ABSTRACT
BACKGROUND: Radiofrequency facet denervation is one of the most frequently performed procedures for chronic low back pain. Although sensory stimulation is generally used as a surrogate measure to denote sufficient proximity of the electrode to the nerve, no study has examined whether stimulation threshold influences outcome. METHODS: We prospectively recorded data in 61 consecutive patients undergoing lumbar facet radiofrequency denervation who experienced significant pain relief after medial branch blocks. For each nerve lesioned, multiple attempts were made to maximize sensory stimulation threshold (SST). Mean SST was calculated on the basis of the lowest stimulation perceived at 0.1-V increments for each medial branch. A positive outcome was defined as a ≥50% reduction in back pain coupled with a positive satisfaction score lasting ≥3 months. The relationship between mean SST and denervation outcomes was evaluated via a receiver's operating characteristic (ROC) curve, and stratifying outcomes on the basis of various cutoff values. RESULTS: No correlation was noted between mean SST and pain relief at rest (Pearson's r=-0.01, 95% confidence interval [CI]: -0.24 to 0.23, P=0.97), with activity (r=-0.17, 95% CI: -0.40 to 0.07, P=0.20), or a successful outcome. No optimal SST could be identified. CONCLUSIONS: There is no significant relationship between mean SST during lumbar facet radiofrequency denervation and treatment outcome, which may be due to differences in general sensory perception. Because stimulation threshold was optimized for each patient, these data cannot be interpreted to suggest that sensory testing should not be performed, or that high sensory stimulation thresholds obtained on the first attempt should be deemed acceptable.