Sensitivity study of H-reflex alterations in idiopathic low back pain patients vs. a healthy population.

Twenty-seven male and 12 female healthy volunteers were tested twice with 2-7 days separation. Hoffman (H) reflexes and muscle (M) activation waves were obtained from the posterior tibial nerves bilaterally. Results were compared to those obtained from patients presenting with a complaint of low back and/or leg pain, without compressive neuropathy. M, F, H latencies and H/Mmax ratio were recorded. H/M ratio and latency comparisons were not significantly different in the control group left to right or test to test. For the low back pain group, 10-14 days following the initial evaluation, each subject returned for a follow-up test. During the interim, the patient was followed conservatively using manipulation and home care. Analysis of variance (ANOVA) testing of ratio values demonstrated a difference in overall mean values (p greater than 0.001) for comparisons between the control (mean = 0.367), pretest (mean = 0.695), and posttest (mean = 0.558) values. Sensitivity in discriminating acute low back pain subjects from healthy controls was tested by determining the distance between mean H/M values for the probability curves of each population, with an arbitrary cutoff value of 0.6 as the upper limit normal. Sensitivity distance was 2.29 with a likelihood ratio of 3.04. This suggests that an H/Mmax ratio greater than or equal to 0.6 will correctly identify two of three patients with idiopathic low back pain.