Opioids compared to placebo or other treatments for chronic low-back pain.

BACKGROUND The use of opioids in the long-term management of chronic low-back pain (CLBP) has increased dramatically. Despite this trend, the benefits and risks of these medications remain unclear. This review is an update of a Cochrane review first published in 2007. OBJECTIVES To determine the efficacy of opioids in adults with CLBP. SEARCH METHODS We electronically searched the Cochrane Back Review Group's Specialized Register, CENTRAL, CINAHL and PsycINFO, MEDLINE, and EMBASE from January 2006 to October 2012. We checked the reference lists of these trials and other relevant systematic reviews for potential trials for inclusion. SELECTION CRITERIA We included randomized controlled trials (RCTs) that assessed the use of opioids (as monotherapy or in combination with other therapies) in adults with CLBP that were at least four weeks in duration. We included trials that compared non-injectable opioids to placebo or other treatments. We excluded trials that compared different opioids only. DATA COLLECTION AND ANALYSIS Two authors independently assessed the risk of bias and extracted data onto a pre-designed form. We pooled results using Review Manager (RevMan) 5.2. We reported on pain and function outcomes using standardized mean difference (SMD) or risk ratios with 95% confidence intervals (95% CI). We used absolute risk difference (RD) with 95% CI to report adverse effects. MAIN RESULTS We included 15 trials (5540 participants). Tramadol was examined in five trials (1378 participants); it was found to be better than placebo for pain (SMD -0.55, 95% CI -0.66 to -0.44; low quality evidence) and function (SMD -0.18, 95% CI -0.29 to -0.07; moderate quality evidence). Transdermal buprenorphine (two trials, 653 participants) may make little difference for pain (SMD -2.47, 95%CI -2.69 to -2.25; very low quality evidence), but no difference compared to placebo for function (SMD -0.14, 95%CI -0.53 to 0.25; very low quality evidence). Strong opioids (morphine, hydromorphone, oxycodone, oxymorphone, and tapentadol), examined in six trials (1887 participants), were better than placebo for pain (SMD -0.43, 95%CI -0.52 to -0.33; moderate quality evidence) and function (SMD -0.26, 95% CI -0.37 to -0.15; moderate quality evidence). One trial (1583 participants) demonstrated that tramadol may make little difference compared to celecoxib (RR 0.82, 95% CI 0.76 to 0.90; very low quality evidence) for pain relief. Two trials (272 participants) found no difference between opioids and antidepressants for either pain (SMD 0.21, 95% CI -0.03 to 0.45; very low quality evidence), or function (SMD -0.11, 95% -0.63 to 0.42; very low quality evidence). The included trials in this review had high drop-out rates, were of short duration, and had limited interpretability of functional improvement. They did not report any serious adverse effects, risks (addiction or overdose), or complications (sleep apnea, opioid-induced hyperalgesia, hypogonadism). In general, the effect sizes were medium for pain and small for function. AUTHORS' CONCLUSIONS There is some evidence (very low to moderate quality) for short-term efficacy (for both pain and function) of opioids to treat CLBP compared to placebo. The very few trials that compared opioids to non-steroidal anti-inflammatory drugs (NSAIDs) or antidepressants did not show any differences regarding pain and function. The initiation of a trial of opioids for long-term management should be done with extreme caution, especially after a comprehensive assessment of potential risks. There are no placebo-RCTs supporting the effectiveness and safety of long-term opioid therapy for treatment of CLBP.