慢性疼痛に対する脊髄刺激療法のランダム化比較試験エビデンスの系統的レビューおよびネットワークメタアナリシス
Sam Eldabe1, Sarah Nevitt2, Corey W Hunter3
1Pain Medicine, James Cook University Hospital, Middlesbrough, UK.
Background/Importance:
Existing systematic reviews and meta-analyses of randomized trials of spinal cord stimulation (SCS) for people with chronic pain have consistently excluded direct comparisons between different types of SCS and are therefore at major risk of publication bias.
Objective:
The aim of this systematic review and network meta-analysis (NMA) was to combine both direct and indirect randomized trial evidence on treatment effects of SCS for pain intensity, physical function, and health-related quality of life (HRQoL).
Evidence Review:
MEDLINE, Cochrane Central Register of Controlled Trials (CENTRAL), Embase, and WikiStim were searched from inception to December 19, 2024 for parallel-arm randomized trials comparing SCS to conventional medical management (CMM), an active intervention (including other types of SCS) or placebo for adults with chronic pain. The measure of treatment effect for the primary outcome (pain intensity) and for secondary outcomes (physical function, HRQoL) was mean difference (MD) and 95% CI, and for responder rate was OR and 95% CI. NMA was performed in a frequentist framework.
Findings:
23 reports of 16 unique randomized trials were included. All but one trial was assessed as having a high risk of bias. Results of NMA show significant reductions in pain intensity scores (0-10) at last follow-up (within the first 24 months) were observed with all types of SCS compared with CMM, with MD range from -2.37 (95% CI -3.54 to -1.20) to -5.55 (95% CI -7.34 to -3.76; 15 trials, 1479 participants; low certainty evidence). Improvements in physical function (MD range from -0.20, 95% CI -10.24 to 9.85) to -31.04 (95% CI -44.36 to -17.33; 8 trials, 837 participants; low certainty evidence) and HRQoL (MD range from 0.15 (95% CI 0.09 to 0.20) to 0.29 (95% CI 0.20 to 0.37; 8 trials, 942 participants; moderate certainty evidence) were observed for SCS compared with CMM at 6 months. Patients were found to have a higher likelihood to be a responder to treatment (ie, ≥50% reduction in pain intensity) with all forms of SCS in comparison to CMM at last follow-up (OR range from 9.75 (95% CI 4.07 to 23.25) to OR 63.4 (18.76 to 212.55; 13 trials, 1887 participants; very low certainty evidence). Clinically meaningful improvements were observed for pain intensity, HRQoL, and in some comparisons for physical function.
Conclusions:
There is low to moderate certainty of evidence that all types of SCS are superior to CMM for the management of chronic pain conditions.
