Introduction

Skeletal muscle is integral to both metabolic and physical function¹. Declines in muscle mass and metabolic capacity (deconditioning) are characteristics of ageing and disease progression that collectively manifest as premature fatigue. ³¹P MRS allows the non-invasive assessment of muscle deconditioning status by quantifying phosphocreatine (PCr) recovery kinetics² following intense exercise with blood flow occlusion, which is entirely mitochondrial dependent³. Bioenergetic failure within skeletal muscle is a well-established cause of skeletal muscle fatigue⁴ and may correlate with fatigue perception in chronic disease, such as Crohn’s disease (CD)⁵. Here, we describe the establishment of a ³¹P-MRS protocol for use during high intensity ischaemic plantar flexion exercise to quantify PCr depletion and its subsequent recovery kinetics as an index of muscle deconditioning. We aim to investigate this purported premature fatigue in CD patients, who present with clinically significant fatigue perception and reduced skeletal muscle performance relative to healthy control volunteers⁵.

Methods

Acquisition: Data was collected on a 3T Philips Achieva scanner. The participants’ dominant limb was chosen for exercise, Volunteers lay in a supine position on the MRI bed with the knee at approximately 30° with their foot secured into an MRI-compatible plantar flexion ergometer (Trispect, Ergospect, Innsbruck Austria). A 14cm Phillips 31P coil was positioned over the medial gastrocnemius and a blood pressure cuff was placed on the lower thigh. First ¹H mDIXON scans were acquired to image the calf. A ~ 12 minute protocol was performed during which ³¹P data was acquired. Following baseline ³¹P assessment for ~ 1 minute, the blood pressure cuff was inflated to 250mmHg for 2-minutes to deplete muscle myoglobin concentrations. Volunteers then performed repeated plantar flexion exercise under ischaemia at either a standardised absolute workload, or at 50% of maximum voluntary contraction (MVC), ascertained / performed prior to data acquisition. Volunteers performed 30 contractions per minute, in time with a metronome. Upon exercise cessation, the cuff pressure was released and ³¹P acquisition continued during exercise recovery. The use of both non-localized pulse-acquire ³¹P-MR spectra and ³¹P Image Selected In Vivo Spectroscopy (ISIS) (localising to the medial gastrocnemius) to assess ³¹P changes during exercise were compared, along with the effects of exercise intensity.

Data Analysis: ³¹P spectra were analysed using jMRUI Beta 6.0, spectra were apodized to 10Hz and zero and first order, phase correction performed. Spectra were frequency aligned and an ER filter applied between 500 -1000Hz. ³¹P spectra peaks were quantified via the AMARES function. Standardised prior knowledge and starting values were applied from ³¹P skeletal muscle spectra. Adjustments were made where necessary for inter-individual variations in peak frequency in order to optimise peak fitting. Peak amplitudes were derived for inorganic phosphate (Pi) and PCr and exercise kinetics plotted. The PCr recovery rate constant, k, was computed using a two parameter monoexponential fit to: PCr_(t) = PCr_initial+(PCr_end – PCr_initial)(1-_^exp(-^k.t)), where t is the time from the start of recovery, PCr_initial and PCr_end is the PCr content at the initial and end of recovery phases respectively. PCr_end was calculated from the mean end-recovery PCr values and then used in a fit to obtain k and PCr_initial.

Results

Figure 1 shows ³¹P ISIS data collected during the ischemic plantar flexion exercise, highlighting the limited temporal resolution and significant phasing issues (Fig 1A) related to motion induced from both the BP cuff inflation and the ischemic exercise protocol (Fig.1B). The advantage a non-localised ³¹P sequence (Fig1Ci) is clear, overcoming motion issues as the ³¹P coil moves with the leg. Figure 2 shows inconsistent PCr depletion in two patient volunteers to the same absolute workload (10W) (coefficient variation of PCr depletion 24.12%). Figure 3 shows the effects of using a relative workload, at 40 % and 50 % of maximum voluntary contraction (MVC). Increasing the ischemic exercise intensity from 40 to 50% MVC increased PCr depletion by 62.24% (56.17 vs 91.13%). Figure 4 shows example data for the finalised protocol using 50% MVC during ischaemic plantar flexion exercise, and associated ³¹P kinetics, mean PCr depletion rate was 85.8 ± 2.0% with a coefficient variation of 4% across three fatigued CD patients. The mean recovery rate constant (k) was 0.57±0.06 min^-1.

Discussion

Here the development and application of a protocol to study ³¹P kinetics has been described. It is shown that ³¹P ISIS has limitations for the study of skeletal muscle PCr recovery kinetics due to its poor temporal resolution (which can be overcome by using a moving average) and sensitivity to motion, induced by both high pressure cuff occlusion and ischemic exercise stimulus. A non-localised ³¹P measure together with application of a high relative exercise intensity (50 % MVC) during ischaemic plantar flexion exercise is shown to provide a robust measure to study PCr depletion and recovery, and has been applied in participants with quiescent Crohn’s disease. This will allow non-invasive quantitation of muscle metabolic quality in CD to ascertain whether premature performance fatigability reported in CD is driven via peripheral muscle deconditioning.

Conclusion

A robust protocol for the assessment of ³¹P kinetics has been developed which allows extensive depletion of PCr and monitoring of recovery.

Acknowledgements

This research is funded by Crohns and Colitis UK (CCUK).