Characterization of preclinical models to investigate spinal cord stimulation for neuropathic pain: a systematic review and meta-analysis

Abstract

Despite advancements in preclinical and clinical spinal cord stimulation (SCS) research, the mechanisms of SCS action remain unclear. This may result from challenges in translatability of findings between species. Our systematic review (PROSPERO: CRD42023457443) aimed to comprehensively characterize the important translational components of preclinical SCS models, including stimulating elements and stimulation specifications. Databases (Embase, PubMed, Web of Science, and WikiStim) were searched on October 5, 2023, identifying 78 studies meeting the search criteria. We conducted a post hoc meta-analysis, including subgroup analyses and meta-regression, to assess SCS efficacy on mechanical hypersensitivity in rats subjected to neuropathic pain. Although monopolar electrodes were predominantly used as stimulating elements until 2013, quadripolar paddle and cylindrical leads gained recent popularity. Most research was conducted using 50 Hz and 200 µs stimulation. Motor threshold (MT) estimation was the predominant strategy to determine SCS intensity, which was set to 71.9% of MT on average. Our analysis revealed a large effect size for SCS (Hedge g = 1.13, 95% CI: [0.93, 1.32]) with similar magnitudes of effect between conventional (≤100 Hz) and nonconventional SCS paradigms while sham SCS had nonsignificant effect size. In addition, different stimulation intensity, frequency, and electrode design did not affect effect size. The risk of bias was assessed using Systematic Review Centre for Laboratory animal Experimentation criteria and was unclear, and only the frequency subgroup analysis showed publication bias. In summary, our review characterizes the critical components of preclinical SCS models and provides recommendations to improve reproducibility and translatability, thereby advancing the scientific foundation for SCS research.

Keywords: Analgesia; Mechanical hypersensitivity; Neuropathic pain; Rat; Spinal cord stimulation.

Figure 1.

The PRISMA flow diagram of the search strategy and study selection process. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analysis.

Figure 2.

(A) Cumulative number of selected studies over the years. (B) Number of selected studies per laboratory (indicated by city and country). (C) Percentage of selected studies reporting specifics of funding source.

Figure 3.

Summary of SYRCLE risk of bias assessment. SYRCLE, Systematic Review Centre for Laboratory animal Experimentation.

Figure 4.

Forest plot from the primary subgroup analysis of the analgesic effect of SCS on neuropathic pain assessed using the von Frey test after conventional, nonconventional, and sham stimulation conditions. 95% CI, 95% confidence interval; N, sample size; post-M, post-SCS mean; post-SD, post-SCS SD; pre-M, pre-SCS mean; pre-SD, pre-SCS SD; SMD, standardized mean difference with small sample size correction (Hedge g).

Figure 5.

Funnel plots for the different subgroup analysis, using a sample size-based precision estimate (1/√n). Black dots represent effect sizes (Hedge g) from studies. The white triangle area represents the 95% confidence interval centered on the estimated standardized mean difference (vertical dashed line). (A) Conventional, nonconventional, and sham comparison. (B) HF, 10 kHz, and burst frequency comparison. There is an asymmetry in the plot, suggesting a publication bias. (C) Low, mid, and high percentage of motor threshold (MT) comparison. (D) Mono, cylindrical, and paddle stimulating element comparison. HF, high frequency.