Does industry funding and study location impact findings from randomized controlled trials of spinal cord stimulation? A systematic review and meta-analysis

TY - JOUR

T1 - Does industry funding and study location impact findings from randomized controlled trials of spinal cord stimulation?

T2 - A systematic review and meta-analysis

AU - Eldabe, Sam

AU - Nevitt, Sarah

AU - Copley, Sue

AU - Maden, Michelle

AU - Goudman, Lisa

AU - Hayek, Salim

AU - Mekhail, Nagy

AU - Moens, Maarten

AU - Rigoard, Phillipe

AU - Duarte, Rui

N1 - © American Society of Regional Anesthesia & Pain Medicine 2023. .

PY - 2023/8/23

Y1 - 2023/8/23

N2 - BACKGROUND/IMPORTANCE: Concerns have been raised that effects observed in studies of spinal cord stimulation (SCS) funded by industry have not been replicated in non-industry-funded studies and that findings may differ based on geographical location where the study was conducted.OBJECTIVE: To investigate the impact of industry funding and geographical location on pain intensity, function, health-related quality of life and adverse events reported in randomized controlled trials (RCTs) of SCS.EVIDENCE REVIEW: Systematic review conducted using MEDLINE, CENTRAL, EMBASE and WikiStim databases until September 2022. Parallel-group RCTs evaluating SCS for patients with neuropathic pain were included. Results of studies were combined in random-effects meta-analysis using the generic-inverse variance method. Subgroup meta-analyses were conducted according to funding source and study location. Risk of bias was assessed using Cochrane RoB 2.0 tool.FINDINGS: Twenty-nine reports of 17 RCTs (1823 participants) were included. For the comparison of SCS with usual care, test for subgroup differences indicate no significant differences (p=0.48, moderate certainty evidence) in pain intensity score at 6 months for studies with no funding or funding not disclosed (pooled mean difference (MD) -1.96 (95% CI -3.23 to -0.69; 95% prediction interval (PI) not estimable, I2=0%, τ2=0)), industry funding (pooled MD -2.70 (95% CI -4.29 to -1.11; 95% PI -8.75 to 3.35, I2=97%, τ2=2.96) or non-industry funding (MD -3.09 (95% CI -4.47 to -1.72); 95% PI, I2 and τ2 not applicable). Studies with industry funding for the comparison of high-frequency SCS (HF-SCS) with low-frequency SCS (LF-SCS) showed statistically significant advantages for HF-SCS compared to LF-SCS while studies with no funding showed no differences between HF-SCS and LF-SCS (low certainty evidence).CONCLUSION: All outcomes of SCS versus usual care were not significantly different between studies funded by industry and those independent from industry. Pain intensity score and change in pain intensity from baseline for comparisons of HF-SCS to LF-SCS seem to be impacted by industry funding.

AB - BACKGROUND/IMPORTANCE: Concerns have been raised that effects observed in studies of spinal cord stimulation (SCS) funded by industry have not been replicated in non-industry-funded studies and that findings may differ based on geographical location where the study was conducted.OBJECTIVE: To investigate the impact of industry funding and geographical location on pain intensity, function, health-related quality of life and adverse events reported in randomized controlled trials (RCTs) of SCS.EVIDENCE REVIEW: Systematic review conducted using MEDLINE, CENTRAL, EMBASE and WikiStim databases until September 2022. Parallel-group RCTs evaluating SCS for patients with neuropathic pain were included. Results of studies were combined in random-effects meta-analysis using the generic-inverse variance method. Subgroup meta-analyses were conducted according to funding source and study location. Risk of bias was assessed using Cochrane RoB 2.0 tool.FINDINGS: Twenty-nine reports of 17 RCTs (1823 participants) were included. For the comparison of SCS with usual care, test for subgroup differences indicate no significant differences (p=0.48, moderate certainty evidence) in pain intensity score at 6 months for studies with no funding or funding not disclosed (pooled mean difference (MD) -1.96 (95% CI -3.23 to -0.69; 95% prediction interval (PI) not estimable, I2=0%, τ2=0)), industry funding (pooled MD -2.70 (95% CI -4.29 to -1.11; 95% PI -8.75 to 3.35, I2=97%, τ2=2.96) or non-industry funding (MD -3.09 (95% CI -4.47 to -1.72); 95% PI, I2 and τ2 not applicable). Studies with industry funding for the comparison of high-frequency SCS (HF-SCS) with low-frequency SCS (LF-SCS) showed statistically significant advantages for HF-SCS compared to LF-SCS while studies with no funding showed no differences between HF-SCS and LF-SCS (low certainty evidence).CONCLUSION: All outcomes of SCS versus usual care were not significantly different between studies funded by industry and those independent from industry. Pain intensity score and change in pain intensity from baseline for comparisons of HF-SCS to LF-SCS seem to be impacted by industry funding.

U2 - 10.1136/rapm-2023-104674

DO - 10.1136/rapm-2023-104674

M3 - Review article

C2 - 37611944

SN - 1098-7339

JO - Regional anesthesia and pain medicine

JF - Regional anesthesia and pain medicine

ER -