Article Text

Download PDFPDF

Effects of Testosterone and Evoked Resistance Exercise after Spinal Cord Injury (TEREX-SCI): study protocol for a randomised controlled trial
  1. Ashraf S Gorgey1,2,
  2. Refka E Khalil1,
  3. Ranjodh Gill3,4,
  4. Laura C O'Brien1,
  5. Timothy Lavis1,2,
  6. Teodoro Castillo1,
  7. David X Cifu1,2,
  8. Jeannie Savas5,
  9. Rehan Khan6,
  10. Christopher Cardozo7,8,
  11. Edward J Lesnefsky9,10,
  12. David R Gater11,
  13. Robert A Adler3,4
  1. 1Spinal Cord Injury and Disorders Center, Hunter Holmes McGuire VAMC, Richmond, Virginia, USA
  2. 2Department of Physical Medicine and Rehabilitation, Virginia Commonwealth University, Richmond, Virginia, USA
  3. 3Endocrinology Service, Hunter Holmes McGuire VA Medical Center, Richmond, Virginia, USA
  4. 4Endocrine Division, Virginia Commonwealth University School of Medicine¸ Richmond, Virginia, USA
  5. 5Surgical Service, Hunter Holmes McGuire VA Medical Center, Richmond, Virginia, USA
  6. 6Radiology Service, Hunter Holmes McGuire VA Medical Center, Richmond, Virginia, USA
  7. 7National Center for the Medical Consequences of Spinal Cord Injury, James J Peters VA Medical Center, Bronx, New York, USA
  8. 8Department of Medicine, Icahn School of Medicine at Mt. Sinai, New York City, New York, USA
  9. 9Cardiology Service, Hunter Holmes McGuire VA Medical Center, Richmond, Virginia, USA
  10. 10Division of Cardiology, Department of Medicine, Pauley Heart Center Virginia Commonwealth University, Richmond, Virginia, USA
  11. 11Department of Physical Medicine and Rehabilitation, Penn State Milton S Hershey Medical Center, Hershey, Pennsylvania, USA
  1. Correspondence to Dr Ashraf S Gorgey; ashraf.gorgey{at}


Introduction Individuals with spinal cord injury (SCI) are at a lifelong risk of obesity and chronic metabolic disorders including insulin resistance and dyslipidemia. Within a few weeks of injury, there is a significant decline in whole body fat-free mass, particularly lower extremity skeletal muscle mass, and subsequent increase in fat mass (FM). This is accompanied by a decrease in anabolic hormones including testosterone. Testosterone replacement therapy (TRT) has been shown to increase skeletal muscle mass and improve metabolic profile. Additionally, resistance training (RT) has been shown to increase lean mass and reduce metabolic disturbances in SCI and other clinical populations.

Methods and analysis 26 individuals with chronic, motor complete SCI between 18 and 50 years old were randomly assigned to a RT+TRT group (n=13) or a TRT group (n=13). 22 participants completed the initial 16-week training phase of the study and 4 participants withdrew. 12 participants of the 22 completed 16 weeks of detraining. The TRT was provided via transdermal testosterone patches (4–6 mg/day). The RT+TRT group had 16 weeks of supervised unilateral progressive RT using surface neuromuscular electrical stimulation with ankle weights. This study will investigate the effects of evoked RT+TRT or TRT alone on body composition (muscle cross-sectional area, visceral adipose tissue, %FM) and metabolic profile (glucose and lipid metabolism) in individuals with motor complete SCI. Findings from this study may help in designing exercise therapies to alleviate the deterioration in body composition after SCI and decrease the incidence of metabolic disorders in this clinical population.

Ethics and dissemination The study is currently approved by the McGuire VA Medical Center and Virginia Commonwealth University. All participants read and signed approved consent forms. Results will be submitted to peer-reviewed journals and presented at national and international conferences.

Trial registration number Pre-result, NCT01652040.


This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See:

View Full Text

Statistics from

Strengths and limitations of this study

  • The trial will investigate the use of surface neuromuscular electrical stimulation-induced resistance training (NMES-RT) to restore muscle size after spinal cord injury (SCI).

  • The trial will provide evidence on the effectiveness of testosterone replacement therapy (TRT) to restore muscle size and lean mass and serve as an alternative approach for those who cannot benefit from NMES.

  • The trial will determine how restoring muscle size and lean mass by RT+TRT or TRT can benefit the metabolic profile after SCI.

  • The study is only limited to men with complete SCI.

  • Surface NMES may not benefit those with full sensation or lower motor neuron denervation.


There are ∼11 000–12 000 new cases of spinal cord injury (SCI) in the USA annually with an overall prevalence of 250 000–400 000.1 ,2 Persons with motor complete injury have loss of sensation and motor function below the level of injury, while incomplete injury is characterised by preserved motor or sensory function. Chronic SCI, defined as 1 year postinjury, is associated with dramatic skeletal muscle atrophy,3–5 increase of fat mass (FM)6–8 and decrease of fat-free mass (FFM).6 ,7 Collectively, these factors put individuals with SCI at risk for metabolic disorders such as type II diabetes and cardiovascular disease.

Previous studies reveal that 60% of individuals with SCI in the USA are either overweight or obese.2 ,9–11 Despite a low body mass index (BMI) in 50% of the SCI population, individuals are likely to have more than 30% of their body mass as FM. Furthermore, person with SCI are 13% fatter per unit BMI than able-bodied individuals. Individuals with SCI also have a redistribution of adipose tissue, with greater trunk FM and visceral adipose tissue (VAT) compared to age and waist circumference matched able-bodied controls.12–14 Adipose tissue, particularly VAT, secretes proinflammatory cytokines including interleukin-6 (IL-6) and tumour-necrosis factor-α (TNF-α). Therefore, the increase in VAT after SCI may contribute to metabolic syndrome by stimulating the hepatic production of C reactive protein (CRP), which is tied to vascular inflammation.15–18 Another type of ectopic adipose tissue, intramuscular fat (IMF), is increased after SCI and has been correlated with reduced insulin sensitivity.19 ,20

Metabolic changes also accompany SCI, with previous studies finding that more than 50% of individuals with SCI are glucose intolerant, while one of five is diabetic.2 ,9–11 ,21 Other studies report that 55% of individuals with SCI are at risk of developing metabolic syndrome.21–23 Individuals with complete tetraplegia are more likely to experience decreased glucose and carbohydrate tolerance and have a higher prevalence of heart disease than those with incomplete injuries.24 ,25 Likewise, depressed HDL-C (<35 mg/dL) and a higher total cholesterol/HDL-C ratio, predictors of coronary heart disease, were noted in those with chronic SCI compared with able-bodied controls.24 ,26 These are not universal findings, however, as a systematic review of carbohydrate and lipid disorders in persons with SCI did not find strong evidence of increased risk beyond that of the general population.27

While previous studies have shown a link between body composition and metabolic profile after SCI, the cellular mechanisms remain unknown. Mitochondria are the site of oxygen consumption and energy production from glucose and lipid metabolism. Unfortunately, mitochondrial function is impaired in a number of diseases including neurodegenerative disease, atherosclerosis, hypertension and cancer.28–31 Fewer and smaller mitochondria are found in skeletal muscle of insulin resistant, obese and type II diabetic individuals.32 Previous studies found decreased muscle oxidative capacity and succinate dehydrogenase activity, complex II of the electron transport chain, after SCI.33–35

One hypothesis is that skeletal muscle mitochondrial function is decreased in metabolic disorders, leading to decreased fatty acid oxidation which contributes to insulin resistance;36 ,37 however, this is controversial.37 Mitochondria are dynamic organelles and undergo biogenesis, remodelling, and degradation. Mitochondrial biogenesis is driven in part through the action of peroxisome proliferator-activated receptor (PPAR) coactivator 1 alpha (PGC-1α).38 ,39 In addition, PGC-1α integrates insulin signalling and lipogenesis in skeletal muscle.40 ,41 PGC-1α is decreased in animal and human models following denervation.42 ,43 Decreased mitochondrial biogenesis and reduced mitochondrial mass may result in decreased energy production and therefore play a significant role in the altered metabolic profile following SCI.44

Body composition and metabolic changes after SCI may be further exacerbated by reduced anabolic hormones including testosterone (T), growth hormone and the growth hormone second messenger insulin-like growth factor-1 (IGF-1).45 ,46 Previous studies have shown that 60% of men with SCI have low T and that testosterone replacement therapy (TRT) increases IGF-1 in men.47–50 In rodent models of SCI, TRT attenuates the loss of muscle.51 ,52 TRT decreases total body fat, increases lean mass53 ,54 and increases the number of proliferating skeletal muscle satellite cells in men.55 These findings suggest that TRT may provide metabolic benefits to individuals with SCI.

Resistance training (RT) improves insulin sensitivity and increases fatty acid and carbohydrate metabolism, as well as attenuates sarcopenia in the elderly and after SCI.56–63 Moreover, RT has been shown to influence body composition by increasing lean mass, decreasing FM and reducing VAT, suggesting that the benefits of RT could overcome the risk of developing insulin resistance.56–59 Functional electrical stimulation (FES) has been shown to improve fatty acid kinetics, carbohydrate metabolism and vascular health after SCI.60–63 Electrically evoked RT using neuromuscular electrical stimulation (NMES-RT) and ankle weights is another form that has been shown to be effective in inducing muscle hypertrophy in individuals with chronic SCI.59 ,64 One study showed a 40% increase in skeletal muscle size and improved glucose tolerance after 12 weeks of training.59 Another study showed that following 12 weeks of NMES-RT, whole thigh, knee extensor and flexor cross-sectional areas (CSAs) increased by 28%, 35% and 16%, respectively. Moreover, the ratio of leg FFM to whole body FFM increased by 20% following intervention. There was 32% decrease in glucose area under the curve adjusted to muscle CSA following 12 weeks of NMES-RT. However, there were only modest effects on whole body composition, as well as a non-significant decrease in VAT.64 It is possible that the limited effects of NMES-RT on parameters of body composition and VAT can be possibly explained by depressed T-level in persons with SCI. Supplementing exogenous T may optimise the outcomes of NMES-RT on parameters of body composition and metabolic profile such as increased basal metabolic rate (BMR).

TRT may be an effective therapy to counterbalance the growing rate of obesity, type II diabetes and cardiovascular disease among individuals with SCI. Moreover, results from the current trial may provide evidence that TRT is an effective intervention for those who cannot effectively benefit from NMES because of lower motor neuron (LMN) denervation or intolerance to electrical stimulation. Therefore, the primary hypothesis is that the addition of TRT will maximise the benefits of electrically evoked RT on parameters of body composition and metabolic profile in men with chronic complete SCI. We, here, report the design of a study for which the major research goal is to investigate the effects of 16 weeks of evoked RT+TRT versus TRT on body composition (primary outcome variables: muscle CSA, VAT, %FM) and metabolic profiles (secondary outcome variables; glucose and lipid metabolism) in individuals with motor complete SCI.

Methods and analysis

Study design

A randomised controlled study was undertaken in which individuals with SCI were randomised to receive RT+TRT or TRT alone for 16 weeks. The study was approved by the McGuire Veteran Affairs Investigation Research Board and the Virginia Commonwealth University (VCU) Office of Research and Innovation. The trial has been registered at (NCT01652040). A member of the research team explained the study and obtained written informed consent. After informed consent, each participant underwent a detailed physical examination at the Hunter Holmes McGuire VA Medical Center (VAMC) by a physiatrist board certified in SCI medicine. This examination included a neurological assessment according to the International Standards for Neurological Classification of SCI (ISNCSCI), including the American Spinal Injury Association (ASIA) Impairment Scale (AIS).65

The study design and procedures are presented in figures 1 and 2. The study visits included estimation of body composition, anthropometry and dual X-ray absorptiometry (DEXA; baselines 1 and 2 and postinterventions 1 and 2). Additionally, MRI scans were obtained for trunk adipose tissue and lower extremity skeletal muscles and IMF CSA (baseline 2, postinterventions 1 and 2). Participants were then escorted to the VCU-CRS unit (VCU Clinical Research Unit) for dinner and remained in the VCU-CRS unit overnight for the four study visits. Muscle biopsies were obtained at baseline 2 and postintervention 1.

Figure 1

Timeline and main procedures of the TEREX-SCI trial for the RT+TRT and TRT groups during baseline 1, baseline 2, postintervention 1 and postintervention 2 for 26 persons with motor complete SCI. RT, resistance training; TEREX-SCI, Testosterone and Evoked Resistance Exercise after Spinal Cord Injury; TRT, testosterone replacement therapy.

Figure 2

A person with T4 motor complete SCI undergoing electrically evoked RT (left panel) and TRT using transdermal patches (right panel) as a part of a 16-week intervention.

Recruitment and randomization

The recruitment process started in July 2012 and ended in June 2015. Data analysis is currently being performed. Recruitment details and randomisation are presented in table 1 and figure 3. Prior to the start of the study, numbers 1–26 were randomised using the n-Query software program by the principal investigator. At the end of the two-day assessment period (baseline 1), the allocation of participants into either group was performed in a blinded fashion by drawing a folded paper with a number (1–26) by the research coordinator. This number was matched with the assignment from the randomisation procedure.

Table 1

Randomisation of individuals with motor complete SCI into either RT+TRT (n=13) or TRT (n=13) using n Query software with a treatment ratio 1:1

Figure 3

Schematic diagram showing the process of recruitment over the 3-year period of the TEREX trial. TEREX, Testosterone and Evoked Resistance Exercise.

Twenty-six participants were recruited to participate in the study. A 4-week delayed entry period was included to obtain baseline measurements, stabilise body weight and educate participants on how to monitor their dietary intake. This allowed participants to serve as their own controls. Four participants withdrew from the trial. At baseline 2, two participants failed to comply with study protocol and withdrew from the study because lack of interest. Nurses failed to locate the veins on the third participant to conduct intravenous glucose tolerance test (IVGTT), and he was asked to withdraw from the trial. The fourth participant was diagnosed with a grade IV pressure ulcer after being admitted into the trial. Following the delayed entry period, participants were randomly assigned to a RT+TRT group (n=13) or a TRT group (n=13). TRT patches (2–6 mg/day) were replaced daily on alternating shoulders at bedtime for 16 weeks. The RT+TRT group received 16 weeks of supervised unilateral progressive RT using surface NMES and ankle weights. Following the intervention the two-day assessment period was repeated (figure 1).

Participants and eligibility criteria

Participants were men between 18 and 50 years old with a BMI of ≤30 kg/m2. The upper limit of age was set to 50 years to avoid unanticipated side effects that may result from TRT. Participants had motor complete SCI C5-L2, ASIA A or B. Participants with pre-existing medical conditions were excluded. These included cardiovascular disease, uncontrolled type II diabetes and those on insulin, pressures sores stage 2 or greater, supra-physiological T level, haematocrit above 50% and urinary tract infection or symptoms.


Resistance training

The first week of RT was conducted with no ankle weights to ensure that the knee extensor muscles could extend the weight of the lower leg against gravity. Once full knee extension was achieved in a sitting position, two pounds were added on a weekly basis with the criteria that full knee extension was achieved before more weight was added. Surface NMES was applied to the knee extensor muscles via surface electrodes (figure 2). One electrode was placed 2–3 cm above the superior aspect of the patella over the vastus medialis muscle, and the other lateral to and 30 cm above the patella over vastus lateralis muscle. Current from the stimulator was manually increased in 5 s intervals to evoke full knee extension with a 3-min rest between sets, 30 Hz, 450 µs pulses and a current sufficient to evoke full knee extension as previously described.64 ,66 ,67 Four sets of 10 repetitions were performed twice weekly for 16 weeks.

Testosterone replacement therapy

Following baseline measurements, T was administered by patches (Androderm, Watson Pharma, Parsippany, New Jersey) that delivered between 2 and 6 mg/day53 (figure 2). Serum T concentration was measured and reviewed in a blinded fashion weekly for the first month and then every 4 weeks by an endocrinologist. Baseline dose was prescribed according to the initial T level. A dose of 6, 4 or 2 mg/day was initially prescribed if the serum baseline T level was <300, 300–600 or above 600, respectively. During the course of the study, the dose was decreased to 2 mg/day if the serum T concentration was more than 1000 ng/dL (34.7 nmol/L) and the participant was re-educated about the use of the patch if the concentration was <250 ng/dL (8.7 nmol/L) above the pretreatment concentration. Patches were returned after use to ensure adherence to the intervention protocol. Participants were instructed to place patches at bedtime and only remove them during showering. If skin irritation became an issue, participants were initially advised to move patches up or down on the shoulder muscles from the irritation site and if the situation was not resolved, a hydrocortisone cream was prescribed.

Detraining after 16-week intervention

Six participants from each group were followed for 16 weeks after the initial intervention. The RT+TRT group continued training once weekly using the same training approach for an additional 16 weeks. For the first week, the maximum ankle weights attained during the training phase were used. The weights were then gradually decreased by two pounds per week until the lowest weight was attained (2 pounds). TRT dose was set at 2 mg/day for the entire detraining phase. Participants from the TRT group were followed for additional 16 weeks without intervention. Following the detraining phase, the two-day assessment period was repeated without performing skeletal muscle biopsy (figure 1). The rationale of the detraining phase is to determine whether once weekly training can maintain skeletal muscle hypertrophy, body composition and metabolic improvements incurred by the 16-week intervention.

Primary outcomes

Data will remain confidential at all times, and any patient identifiers will be removed prior to data analysis. Analysis for all study procedures will be performed in a blinded fashion ensuring full concealment until complete data analysis.

Anthropometrics and body composition assessments

Height of each participant was determined while lying in the supine position. Two smooth wooden boards were placed at the participant's head and heels, and the distance between them was measured to the nearest centimeter. Measurement of waist circumference was determined in triplicate by identifying the narrowest region of the trunk from sitting and supine positions. Three-site skin fold assessment was conducted in triplicate for suprailiac, abdominal and thigh.

A Lunar Prodigy Advance (Lunar, Madison, Wisconsin) bone densitometer was used to measure total body and regional (lumbar spine, proximal femur and forearm) FM and FFM. Testing was performed after lower extremity elevation for at least 20 min to minimise fluid shift. The participant was assisted to lie on a padded table, and both legs were strapped proximal to the knees and ankles. The arms and legs were positioned to ensure proper alignment.

Magnetic resonance imaging

MRI was performed at the VAMC Hospital using a 1.5 Tesla magnet (GE) as previously described.4 ,19 ,59 ,68 ,69 Transaxial images, 10 mm thick and 10 mm apart, were taken from the hip joint to the knee joint and from knee to the ankle using the whole body coil. The location of the scan was identified by placing a mark 6 inches proximal to and distal to the patella and matched on follow-up scans. To analyse VAT and subcutaneous adipose tissue (SAT) transverse slices (0.8 cm thickness) were acquired every 0.4 cm gap from the xyphoid process to the femoral heads. Images were acquired in series of two stacks with L4–L5 used as a separating point. TRT patches were removed 48–72 hours prior to MRI scans to avoid skin burn.

Analyses will be performed using commercial available software (X-vessel) as previously described.4 ,19 ,68 ,69 Briefly, the thigh and leg images will be segmented into fat (high intensity), skeletal muscle (mid intensity) and background/bone (low intensity). Manual selection of a pixel of skeletal muscle will highlight all skeletal muscle pixels and provide the total number of skeletal muscle pixels while excluding fat. VAT and SAT will be measured by manually tracing around the anatomical borders. The number of pixels in the highlighted region will be multiplied by the matrix size to measure VAT and SAT CSA (cm2).

Skeletal muscle torque and specific tension

Torque of the knee extensor muscle group was evaluated using a Biodex isokinetic dynamometer (Shirley, New York). Measurements were performed 72 hours after the muscle biopsy to prevent acute effects on protein expression. Participants were seated with the trunk-thigh angle and the knee-thigh angle at 90°. Each participant was securely strapped to the test chair by a crossover shoulder harnesses and a belt across the hip joint. The axis of the dynamometer was aligned to the anatomical knee axis, and the lever arm was attached 2–3 cm above the lateral malleolus. Before measuring isometric torque, passive tension of the right knee extensor muscle group was measured at 5°, 30°, 60°, 90°,180°, 270°/s as an index of spasticity. Isometric torque was measured using current amplitude of 50 and 100 mA at a frequency of 30 Hz and a pulse duration 450 µs. Isokinetic torque was measured at 60°, 90° and 180°/s using the same stimulation protocol.

Serum and plasma analysis

All metabolic profile analysis is presented in table 2. Blood samples were collected after an overnight fast. Total T was measured by liquid chromatography with isotope dilution mass spectrometry detection after supported liquid extraction. Free T concentration was calculated using sex hormone binding globulin and albumin concentrations ( Serum IGF-I concentration was measured by an immunoluminometric assay (Quest Diagnostics, Madison, New Jersey, USA). Fasting lipid profiles (HDL-C, LDL-C, total cholesterol and triglycerides) were determined as previously described.8 ,20 Inflammatory biomarkers CRP, IL-6, TNF-α and free fatty acids were determined by commercially available ELISA kits (ALPCO; Salem, NH).

Table 2

Metabolic health variables measured at baseline 1, baseline 2, postintervention 1 and postintervention 2

Energy expenditure

After an overnight fast for 10–12 hours, participants were kept in a dark room for 20–30 min to attain a resting state during which BMR was measured as previously described.8 Briefly, while in a supine position a canopy was placed over the participant's head. Each participant was allowed 2–3 min before starting the test to ensure no signs of apnea or claustrophobic episodes. All participants were instructed to stay awake during the entire test and to breathe normally. The canopy was then attached to a vacuum to draw the expired gases to the flow meter of the metabolic unit (COSMED KB42). Prior to the test, the metabolic unit was calibrated using the standard procedures identified by the manufacturer. Carbon dioxide and oxygen output was used to calculate the respiratory exchange ratio, and BMR (kcal/day) was calculated using the average of the last 15 min of the test. This was used to measure the percentage of substrate usage (% fat vs % carbohydrate).8 ,71

To determine whether NMES-RT improves exercise performance, testing was performed using a FES bike (Restorative Therapies, RTI-300) against progressive resistance protocol until fatigue. The protocol started with 3 min resting, 3 min warm-up (35–37 RPM) using the servomotor and then a 2 min incremental progressive resistance protocol (1, 3, 5 N m, etc) until fatigue. After fatigue, a 1-min cool down period was allowed followed by 5 min of rest. Energy expenditure and cardiovascular performance (VO2 (L/min), blood pressure and heart rate) was collected at baseline 2 and postinterventions 1 and 2.

Each participant met with a dietician at the start of the study and was asked to maintain a 5-day food dietary log monitoring their caloric intake for the duration of the study. Participants were instructed to record all liquid and food consumption, and no nutritional advice was given on the size or the portion of the food. Dietary logs were analysed on a weekly basis using a nutritional software package (Nutrition Data System for Research V.2014) under the supervision of a registered dietitian. After analysis was completed, the average caloric intake (kcal) and percentage macronutrients (carbohydrates, fat and protein) were calculated. Each participant received monthly feedback via phone call with the registered dietician on how to maintain appropriate dietary habits based on his BMR and percentage macronutrients (45% carbohydrates, 30% fat and 25% protein). Every effort was made to balance the dietary habits between the RT+TRT and the TRT group.

Intravenous glucose tolerance test

An IVGTT was used to determine insulin sensitivity and glucose effectiveness before training and 48 hours after the last exercise bout. After an overnight fast, an intravenous line was placed to facilitate infusion of glucose and blood sampling. Blood samples were taken every 2–3 min before and after glucose injection (0.3 g/kg intravenously over 30 s), followed by 5–10 min sampling ending at 180 min. Twenty minutes after the glucose injection, a bolus of insulin (0.02 U/kg) was injected to determine insulin sensitivity. Plasma glucose was measured by the AutoAnalyzer glucose oxidase method, and plasma insulin concentrations were determined by commercial radioimmunoassay (table 2). The glucose disposal rate per unit of secreted insulin per unit time and glucose-mediated glucose disposal rate were calculated from a least-squares fitting of the temporal pattern of glucose and insulin throughout the IVGTT using the MINMOD program.72 A representative analysis of IVGTT is presented in figure 4. The acute insulin response to intravenous glucose was calculated as the mean rise in plasma insulin above baseline at 3, 4 and 5 min after intravenous glucose administration. KG, a measure of glucose tolerance, was calculated as the least square slope of the natural log of absolute glucose concentration between 5 and 20 min after the glucose bolus.73 The homeostatic model of assessment of insulin resistance (HOMA-IR) was calculated, and insulin sensitivity was determined using the Matsuda and Defronzo formula.73 ,74

Figure 4

A representative figure of analysis for IVGTT in a person with SCI after infusion of dextrose followed by insulin 20 min later. The dextrose concentration is calculated by multiplying 0.3 g × body weight (kg) in 50% solution. Insulin concentration is determined by multiplying 0.02 units × body weight (kg). Blue line represents fasting glucose concentration, resting, following infusion and over 120 min. Red line represents the line of best fit of glucose concentration. Green line represents fasting insulin level, spike following 20 min of dextrose infusion and over 120 min. Black line represents the line of best fit for insulin concentration. IVGTT, intravenous glucose tolerance test.

Secondary outcomes

Skeletal muscle biopsy

Biopsy samples of vastus lateralis muscle (∼50–100 mg wet weight total) were obtained by a 14-gauge tru-cut biopsy needle, immediately prior to and 72 hours after the 16 weeks of RT+TRT or TRT interventions. The 72 hours postintervention was to avoid any acute effects from the last training bout on muscle protein expression to ensure that changes are due to training effect. There was no muscle biopsy during the detraining phase. The biopsy samples were quickly frozen in liquid nitrogen and stored at −80°C until further analysis. One sample was split into two halves and used for measuring activities of mitochondrial enzymes. The second sample was used for western blot analysis. The third sample was used for immunohistochemistry.

Mitochondrial electron transport chain activities

Electron transport chain enzyme activities were measured spectrophotometrically in skeletal muscle homogenates as previously described.75 Rotenone-sensitive NADH cytochrome c reductase measured complexes I and III. Decylubiquinol-cytochrome c oxidoreductase was measured as the antimycin-sensitive reductase to assess complex III. Citrate synthase was measured as an estimate of mitochondrial mass as previously described.75

Protein content

Proteins were resolved by SDS-PAGE then transferred to a polyvinylidene diflouride (PVDF) membrane (Trans-blot; Bio-Rad). Equal protein loading was confirmed with a Ponceau S stain after the transfer. After blocking for 1 hour, membranes were incubated overnight at 4°C with primary antibody diluted 1:1000. Primary antibodies included glucose transpoter-4, focal adhesion kinase, PGC-1α (Santa Cruz Biotechnology), total protein kinase B (AKT), phosphorylated AKT, total mammalian target of rapamycin (mTOR) and phosphorylated mTOR (Cell Signaling). Membranes were rinsed and placed in a 1:2000 solution of a horseradish peroxidase-conjugated secondary antibody (Cell Signaling) for 1 hour at room temperature. Membranes were then rinsed and incubated with a horseradish peroxidase chemiluminescence detection reagent (GE Healthcare) for 5 min. Proteins were visualised using an Amersham Imager 600 (GE Healthcare). Optical densities were measured using iQuant software, and all samples were normalised to the baseline values for that participant.

Histological analysis

Immediately after muscle biopsy, samples were mounted on tongue blades by using a medium of OCT compound and tragacanth gum and stored at −70°C until analysis. Serial cross sections (8–10 µM) were collected on glass slides and frozen at −20°C until analysis. Fibre type and CSA will be determined by histochemical staining for myosin ATPase (preincubation at pH 4.3 or 9.4) as previously described.76 Type I fibres will be identified by dark staining after acid preincubation, type IIA fibres light staining and type IIX intermediate. At pH 9.4, the staining pattern was the opposite. H&E staining was performed according to conventional histological procedures. Mitochondrial complex II and IV activity was estimated by the activity of succinate dehydrogenase and cytochrome c oxidase activity, respectively, as previously described.77 ,78 Stained muscle sections will be observed using an Olympus BX-51 fluorescent microscope (Olympus, Tokyo, Japan) and analysed using ImageJ software.

Statistical analyses

Paired t-tests will be used to determine differences in body composition and metabolic profile between baselines 1 and 2. To determine the effect of interventions and detraining, a repeated measures ANOVA will be used. Simple linear and multivariate regression analyses will be used to examine the relationship between body composition and metabolic profile variables. For protein expression, we will use paired t-tests to examine the effects of each intervention.

The effect size was calculated based on the effects of RT on body composition and metabolic profiles that were previously published.62 The number of participants necessary to find statistical differences in the major variables (muscle size, VAT and insulin concentration) of this study was found to be 10 participants per group. One participant from the RT+TRT group withdrew after developing side effects to TRT patches, and the medical monitor personnel recommended him to withdraw from the trial in week 8. However, we will perform intent-to-treat analysis on his data, which means that despite his early withdrawal from the study his data will be included in the final analysis. This will allow extrapolation of his postintervention data using the SPSS missing values option. We anticipate that we will collect sufficient data to determine the effects of rehabilitation interventions on protein expression, mitochondrial enzymatic and mitochondrial electron transport chain activities in individuals with SCI. Statistical analysis will be performed using SPSS V.23.0 (Chicago, Illinois, USA) with a level of significance set at p<0.05.


All participants read and signed approved consent forms prior to baseline assessment. Results of the study will be published in a peer-reviewed journal and presented at national and international conferences.

Data monitoring

The research staff oversaw and monitored the study to ensure data quality and participant compliance. There were no adverse events. Only members of the research team will have access to data.


Individuals with SCI experience profound skeletal muscle atrophy, deterioration in body composition and abnormal metabolic profile. Within few weeks of injury, there is a significant decrease in whole body FFM, particularly lower extremity skeletal muscle mass, and a subsequent increase in FM. These changes predispose this population to the risk of glucose intolerance, insulin resistance, dyslipidaemia and the development of type II diabetes and cardiovascular disease. The main purpose of this study is to investigate the effects of 16 weeks of evoked RT+TRT or TRT alone on body composition (muscle CSA, IMF, VAT, FM, FFM) and metabolic profile (glucose, lipid and BMR) in individuals with motor complete SCI.

Ectopic adipose tissue accumulation, IMF and VAT, has been strongly associated with altered metabolic profile after SCI.19 ,20 IMF has been determined to account for a 70% reduction in glucose tolerance in individuals with complete SCI.19 VAT is independently associated with impairment in glucose tolerance, insulin resistance and dyslipidaemia after SCI.20 Edwards et al noted significant positive association between VAT and insulin resistance and a negative correlation between VAT:SAT ratio and HDL-C.13 An increase in VAT is also related to leptin and plasminogen activator inhibitor-1 concentrations.14 It is possible that increases in FM are associated with inflammatory biomarkers that trigger metabolic syndrome after SCI.22 Ectopic adipose tissue has been demonstrated to secrete proinflammatory cytokines, including IL-6 and TNF-α. This stimulates hepatic production of CRP which is suggestive of vascular inflammation.15–19

RT is an important type of exercise that has been shown to induce positive physiological adaptations such as increasing lean mass and reducing the incidence of metabolic disorders in other clinical populations. Previous work suggests that twice weekly NMES-RT can induce favourable body composition and metabolic adaptations. Twelve weeks of NMES-RT has shown to increase thigh muscle CSA by 35–40% as measured by MRI.59 Moreover, there was a reduction in %leg FM and a trend towards decrease in VAT CSA after 12 weeks of NMES-RT.64 The favourable adaptations in body composition were associated with a decrease in plasma insulin area under the curve and plasma triglycerides.59 These findings were attributed to an increase in plasma IGF-1. However, the effect of training appears to be limited to the trained muscle and only modestly impacted whole body composition. It is unclear whether a RT programme longer than 12 weeks may provide additional benefits to individuals with SCI.

The cellular changes underlying the alterations in skeletal muscle glucose usage and energy metabolism after SCI are unclear. Previous work indicated that expression of AMP-activated protein kinase (AMPK), a key regulator of energy homeostasis for lipid and carbohydrate usage, was altered in persons with SCI compared to BMI-matched able-bodied controls.79 Another study revealed decreased expression of genes involved in glucose and lipid metabolism.80 Despite these abnormalities, one study reported that leg glucose uptake during cycling was increased in individuals with SCI compared to able-bodied controls.61 Another study showed similar glucose uptake of isolated muscle fibres from SCI and able-bodied individuals in vitro.81 Benefits of other forms of functional electrical stimulation lower extremity cycling (FES-LEC) have included improvements in body composition, carbohydrate and lipid metabolism and muscle fibre type composition.60–63 ,82–84 Similarly, we have recently shown that 16 weeks of FES-LEC increased the protein abundance of GLUT-4, PGC-1α and AMPK by 3.8-fold, 2.3-fold and 3.4-fold, respectively, in the vastus lateralis muscle in persons with motor complete SCI.85

TRT has been shown to increase skeletal muscle mass in hypogonadal men, men with chronic illness and older men.53–55 Androgen deficiency in men is associated with a loss of FFM and an increase in FM.53 In epidemiologic studies, men with decreased free T had lower appendicular skeletal muscle mass than those with normal T levels.86 Previous work documented that TRT increases muscle mass with a reciprocal decrease in total body FM.53 ,54 This reciprocal action has been suggested due to a switch from adipogenic towards myogenic differentiation of mesenchymal stem cells.86 In a randomised controlled double-blinded clinical trial, TRT was shown to improve insulin sensitivity, CRP and reduce VAT.54 Sixty percent of men with SCI have low T level, and levels are associated with time since injury.49 ,50 TRT has also been reported to increase IGF-1, and several molecular mechanisms related to the protective pathways have been recently elucidated.47 ,48 ,51 ,87 Therefore, enhancing the decline in anabolic homeostasis by providing TRT may provide additional benefits as was previously demonstrated by increasing lean mass and metabolic rate in individuals with SCI.

It was very important to highlight that this protocol has three different phases including 4 weeks of delayed entry, 16 weeks of intervention and 16 weeks of detraining. The delayed entry period was included to allow each participant to serve as his own control. Moreover, we were successful in retaining 12 participants (n=6/group) to complete the detraining phase. This means that we had six participants who agreed to stick to our exercise programme and the use of TRT patches up to 9 months. This may reflect on the study protocol frequency that ensured long-term adherence despite the length of the study. A very important point that is worth highlighting is that our study protocol was designed to include three levels of outcome measurements including body composition, metabolic profile and cellular changes. This design is likely to provide mechanistic explanations to changes that occur at the body composition and metabolic levels.


The current study was limited to those who were ≤50 years old. Because of advances in healthcare, many individuals with SCI have a near-normal life span which may make the results of this study less generalisable. However, this age limit was implemented because the effect of TRT on cardiovascular health has been controversial. There are data showing that hypogonadism is a risk for cardiovascular disease.88 Some replacement studies show increased risk, but another study showed decreased mortality in men receiving testosterone.89 Another study in older men reported that injectable testosterone may be associated with increased cardiovascular risk but topical testosterone was not.90 Therefore, we set the inclusion criteria of ≤50 years old to reduce the likelihood of developing cardiovascular complications.

Currently, NMES-RT is not readily available to the majority of SCI patients. Women were not included because administering TRT is not either appropriate or safe, because women are at risk of virilisation by testosterone. Thus, the trial was limited only to males with SCI. Moreover, only 21% of individuals with motor complete SCI are women based on the Model System Data.

In summary, we anticipate that this trial will provide important insights into the body composition and metabolic benefits of 16 weeks of evoked RT+TRT or TRT. If beneficial, this may be a feasible strategy for the rehabilitation of individuals with chronic SCI and increase the health of this and other clinical populations. Additionally, TRT alone may provide an alternative intervention for those who cannot benefit from training using surface NMES, because of LMN denervation or intolerance to applications of electrical stimulation. Although the current study did not include any cauda equina participants, further research is warranted to examine the effects of TRT on muscle atrophy following LMN denervation. This is important because skeletal muscle following LMN denervation may respond differently than innervated muscle. The study will also shed light on several molecular pathways that have been suggested to influence body composition and metabolic profile.


The authors thank Rodney Wade, Sakita Sistrun and Liron Segal for helping in the processes of data capturing and analysis, as well as the study participants for their time and effort. They are thankful to Hunter Holmes McGuire Research Institute and Spinal Cord Injury Services and Disorders for providing the environment to conduct clinical human research trials. They are also thankful to VCU Clinical Research Service for providing space and staff to perform the blood work procedures. The work is supported by the Department of Veteran Affairs, Veteran Health Administration, Rehabilitation Research and Development Service (B7867-W).


View Abstract


  • Contributors ASG supervised all aspects of the trial including all interventions and measurements and secured funding for the trial procedures. ASG, REK and LCO drafted the manuscript. REK, JS, RK and LCO contributed to data collection and analysis. RG, TL, TC, DXC, CC, RAA, EJL and DRG are research physicians who contributed to patient monitoring and study design. DXC and RAA will provide guidance during data analysis and manuscript preparation. All authors affirm that the manuscript is an honest, accurate and transparent account of the study being reported; that no important aspects of the study have been omitted and that any discrepancies from the study as planned have been explained.

  • Funding This work was supported by VA-RRD CDA2 (to ASG).

  • Competing interests None declared.

  • Ethics approval Hunter Holmes McGuire VA Medical Center and VCU.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Trial status Enrollment into the study started in July 2012 and as of April 2016 all participants have completed the study. Data collection and data analysis are expected to be completed in December 2016. The study is expected to be closed in June 2017.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.