The reliability of (31)P-MRS and NIRS measurements of spinal muscle function

While phosphorous magnetic resonance spectroscopy ((31)P-MRS) and near-infrared spectroscopy (NIRS) provide methods for measuring spinal muscle function non-invasively, their reliability is not established. The aim of this study was assess the reliability (ICC) and error magnitude (CV%) of measurements of muscle phosphocreatine (PCr), tissue oxygenation index (TOI) and muscle deoxyhaemoglobin (HHb) acquired during fatigue and in recovery after 24 s of exercise in the lumbar muscles. 10 healthy participants (19-25 years, 5 male, 5 female) performed exercise that involved holding the upper body unsupported in slight extension until fatigue and then, after 30 min of rest, for repeated bursts of 24 s. ICCs indicated good to excellent reliability of baseline measures (TOI: 0.75) and of amplitude changes during fatigue (PCr: 0.73, TOI: 0.69, HHb: 0.80) and recovery (HHb: 0.96), and poor to fair reliability for time constants describing rates of change during fatigue (PCr: 0.11) and recovery (PCr: 0.31, HHb: 0.47). CV% indicated varying relative measurement error across baseline measures (TOI: 5%), amplitude changes during fatigue (PCr: 7%, TOI: 38%, HHb: 31%) and recovery (HHb: 31%), and in time constants for fatigue (PCr: 39%) and recovery (PCr: 20%, HHb: 37%). The results suggested that reliability would be sufficient for future studies on spinal muscle function, but that measurement error may be too large to evaluate individuals.