Comparison Of Different Epidural Injection Approaches In The Treatment Of Lumbosacral Radicular Pain:A Bayesian Network Meta-analysis

Objectives:To explore the effectiveness of different epidural injection approaches in adult patients with lumbosacral radicular pain.Methods:The present review was conducted using Bayesian model for network meta-analysis.An electronic literature search was performed in the Pubmed,Embase,Cochrane Library,and Web of Science databases to identify relevant studies that were published up to March 15 2021.Two authors independently performed data extraction and quality assessment.Meta-analysis was performed by Review Manager 5.3software,and network meta-analysis was performed by Stata 14.1 and R 3.6.2software.Bayesian random effects model was conducted to incorporate the estimates of direct and indirect treatment comparisons and rank the interventions in order.Effect estimates from Bayesian network meta-analysis were presented as mean difference（MD）with 95%credibility interval（CI）.Network diagrams were drawn,and the curative effects of different approaches were ranked.Results:Our meta-analysis displayed four epidural steroids injection approaches,including caudal（C）,midline interlaminar（MIL）,parasagittal interlaminar（PIL）and transforaminal（TF）approaches from 7 trials.The outcome measures were pain relief and functional status improvement.In pairwise meta-analysis,significant treatment difference for pain relief was reported for MIL vs TF approaches（MD:1.05;95%CI:0.69 to 1.40;P<0.00001;I~2=0%）in short-term,MIL vs TF approaches（MD:1.29;95%CI:0.95 to 1.64;P<0.00001;I~2=76%）and C vs TF approaches（MD:1.80;95%CI:1.36 to 2.24;P<0.00001;I~2=97%）in intermediate-term.For functional status improvement,significant treatment difference was reported for MIL vs TF approaches（MD:7.28;95%CI:5.52 to 9.04;P<0.00001;I~2=76%）in short-term,MIL vs TF approaches（MD:5.38;95%CI:3.85 to 6.91;P<0.00001;I~2=97%）and C vs TF approaches（MD:3.14;95%CI:1.76 to 4.52;P<0.00001;I~2=96%）in intermediate-term.In network meta-analysis,PIL and TF approaches were ranked in the top two for pain relief and functional improvement,both in short-term or intermediate-term follow-up.Moreover,MIL vs PIL approaches in pain relief（MD:1.16;95%CI:[0.31 to 2.06]）and function improvement（MD:9.9;95%CI:[0.64to 19.94]）,MIL vs TF approaches in pain relief（MD:1.12;95%CI:[0.51 to 1.85]）and functional improvement（MD:1.08;95%CI:[1.08 to 17.08]）shown statistically significant in short-term follow up.In terms of pain relief and functional improvement,PIL approach took the top spot followed by TF,C and MIL approaches in short-term follow up.In intermediate-term follow up,PIL was followed by TF,MIL and C in pain relief,and the ranking of functional improvement was the same as short-term follow up.Conclusion:This network meta-analysis demonstrated that in the epidural steriod injection for the treatment of lumbosacral radicular pain caused by LDH,PIL approach has the highest probability of pain relief and functional improvement,but there was no significant difference between the PIL approach and the TF approach.However,further high-quality hand-to-hand RCTs with larger sample size and longer trial period are required to confirm and update these results.