Introduction

Chronic low back pain (CLBP) affects more than one billion people worldwide. Despite the high prevalence, many treatments for CLBP only display small-to-moderate effects on pain and disability. Subchondral vertebral bone marrow non-neoplastic lesions, known as Modic changes (MCs), have been associated with CLBP. It is hypothesized that MCs are related to low-grade discitis caused by Cutibacterium acnes. Although some trials reported that three months of Amoxicillin-clavulanate treatment was effective in treating patients with CLBP, the findings remain controversial. Therefore, this systematic review aimed to summarize evidence regarding the effectiveness of oral antibiotic intervention for CLBP.

Methods

Five databases (AMED, CINAHL, Cochrane Library, Embase, and Medline) were searched from inception to November 15, 2021. Randomized controlled trials (RCTs) or non-RCTs that investigated the effectiveness of oral antibiotics in treating patients with CLBP were eligible for inclusion. Two independent reviewers screened titles, abstracts, and full-text articles, as well as extracted data. They also independently evaluated the methodological quality of the included RCTs and non-RCTs using RoB 2 and ROBINS-I, respectively. The quality of evidence for the treatment effectiveness was appraised by GRADE.

Results

A total of 148 potential articles were identified. After removing duplicates, 135 abstracts and 13 full-text articles were screened. Five RCT articles (from three cohorts) and four case series were included. Strong evidence supported that Amoxicillin-clavulanate or Amoxicillin was significantly better than placebo in reducing Roland Morris Disability Questionnaire or Oswestry Disability Index scores, and improving quality of life in patients with CLBP and concomitant type 1 MCs (as determined by T1/T2 or STIR sequences) at the 1-year follow-up. Limited evidence from non-RCTs suggested that three months of oral Amoxicillin-clavulanate significantly improved LBP and leg pain intensity, number of days with LBP, and LBP-related disability. Likewise, moderate evidence from a RCT suggested that three months of oral Amoxicillin-clavulanate was significantly better than placebo in improving pain and global perceived health at the 1-year follow-up. Conversely, moderate evidence found that three months of oral Amoxicillin alone was not significantly better than placebo in improving LBP intensity or global perceived health in patients with CLBP at the 1-year follow-up. A cost-effectiveness trial revealed that the healthcare cost of the Amoxicillin group was at least double that of the placebo group. Some patients receiving oral antibiotics also reported significantly more adverse effects.

Discussion

This is the first systematic review, to our knowledge, performed on this topic. While strong evidence supports that oral Amoxicillin-clavulanate or Amoxicillin are superior to placebo in reducing LBP-related disability in a subgroup of patients with CLBP and concomitant type 1 MC, it remains uncertain whether oral antibiotics can improve clinical symptoms in patients with CLBP. Since antibiotics-related side effects might cause participants’ bias and different magnetic resonance imaging sequences might affect how MCs phenotype was assessed, the existing findings should be interpreted with caution. Given the high-cost and potential adverse effects/antibiotic resistance of antibiotics, future high-quality RCTs should replicate the results before changing clinical practice and guidelines.