Comment on Abdel Shaheed etal.: "Efficacy and tolerability of muscle relaxants for low back pain: Systematic review and meta-analysis".

Abstract

To the Editor: I read with interest the systematic review by Abdel Shaheed and colleagues published on June 22, 2016 (Abdel Shaheed et al., 2016). In their systematic review and meta-analysis on the efficacy and tolerability of muscle relaxants for low back pain (LBP), the authors reported that muscle relaxants provide clinically significant pain relief in patients with acute LBP. Similar observation was reported in a systematic literature review on eperisone (a centrally acting muscle relaxant) in acute LBP patients (Bavage et al., 2016). I have few queries on the methodology and main conclusion of the systematic review, and I request the authors to address the same. First, the systematic review included published randomized controlled trials (RCTs); the authors could have also searched for grey RCTs which meet the inclusion criteria of this systematic review (NCT00671879 and NCT00671502, Table 1). Furthermore, the authors could have also requested sponsors/original investigators for data of the completed trials without results in the US National Institutes of Health database (NCT02646124, NCT00679146, NCT01300312, and NCT01806818, Table 1) (Young and Hopewell, 2011). Including unpublished trials in the meta-analysis are useful to assess the presence and magnitude of publication bias (McAuley et al., 2000; Conn et al., 2003; Hopewell et al., 2007; Furlan et al., 2015). The authors, however, reported this as one of the major limitations of their systematic review which possibly included potential publication bias. Second, under study selection the authors stated that only placebo-controlled RCTs and RCTs comparing two drugs from the same class or different doses of the same drug were eligible for inclusion. Furthermore, they reported that no eligible RCTs evaluating benzodiazepines in LBP were identified. However, there is a RCT conducted by Rusinyol and colleagues which meets the inclusion criteria (Rusinyol et al., 2009). This trial evaluated the efficacy and tolerability of two different doses of eperisone (150 and 300 mg/day) vs. diazepam (15 mg/day) in acute LBP patients and also performed intergroup comparison between two different doses of eperisone. Exclusion of this RCT might have also contributed to publication bias. Third, the meta-analysis was performed on pain and disability outcomes (primary endpoints) converted to a common 0–100 scale (Figures 2 and 4; Table S7) and adverse event rates for muscle relaxant vs. placebo trials (Table S5) (Abdel Shaheed et al., 2016). In addition, the authors could have also meta-analysed global measure (overall improvement, percentage of patients reporting subjective improvement) and physiological outcomes (e.g. hand-to-floor distance, Lasegue test) which also play important role in assessing the efficacy of muscle relaxants for LBP (van Tulder et al., 2003; Furlan et al., 2015; Gagnier et al., 2016). Last, in the abstract, what does this review add? and discussion, the authors reported that muscle relaxants provide clinically significant short-term pain relief for people with acute LBP. However, under Conclusion the authors quoted that “muscle relaxant drugs do not provide clinically significant pain relief in the short term for people with acute LBP.” I am eager to know the authors’ view on this. Randomized, double-blind, double-dummy, placebo-controlled, parallel-group study N = 830 and N = 840 Carisoprodol SR tablet 700 mg BID Carisoprodol SR tablet 500 mg BID Placebo Primary outcome Subject rated change relief from starting backache of pain on a 100-point VAS Secondary outcomes Subject functional assessment based on the RMDQ Adverse event assessment (serious AE and other AEs) Randomized, double-blind, parallel-group study N = 116 Naproxen 500 mg po BID+placebo capsules Naproxen 500 mg po BID+diazepam 5 mg capsules, 1-2 tabs po BID Primary outcome RMDQ Secondary outcomes Pain Analgesic use RMDQ Satisfaction with treatment Days of impairment Randomized, double-blind, parallel-group study N = 334 1 tablet TCC 8 mg+ketoprofen 100 mg BID+2 tablets TCC placebo BID 2 tablets TCC 4 mg BID+1 tablet of FDC placebo BID Primary outcome Average pain within the last 24 hours (VAS) Secondary outcome Average pain within the last 24 hours (VAS) An observer blind, parallel-group, prospective, randomized, controlled study N = 239 Patient of either sex between 18 and 60 years of age with confirmed diagnosis of acute musculoskeletal spasm with LBP FDC of eperisone hydrochloride 50 mg+diclofenac sodium 50 mg TID Eperisone hydrochloride 50 mg tablet TID Primary outcome Hand-to-floor distance Secondary outcomes Lasegue's sign is the lumbar pain (objective efficacy) Lumbar cinesalgia (subjective efficacy) Randomized, double-blind, parallel-group, placebo-controlled, multi-centre study N = 98 Primary outcome VAS improvement of pain at rest, on movement and at night Secondary outcomes Change in modified Schober and lateral body bending tests, and Oswestry Disability Index Change in EuroQoL by patients Overall pain improvement by investigator