Reliability of Cerebral and Muscle Oxygenation/Blood Volume During Hand-Gripping Measured Simultaneously by Near Infrared Spectroscopy

Abstract

The stimulus to contract a muscle originates in the cerebral motor cortex and is transmitted to the motor unit via the spinal cord. Near infrared spectroscopy (NIRS) has been used to measure both cerebral and muscle oxygenation (Cox, Mox) /blood volume (Cbv, Mbv) changes during exercise. Although the reliability of these measurements has been reported separately in both tissues, their reproducibility in these two tissues has not been examined simultaneously. PURPOSE: To examine the reliability of the Cox/Mox/Cbv/Mbv measurements during handgrip exercise in healthy subjects. METHODS: Informed consent was obtained from six volunteers (mean age = 28 ± 13 yr, grip strength = 30.4 ± 10.8 kg) who completed two identical testing sessions 24h apart. After a 2min baseline, each subject performed maximal handgrip contractions on a dynamometer (Takei, Japan) with their dominant hand continuously for 15sec, and intermittently for 60sec at a contraction rate of 10/min without hyperventilation during the same session. A 2min rest was allowed between the two protocols. NIRS responses were recorded simultaneously from the extensor carpi ulnaris of the dominant forearm and the contralateral frontal lobe using a dual wavelength spectrometer (MicroRunman, NIM Inc., PA). Delta values were calculated as the difference between the resting value just before initiation of contraction and the maximal value observed during the test and/or recovery period. Values for the two protocols were pooled (N = 12) for statistical analysis. RESULTS: During trials 1 and 2, Mox and Mbv showed a consistent decrease (−0.306±0.069, −0.270±0.082, −0.300±0.086 and −0.315±0.054 respectively), whereas Cox and Cbv demonstrated a continual increase (0.069±0.027, 0.076±0.036, 0.076±0.036 and 0.033±0.015 respectively). The intraclass correlation coefficients for the Cox, Cbv, Mox and Mbv were 0.86 (p=0.001), 0.82 (p=0.004), 0.60 (p=0.085) and 0.32 (p=0.281) respectively. No significant differences were observed between the mean values of the two sessions for each of the variables (p=0.106, 0.372, 0.639, and 0.216 for Cox, Cbv, Mox, and Mbv). Bland-Altman analysis indicated no outliers for the cerebral measurements, whereas there was one outlier for the Mbv. The half recovery time for Mox (an index of muscle aerobic capacity) demonstrated moderate reliability (ICC = 0.72, p=0.023). CONCLUSIONS: The evidence indicates that: (1) the Cox, Cbv and Mox responses, as well as half recovery time for Mox, measured by NIRS are reproducible, and (2) the consistency of the measurements is higher in cerebral compared to muscle tissue. Funding: First author was supported by Ministério da Educação / CAPES, Brazil.