Comparison of spinal mobility and isometric trunk extensor forces with electromyographic spectral analysis in identifying low back pain.

This study compared conventional clinical measurements with electromyographic (EMG) spectral measurements for identification of individuals with low back pain (LBP). Twenty freshman sweep rowers were subjects for this study. Range-of-motion (ROM) measurements were taken for forward bending (FB), backward bending (BB) (double inclinometers), lateral bending (LB) (tape measure), and rotation (double-arm goniometer). Intratester reliability for ROM was also assessed. The Back Analysis System was used to determine static trunk extensor strength (ie, maximal voluntary contraction [MVC]) and to compute EMG spectral parameters from a paraspinal multi-electrode array. A two-group stepwise discriminant-analysis procedure for the ROM and MVC variables correctly identified 57% of the rowers with LBP and 63% of the rowers without LBP. A similar discriminant-analysis procedure for EMG spectral parameters correctly identified either 88% of the rowers with LBP and 100% of the rowers without LBP or 100% of the rowers with LBP and 88% of the rowers with LBP, depending on whether EMG measurements of recovery were calculated at 1 minute or at 2 minutes into the recovery period. Sensitivity (66%) and specificity (71%) results from the more traditional tests suggest that these techniques may be of limited usefulness for LBP screening or diagnosis.