Indirect Comparison of 10 kHz Spinal Cord Stimulation (SCS) versus Traditional Low-Frequency SCS for the Treatment of Painful Diabetic Neuropathy: A Systematic Review of Randomized Controlled Trials

Abstract

Spinal cord stimulation (SCS) is increasingly used to treat painful diabetic neuropathy (PDN). At the time of a recent meta-analysis in this field, data were only available from randomized controlled trials (RCTs) of traditional low-frequency SCS (LF-SCS). However, outcomes from high-frequency 10 kHz SCS treatment are now available. Our study aimed to systematically review the contemporary evidence for SCS in patients with lower limb pain due to PDN and include an indirect comparison of the high- and low-frequency modalities. We searched the PubMed/CENTRAL databases up to 18 August 2022, for peer-reviewed RCTs of SCS that enrolled PDN patients with lower limb pain symptoms. The quality of the evidence was assessed with the Cochrane Risk of Bias tool. Using SCS treatment arm data from the RCTs, we indirectly compared the absolute treatment effect of 10 kHz SCS and LF-SCS. Results are presented in tables and forest plots. This systematic review was reported according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 guidelines. Three RCTs met our eligibility criteria, including the recent 10 kHz SCS RCT (N = 216, 90 implanted) and 2 others that examined LF-SCS (N = 36, 17 implanted; N = 60, 37 implanted). Our analysis of 6-month data found clinically meaningful pain relief with each SCS modality. However, significantly greater pain reduction was identified for 10 kHz SCS over LF-SCS: average pain reduction in the 10 kHz SCS cohort was 73.7% compared with 47.5% in the pooled LF-SCS group (p < 0.0001). In the permanent implant subset, the 50% pain reduction responder rate was 83.3% in the 10 kHz SCS cohort versus 63.0% in the pooled LF-SCS group (p = 0.0072). The overall risk of bias of each included RCT was deemed high, mainly due to the absence of patient blinding. Our analysis indicates that paresthesia-free 10 kHz SCS can provide superior pain relief and responder rate over LF-SCS for managing PDN patients refractory to conventional medical management.

Keywords: 10 kHz SCS; diabetes; meta-analysis; neuropathic pain; painful diabetic neuropathy; peripheral neuropathy; spinal cord stimulation; systematic review.

Figure 1

Flow chart of the systematic literature search according to PRISMA guidelines. Registration number: INPLASY202290056. DOI: 10.37766/inplasy2022.9.0056.

Figure 2

Risk of bias assessment using the Cochrane RoB—2 tool. Study ID: Petersen et al. (2022) [38], Slangen et al. (2014) [39], de Vos et al. (2014) [40].

Figure 3

Forest plot comparison of the mITT population mean pain intensity reductions from baseline at 6 months with 95% CI (p-value is for 10 kHz SCS vs. LF-SCS group). The mITT population comprised randomized individuals who entered the SCS trial phase. Abbreviations: CI, Confidence Interval; LF-SCS, Low-Frequency Spinal Cord Stimulation; mITT, Modified Intention-to-Treat; SCS, Spinal Cord Stimulation.

Figure 4

Forest plot comparison of the mITT population responder rates at 6 months, showing RR with 95% CI. The mITT comprised randomized individuals who entered the SCS trial phase. Abbreviations: CI, Confidence Interval; LF-SCS, Traditional Low-Frequency Spinal Cord Stimulation; mITT, Modified Intention-to-Treat; RR, Relative Ratio; SCS, Spinal Cord Stimulation.

Figure 5

Forest plot comparison of the permanent implant population responder rates at 6 months, showing RR with 95% CI. The permanent implant population comprised subjects who completed the SCS trial phase and received a permanent system. Abbreviations: CI, Confidence Interval; LF-SCS, Traditional Low-Frequency Spinal Cord Stimulation; RR, Relative Risk; SCS, Spinal Cord Stimulation.