Conservative treatments for chronic non-specific low back pain:a time course network meta-analysis

<div><b>Objective</b>: </div><div>Examine efficacy of conservative (non-surgical) treatments including acupuncture, education or advice, electrotherapy, exercise training, manual therapies/manipulation, massage, McKenzie, pharmacotherapy, psychological therapies, traction, physical therapy, placebo, multidisciplinary pain management, usual care and, no treatment for chronic radicular and non-specific back pain (CLBP) via time-course network meta-analysis (NMA). </div><div><br/></div><div/><div><b>Design</b>: </div><div>Systematic review with time-course NMA.</div><div><br/></div><div><b>Data</b> <b>sources</b>: </div><div>Six electronic databases and 302 prior systematic reviews. </div><div><br/></div><div><b>Eligibility</b> <b>criteria</b>: </div><div>Randomised controlled trials examining conservative treatments for CLBP. </div><div><br/></div><div><b>Data extraction and synthesis</b>: </div><div>Cochrane Risk of Bias version 1 was used. Bayesian time-course NMA was conducted. RCTs of non-specific and radicular CLBP were separated in sub-group analyses. Conclusions were drawn via the minimally contextualized GRADE framework for NMA with a minimal clinically important difference [MCID] of 0.5 SD.</div><div>Main outcomes and measures: back pain, leg pain, disability, mental health at immediate (&lt;1 day), short-term (≥1 day but ≤3 months), intermediate-term (&gt;3 but &lt;12 months) and long-term (≥12 months) time-points.</div><div><br/></div><div><b>Results</b>: </div><div>581 reports of 551 studies (71126 patients) were included. 510 trials included people with non-specific CLBP and 41 people with radicular CLBP. For back pain (0-100 scale), acupuncture (mean difference: -20.91, CrI:[-24.00; -11.95]), electrotherapy (-18.98[-21.84;-10.95]), exercise (-15.59[-17.51;-10.05]), manual therapy (-19.48[-22.17;-11.74]), massage (-25.61[-30.42;-10.91]), multidisciplinary pain management (-18.96[-22.26;-9.58]) exceeded MCID at short-term. For disability (0-100 scale), acupuncture (-10.52[-11.84;-6.59]), massage (-9.95[-11.45;-5.50]) and multidisciplinary pain management (-12.56[-13.91;-8.55]) were effective at short-term; in the immediate- and intermediate-term only McKenzie and massage, respectively, exceeded MCID. Long-term, no treatment yielded clinically important effects. While two of 14 treatments for back pain and nine of 14 for disability showed statistically significant benefits compared to no treatment, all effects remained below the MCID. Certainty was low (1.4%) to very-low (98.6%) across interventions and time-points. Findings for massage and McKenzie were not stable in sensitivity analyses. Treatment effects for radicular CLBP did not appear to differ to non-specific CLBP.</div><div><br/></div><div><b>Conclusions</b>: </div><div>Some treatments are effective for pain and function in non-specific CLBP, but improvements do not persist long-term. The majority of evidence was available for non-specific CLBP; the evidence base for radicular CLBP is limited. Whilst sensitivity analyses did not provide evidence for a different response in radicular CLBP, there remains an evidence gap for this sub-population. We argue future work could explore strategies to establish long term efficacy via lifestyle- and behavioural-modification.</div><div><br/></div><div><b>Registration</b>: CRD42020182039</div><div><br/></div><div><b>Funding</b>: Nil</div>