A Meta-analysis Exploring the Efficacy of Neuropathic Pain Medication for Low Back Pain or Spine-Related Leg Pain: Is Efficacy Dependent on the Presence of Neuropathic Pain?

BACKGROUND AND OBJECTIVE: Highly variable pain mechanisms in people with low back pain or spine-related leg pain might contribute to inefficacy of neuropathic pain medication. This meta-analysis aimed to determine how neuropathic pain is identified in clinical trials for people taking neuropathic pain medication for low back pain or spine-related leg pain and whether subgrouping based on the presence of neuropathic pain influences efficacy. METHODS: EMBASE, MEDLINE, Cochrane Central, CINAHL [EBSCO], APA PsycINFO, ClinicalTrials.gov, and the World Health Organization International Clinical Trials Registry were searched from inception to 14 May, 2024. Randomized and crossover trials comparing first-line neuropathic pain medication for people with low back pain or spine-related leg pain to placebo or usual care were included. Two independent authors extracted data. Random-effects meta-analyses of all studies combined, and pre-planned subgroup meta-analyses based on the certainty of neuropathic pain (according to the neuropathic pain Special Interest Group [NeuPSIG] neuropathic pain grading criteria) were completed. Certainty of evidence was judged using the grading of recommendations assessment development and evaluation [GRADE] framework. RESULTS: Twenty-seven included studies reported on 3619 participants. Overall, 33% of studies were judged unlikely to include people with neuropathic pain, 26% remained unclear. Only 41% identified people with possible, probable, or definite neuropathic pain. For pain, general analyses revealed only small effects at short term (mean difference [MD] - 9.30 [95% confidence interval [CI] - 13.71, - 4.88], I2 = 87%) and medium term (MD - 5.49 [95% CI - 7.24, - 3.74], I2 = 0%). Subgrouping at short term revealed studies including people with definite or probable neuropathic pain showed larger effects on pain (definite; MD - 16.65 [95% CI - 35.95, 2.65], I2 = 84%; probable; MD - 10.45 [95% CI - 14.79, - 6.12], I2 = 20%) than studies including people with possible (MD - 5.50 [95% CI - 20.52, 9.52], I2 = 78%), unlikely (MD - 6.67 [95% CI - 10.58, 2.76], I2 = 0%), or unclear neuropathic pain (MD - 8.93 [95% CI - 20.57, 2.71], I2 = 96%). Similarly, general analyses revealed negligible effects on disability at short term (MD - 3.35 [95% CI - 9.00, 2.29], I2 = 93%) and medium term (MD - 4.06 [95% CI - 5.63, - 2.48], I2 = 0%). Sub-grouping at short term revealed larger effects in studies including people with definite/probable neuropathic pain (MD - 9.25 [95% CI - 12.59, - 5.90], I2 = 2%) compared with those with possible/unclear/unlikely neuropathic pain (MD -1.57 [95% CI - 8.96, 5.82] I2 = 95%). Medium-term outcomes showed a similar trend, but were limited by low numbers of studies. Certainty of evidence was low to very low for all outcomes. CONCLUSIONS: Most studies using neuropathic pain medication for low back pain or spine-related leg pain fail to adequately consider the presence of neuropathic pain. Meta-analyses suggest neuropathic pain medication may be most effective in people with low back pain or spine-related leg pain with a definite/probable neuropathic pain component. However, the low to very low certainty of evidence and poor identification of neuropathic pain in most studies prevent firm recommendations.