Haig A J
Department of Physical Medicine and Rehabilitation, University of Michigan, Ann Arbor 48109-0042, USA.
Arch Phys Med Rehabil. 1997 Nov;78(11):1177-84. doi: 10.1016/s0003-9993(97)90328-2.
To assess the extent of denervation of the paraspinal muscles in spinal disorders.
Nonrandomized prospective trial.
Electrodiagnostic laboratory of a university spine center and of a private practice in a small community.
One hundred fourteen consecutive persons referred for electrodiagnosis of spinal or lower extremity disorders.
The paraspinal mapping (PM) electromyography (EMG) protocol along with codified history, physical examination, extremity EMG, and in 44 cases, radiologic imaging results.
PM scores compared with recently established norms (95% of normal subjects scoring less than < 6), lower extremity (LE) EMG findings, and imaging results.
Fifty-eight subjects had normal PM scores; 62 were abnormal. Compared to imaging studies, false positive rate was 8% (1/13). False negative rate (normal PM with definite nerve involvement on imaging) was 33% (6/18), but this decreased to 5% when PM was combined with LE EMG. Four of 4 high lumbar lesions were detected by PM. Of 14 apparent false negatives (13%) compared to LE EMG, 6 had S1 root lesions and 5 had nonspinal lesions. All 7 subjects with isolated S1 radiculopathy had PM scores within normal limits. Occasional polyneuropathies presented with normal PM and abnormal distal findings or with abnormal PM but normal distal findings. No patient with isolated paraspinal findings had evidence of malignancy on follow-up. This report also describes the distribution of denervation in the multifidus, longissimus, and iliocostalis in various disorders.
PM related well to imaging studies and LE EMG. It detects uncommon high lumbar root lesions. S1 may not innervate the paraspinal muscles. There are isolated paraspinal findings in persons without malignancy. The effects of specific muscle denervation on biomechanics, rehabilitation exercise, and prognosis are not known.