      and the same placement (see below) as for the training sessions. In several cases, gel- and saline-coated electrodes were used during testing with no appreciable differences in the results. The signal from the Cybex was digitized at 166Hz and stored on computer using Axotape 2.
      f. Axon Instrument, 1101 Chess Dr, Foster City, CA 94404-1102.

       Training and compliance

      Training was conducted in the laboratory 5 days per week for 24 weeks, and consisted of stimulating both quadriceps muscles with the same parameters. The stimulation, 300-μsec rectangular pulses delivered at 25Hz with a 5-sec on/5-sec off duty cycle, was applied with an isolated stimulator
      g. Quadstim, Biomotion, 1503, 10010-119 St, Edmonton, Alberta, Canada T5K 1Y8.
      through multiweek electrodes. The electrodes were placed at 5cm and 15cm proximal to the superior patellar margin. These sites were chosen because they were easily accessible and repeatable, but more important, they were relatively close to the motor point, thereby requiring less stimulation current to produce a smooth contraction. The stimulation amplitude was adjusted (0 to 150mA) to produce a contraction that could be maintained for the duration of the stimulus. The stimulus intensity was increased if the leg began to drop more than 15° during the stimulation.
      A within-subject, contralateral limb design was used to evaluate the effects of resistance during training. On the right side, the stimulated quadriceps muscles extended against no resistance other than gravity (unresisted), while on the left side the limb extended against an added resistance (resisted). An isokinetic leg extension/flexion device (either Hydragym
      h. Hydragym, Edmonton, Alberta, Canada.
      or Cybex
      c. Cybex, Division of Lumex Inc, Ronkonkoma, NY 11799-0903.
      ) was used to apply this load. Subjects trained for 1 hour per day or until a fatigue criterion (less than 30° of extension from 90° in the unloaded limb) was met. The resistance on the loaded limb was increased when full extension could be produced for 15 minutes or more at a set load. Resistance was added either by changing the damping on the Hydragym apparatus (eg, going from level 1 to level 2 on a 6-level scale) or by decreasing the set velocity on the Cybex by 5°/sec intervals (eg, going from 30°/sec to 25°/sec).
      The training program was set up to try to initially increase the strength so that the subject could stand when assisted with FES. Based on studies by Kralj and Bajd,
      • Kralj AR
      • Bajd T
      Functional electrical stimulation: Standing and walking after spinal cord injury.
      standing requires that the knee extensors be able to generate a torque of more than 50Nm. More important, the training with low resistance for 24 weeks was designed to increase endurance. This type of training should allow SCI subjects to stand or walk with FES for long periods.
      Most of the subjects who embarked on the training program continued until the end of the 24-week period. The compliance, measured as the attendance of subjects for the sessions, was remarkably high, ranging from 65% to 99%, with a mean and standard deviation (SD) of 93.4% ± 5.6%. The average is equivalent to missing only 8 of 120 sessions.

       Data and statistical analysis

      A final BMD evaluation was performed on completion of the training program and compared with the initial evaluation. Areas from 1.44cm2 to 12.6cm2 were used to measure the BMD for the distal femur, proximal tibia, and mid-tibia. Analysis of variance (ANOVA) was used to determine the differences between the control and SCI subjects for the three areas of interest before and after FES training. Tukey tests were used to determine significant changes.
      To study the strength and endurance, three peak torque values were averaged for every 30-second interval over the 4-minute fatigue test. The peak torque was measured from a stable portion of the signal, between 100msec and 300msec (ie, the initial overshoot or superimposed spasms were omitted). Because the test lasted 4 minutes, the last average occurred after the 3.5-minute mark. The initial (maximal) values were used to determine strength gains. The final values were normalized with respect to the initial values to indicate the changes in the fatigability (index of endurance = 100% × torque values at 3.5min/torque values at 0min) of the muscle as training progressed. In some cases, it was possible to generate a muscle twitch in the quadriceps muscle right from the onset of training using a 1-msec rectangular pulse and the same electrode placement as previously outlined. It was then possible to measure the contraction time from the recorded torque curve (ie, from the onset of torque change to the peak torque value), thereby yielding an indication of the changes in muscle characteristics.
      Regressions were used to determine rates of change in both torque and endurance using either linear or nonlinear least-mean squares algorithms. A repeated-measures ANOVA was used to compare initial and final strengths for both limbs.

      Results

       Bone density

      The mean (±standard error) for the bone density of the control and SCI subjects (before training) are illustrated in figure 1.
      Figure thumbnail gr1
      Fig. 1Absolute bone mineral density (BMD) for control (■) and for the left (□) and right (▨) sides of subjects with SCI before training. The three regions are statistically different from each other, both for the control and SCI subjects (p < .05). The lower bone densities in the SCI subjects are also significantly different from control subjects in all three regions (indicated by the asterisk above the bars). The BMD expressed as a percentage of control values are indicated at the bottom of the bars.
      The pretraining BMD values of the SCI subjects for all three regions were lower than those of age- and sex-matched control subjects. BMD loss was approximately the same at all three sites, with the declines ranging from 25.8% to 44.4%. The same order in density was also seen between the regions in both SCI and control individuals (mid-tibia > distal femur > proximal tibia, p < .05). There was no statistical difference between the left (resisted) and right (unresisted) sides in the SCI subjects (p > .05).
      Figure 2 shows the BMD for the three regions for each subject with respect to the time after the spinal cord lesion.
      Figure thumbnail gr2
      Fig. 2Bone mineral density (BMD) of the (A) distal femur, (B) proximal tibial, and (C) mid-tibia for controls (symbols having means ± standard error at years n = 0) and for subjects with SCI as a function of time after the SCI. The other points represent the pooled data for the right and left side for each of the 12 SCI subjects. (D) The BMD for the 3 regions are expressed as a percentage of the initial values obtained from the regression equations for the SCI subjects.
      The data were fitted with the exponential equation y = be−ax, yielding y = .71 × e−.06x (correlation coefficient [R] = .68) for distal femur (fig 2A), y = .64 × e−.08x (R = .70) for the proximal tibia (fig 2B), and y = 1.37 × e−.03x (R = .75) for mid-tibia (fig 2C). The time constants for decay (τ = 1/a) were relatively similar for the 3 regions (between 13 and 33yrs), with the only significant difference being between proximal tibia and mid-tibia (p < .05); the relative changes for all 3 regions are plotted in figure 2D. Because of the long time constant, bone loss in these regions is relatively small, less than 10% after 1 year. After 10 years, the bone density had decayed by about half for the distal femur (43.2%) and proximal tibia (54.6%), but only about a quarter for mid-tibia (24.1%). The data were also fitted with the exponential equation y = c + be−ax, but either c was not significantly different from 0, or the fit improved by less than 1%. Thus, bone density will decay to almost zero with sufficient time.
      Training increased BMD significantly, but the type of training had no effect, ie, there was no difference between the resisted (left) and unresisted (right) sides. Because there was no side difference before training and the type of training had no effect (no left-right difference), the data were collapsed to simply examine regional changes. There was a significant increase in the BMD of the distal femur (.082g/cm2) and proximal tibia (.052g/cm2), but not in the mid-tibia (.001 g/cm2) (fig 3A).
      Figure thumbnail gr3
      Fig. 3Change in bone mineral density (BMD) (mean ± standard error) as a result of (A) training in absolute units (g/cm2) and (B) as a percentage of the values in control subjects. The asterisks indicate significant differences at p < .05.
      The difference between distal femur and proximal tibia was not significant, but the changes in BMD for both were significantly different from that of the mid-tibia (p < .05). When expressed as a percentage of the control values, these changes amount to 11.1%, 9.7%, and 0.0% for the distal femur, proximal tibia, and mid-tibia, respectively (fig 3B). More important, these values represent a recovery of the BMD lost after SCI of 28.7% for distal femur and 28.0% for proximal tibia. In contrast, the mid-tibia showed no recovery.
      Interestingly, there was no correlation between the initial BMD and the BMD change with training (fig 4A).
      Figure thumbnail gr4
      Fig. 4The change in bone density with functional electrical stimulation training is plotted (A) as a function of the value before training, and (B) as a function of the time after SCI when training was begun. The points are the bone mineral density for the distal femur and proximal tibia of both limbs. The three lines in (B) are the regression line (bold) and the 95% confidence intervals.
      The BMD change was weakly correlated with the years since SCI (fig 4B). The linear regression line (y = .01-.006x; R = .30) is also shown in figure 4B, with the 95% confidence intervals. The confidence intervals indicate that no significant gain was seen when training began more than 13.5 years after SCI. Unfortunately, these subjects tended to have the lowest bone densities (fig 2), and thus the greatest need for improvement.

       Muscle force

      Figure 5 illustrates the left knee torques generated by subject 3 at the beginning and after 24 weeks of FES.
      Figure thumbnail gr5
      Fig. 5Curves of the torques at the beginning (solid line) and end (dashed and dotted lines) of the 24-week training period. The downward-pointing arrow shows the onset of a spasm in one of the first contractions of the last recording session. The double-ended arrow at the bottom of the graph indicates the region where the torque measurements were taken (ie, avoiding the initial overshoot and the spasm). The small vertical lines represent the stimuli (ie, 40Hz for 2sec). It should be noted that a 0-Nm torque is recorded when the muscle contraction is not strong enough to move the leg at the set speed of the isokinetic dynamometer (eg, the torque trace at the beginning of training is null beyond 500msec).
      The torque generated by FES increased by 40Nm in this subject by the end of training. The maximal torque could sometimes be even larger (by about 25Nm), if a spasm was triggered on top of the contraction (see the difference between the dashed and dotted lines after the time indicated by the single-ended arrow, fig 5). In this example, the spasm could easily be discerned because there was a delay in its onset. However, in other cases, the increase in torque was much smoother, making it harder to detect. In addition to the large increase in amplitude, the torque could also be maintained for the duration of the tetanic stimulation (ie, 2sec). At the onset of training, there was extreme weakness and the torque decayed rapidly despite continued stimulation (ie, the contraction was not strong enough for the limb to keep pushing against the dynamometer).
      The increase in torque with training for all subjects is shown in figure 6.
      Figure thumbnail gr6
      Fig. 6Increase in strength in the resisted (solid lines) and unresisted limbs (dashed lines) as a function of the duration of training. The thick lines represent the regression lines, the thin lines give the 95% confidence intervals.
      For clarity, only the regression lines and the 95% confidence intervals are plotted. Clearly, FES training increases strength dramatically, even after the extra torque produced by the occasional spasms (fig 5) was excluded. The data were fitted with parabolas, which had values of y = 98.1 + 8.1x − .09x2 (R = .57, resisted) and y = 95.6 + 4.5x − .06x2 (R = .54, unresisted). Thus, the initial strength gain was substantially faster on the legs that worked against resistance (8.1% per week) compared with the unresisted side (4.5% per week). The rate of increase in strength did slow down with time and appeared to be approaching asymptotic values by 24 weeks. These values represented gains of nearly 75% (unresisted) and 150% (resisted).
      The torque gain as a function of the initial torque values is presented in figure 7.
      Figure thumbnail gr7
      Fig. 7The change in torque as a result of training is plotted against the torque at the start of training for the resisted (solid line) and unresisted (dashed line) sides (A) and as a percentage of the initial torque (B). The lines in (A) represent the mean change. The regression equation in (B) is described in the text.
      There was no significant trend for either the resisted or unresisted side, which indicates that the absolute increase is independent of the starting torque value (fig 7A). However, the relative increase will then be much greater for subjects with low initial strength (fig 7B). If the change in torque Δy = c, a constant, then Δy/y = c/y. The curves fitted to this equation have values of c equal to 4809 (R = .78, resisted) and 2995 (R = .90, unresisted).
      An index of fatigability was calculated for each week of training, but the fatigability did not change significantly for either the resisted or the unresisted legs. Thus, the endurance did not improve. This was corroborated by measuring the contraction time of the muscle twitch. Again, no significant change was observed over the 24-week training period.
      Some individuals showed greater muscle definition and an apparent increase in muscle size, but the limb girth of the whole population did not change significantly over the training period on either the resisted or the unresisted sides. Finally, the skinfold measurements from 3 of the subjects were also inconclusive.

      Discussion

      This study demonstrates that 1 hour of daily FES training can substantially reverse the loss of bone density after SCI. Likewise, FES produced a dramatic increase in muscle strength that depends on the type of training applied (resisted vs unresisted). These changes occurred without a significant change in fatigability or limb size, as measured by the thigh girth.

       Bone density

      After SCI, there is a large and continuing loss of BMD
      • Rodgers MM
      • Glaser RM
      • Figoni SF
      • Hooker SP
      • Ezenwa BN
      • Collins SR
      • et al.
      Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training.
      (fig 2). However, our data show that nearly 30% of the lost BMD can be recovered in the distal femur and proximal tibia with FES training. Previous studies, using similar FES techniques that produce mechanical stress and loading of bone, were largely unsuccessful. Leeds
      • Leeds EM
      • Klose KJ
      • Ganz W
      • Serafini A
      • Green BA
      Bone mineral density after bicycle ergometry training.
      and BeDell
      • BeDell KK
      • Scremin AME
      • Perell KL
      • Kunkel CF
      Effects of functional electrical stimulation-induced lower extremity cycling on bone density of spinal cord–injured patients.
      and their colleagues found no change in BMD after FES-assisted cycling training. Rodgers and associates
      • Rodgers MM
      • Glaser RM
      • Figoni SF
      • Hooker SP
      • Ezenwa BN
      • Collins SR
      • et al.
      Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training.
      showed a decrease in the rate of bone loss in the proximal tibia after knee extension training, but were unable to demonstrate any reversal of osteopenia. Recently, Mohr and colleagues
      • Mohr T
      • Podenphant J
      • Biering-Sorensen F
      • Galbo H
      • Thamsborg G
      • Kjaer M
      Increased bone mineral density after prolonged electrically induced cycle training of paralyzed limbs in spinal cord injured man.
      showed a 10% increase in proximal tibial BMD after 1 year of FES-assisted cycling training. The difference between our study and the previous ones may lie in the amount of training and loading. For example, Hangartner and associates
      • Hangartner TN
      • Rodgers MM
      • Glaser RM
      • Barre PS
      Tibial bone density loss in spinal cord injured patients: effects of FES exercise.
      had subjects train 3 times per week with 2 sets of 30 contractions and 1 set of 60 repetitions, for a total of approximately 360 contractions per week. In our study, subjects trained for up to 1 hour per day at a rate of 12 contractions per minute, which totals 720 contractions per day. Moreover, subjects trained 4 times per week and were tested once (12 contractions/min for 4min), for a total of 2928 contractions per week. This is more than 8 times the amount of training and loading as in the study by Hangartner et al.
      • Hangartner TN
      • Rodgers MM
      • Glaser RM
      • Barre PS
      Tibial bone density loss in spinal cord injured patients: effects of FES exercise.
      Our data certainly support the “mechanostat theory” of Frost,
      • Frost HM
      Bone “mass” and the “mechanostat”: a proposal.
      which claims that bone will only respond to certain levels of loading. Induced strains must be both above a lower threshold and below an upper threshold level for bone to have an adaptive response.
      • Bailey DA
      • Faulkner RA
      • McKay HA
      Growth, physical activity, and bone mineral acquisition.
      Why is it that individuals who have a lot of muscle spasms do not maintain high level of BMD? We showed previously
      • Stein RB
      • Gordon T
      • Jefferson J
      • Sharfenberger A
      • Yang JF
      • Totosy de Zepetnek J
      • et al.
      Optimal stimulation of paralyzed muscle after human spinal cord injury.
      that even SCI subjects with spasms generate on average only about 20% as much electromyography over a 24-hour period as control subjects. Perhaps the number of spasms or the intensity of the resulting contractions are not sufficient to stimulate osteogenesis.
      Another possible explanation for the discrepancies with some previous studies may be the skeletal regions being evaluated. Several groups
      • Leeds EM
      • Klose KJ
      • Ganz W
      • Serafini A
      • Green BA
      Bone mineral density after bicycle ergometry training.
      • BeDell KK
      • Scremin AME
      • Perell KL
      • Kunkel CF
      Effects of functional electrical stimulation-induced lower extremity cycling on bone density of spinal cord–injured patients.
      measured BMD changes in the femoral neck, in Ward's triangle (a triangular radiolucency formed between the primary trabecular patterns within the femoral neck), and in the trochanter, although loading was applied mainly lower in the legs by FES cycling exercises. Bloomfield and colleagues
      • Bloomfield SA
      • Mysiw WJ
      • Jackson RD
      Bone mass and endocrine adaptations to training in spinal cord injured individuals.
      also reported no change in BMD in the femoral neck, distal femur, and proximal tibia for the group as a whole. However, in a subset of subjects, the FES cycle ergometry training increased the BMD in the distal femur. Similarly, Mohr and associates
      • Mohr T
      • Podenphant J
      • Biering-Sorensen F
      • Galbo H
      • Thamsborg G
      • Kjaer M
      Increased bone mineral density after prolonged electrically induced cycle training of paralyzed limbs in spinal cord injured man.
      observed an increase in the proximal tibia BMD, but not in the femoral neck and lumbar spine, which suggests site-specific changes. Our results support this notion of site-specificity, because BMD increased in the distal femur and proximal tibia but not in the mid-tibia. The regional effects are likely the result of the specificity of the loading, because the FES knee-extension exercise used in this study produced much greater loading around the knee joint (distal femur and proximal tibia) than in the tibial shaft (mid-tibia). Along with this loading, contraction of the quadriceps muscles and movements of the knee joint could increase local blood flow and augment the circulation of material necessary for bone formation in the distal femur and proximal tibia.
      • Phillips W
      • Burkett LN
      • Munro R
      • Davis M
      • Pomeroy K
      Relative changes in blood flow with functional electrical stimulation during exercise of the paralyzed lower limbs.
      In contrast, the mid-tibia would not benefit from such local increases in circulation.
      Why did the BMD not increase more in the leg that worked against resistance? One explanation may be the duration of the stimulation and the setting of the isokinetic device. The quadriceps were stimulated for 5 seconds, while the Cybex was set at 30°/sec. Thus, after approximately 3 seconds, the knee joint would be fully extended and the muscles would then contract isometrically for about 2 seconds. The unresisted leg would contract more quickly and have a somewhat longer isometric period. Of course, the speed of the Cybex could have been reduced to 18°/sec, thereby eliminating this isometric contraction on the resisted side but not on the unresisted side, because it was free to move. However, this would have placed the subjects at greater risk of fracture, particularly at the onset of training, because reducing the speed has the effect of increasing the resistance. Both sides had the weight of the leg as an initial load, so the differences between the two may not have been great enough to influence bone formation or loss. Another possible explanation for the lack of difference between the sides may be cross-training effects. Vuori and colleagues
      • Vuori I
      • Heinonen A
      • Sievanen H
      • Kannus P
      • Pasanen M
      • Oja P
      Effects of unilateral strength training and detraining on bone mineral density and content in young women: a study of mechanical loading and deloading on human bones.
      reported that unilateral leg presses in young women without SCI produced a small but systematic increase in BMD in the contralateral limb.
      Several studies have demonstrated varied level of bone loss for different regions of the body including the lumbar spinal, the femoral neck, Ward's triangle, and the proximal and distal tibia.
      • Rodgers MM
      • Glaser RM
      • Figoni SF
      • Hooker SP
      • Ezenwa BN
      • Collins SR
      • et al.
      Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training.
      • Garland DE
      • Stewart CA
      • Adkins RH
      • Hu SS
      • Rosen C
      • Liotta FJ
      • et al.
      Osteoporosis after spinal cord injury.
      • Wilmet E
      • Ismail AA
      • Heilporn A
      • Welraeds D
      • Bergmann P
      Longitudinal study of the bone mineral content and of soft tissue composition after spinal cord section.
      • Szollar SM
      • Martin EM
      • Parthemore JG
      • Sartoris DJ
      • Deftos LJ
      Demineralization in tetraplegic and paralegic man over time.
      • Hangartner TN
      • Rodgers MM
      • Glaser RM
      • Barre PS
      Tibial bone density loss in spinal cord injured patients: effects of FES exercise.
      • Mohr T
      • Podenphant J
      • Biering-Sorensen F
      • Galbo H
      • Thamsborg G
      • Kjaer M
      Increased bone mineral density after prolonged electrically induced cycle training of paralyzed limbs in spinal cord injured man.
      • Szollar SM
      • Martin EM
      • Parthemore JG
      • Sartoris DJ
      • Deftos LJ
      Densitometric patterns of spinal cord injury associated bone loss.
      • Leslie WD
      • Nance PW
      Dissociated hip and spine demineralization: a specific finding in spinal cord injury.
      • Garland D
      • Ashford R
      • Adkins R
      Regional osteoporosis in females with spinal cord injury.
      When comparing our initial BMD data for the proximal tibia with previous studies, our BMD (65.5%) was slightly larger than that reported by Mohr and associates
      • Mohr T
      • Podenphant J
      • Biering-Sorensen F
      • Galbo H
      • Thamsborg G
      • Kjaer M
      Increased bone mineral density after prolonged electrically induced cycle training of paralyzed limbs in spinal cord injured man.
      (50%) and Hangartner and coworkers
      • Hangartner TN
      • Rodgers MM
      • Glaser RM
      • Barre PS
      Tibial bone density loss in spinal cord injured patients: effects of FES exercise.
      (34.5% after 7yrs). Moreover, the rate of loss for BMD in our study was slightly less than that reported by Hangartner.
      • Hangartner TN
      • Rodgers MM
      • Glaser RM
      • Barre PS
      Tibial bone density loss in spinal cord injured patients: effects of FES exercise.
      This difference is most likely the result of the technique used. In the study by Hangartner,
      • Hangartner TN
      • Rodgers MM
      • Glaser RM
      • Barre PS
      Tibial bone density loss in spinal cord injured patients: effects of FES exercise.
      the BMD was measured repeatedly over the time course of SCI for each individual, while our data were obtained from each subject at a specific time after SCI lesion. With their approach, they were able to show that bone loss is quite rapid in the initial months, followed by less steep changes over the next several years. On the other hand, both our data and those of Hangartner
      • Hangartner TN
      • Rodgers MM
      • Glaser RM
      • Barre PS
      Tibial bone density loss in spinal cord injured patients: effects of FES exercise.
      have no clear asymptote, indicating that BMD will eventually decay to almost zero.
      Different rates of loss have also been observed for different bone tissues.
      • Rodgers MM
      • Glaser RM
      • Figoni SF
      • Hooker SP
      • Ezenwa BN
      • Collins SR
      • et al.
      Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training.
      • Wilmet E
      • Ismail AA
      • Heilporn A
      • Welraeds D
      • Bergmann P
      Longitudinal study of the bone mineral content and of soft tissue composition after spinal cord section.
      • Hangartner TN
      • Rodgers MM
      • Glaser RM
      • Barre PS
      Tibial bone density loss in spinal cord injured patients: effects of FES exercise.
      For example, Wilmet and associates
      • Wilmet E
      • Ismail AA
      • Heilporn A
      • Welraeds D
      • Bergmann P
      Longitudinal study of the bone mineral content and of soft tissue composition after spinal cord section.
      report a loss of about 4% per month in trabecular bone and 2% per month in cortical bone. Our data support this differential rate of bone loss (fig 2) with higher values for distal femur and proximal tibia (mainly trabecular bone), compared with mid-tibia (mostly cortical bone). Presumably, the differential rates also apply to the gain, which would help explain the BMD increases in distal femur and proximal tibia and the lack of change in the mid-tibia.

       Muscle

      Because of the contralateral limb design used in this study, the effect of FES-assisted resistance training on the muscle force could be studied. The results clearly show that muscle strength increased on the resisted side at almost twice the rate of that on the unresisted side. Both sides began to level off near the end of the 24 weeks of training. When looking at the data for the unresisted side, our results agree with several studies
      • Kralj AR
      • Bajd T
      Functional electrical stimulation: Standing and walking after spinal cord injury.
      • Kralj AR
      • Bajd T
      • Turk R
      Electrical stimulation providing functional use of paraplegic patients muscles.
      • Peckham PH
      • Mortimer JT
      • Marsolais EB
      Alteration in the force and fatigability of skeletal muscle in quadriplegic humans following exercise induced by chronic electrical stimulation.
      • Peckham PH
      Functional electrical stimulation: current status and future prospects of applications to the neuromuscular system in spinal cord injury.
      but differ from one of our previous studies.
      • Stein RB
      • Gordon T
      • Jefferson J
      • Sharfenberger A
      • Yang JF
      • Totosy de Zepetnek J
      • et al.
      Optimal stimulation of paralyzed muscle after human spinal cord injury.
      Kralj and colleagues
      • Kralj AR
      • Bajd T
      Functional electrical stimulation: Standing and walking after spinal cord injury.
      • Kralj AR
      • Bajd T
      • Turk R
      Electrical stimulation providing functional use of paraplegic patients muscles.
      reported that FES isotonic training with no resistance, except for the weight of the limb, produced strength gains in the quadriceps muscles. Peckham and colleagues
      • Peckham PH
      • Mortimer JT
      • Marsolais EB
      Alteration in the force and fatigability of skeletal muscle in quadriplegic humans following exercise induced by chronic electrical stimulation.
      • Peckham PH
      Functional electrical stimulation: current status and future prospects of applications to the neuromuscular system in spinal cord injury.
      also showed an increase in muscle force in the hand muscles after unresisted FES training. In contrast, Stein and associates
      • Stein RB
      • Gordon T
      • Jefferson J
      • Sharfenberger A
      • Yang JF
      • Totosy de Zepetnek J
      • et al.
      Optimal stimulation of paralyzed muscle after human spinal cord injury.
      reported that stimulation of the tibialis anterior muscle under unloaded conditions for up to 8 hours a day increased the fatigue resistance, but had no effect on the strength or myosin–adenosine triphosphatase activity. The stimulation parameters and the allowance of muscle shortening were similar in that study and the present one, but the muscles studied are different and the muscle loading in the present study is greater than in the previous study.
      Loading the extremities during FES-assisted contractions has been demonstrated to be effective in increasing strength.
      • Kralj AR
      • Bajd T
      Functional electrical stimulation: Standing and walking after spinal cord injury.
      • Kralj AR
      • Bajd T
      • Turk R
      Electrical stimulation providing functional use of paraplegic patients muscles.
      • Petrofsky JS
      • Phillips CA
      The use of functional electrical stimulation for rehabilitation of spinal cord injured patients.
      The most impressive results in strength gains have been obtained under progressive overload and gradual increases in training volume. Rodgers and coworkers
      • Rodgers MM
      • Glaser RM
      • Figoni SF
      • Hooker SP
      • Ezenwa BN
      • Collins SR
      • et al.
      Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training.
      reported that such training greatly improved quadriceps muscle performance, but do not show any leveling off near the end of training. Perhaps this could be explained by the amount of training, because their subjects only trained 3 times per week, compared with 4 times plus a testing session in this study. The measure used may also explain the difference. Rodgers
      • Rodgers MM
      • Glaser RM
      • Figoni SF
      • Hooker SP
      • Ezenwa BN
      • Collins SR
      • et al.
      Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training.
      measured the weight lifted and multiplied it by the number of repetitions that the subject achieved at that weight. In contrast, only the maximal knee torque (average of 3 contractions) produced at the onset of each testing session was used in our study.
      Our fatigue index data demonstrated that endurance did not improve as a result of 1 hour/day of FES-assisted training. Whether on the resisted or unresisted sides, the fatigue index remained around 50%. Moreover, the twitch contraction times were unmodified by the training. Kralj and Bajd
      • Kralj AR
      • Bajd T
      Functional electrical stimulation: Standing and walking after spinal cord injury.
      reported that FES-assisted training was ineffective in increasing muscle endurance for the quadriceps muscles, and 3 of their 5 subjects even showed more fatigue. In contrast, other authors have shown increased endurance for the hand
      • Peckham PH
      • Mortimer JT
      • Marsolais EB
      Alteration in the force and fatigability of skeletal muscle in quadriplegic humans following exercise induced by chronic electrical stimulation.
      and another leg muscle.
      • Stein RB
      • Gordon T
      • Jefferson J
      • Sharfenberger A
      • Yang JF
      • Totosy de Zepetnek J
      • et al.
      Optimal stimulation of paralyzed muscle after human spinal cord injury.
      This latter study demonstrated that to increase endurance markedly, one should stimulate about 2 hours a day. Two hours per day was also enough time to induce changes in the twitch contraction times, bringing them closer to control subjects. The goal of our study was to improve muscle strength, but not necessarily muscle fatigue. Because a large time commitment affects the willingness of subjects to participate, the training in the present study was limited to 1 hour/day.
      Because the muscle is much stronger at the end of training, it can maintain a greater fatigued force level than at the onset of training. This force level often exceeded the maximal initial force measured at the onset of training. For most subjects (7 of 12), the fatigued knee torque level on the resisted side approached or exceeded 50Nm, which, according to Kralj and Bajd,
      • Kralj AR
      • Bajd T
      Functional electrical stimulation: Standing and walking after spinal cord injury.
      is sufficient for standing and walking in patients with complete paraplegia. Thus, the quadriceps muscles could maintain enough torque to be functionally useful while using a FES system. Moreover, such muscle torque would also be quite useful in SCI persons with incomplete motor function loss.
      The thigh girth was not altered on either side by FES-assisted training in the present study. On the other hand, an increase in girth, particularly on the resisted side, was noted in several individuals. Also, a number of subjects stated that their thigh appeared firmer at the end of training. This suggests that muscle may have replaced adipose tissue without a change in girth. Similar suggestions were made by Rodgers and coworkers,
      • Rodgers MM
      • Glaser RM
      • Figoni SF
      • Hooker SP
      • Ezenwa BN
      • Collins SR
      • et al.
      Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training.
      but they also had an insufficient number of subjects to reach statistical significance. Interestingly, decreasing subcutaneous fat can reduce the stimulation intensity needed.
      • Burger H
      • Valencic V
      • Kogovsek N
      • Marincek C
      This, along with better muscle efficiency, can help explain why less intense stimulation was required to produce a maximal contraction at the end of training.

       Clinical notes

      The FES training in this study offers several interesting clinical benefits. Because bone and muscle strength increased on the unresisted side, training can be done at home without any specialized equipment. Because muscle strength increases at a slower rate with no resistance, the risk of fracture is lessened by training without resistance (fig 6). Because the bone can be quite frail before the onset of FES training, allowing the limb to move, as opposed applying to isometric training, can also lessen the risk of fracture. Several bone fractures have occurred during FES training under isometric conditions (Hartkopp and colleagues
      • Hartkopp A
      • Murphy RJ
      • Mohr T
      • Kjaer M
      • Biering-Sorensen F
      Bone fracture during electrical stimulation of the quadriceps in a spinal cord injured subject.
      and personal communications). To minimize risk, training should start in the first years after injury, before BMD has reached very low levels. Also, no significant improvement was seen in subjects who were injured more than 13.5 years before beginning training.
      Although fractures in or compression of the lumbar vertebrae (L2-L5) are very uncommon, this is one of the sites most often used to evaluate bone loss in SCI patients.
      • Garland DE
      • Stewart CA
      • Adkins RH
      • Hu SS
      • Rosen C
      • Liotta FJ
      • et al.
      Osteoporosis after spinal cord injury.
      • Wilmet E
      • Ismail AA
      • Heilporn A
      • Welraeds D
      • Bergmann P
      Longitudinal study of the bone mineral content and of soft tissue composition after spinal cord section.
      • BeDell KK
      • Scremin AME
      • Perell KL
      • Kunkel CF
      Effects of functional electrical stimulation-induced lower extremity cycling on bone density of spinal cord–injured patients.
      • Szollar SM
      • Martin EM
      • Parthemore JG
      • Sartoris DJ
      • Deftos LJ
      Densitometric patterns of spinal cord injury associated bone loss.
      • Leduc BE
      • Lefebvre B
      • Puig P
      • Daigneault L
      L'ostéoporose sous-lésionnelle chez le blessé médullaire évaluée par ostéodensitométrie.
      In addition, Roberts and colleagues
      • Roberts D
      • Lee W
      • Cuneo RC
      • Wittmann J
      • Ward G
      • Flatman R
      • et al.
      Longitudinal study of bone turnover after acute spinal cord injury.
      have reported that biochemical markers showed little change for the hip, lumbar spine, and radius, but demonstrated decrements for the entire lower limbs. More common are fractures of long bones such as the femur or tibia, particularly near their ends.
      • Freehafer AA
      Limb fractures in patients with spinal cord injury.
      Thus, when measuring and reporting bone loss in patients with SCI, sites that are at risk should be evaluated. Freehafer
      • Freehafer AA
      Limb fractures in patients with spinal cord injury.
      reported that fractures of the distal femur and proximal tibia account for 33.9% of all fractures in individuals with SCI and that these fractures were related to bone density loss. Interestingly, these are just the regions that show marked strengthening with FES training.
      Several therapeutic methods, which have been shown to be effective in aging and in endocrine disorders related to osteoporosis, have been tried in individuals with SCI. Pharmacologic agents, such as growth hormone, estrogen, parathyroid hormone, fluoride, calcitonin, bisphosphonate, and vitamin D have been used with limited success and with some side-effects in these individuals.
      • Nance PW
      • Pearson E
      • Richards C
      • Noreau L
      • Bélanger M
      • Andrews B
      • et al.
      Osteoporosis after spinal cord injury.
      The results of this study certainly demonstrate that FES exercises that cause loading of bone and muscle contractions can be effective in bone formation. Combining this type of intervention with some of the therapeutic methods could lead to greater bone recovery and even less risk of fracture from minimal trauma.
      Finally, there are positive side-effects of FES training that have been reported by some of our subjects and in the literature. They include improved appearance of the legs, decreased spasms, increased blood flow, and better skin quality.
      • Kralj AR
      • Bajd T
      Functional electrical stimulation: Standing and walking after spinal cord injury.
      • Figoni SF
      • Glaser RM
      • Rodgers MM
      • Hooker SP
      • Ezenwa BN
      • Collins SR
      • et al.
      Acute hemodynamic responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise.
      • Hooker SP
      • Figoni SF
      • Rodgers MM
      • Glaser RM
      • Mathews T
      • Suryaprasad AG
      • et al.
      Physiologic effects of electrical stimulation leg cycle exercise training in spinal cord injured persons.
      • Arnold PB
      • McVey PP
      • Farrell WJ
      • Deurloo TM
      • Grasso AR
      Functional electric stimulation: its efficacy and safety in improving pulmonary function and musculoskeletal fitness.
      • Seib TP
      • Price R
      • Reyes MR
      • Lehmann JF
      The quantitative measurement of spasticity: effect of cutaneous electrical stimulation.
      Thus, along with improved BMD and muscle strength, FES training has the possibility of improving the quality of life.

      Acknowledgements

      The authors acknowledge the support of imaging technicians Carole Charbonneau, Norman Chevrier, and Nigel Gann. Helpful technical assistance was also provided by Carole Roy, Julie Dugal, and Monique Boivin.

      References

        • Lieber RL
        Skeletal muscle adaptability. II. Muscle properties following spinal-cord injury.
        Dev Med Child Neurol. 1986; 28: 533-542
        • Lieber RL
        • Friden JO
        • Hargens AR
        • Feringa ER
        Long-term effects of spinal cord transection on fast and slow rat skeletal muscle. II. Morphometric properties.
        Exp Neurol. 1986; 91: 435-448
        • Lieber RL
        • Johansson CB
        • Vahlsing HL
        • Hargens AR
        • Feringa ER
        Long-term effects of spinal cord transection on fast and slow rat skeletal muscle. I. Contractile properties.
        Exp Neurol. 1986; 91: 423-434
        • Gordon T
        Fatigue in adapted systems. Overuse and underuse paradigms.
        Adv Exp Med Biol. 1995; 384: 429-456
        • Gordon T
        • Mao J
        Muscle atrophy and procedures for training after spinal cord injury.
        Phys Ther. 1994; 74: 50-60
        • Needham-Shropshire BM
        • Klose KJ
        • Tucker ME
        • Thomas CK
        Manual muscle test score and force comparisons after cervical spinal cord injury.
        J Spinal Cord Med. 1997; 20: 324-330
        • Gefen JY
        • Gelmann AS
        • Herbison GJ
        • Cohen ME
        • Schmidt RR
        Use of shoulder flexors to achieve isometric elbow extension in C6 tetraplegic patients during weight shift.
        Spinal Cord. 1997; 35: 308-313
        • Cameron T
        • Calancie B
        Mechanical and fatigue properties of wrist flexor muscles during repetitive contractions after cervical spinal cord injury.
        Arch Phys Med Rehabil. 1995; 76: 929-933
        • Ragnarsson KT
        • Sell GH
        Lower extremity fractures after spinal cord injury: a retrospective study.
        Arch Phys Med Rehabil. 1981; 62: 418-423
        • Rodgers MM
        • Glaser RM
        • Figoni SF
        • Hooker SP
        • Ezenwa BN
        • Collins SR
        • et al.
        Musculoskeletal responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise training.
        J Rehab Res Dev. 1991; 28: 19-26
        • Garland DE
        • Stewart CA
        • Adkins RH
        • Hu SS
        • Rosen C
        • Liotta FJ
        • et al.
        Osteoporosis after spinal cord injury.
        J Orthop Res. 1992; 10: 371-378
        • Keating JF
        • Kerr M
        • Delargy M
        Minimal trauma causing fractures in patients with spinal cord injury.
        Disabil Rehabil. 1992; 14: 108-109
        • Saltzstein RJ
        • Hardin S
        • Hastings J
        Osteoporosis in spinal cord injury: using an index of mobility and its relationship to bone density.
        J Am Paraplegia Soc. 1992; 15: 232-234
        • Wilmet E
        • Ismail AA
        • Heilporn A
        • Welraeds D
        • Bergmann P
        Longitudinal study of the bone mineral content and of soft tissue composition after spinal cord section.
        Paraplegia. 1995; 33: 674-677
        • Chow YW
        • Inman C
        • Pollintine P
        • Sharp CA
        • Haddaway MJ
        • el-Masry W
        • et al.
        Ultrasound bone densitometry and dual energy X-ray absorptiometry in patients with spinal cord injury: a cross-sectional study.
        Spinal Cord. 1996; 34: 736-741
        • Szollar SM
        • Martin EM
        • Parthemore JG
        • Sartoris DJ
        • Deftos LJ
        Demineralization in tetraplegic and paralegic man over time.
        Spinal Cord. 1997; 35: 223-228
        • Leeds EM
        • Klose KJ
        • Ganz W
        • Serafini A
        • Green BA
        Bone mineral density after bicycle ergometry training.
        Arch Phys Med Rehabil. 1990; 71: 207-209
        • Hangartner TN
        • Rodgers MM
        • Glaser RM
        • Barre PS
        Tibial bone density loss in spinal cord injured patients: effects of FES exercise.
        J Rehabil Res Dev. 1994; 31: 50-61
        • BeDell KK
        • Scremin AME
        • Perell KL
        • Kunkel CF
        Effects of functional electrical stimulation-induced lower extremity cycling on bone density of spinal cord–injured patients.
        Am J Phys Med Rehabil. 1996; 75: 29-34
        • Mohr T
        • Podenphant J
        • Biering-Sorensen F
        • Galbo H
        • Thamsborg G
        • Kjaer M
        Increased bone mineral density after prolonged electrically induced cycle training of paralyzed limbs in spinal cord injured man.
        Calcif Tissue Int. 1997; 61: 22-25
        • Pilonchery G
        • Minaire P
        • Milan JJ
        • Revol A
        Urinary elimination of glycosaminoglycans during the immobilization osteoporosis of spinal cord injury patients.
        Clin Orthop. 1983; 174: 230-235
        • Nottage WM
        A review of long-bone fractures in patients with spinal cord injuries.
        Clin Orthop. 1981;Mar-Apr; : 65-70
        • Bohnor JA
        • Dilworth BB
        • Sullivan KM
        Exercise and osteoporosis: a critique of the literature.
        Physiother Can. 1988; 40: 146-155
        • Snow-Harter C
        • Marcus R
        Exercise, bone mineral density, and osteoporosis.
        Exerc Sport Sci Rev. 1991; 19: 351-388
        • Schwarz P
        • Bulow JB
        • Kjaer M
        Bone metabolism and physical training] [Danish.
        Ugeskr Laeger. 1999; 161: 2522-2525
        • Pettersson U
        • Nordstrom P
        • Lorentzon R
        A comparison of bone mineral density and muscle strength in young male adults with different exercise level.
        Calcif Tissue Int. 1999; 64: 490-498
        • Ayalon J
        • Simkin A
        • Leichter I
        • Raifmann S
        Dynamic bone loading exercises for postmenopausal women: effect on the density of the distal radius.
        Arch Phys Med Rehabil. 1987; 68: 280-283
        • de Bruin ED
        • Frey-Rindova P
        • Herzog RE
        • Dietz V
        • Dambacher MA
        • Stussi E
        Changes of tibia bone properties after spinal cord injury: effects of early intervention.
        Arch Phys Med Rehabil. 1999; 80: 214-220
        • Doyle F
        • Brown J
        • Lachance C
        Relation between bone mass and muscle weight.
        Lancet. 1970; 1: 391-393
        • Hodgson SF
        • Johnston CC
        • Avioli LV
        • Heath III, HH
        • Khosla S
        • Kleerekoper M
        • et al.
        AACE Clinical practice guidelines for the prevention and treatment of postmenopausal osteoporosis.
        American Association of Clinical Endocrinologists; American College of Endocrinology, Jacksonville (FL)Mar 1996
        • Winter DA
        • Wells RP
        • Orr GW
        Errors in the use of isokinetic dynamometers.
        Eur J Appl Physiol. 1981; 46: 397-408
        • Stein RB
        • Gordon T
        • Jefferson J
        • Sharfenberger A
        • Yang JF
        • Totosy de Zepetnek J
        • et al.
        Optimal stimulation of paralyzed muscle after human spinal cord injury.
        J Appl Physiol. 1992; 72: 1393-1400
        • Kralj AR
        • Bajd T
        Functional electrical stimulation: Standing and walking after spinal cord injury.
        CRC Press, Boca Raton (FL)1989
        • Frost HM
        Bone “mass” and the “mechanostat”: a proposal.
        Anat Rec. 1987; 219: 1-9
        • Bailey DA
        • Faulkner RA
        • McKay HA
        Growth, physical activity, and bone mineral acquisition.
        Exerc Sport Sci Rev. 1996; 24: 233-266
        • Bloomfield SA
        • Mysiw WJ
        • Jackson RD
        Bone mass and endocrine adaptations to training in spinal cord injured individuals.
        Bone. 1996; 19: 61-68
        • Phillips W
        • Burkett LN
        • Munro R
        • Davis M
        • Pomeroy K
        Relative changes in blood flow with functional electrical stimulation during exercise of the paralyzed lower limbs.
        Paraplegia. 1995; 33: 90-93
        • Vuori I
        • Heinonen A
        • Sievanen H
        • Kannus P
        • Pasanen M
        • Oja P
        Effects of unilateral strength training and detraining on bone mineral density and content in young women: a study of mechanical loading and deloading on human bones.
        Calcif Tissue Int. 1994; 55: 59-67
        • Szollar SM
        • Martin EM
        • Parthemore JG
        • Sartoris DJ
        • Deftos LJ
        Densitometric patterns of spinal cord injury associated bone loss.
        Spinal Cord. 1997; 35: 374-382
        • Leslie WD
        • Nance PW
        Dissociated hip and spine demineralization: a specific finding in spinal cord injury.
        Arch Phys Med Rehabil. 1993; 74: 960-964
        • Garland D
        • Ashford R
        • Adkins R
        Regional osteoporosis in females with spinal cord injury.
        J Spinal Cord Med. 1995; 18: 115
        • Kralj AR
        • Bajd T
        • Turk R
        Electrical stimulation providing functional use of paraplegic patients muscles.
        Med Prog Technol. 1980; 7: 3-9
        • Peckham PH
        • Mortimer JT
        • Marsolais EB
        Alteration in the force and fatigability of skeletal muscle in quadriplegic humans following exercise induced by chronic electrical stimulation.
        Clin J Sport Med. 1976; 114: 326-334
        • Peckham PH
        Functional electrical stimulation: current status and future prospects of applications to the neuromuscular system in spinal cord injury.
        Paraplegia. 1987; 25: 279-288
        • Petrofsky JS
        • Phillips CA
        The use of functional electrical stimulation for rehabilitation of spinal cord injured patients.
        Cent Nerv Syst Trauma. 1984; 1: 57-74
        • Burger H
        • Valencic V
        • Kogovsek N
        • Marincek C
        Is electrical stimulation of muculus gluteus maximus influenced by body fat? The Ljubljana FES conference, Ljubljana, Slovenia1993: 105-107
        • Hartkopp A
        • Murphy RJ
        • Mohr T
        • Kjaer M
        • Biering-Sorensen F
        Bone fracture during electrical stimulation of the quadriceps in a spinal cord injured subject.
        Arch Phys Med Rehabil. 1998; 79: 1133-1136
        • Leduc BE
        • Lefebvre B
        • Puig P
        • Daigneault L
        L'ostéoporose sous-lésionnelle chez le blessé médullaire évaluée par ostéodensitométrie.
        Ann Réadaptation Med Phys. 1991; 34: 199-204
        • Roberts D
        • Lee W
        • Cuneo RC
        • Wittmann J
        • Ward G
        • Flatman R
        • et al.
        Longitudinal study of bone turnover after acute spinal cord injury.
        J Clin Endocrinol Metab. 1998; 83: 415-422
        • Freehafer AA
        Limb fractures in patients with spinal cord injury.
        Arch Phys Med Rehabil. 1995; 76: 823-827
        • Nance PW
        • Pearson E
        • Richards C
        • Noreau L
        • Bélanger M
        • Andrews B
        • et al.
        Osteoporosis after spinal cord injury.
        in: Annual General Meeting: NCE Network for Neural Regeneration and Recovery; 1994 Jun 9-12; Toronto Neuroscience Network (Network of Centres of Excellence), Toronto1994
        • Figoni SF
        • Glaser RM
        • Rodgers MM
        • Hooker SP
        • Ezenwa BN
        • Collins SR
        • et al.
        Acute hemodynamic responses of spinal cord injured individuals to functional neuromuscular stimulation-induced knee extension exercise.
        J Rehabil Res Dev. 1991; 28: 9-18
        • Hooker SP
        • Figoni SF
        • Rodgers MM
        • Glaser RM
        • Mathews T
        • Suryaprasad AG
        • et al.
        Physiologic effects of electrical stimulation leg cycle exercise training in spinal cord injured persons.
        Arch Phys Med Rehabil. 1992; 73: 470-476
        • Arnold PB
        • McVey PP
        • Farrell WJ
        • Deurloo TM
        • Grasso AR
        Functional electric stimulation: its efficacy and safety in improving pulmonary function and musculoskeletal fitness.
        Arch Phys Med Rehabil. 1992; 73: 665-668
        • Seib TP
        • Price R
        • Reyes MR
        • Lehmann JF
        The quantitative measurement of spasticity: effect of cutaneous electrical stimulation.
        Arch Phys Med Rehabil. 1994; 75: 746-750