Influence of upper-body continuous, resistance or high-intensity interval training (CRIT) on postprandial responses in persons with spinal cord injury: study protocol for a randomised controlled trial

Traditional forms of Moderate Intensity Continuous Training (MICT) can provide some therapeutic benefits and reduce some components of cardiometabolic risk in persons with SCI [8]. Sixteen weeks of upper-body MICT, with minimal resistance training, reduces total body fat mass and visceral adipose tissue content in persons with SCI [9]. While just 6 weeks of MICT can also enhance fasting markers of hepatic insulin sensitivity in persons with paraplegia, the effects on postprandial markers of peripheral insulin sensitivity are negligible [10]. Indeed, neither of these training studies were able to demonstrate a significant benefit of MICT on cardiometabolic components related to postprandial insulin sensitivity, hyperlipidaemia or systemic inflammation. This has led to a call for high-quality randomised controlled trials to assess the efficacy of higher-intensity forms of exercise and assess the acute post-exerciseresponses in systemic biomarkers and energy homeostasis and to novel exercise challenges [11]. Currently, relatively little is known about the acute regulation of energy homeostasis at rest, during or post exercise in the SCI population.

When compared to non-injured controls performing voluntary leg exercise, persons with SCI have markedly reduced mobilisation, delivery, and limb uptake of free fatty acids (FFA) during electrically stimulated leg exercise [12]. This is most likely the result of reduced sympathoadrenal β-adrenergic stimulation and/or limited neural activity in motor centers and afferent nerves from contracting skeletal muscles, depending on the level of injury [12]. The limited availability of FFA during exercise leads to a heavy reliance on carbohydrate (CHO), with a concomitant limited contribution from fat as a fuel source [13, 14]. These observations appear consistent across a variety of modes and intensities of exercise [15, 16]. Increased reliance on carbohydrate as a fuel source during exercise may affect the metabolic handling of systemic glucose and lipids, both during and post exercise.

In non-injured individuals, human energy expenditure (EE) is elevated during recovery from exercise [17]. These increases in EE occur in a dose-dependent manner, which is predominantly related to exercise intensity and modality, as opposed to exercise duration [17]. Only three studies have compared changes in EE following isocaloric arm and leg cycling in non-injured humans and have demonstrated only modest increases in post-exercise EE in response to arm cycling [18‐20]. Arm cycling resulted in a lower cumulative post-exercise EE compared to isocaloric leg cycling [19], suggesting that upper-extremity exercise per se has a limited ability to elevate post-exercise EE. Only one study has examined these responses to arm cycling in persons with SCI [21], demonstrating only a modest increase in post-exercise EE, which was similar to that observed in non-injured humans. This may reflect a reduced ability of the upper-body skeletal muscles to tolerate sustained anaerobic metabolism and accumulate oxygen debt. However, it remains to be determined if exercise mode and/or intensity have the ability to modulate post-exercise EE in persons with SCI.

Associated with the increases in EE at rest, both CHO [22] and fat oxidation rates [23‐25] are significantly increased during recovery from exercise. This effect of exercise on substrate partitioning occurs in a manner dependent more on the total energy cost of exercise and less on exercise intensity or metabolic rate [23]. While the increase in post-exercise fat oxidation has been demonstrated in non-injured humans [20], to date, no studies have examined this response in persons with SCI. Our pilot data [26] demonstrate that fat oxidation is elevated for at least 120 min following ~ 50 min of circuit resistance training (Continuous Resistance Training; CRT) in persons with SCI.

Several studies in persons with SCI have reported exaggerated postprandial lipaemia (PPL) [27, 28] and glycaemia (PPG) [29], which is of some concern, as such chronically elevated responses are component risks for cardiometabolic disease [30]. In persons without SCI, a juxta-meal exercise bout has a dramatic effect on the metabolic handling and disposal of macronutrients from a meal. Specific to CHO and fat disposal, pre-meal exercise attenuates PPL [25] and PPG [22] even in persons with known disorders of energy homeostasis [22, 29]. Interestingly, both exercise modality [17, 24] and intensity [17, 31] have been shown to modify the effect of exercise on post-exercise energy homeostasis, independent of exercise-related EE. However, to our knowledge, no studies have examined the acute effects of different forms of exercise on postprandial systemic metabolite, hormone or biomarker concentrations or whole-body substrate oxidation rates in persons with SCI.

This study is a partially randomised, repeated-measures, counter-balanced, crossover design. Participants will attend two preliminary sessions including baseline assessments and a HIIT familiarisation session before completing the four experimental conditions. Each participant will complete the CRT condition first in order that the intensity and/or duration of the HIIT and MICT protocols can be adjusted to deliver an isocaloric exercise challenge. The CON, MICT and HIIT conditions will be completed in a randomised order, at least 3 days apart. The planned experimental design is summarised below (Fig. 1) and is consistent with current Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidelines [32, 33]. The study protocol has been approved by the Human Subjects Research Office, Miller School of Medicine, University of Miami (Institutional Review Board No. 20171114, Version 3, dated 5 February 2018) and the trial has been registered as a current controlled trial (ClinicalTrials.gov, ID: NCT03545867 on 1 June 2018).

All human testing will take place in the Lois Pope Life Center at the Miami Project to Cure Paralysis at the University of Miami’s Miller School of Medicine. All biochemical analyses will take place at the Diabetes Research Institute, Division of Endocrinology, Diabetes and Metabolism at the University of Miami Miller School of Medicine.

It has been reported that participant recruitment to research studies, especially in the absence of direct access to a clinical population, requires considerably more resources and time than initially anticipated in order to achieve adequate enrolment [34]. Patient barriers are reported to include [35]: (1) additional demands of the trial increasing participant burden and (2) patient preferences for or against a particular treatment. Barriers to research participation are perhaps even more exaggerated for people with disabilities due to complex health problems [36], lack of transportation [37], cognitive impairments and financial stress [38].

In order to overcome some of these barriers and challenges, The Miami Project maintains and supports the running of a clinical exercise programme, where individuals with SCI who have volunteered to participate in a Miami Project study, can access a fully equipped gymnasium free-of-charge. Associated with this, a user database is maintained, containing basic descriptive information on persons with SCI who have volunteered to participate in previous studies and volunteered ongoing participation. Upon initial contact with individuals from this user community (i.e. email or telephone), the study will be described and those interested in enrolling will be asked to read a detailed participant information sheet and complete a health screening questionnaire. Interested individuals will be screened via a telephone conversation 48 h after their initial contact to ascertain eligibility. A focus will be placed on establishing a good rapport at this initial contact and also as participants progress through the enrolment process (Fig. 1). Written consent will be obtained during the initial laboratory visit and participants will be informed that they can withdraw from the study at any time without consequence. Participants will be paid for their participation in this study, following completion of all assessments. This is primarily designed to reimburse travel expenses and thank them for any inconvenience caused.

After completion of baseline laboratory testing, the HIIT familiarisation and the CRT condition, participants will complete the remaining three experimental conditions in random order. Randomisation will be performed by an independent senior researcher (MN) using a list generated with a web-based platform (http://​www.​randomization.​com), concealed from those involved in participant management (DM, JM and JB) to prevent biased allocation [39, 40].

To address our hypothesis, a cohort of 10 healthy men with chronic SCI, who satisfy the inclusion and exclusion criteria outlined below, will be recruited via advertisement and direct contact with the local SCI community.

Inclusion criteria: (1) Male; (2) Aged ≥ 18 years old; (3) Neurologically stable SCI (ASIA Impairment Scale A-C) at T6 and lower spinal levels for > 1 year; (4) Able and willing to comply with study procedures; (5) Have the ability to understand written and spoken English; and (6) Have the capacity to provide informed consent.

During their second preliminary visit to the laboratory, participants will be fitted with a Hans-Rudolph Softmask and expired gases will be collected and analysed throughout exercise (as described above). Participants will conduct ~ 50 min of arm-crank ergometry (ACE) on the same device/position as described above during the GXT. The cycle ergometer will be programmed to vary power output so that a warm-up and cool-down (2 min) and the active recovery intervals will be completed at 10% peak power output (PO_peak), and the working intervals completed at 70% PO_peak. The ratio of work:recovery intervals will be 1:1. The EE data will later be used to calculate the duration of HIIT required to elicit an isocaloric challenge to CRT.

The same experimental procedures will be completed on all four main trial days (Fig. 1). These procedures are also reflected in the Standardised Protocol Items: Recommendations for Interventional Trials (SPIRIT) Figure (Fig. 2). The associated SPIRIT Checklist is available online (Additional file 1). Twenty-four hours prior to each main laboratory trial, participants will abstain from caffeine (tea/coffee) and alcohol. On the morning of the main trials, participants will be instructed to consume ~ 10 ml.kg^− 1 of water on waking and report to the laboratory at 0800 hours ± 0.5 h following an overnight fast (≥ 10 h). Following entry to the laboratory participants will be fitted with the mask for indirect calorimetry (as described above) and will remain seated prone in their wheelchair for ~ 10 min to assess resting EE (REE). Immediately after this, an initial 10-ml venous blood sample will be drawn. For the next ~ 50 min, expired gases and heart rate values will continue to be collected while the participants rest (CON) or exercise (MICT, HIIT or CRT). Immediately after this period, an indwelling cannula will be inserted in to an antecubital vein and kept patent with sterile saline. An initial sample will be drawn before participants consume a 600-kcal liquid meal test meal (35% fat, 50% CHO, 15% protein). Further 10-ml venous blood samples will be drawn at 0, 15, 30, 60, 90 and 120 min post meal. Expired gases will continue to be collected throughout the postprandial period.

Participants will be monitored for the following both during and after treatments: headaches, pain, lightheadedness, dizziness, altered vision, respiratory distress, cyanosis, spasms, or change in function. We will also monitor cardiovascular responses to treatment to ensure that participants do not exceed lower (< 40 bpm) and upper (> 180 bpm) limits for heart rate or systolic blood pressure (i.e. < 85 mmHg and > 200 mmHg).

The skin located near pressure points will be inspected after every exercise session. Exercise trials will be stopped if the participants experience any of the following: chest pain, dyspnoea, diaphoresis, or a pale or ashen appearance. All testing is being performed at a medical centre and is, therefore, located near an emergency room. Security personnel at the Miami Project are all trained ‘first responders’. The laboratory has approved policies and procedures for emergencies and all staff have current certification for cardio-pulmonary resuscitation.

All adverse events will be reported to the Institutional Review Board (IRB) within the mandated time period. For any adverse event the principal investigator will immediately notify and consult with the study physician, who will come to an opinion as to whether the event(s) is/are related to the study procedures described in the protocol. Adverse events will be evaluated by the study physician using the following criteria:

For the first two grades the investigators will observe the participant and as necessary institute standard medical or therapeutic care. Repeated occurrence of grade 1 and 2 events may be cause for notification of the IRB by the investigators that stoppage is appropriate. The investigators may take this action and the IRB will be so notified. Grade 3 and 4 events will be individually evaluated. Any occurrence of grade 3 and 4 events may be cause for notification of the IRB by the investigators that stoppage is appropriate. The investigators may take this action and the IRB will be so notified. Otherwise, upon on an IRB determination that the serious adverse event was related to the protocol, the study will stop and undergo evaluation for continuation.

In order to simplify data analysis and facilitate the interpretation of a complex dataset [51, 52], serial measurements of glucose and insulin responses at baseline, post exercise and in response to the rest/exercise challenge will be converted into simple summary statistics [53], such as incremental area under the curve (iAUC) [54] and insulin sensitivity index (ISIMatsuda) [55]. The Homeostasis Model Assessment (HOMA) calculator, incorporating the updated HOMA-2 model [56], will be used to derive fasting estimates of pancreatic β-cell function, insulin resistance and sensitivity, both at rest and post exercise. A full lipid profile (i.e. triglycerides, total cholesterol, non-esterified fatty acids (NEFA), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C)) will also be assessed to establish the influence of exercise protocol on post-exercise and postprandial responses. From indirect calorimetric measurements, EE, fat and carbohydrate oxidation rates will be determined and compared, primarily during the postprandial phase. The primary and secondary outcome variables are presented in Table 1.

We plan to recruit 11 male participants with chronic (> 1 year) SCI, by engaging users from the Miami Project to Cure Paralysis community. The primary outcome variables are related to fasting and postprandial glycaemic control, but specifically postprandial insulin area under curve (iAUC). While there are only two previous studies that have compared the effects of MICT and HIIT on biomarkers of metabolic regulation in non-disabled humans [57, 58], both suggest no effect of MICT (vs. CON) and an enhanced effect of HIIT. We therefore hypothesised that there will be no differences in postprandial glycaemic control between CON and MICT and a similarly enhanced effect of HIIT and CRT. Using data from these previous studies, the mean predicted effect size for postprandial iAUC is ~ 1 [57] and for postprandial triglyceride concentrations is 0.97 [59]. We therefore estimated that a final sample size of 11 will be required for this repeated measures study, to provide approximately 90% power to detect a significant difference in ∆ insulin iAUC at an alpha level of 0.05. Recruitment will proceed on a rolling basis until the adequate sample size is reached; emphasis will be placed on considering the demands of the study before enrolling to reduce drop-out during the study.

All data from participants who successfully complete the study will be included in the analysis. The differences in key outcome variables between experimental conditions (CON, MICT, CRT and HIIT) and time (dependent on variable) will be analysed using a two-way (condition × time) fully repeated measures analysis of variance (ANOVA). Where significant interactions are observed, multiple t tests will determine the location of variance. Significant post-hoc effects will be subjected to a Holm-Bonferroni stepwise adjustment. Standardised effect sizes (Cohen’s d) will also be calculated. This will provide a practical interpretation of the size of the effects of each experimental condition relative to CON. For all the above statistical approaches, statistical significance will be set at an alpha level of p ≤ 0.05. We will explore the use of confidence intervals and magnitude-based inferences to assess the clinical significance of the effect.