Advertisement

Methodological Considerations Which Could Improve Spinal Cord Injury Research

  • Gabriel Zieff
  • Sabina Miller
  • Daniel Credeur
  • Lee StonerEmail author
Commentary
  • 122 Downloads

Abstract

Advances in spinal cord injury-based research in the last 50 years have resulted in significant improvements to therapy options. However, the efficacy of such research could be further enhanced if threats to internal and external validity were addressed. To provide perspective, a sample topic was identified: the effects of acute and chronic exercise on clinical and sub-clinical markers of cardiovascular health. The intention was not a systematic review, nor a critique of exercise-based research, but rather a means to generate further discussion. Thirty-one articles were identified, and four common issues were found relating to: (1) sampling; (2) study design; (3) control group; and (4) clinical inference. These concerns were largely attributed to insufficient resources, and challenges associated with recruiting individuals with spinal cord injury. Overcoming these challenges will be difficult, but some opportunities include: (1) implementing multi-center trials; (2) sampling from subject groups appropriate to the research question; (3) including an appropriate control group; and (4) clearly defining clinical inference. These opportunities are not always feasible, and some easier to implement than others. However, addressing these concerns may assist in progressing spinal cord injury-based research, thereby helping to ensure steady advancement of therapy options for persons with spinal cord injury.

Keywords

Internal validity External validity Generalizability Clinical inference 

Introduction

Prior to the 21st Century, a traumatic spinal cord injury (SCI) was seen as a terminal injury with limited treatment options, and poor prognosis [21]. In fact, during World War I, 80% of individuals suffering a SCI died within the first 2 weeks of injury [21]. It was not until the 1950s that the prognosis for people with SCI improved, concomitant with improvements in therapy and rehabilitation [21]. Arguably, while SCI-based research has substantially contributed to improvements in therapy, the efficacy of such research could be further enhanced if a number of threats to internal and external validity were addressed. Internal validity refers to “the validity of inferences as they pertain to the members of the source population.” [27] If the study is properly conducted without major methodological problems (i.e. has high internal validity) [27], external validity refers to “the validity of the inferences as they pertain to people outside that [specific] population.” [27] External validity is often referred to as generalizability, and in the context of SCI research, refers to the SCI population as a whole.

In order to provide perspective regarding the threats to internal and external validity in SCI research, articles were reviewed based on a sample topic: the effects of acute and chronic exercise on clinical and sub-clinical markers of cardiovascular health. Clinical and sub-clinical markers of cardiovascular health can be defined as factors affecting cardiac, vascular, inflammatory, and autonomic markers that are known to affect cardiovascular health outcomes, such as stroke or myocardial infarction. The intention is not a systematic review, nor a specific critique of exercise-based research; rather, the intention of this brief report is to generate further discussion. The exemplar topic was chosen for several reasons. First, while discussion of suitable tools for monitoring cardiovascular health in the SCI population is beyond the scope of the article (readers are referred to the review by Stoner et al.) [31], the current authors have extensive expertise in assessing cardiovascular health, including SCI populations [31]. Second, and more importantly, cardiovascular complications are a major concern in the SCI population due to a confluence of factors including lower limb immobility, sedentary behavior, and impaired autonomic function [7, 8, 17, 36]. Thus, cardiovascular outcomes (ranging from blood pressure to arterial stiffness) known to impact morbidity, mortality, and quality of life in SCI persons were included in this review and are described in Table 1. Inclusion criteria were specified relative to PICO (Population, Intervention, Control, Outcome), and consisted of: (1) population: SCI; (2) intervention: acute or chronic exercise; (3) comparison (control): optional; (4) outcomes: cardiovascular; published between April 2006 and January 2018. The search was performed using PubMed and the broad search terms: “spinal cord injury[Title] AND (exercise[Title] OR physical activity[Title]), coupled with a manual search of reference lists. Following careful examination, 31 articles met the inclusion criteria and were described in Table 1. Four common issues were identified, relating to: (1) sampling issues; (2) inadequate study design; (3) the lack of an appropriate control group; and (4) limited clinical inference. Of note, four of the 31 articles (13%) included were case studies [28, 29, 37, 44]. To acknowledge the methodological strengths and limitations of case-studies, the prevalence of the commonly identified issues are reported both as a percentage of all 31 studies, as well as relative only to the 27 non-case studies.
Table 1

Summary of included studies

Study

Intervention

Design

Participants

Control condition

Control matched

Sample method

Power calculation

Clinical Inference

CV Outcome

Results*

Alves

3 bouts: VT1, 15% below VT1, 15% above VT1; 24−34 min,

Quasi-experimental

18 (9 WC basketball SCI)

9 AB

Not available

Not available

Not available

Not available

Glucose, lipids

Inc in TG, LDL post-exercise in SCI vs. CON

Bakkum

Hybrid cycle, hand cycle 2 times/week 16 weeks

RCT

19 SCI

Hybrid vs. ACE

Sex, age, injury duration, lesion level, height, mass, BMI, PASID

Convenience

Not available

Not available

BP, lipids, cytokines

Dec DBP, CRP, and IL-6/10 ratio

Brurok

7 weeks CON period ADL followed by 8 weeks, 3 times/week EXP period of HIIT

Cross-over

6 SCI

N/A

Crossover

Convenience

Not available

Not available

SV

Dec SBP post-intervention

Claydon

Acute graded ACE

Quasi-experimental

27 SCI (19 cervical SCI, 8 thoracic SCI)

N/A

Not available

Not available

Not available

Not available

BP

Inc MAP in thoracic SCI but dec in cervical SCI

Currie

Max cycling, sit-up test

Quasi-experimental

21 SCI

8 Non-athletic SCI

Injury duration, sex

Convenience

Not available

Not available

BP

Non-athletic group had greater dec in SBP and DBP after sit-up test

Gorgey

5 days/week 16 weeks arm cycling or FES

RCT

11 SCI

5 SCI no exercise

Age, sex, waist, hip, and abdominal circumferences

Convenience

Not available

Not available

HDL, LDL, TC, TG

No change

Griffin

FES cycling, 2−3 times/week, 10 weeks,

Quasi-experimental

18 SCI

N/A

Not available

Convenience

Not available

Not available

Lipids

Dec IL-6, TNF-α, and CRP after intervention

Han

30 min, 3 times/week, 12 weeks, at AT

Quasi-experimental

27 (11 SCI)

16 AB

Age, sex

Convenience

Not available

Not available

BP

No change

Hasnan

Acute ACE, FES-LCE, combined ACE + FES-LCE, and hybrid FES tricycle

Cross-over

9 SCI

N/A

Not available

Convenience

Not available

Not available

Q, SV

Greater dec in Q during FES-LCE vs other modalities

Horiuchi

4 sessions/week, 10 weeks; 25 min ACE

Quasi-experimental

9 SCI

9 AB (Only assessed at baseline)

Age, height, BMI

Not available

Reports power calculation

Partial eta squared

PAI-1, BP, lipids

Dec SBP, PAI

Kinoshita

W/C basketball game

Quasi-experimental

5 SCI W/C Players

N/A

Sex

Convenience

Reports power calculation

Not available

IL-6, TNF-α, CRP

Inc IL-6 after game

Kouda

20 min ACE at 60% VO2max

Quasi-experimental

8 SCI

8 AB

Not available

Convenience

Not available

Not available

IL-6, PE2

SCI: no change in IL-6 wit exercises, inc IL-6 in AB

Lee

Graded ACE

Quasi-experimental

23 (15 SCI)

8 AB

Not available

Not available

Not available

Not available

BP

MAP lower in non-active SCI group vs. active SCI group

Machac

Max ACE

Quasi-experimental

47 (20 SCI)

27 AB

Race, age, weight, %BF

Not available

Not available

Cohen’s d

BP

Inc in AB BP, but not SCI BP

Mitsui

2 h ACE 60% VO2

RCT

17 (9 SCI)

8 AB

Not matched

Convenience

Not available

Not available

LDL

Inc in AB LDL, but no change in SCI LDL

Rosety-Rodriguez (2015)

ACE mod intensity 12 weeks; 3 sessions/week

RCT

17 sedentary SCI

8 non-exercise

Age, duration of injury, anthrop, fitness, lesion level

Not available

Not available

Not available

Endo-1, sICAM-1, sVCAM-1

Dec endothelin-1 and SICAM-1

Rosety-Rodriguez (2014)

ACE mod intensity 12 weeks; 3 sessions/week

RCT

17 sedentary SCI

8 non-exercise

Age, duration of injury, anthrop, fitness, lesion level

Not available

Not available

Cohen’s d

TNF-α, IL-6

Dec TNF-α and IL-6 after intervention

Sasso

12 weeks home-based RT and aerobic exercise

Case study

1 SCI (male)

N/A

Not available

Not available

Not available

Case study

TC

No change

Silva

4 sessions: max ACE, anodal tDCS,, sham tDCS, or control

Case study

1 SCI (male)

N/A

Not available

Not available

Not available

Case study

HRV

Inc time to HRV threshold with anodal tDCS

Spengler

Acute ACE and W/C exercise

Quasi-experimental

18 males (9 SCI)

9 AB

Age, weight, physical activity

Not available

Reports power calculations

Not available

SV

Max Q in AB higher than SCI

Stoner

18 weeks, 2 times/week, LB RT

Quasi-experimental

5 SCI

N/A

Not available

Not available

Not available

Not available

FMD

Inc FMD

Ter Woerds

10 min passive LCE

Quasi-experimental

16 (8 SCI)

8 AB

Not available

Not available

Reports power calculation

Not available

BP

No changes

Thijssen

6 weeks, 2 times/week Hybrid FES cycling

Quasi-experimental

7 SCI

N/A

Not available

Not available

Reports power calculation

Not available

FMD

Dec FMD at 2 weeks

Tordi

6 weeks, 3 times/week, 30 min wheelchair exercise

Case study

1 SCI (male)

N/A

Not available

Not available

Not available

Case study

PWV

Dec upper and lower limb PWV

Totosy

16 weeks, 2 times/week, ≥ 20 min AE and 8–10 repetitions UB RT

RCT

17 SCI

Normal physical activity

Not available

Not available

Not available

Cohen’s d

PWV, FMD, Dist

Inc carotid dist in EXP group

Turiel

6 weeks, body-weight supported treadmill training

Quasi-experimental

14 SCI

N/A

Not available

Not available

Not available

Not available

LVFx, CFR, FMD, CRP

Improved LVFx, CFR, and FMD; dec CRP

Umemoto

2 h 60% ACE

Quasi-experimental

13 (6 SCI)

7 AB

Age, anthrop, cardiorespiratory fitness

Not available

Not available

Not available

Il-6, TNF-α, CRP

Inc IL-6 in both groups

Vasiliadis

30 min 60% ACE

Quasi-experimental

16 (8 SCI)

8 AB

Age, anthrop, HR, MAP

Not available

Not available

Partial eta squared

VEGF-A, sVEGFr-1, VEGFr-2, MMP-2, Endo

Inc VEGF, sVEGFr-1, sVEGFr-2, MMP-2 and endo in both groups

Wecht

90 min peak LCE

Quasi-experimental

18 SCI (9 fit, 9 unfit)

9 Unfit

Age, anthrop, injury duration

Not available

Not available

Not available

HRV

Inc HF recovery, dec LF recovery and LF/HF ratio in fit vs. unfit

West

6 weeks UB RT and arm/passive leg cycle

Case Study

1 SCI (male)

N/A

Not available

Not available

Not available

Not available

IMT, SV

10% inc in SV

Zbogar

3 months 3 times/week 30 min FES-LCE

Quasi-experimental

4 SCI

N/A

Not available

Not available

Not available

Not available

Comp

Inc in small artery comp

AB Able-bodied, ACE Arm cycle ergometer, ADL Activities of daily living, AT Anaerobic threshold, BMI Body Mass Index, BP Blood pressure, CON Control (group), CRP C-Reactive Protein, DBP Diastolic blood pressure, EDD End diastolic diameter, ESD End systolic diameter, ESV End systolic volume, EXP Experimental(group), FES Functional electrical stimulation, HIIT High intensity interval training, HRR Heart rate reserve, HRV Heart rate variability, IMT Intima-media thickness, LDL Low-density lipoprotein, LCE Leg cycle ergometer, MMP-2 Matrix metalloproteinase-2, PAI-1 Plasminogen activator inhibitor-1, PASID Physical activity scale for individuals with disability, PWV Pulse wave velocity, Q Cardiac output, RCT Randomized control trial, SBP Systolic blood pressure, SCI Spinal cord injury, sICAM-1 Soluble intercellular adhesion molecule-1, SV Stroke volume, sVCAM-1 Soluble vascular adhesion molecule-1, sVEGFr-1 Soluble vascular endothelial growth factor receptor-1, TC Total cholesterol, tDCS Transcranial direct current stimulation, TG Triglycerides, TNF-α Tumor necrosis factor alpha, UB Upper body, VEGF-A Vascular endothelial growth factor-A, VEGFr-2 Vascular endothelial growth factor receptor 2, VT1 Ventilatory threshold 1, W/C Wheelchair

Sampling Issues

Only 16% (n = 5) of all of the studies (and 19% of non-case studies), all of which had sample sizes less than 20 individuals, were based upon a power calculation [15, 18, 30, 34, 35]. Three of these studies (60%) reported an a priori power calculation [18, 34, 35]. Further, only three (60%) reached sufficient statistical power [30, 34, 35]. Considering that the majority of studies did not report a power calculation, an arbitrary sample size of 20 was used to compartmentalize the literature into “small” and “large” samples. However, it should be clearly stated that cardiovascular outcome effect sizes may be large in SCI populations, and that small samples may yield adequate power. Our search revealed that 84% (n = 26) of all 31 studies studies [1, 4, 9, 10, 14, 15, 18, 19, 23, 25, 26, 28, 29, 30, 32, 33, 34, 35, 37, 38, 39, 40, 41, 43, 44, 46], and 81% of non-case studies, had sample sizes fewer than 20 individuals. Small sample sizes in research result in decreased statistical power, which increases the likelihood of a type II error and limits internal validity (i.e., ability to detect differences as a result of intervention).

In addition to low sample size, convenience sampling is another issue potentially limiting SCI studies. A convenience sample is when participants are chosen based off of convenience, such as location or availability. With convenience sampling, individuals may not be representative of the larger population, thus, affecting both internal and external validity. Of the studies included, 32% (n = 10) used convenience sampling [3, 4, 6, 9, 10, 12, 14, 18, 19, 23] (37% of non-case studies) while the remaining 68% (n = 21) did not report their sampling method (63% of non-case studies) [1, 5, 15, 20, 22, 25, 26, 28, 29, 30, 32, 34, 35, 37, 38, 39, 40, 41, 43, 44, 46]. Collectively, low sample size and convenience sampling limit the generalizability of findings to the SCI population as a whole.

Many studies likely recruit a small sample, and use a convenience sample due, at least in part, to insufficient funding or resources, and challenges associated with recruiting people with SCI, which include environmental and socioeconomic barriers [2]. While not always feasible, one solution to address the sampling issues—and one which several SCI rehabilitation studies have utilized [13, 45], albeit they were not within our specific example topic, is a multi-center or multi-hospital study. Though multi-center studies have challenges of their own, such as variance in intervention standardization, outcomes, and clinical practices, carefully implemented multi-center studies would be useful in advancing SCI research. In particular, multi-center coordination provides a greater sampling pool for a study to pull participants from, which would in turn increase the generalizability of the study’s findings.

Inadequate Study Design

Randomized control trials (RCT) are considered the “gold standard” for clinical-based research. However, RCTs may not always be feasible or practical for SCI-based research. In particular, RCTs may disfavor the continuity of research due to concerns related to cost, and access to appropriate resources and a sufficient sample pool. However, when feasible, RCTs enable superior generalization to the general SCI population by ensuring that allocation to exposure is not determined by a third variable, which itself may influence the outcome [24]. Ideally, a RCT would include both random sampling and random allocation. Random sampling increases generalizability to SCI populations, while random allocation to the exposure tries to ensure that exposure status is independent of variables that might influence both exposure and outcome (confounders), assuming sufficient sample size.

Of the studies reviewed, only 19% (n = 6), and 22% of non-case studies, were classified as a RCT [3, 9, 23, 25, 26, 38]. That is, these 6 studies were the only ones in which participants were randomly exposed to a control or experimental condition. In contrast, the majority of the remaining studies were quasi-experimental and/or utilized cross-over designs. In these studies, order of experimental condition was not randomized, group placement was not randomized, or there was only a single group that was measured pre- and post-intervention. Ensuring control and experimental groups are both randomly sampled and allocated reduces possible threats to external validity, which in turn provides a sound basis for statistical inference [16].

Despite the challenges associated with conducting a RCT, several practices exist, which may aid in implementing this study design. First, the process of officially registering a study as a RCT may assist researchers by providing guidelines, benchmarks, and best practices for RCT implementation, particularly for those who may not have previously carried out RCTs. Second, officially registering the RCT may provide an investigator with a network of other researchers in the SCI field, who may be able to provide further support, guidance, or recommendations regarding conducting methodologically rigorous SCI research. Last, we highlight a succinct and informative paper [11] providing recommendations for RCT implementation, These recommendations are particularly relevant for instances in which an ideal study scenario is not feasible (e.g. outlines next best practices when study is not able to be double-blinded, which would likely be the case in SCI research).

Lack of an Appropriate Control Group

As discussed above, a RCT should accommodate a control group which is: (1) randomly sampled from the population of interest, and (2) randomly allocated. Further, to reduce the likelihood of confounding, and to increase the likelihood of generalizability, the control group and experimental group should ideally be as similar as possible with respect to important factors related to SCI (examples: age, sex, years post-injury, injury type, anthropometrics, etc.) [16]. It is important to note that barriers to SCI research participation as well as variance in clinical presentation make the prospect of having an adequate control group challenging. Nevertheless, some strategies include (1) matching groups by injury level (e.g. both groups have equal breakdown of individuals with cervical, thoracic, and lumbar injuries if both groups are to consist of SCI individuals), (2) ensuring both groups have similar level of baseline cardiovascular risk (particularly if control group consists of able-bodied individuals), and (3) using multiple research sites to increase the ability to select a large enough sample in which appropriate matching of experimental and control groups can occur.

Ultimately, valid control groups strongly effect inferences drawn from a study. Specifically, adequate control groups minimize the degree to which bias is involved in conducting and analyzing a study, affect the endpoints that can be studied, and ensure that the interventional effects are beyond normal variation to prevent under- or over-estimation of study effects [16]. Only 58% (n = 18) of the studies examined, and 67% of non-case studies, included a comparison group [1, 3, 6, 9, 12, 15, 20, 22, 23, 25, 26, 30, 33, 34, 38, 40, 41, 43]. Additionally, only 67% (n = 12) of studies that included a comparison group were appropriately matched (examples: age, injury duration, etc.) to the experimental group [3, 6, 9, 12, 15, 22, 25, 26, 30, 40, 41, 43].

Limited Clinical Inference

Clinical inference of study results refers to whether a “significant” result is clinically beneficial or harmful. The American Statistical Association recently declared that the concept of statistical significance (and associated use of P value) has been “commonly misused and misinterpreted.” [42] P values provide insight into whether or not the data are statistically significant within the context of the target population, i.e., the probability that the observed magnitude of difference for the sample is different from that of the intended target population. The P value does not provide information about clinical significance, i.e., whether the findings are meaningful [42]. As a hypothetical example, a study may find that a novel exercise therapy decreases systolic blood pressure by 1 mmHg, with a P value of 0.04. The sample size could be expanded, and the P value decreases to 0.001, but the systolic blood pressure decrease remains at 1 mmHg, and the effect size (e.g., partial eta squared) indicates a negligible effect. Using these findings, a literature search should be conducted to determine whether a 1 mmHg decrease is clinically meaningful, so that the research consumers can make an informed decision as to whether to apply valuable resources to this novel exercise therapy.

Of the studies reviewed, 71% (n = 22) of all of the included studies and 81% of the non-case studies relied on P-values for accepting or rejecting null hypotheses [1, 3, 4, 5, 6, 9, 10, 12, 14, 15, 18, 19, 20, 23, 25, 26, 30, 32, 34, 35, 38, 39, 40, 41, 43, 46], while only 16% (n = 5) of all included studies and 19% of non-case studies included additional measurements of clinical inference such as effect size, and attempted to infer whether the magnitude of change was beneficial or harmful (Table 1) [15, 22, 26, 38, 41]. Additionally, there was one example of a case study inappropriately reporting P values on a single case (i.e., sample size n = 1) [37]. Common measures of effect size include partial eta squared and Cohen’s d. Investigators are directed to a short review of effect size indices [33], which may increase awareness regarding the importance, and proper implementation of such indices. The practice of including statistical and clinical significance (when sample size allows) will allow scientists to better infer the clinical and practical viability of study findings, which in turn will assist in translating the findings.

Conclusion

The purpose of the current article was to examine threats to internal and external validity in SCI-based research. To provide context, articles were reviewed on a specific sample topic: the effects of acute and chronic exercise on subclinical markers of cardiovascular health outcomes in individuals with SCI. Four common issues were identified related to: (1) inadequate sampling; (2) inadequate study design; (3) lack of an appropriate control group; and (4) limited clinical inference. These concerns are likely attributable, at least in part, to insufficient funding or resources, and challenges associated with recruiting SCI patients. Overcoming these challenges will be difficult, but some opportunities include: (1) implementing multi-center trials; (2) sampling from patient groups appropriate to the research question; (3) including a control group appropriate to the research question; and (4) ensuring clinical inference. These are not absolute recommendations, but rather suggestions meant to stimulate further discussion. In line with this, while researchers and reviewers need to be aware that addressing each of the highlighted concerns may be idealistic and, at times, even inappropriate (e.g., multi-center trials may lead to inconsistent data collection if researchers have differing levels of training), addressing these methodological issues, when feasible, will assist in progressing SCI-based research, and help to ensure steady advancement of treatment options for persons with SCI.

References

  1. 1.
    Alves ES, Santos RV, Ruiz FS, Lira FS, Almeida AA, Lima G, Goni TC, Oyama L, Edwards KM, Tufik S, De Mello MT. Physiological and lipid profile response to acute exercise at different intensities in individuals with spinal cord injury. Spinal Cord Ser Cases. 2017;3:17037.  https://doi.org/10.1038/scsandc.2017.37.CrossRefPubMedPubMedCentralGoogle Scholar
  2. 2.
    Anderson KD, Cowan RE, Horsewell J. Facilitators and barriers to spinal cord injury clinical trial participation: multi-national perspective of people living with spinal cord injury. J Neurotrauma. 2016;33(5):493–9.  https://doi.org/10.1089/neu.2015.4064.CrossRefPubMedGoogle Scholar
  3. 3.
    Bakkum A, Paulson T, Bishop N, Goosey-Tolfrey VL, Stolwijk-Swüste JM, van Kuppevelt DJ, de Groot S, Janssen TW. Effects of hybrid cycle and handcycle exercise on cardiovascular disease risk factors in people with spinal cord injury: a randomized controlled trial. J Rehabil Med. 2015;47(6):523–30.  https://doi.org/10.2340/16501977-1946.CrossRefPubMedGoogle Scholar
  4. 4.
    Brurok B, Helgerud J, Karlsen T, Leivseth G, Hoff J. Effect of aerobic high-intensity hybrid training on stroke volume and peak oxygen consumption in men with spinal cord injury. Am J Phys Med Rehabil. 2011;90(5):407–14.  https://doi.org/10.1097/PHM.0b013e31820f960f.CrossRefPubMedGoogle Scholar
  5. 5.
    Claydon VE, Hol AT, Eng JJ, Krassioukov AV. Cardiovascular responses and postexercise hypotension after arm cycling exercise in subjects with spinal cord injury. Arch Phys Med Rehabil. 2006;87(8):1106–14.  https://doi.org/10.1016/j.apmr.2006.05.011.CrossRefPubMedGoogle Scholar
  6. 6.
    Currie KD, West CR, Hubli M, Gee CM, Krassioukov AV. Peak heart rates and sympathetic function in tetraplegic nonathletes and athletes. Med Sci Sports Exerc. 2015;47(6):1259–64.  https://doi.org/10.1249/MSS.0000000000000514.CrossRefPubMedGoogle Scholar
  7. 7.
    DeVivo MJ, Kartus PL, Stover SL, Rutt RD, Fine PR. Cause of death for patients with spinal cord injuries. Arch Intern Med. 1989;149(8):1761–1766. http://www.ncbi.nlm.nih.gov/pubmed/2669663.CrossRefGoogle Scholar
  8. 8.
    DeVivo MJ, Black KJ, Stover SL. Causes of death during the first 12 years after spinal cord injury. Arch Phys Med Rehabil. 1993;74(3):248–54.PubMedGoogle Scholar
  9. 9.
    Gorgey AS, Martin H, Metz A, Khalil RE, Dolbow DR, Gater DR. Longitudinal changes in body composition and metabolic profile between exercise clinical trials in men with chronic spinal cord injury. J Spinal Cord Med. 2016;39(6):699–712.  https://doi.org/10.1080/10790268.2016.1157970.CrossRefPubMedPubMedCentralGoogle Scholar
  10. 10.
    Griffin L, Decker MJ, Hwang JY, Wang B, Kitchen K, Ding Z, Ivy JL. Functional electrical stimulation cycling improves body composition, metabolic and neural factors in persons with spinal cord injury. J Electromyogr Kinesiol. 2009;19(4):614–22.  https://doi.org/10.1016/j.jelekin.2008.03.002.CrossRefPubMedGoogle Scholar
  11. 11.
    Guerrera F, Renaud S, Tabbò F, Filosso PL. How to design a randomized clinical trial: tips and tricks for conduct a successful study in thoracic disease domain. J Thorac Dis. 2017.  https://doi.org/10.21037/jtd.2017.06.147.CrossRefPubMedPubMedCentralGoogle Scholar
  12. 12.
    Han D-S, Hsiao M-Y, Wang T-G, Chen S-Y, Yang W-S. Association of serum myokines and aerobic exercise training in patients with spinal cord injury: an observational study. BMC Neurol. 2016;16(1):142.  https://doi.org/10.1186/s12883-016-0661-9.CrossRefPubMedPubMedCentralGoogle Scholar
  13. 13.
    Harvey LA, Dunlop SA, Churilov L, Galea MP. Spinal cord injury physical activity (SCIPA) hands on trial collaborators Early intensive hand rehabilitation is not more effective than usual care plus one-to-one hand therapy in people with sub-acute spinal cord injury (‘Hands On’): a randomised trial. J Physiother. 2017;63(4):197–204.  https://doi.org/10.1016/j.jphys.2017.08.005.CrossRefPubMedGoogle Scholar
  14. 14.
    Hasnan N, Ektas N, Tanhoffer AIP, Tanhoffer R, Fornusek C, Middleton JW, Husain R, Davis GM. Exercise responses during functional electrical stimulation cycling in individuals with spinal cord injury. Med Sci Sports Exerc. 2013;45(6):1131–8.  https://doi.org/10.1249/MSS.0b013e3182805d5a.CrossRefPubMedGoogle Scholar
  15. 15.
    Horiuchi M, Okita K. Arm-cranking exercise training reduces plasminogen activator inhibitor 1 in people with spinal cord injury. Arch Phys Med Rehabil. 2017;98(11):2174–80.  https://doi.org/10.1016/j.apmr.2017.02.007.CrossRefPubMedGoogle Scholar
  16. 16.
    International Conference on Harmonisation of technical requirements for registration of pharmaceuticals for human use. ICH harmonized tripartite guideline: Guideline for Good Clinical Practice. J Postgrad Med. 47(1):45–50.Google Scholar
  17. 17.
    Karlsson A-K. Overview: Autonomic dysfunction in spinal cord injury: clinical presentation of symptoms and signs. Progress Brain Res. 2006;152:1–8.  https://doi.org/10.1016/s0079-6123(05)52034-x.CrossRefGoogle Scholar
  18. 18.
    Kinoshita T, Nakamura T, Umemoto Y, Kojima D, Moriki T, Mitsui T, Goto M, Ishida Y, Tajima F. Increase in interleukin-6 immediately after wheelchair basketball games in persons with spinal cord injury: preliminary report. Spinal Cord. 2013;51(6):508–10.  https://doi.org/10.1038/sc.2013.4.CrossRefPubMedGoogle Scholar
  19. 19.
    Kouda K, Furusawa K, Sugiyama H, Sumiya T, Ito T, Tajima F, Shimizu K. Does 20-min arm crank ergometer exercise increase plasma interleukin-6 in individuals with cervical spinal cord injury? Eur J Appl Physiol. 2012;112(2):597–604.  https://doi.org/10.1007/s00421-011-2004-2.CrossRefPubMedGoogle Scholar
  20. 20.
    Lee YH, Lee JH, Kim SH, Yi D, Oh KJ, Kim JH, Park TJ, Kim H, Chang JS, Kong ID. Hemodynamic adaptations to regular exercise in people with spinal cord injury. Ann Rehabil Med. 2017;41(1):25.  https://doi.org/10.5535/arm.2017.41.1.25.CrossRefPubMedPubMedCentralGoogle Scholar
  21. 21.
    Lifshutz J, Colohan A. A brief history of therapy for traumatic spinal cord injury. Neurosurg Focus. 2004;16(1):E5.  https://doi.org/10.3171/foc.2004.16.1.6.CrossRefPubMedGoogle Scholar
  22. 22.
    Machač S, Radvanský J, Kolář P, Kříž J. Cardiovascular response to peak voluntary exercise in males with cervical spinal cord injury. J Spinal Cord Med. 2016;39(4):412–20.  https://doi.org/10.1080/10790268.2015.1126939.CrossRefPubMedPubMedCentralGoogle Scholar
  23. 23.
    Mitsui T, Nakamura T, Ito T, Umemoto Y, Sakamoto K, Kinoshita T, Nakagawa M, Tajima F. Exercise significantly increases plasma adrenaline and oxidized low-density lipoprotein in normal healthy subjects but not in persons with spinal cord injury. Arch Phys Med Rehabil. 2012;93(4):725–7.  https://doi.org/10.1016/j.apmr.2011.08.046.CrossRefPubMedGoogle Scholar
  24. 24.
    Norman GR, Streiner DL. PDQ (pretty darned quick) statistics. 3rd ed. Lewiston: BC Decker; 2003.Google Scholar
  25. 25.
    Rosety-Rodriguez M, Camacho A, Rosety I, Fornieles G, Rosety MA, Diaz AJ, Bernardi M, Rosety M, Ordonez FJ. Low-grade systemic inflammation and leptin levels were improved by arm cranking exercise in adults with chronic spinal cord injury. Arch Phys Med Rehabil. 2014;95(2):297–302.  https://doi.org/10.1016/j.apmr.2013.08.246.CrossRefPubMedGoogle Scholar
  26. 26.
    Rosety-Rodriguez M, Camacho-Molina A, Rosety I, Fornieles G, Rosety MA, Ordonez FJ. Reduction of exercise-mediated endothelial dysfunction markers in sedentary adults with chronic spinal cord injury. Med Clin (Barc). 2015;144(2):59–61.  https://doi.org/10.1016/j.medcli.2013.10.021.CrossRefPubMedGoogle Scholar
  27. 27.
    Rothman K, Greenland S, Lash T. Modern Epidemiology. Third. Lippincott Williams & Williams. 2008.Google Scholar
  28. 28.
    Sasso E, Backus D. Home-based circuit resistance training to overcome barriers to exercise for people with spinal cord injury. J Neurol Phys Ther. 2013;37(2):65–71.  https://doi.org/10.1097/NPT.0b013e31829247a9.CrossRefPubMedGoogle Scholar
  29. 29.
    Silva FTG, Rêgo JTP, Raulino FR, Silva MR, Reynaud F, Egito ES, Dantas PM. Transcranial direct current stimulation on the autonomic modulation and exercise time in individuals with spinal cord injury. A case report. Auton Neurosci. 2015;193:152–5.  https://doi.org/10.1016/j.autneu.2015.08.007.CrossRefPubMedGoogle Scholar
  30. 30.
    Spengler C, Hostettler S, Leuthold L, Brechbühl J, Mueller G, Illi S. Maximal cardiac output during arm exercise in the sitting position after cervical spinal cord injury. J Rehabil Med. 2012;44(2):131–6.  https://doi.org/10.2340/16501977-0913.CrossRefPubMedGoogle Scholar
  31. 31.
    Stoner L, Credeur D, Dolbow DR, Gater DR. Vascular health toolbox for spinal cord injury: recommendations for clinical practice. Atherosclerosis. 2015;243:373–82.  https://doi.org/10.1016/j.atherosclerosis.2015.10.013.CrossRefPubMedGoogle Scholar
  32. 32.
    Stoner L, Sabatier MJ, Mahoney ET, Dudley GA, McCully KK. Electrical stimulation-evoked resistance exercise therapy improves arterial health after chronic spinal cord injury. Spinal Cord. 2007;45(1):49–56.  https://doi.org/10.1038/sj.sc.3101940.CrossRefPubMedGoogle Scholar
  33. 33.
    Sullivan GM, Feinn R. Using Effect Size-or Why the P Value Is Not Enough. J Grad Med Educ. 2012.  https://doi.org/10.4300/jgme-d-12-00156.1.CrossRefPubMedPubMedCentralGoogle Scholar
  34. 34.
    Ter Woerds W, De Groot PCE, van Kuppevelt DHJM, Hopman MTE. Passive leg movements and passive cycling do not alter arterial leg blood flow in subjects with spinal cord injury. Phys Ther. 2006;86(5):636–645. http://www.ncbi.nlm.nih.gov/pubmed/16649888. Accessed Oct 10 2017.
  35. 35.
    Thijssen DH, Ellenkamp R, Smits P, Hopman MT. Rapid vascular adaptations to training and detraining in persons with spinal cord injury. Arch Phys Med Rehabil. 2006;87(4):474–81.  https://doi.org/10.1016/j.apmr.2005.11.005.CrossRefPubMedGoogle Scholar
  36. 36.
    Thijssen DHJ, Maiorana AJ, O’Driscoll G, Cable NT, Hopman MTE, Green DJ. Impact of inactivity and exercise on the vasculature in humans. Eur J Appl Physiol. 2010;108(5):845–75.  https://doi.org/10.1007/s00421-009-1260-x.CrossRefPubMedGoogle Scholar
  37. 37.
    Tordi N, Mourot L, Chapuis A, Parratte B, Regnard J. Effects of a primary rehabilitation programme on arterial vascular adaptations in an individual with paraplegia. Ann Phys Rehabil Med. 2009;52(1):66–73.  https://doi.org/10.1016/j.rehab.2008.11.002.CrossRefPubMedGoogle Scholar
  38. 38.
    Totosy de Zepetnek JO, Pelletier CA, Hicks AL, MacDonald MJ. Following the physical activity guidelines for adults with spinal cord injury for 16 weeks does not improve vascular health: a randomized controlled trial. Arch Phys Med Rehabil. 2015;96(9):1566–75.  https://doi.org/10.1016/j.apmr.2015.05.019.CrossRefPubMedPubMedCentralGoogle Scholar
  39. 39.
    Turiel M, Sitia S, Cicala S, Magagnin V, Bo I, Porta A, Caiani E, Ricci C, Licari V, De Gennaro Colonna V, Tomasoni L. Robotic treadmill training improves cardiovascular function in spinal cord injury patients. Int J Cardiol. 2011;149(3):323–9.  https://doi.org/10.1016/j.ijcard.2010.02.010.CrossRefPubMedGoogle Scholar
  40. 40.
    Umemoto Y, Furusawa K, Kouda K, Sasaki Y, Kanno N, Kojima D, Tajima F. Plasma IL-6 levels during arm exercise in persons with spinal cord injury. Spinal Cord. 2011;49(12):1182–7.  https://doi.org/10.1038/sc.2011.74.CrossRefPubMedGoogle Scholar
  41. 41.
    Vasiliadis AV, Zafeiridis A, Dipla K, Galanis N, Chatzidimitriou D, Kyparos A, Nikolaidis MG, Vrabas IS. Circulating angiogenic biomolecules at rest and in response to upper-limb exercise in individuals with spinal cord injury. J Spinal Cord Med. 2014;37(2):226–32.  https://doi.org/10.1179/2045772313Y.0000000141.CrossRefPubMedPubMedCentralGoogle Scholar
  42. 42.
    Wasserstein RL, Lazar NA. The ASA’s statement on p-values: context, process, and purpose. Am Stat. 2016;1305:0–17.  https://doi.org/10.1080/00031305.2016.1154108.CrossRefGoogle Scholar
  43. 43.
    Wecht JM, Marsico R, Weir JP, Spungen AM, Bauman WA, De Meersman RE. Autonomic recovery from peak arm exercise in fit and unfit individuals with paraplegia. Med Sci Sports Exerc. 2006;38(7):1223–8.  https://doi.org/10.1249/01.mss.0000227306.34149.ba.CrossRefPubMedGoogle Scholar
  44. 44.
    West CR, Currie KD, Gee C, Krassioukov AV, Borisoff J. Active-arm passive-leg exercise improves cardiovascular function in spinal cord injury. Am J Phys Med Rehabil. 2015;94(11):e102–6.  https://doi.org/10.1097/PHM.0000000000000358.CrossRefPubMedGoogle Scholar
  45. 45.
    van der Woude LHV, de Groot S, Postema K, Bussmann JB, Janssen TW; ALLRISC, Post MW. Active life style rehabilitation interventions in aging spinal cord injury (ALLRISC): a multicentre research program. Disabil Rehabil. 2013;35(13):1097–103.  https://doi.org/10.3109/09638288.2012.718407.CrossRefPubMedGoogle Scholar
  46. 46.
    Zbogar D, Eng JJ, Krassioukov AV, Scott JM, Esch BTA, Warburton DER. The effects of functional electrical stimulation leg cycle ergometry training on arterial compliance in individuals with spinal cord injury. Spinal Cord. 2008;46(11):722–6.  https://doi.org/10.1038/sc.2008.34.CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Beijing Sport University 2019

Authors and Affiliations

  1. 1.Department of Exercise and Sport ScienceThe University of North CarolinaChapel HillUSA
  2. 2.School of KinesiologyUniversity of Southern MississippiHattiesburgUSA

Personalised recommendations