Intermittent isometric fatigue study of the lumbar multifidus muscle in four-point kneeling: an intra-rater reliability investigation.

Back muscle endurance is considered important in low back pain (LBP) rehabilitation. Specific training of multifidus may also be necessary to restore normal low back function. The reliability of surface electromyogram (EMG) to assess endurance of the multifidus muscle during intermittent isometric exercise was evaluated. Multifidus endurance was monitored in the four-point kneeling exercise position using the power spectral analysis method. Twenty healthy volunteers were tested on three separate occasions. Subjects performed repeat 10 s high-intensity voluntary contractions of multifidus for 3 min. The median frequency (MF) and the integrated-rectified (I-R) EMG signal displayed the fatigue pattern of multifidus. Intraclass correlation coefficients indicated fair-good reproducibility for MF (0.48-0.67) but poor reliability for IR-EMG. In conclusion, problems concerning functional testing protocols for the back muscles remain and careful development is necessary for more realistic rehabilitation monitoring.

Paraspinal muscle EMG fatigue testing with two methods in healthy volunteers. Reliability in the context of clinical applications.

OBJECTIVE: Comparison of the accuracy of surface electromyogram for back muscle endurance assessment with two different tests. DESIGN: Test-retest measurements in 16 healthy volunteers on two separate occasions for each test under controlled conditions. BACKGROUND: Back muscle endurance is considered important in low back pain rehabilitation. Reliability of paraspinal muscle endurance assessment is a pre-requisite for accurate and meaningful clinical applications of the technique. METHODS: All participants performed each test twice. A direct comparison was made between two popular fatigue testing methods, the modified Biering-Sørensen and a 60% maximum voluntary isometric contraction in the upright position during which time fatigue was assessed from the electromyogram spectral and amplitude analysis. RESULTS: Reproducibility of initial median frequency was excellent for both tests. Normalised median frequency slope values were more reliable with the 60% maximum voluntary contraction upright test. The clinical applicability of these measures in detecting significant differences after patient rehabilitation is recommended. Root mean square had very large between-day error for both tests.