Low back pain is a particularly frustrating condition that is common, poorly understood, and difficult to treat. Could a long course of antibiotics be the answer for some patients? A recent study from Denmark suggests that it might be: “Antibiotic treatment in patients with chronic low back pain and vertebral bone edema (Modic type 1 changes): a double-blind randomized clinical controlled trial of efficacy” by Albert, Sorensen, Christensen and Manniche. Is this a crazy idea like long-term antibiotics for “chronic Lyme disease” or will it pan out like antibiotic eradication of H. pylori in patients with ulcers? Time will tell. This was a rigorous, well-done study, but we can never rely on the results of a single study until it is replicated or confirmed elsewhere.
Background
Modic type 1 changes represent edema in the bone and can be seen only on MRI. They are present in 6% of the general population and 35-40% of the low back pain population. Infection is one of the hypothetical causes of bone edema, and several studies have found bacteria in material removed during disc surgery. Low virulence anaerobic organisms (Proprionibacterium acnes and Corynebacterium propinquum) were found in 53% of patients in one study. Another study found bacteria in 37% of disc material from patients with herniated discs and 0% of disc material from patients with other spinal disorders. It is thought that these skin and mouth organisms spread to the disc during normal bacteremias as a result of neovascularization associated with disc disease; the Modic changes may be a side effect manifested in the adjacent bone rather than a bone infection per se.
An uncontrolled pilot study of 32 patients had shown promising results with antibiotic treatment. This new double blind study confirmed those preliminary results.
Study Methods
Subjects were 162 chronic back pain patients recruited from spine centers. They had to have MRI-confirmed disc herniation within the previous 6-24 months, with Modic type 1 changes on MRI. Subjects had had either conservative or surgical treatment; patients with and without sciatica and neuropathic pain were included. Randomization and blinding were rigorous. The placebo was a calcium carbonate pill indistinguishable from the active treatment. The antibiotic, amoxicillin-clavulanate (500mg/125mg), was given 3 times a day for 100 days, consistent with the recommendations of infectious disease experts based on bacterial culture results of previous studies. There were 4 treatment groups: single and double dose antibiotics, single and double dose placebos. MRIs, symptom questionnaires, clinical exams, and blood tests were done at baseline, at end of treatment, and at a 12-month follow-up.
Results
The antibiotic group improved on the primary outcome measures of disease-specific disability and lumbar pain, and also on the secondary outcome measures of global perceived effect, leg pain, hours of LBP during the last 4 weeks, the EQ-5D Thermometer (a measure of quality-adjusted health status), days with sick leave, bothersomeness, constant pain, MRI Modic grading, and physical exam measures (i.e. pain with Valsalva maneuver). Improvement in all these factors was clinically as well as statistically significant at p levels from .05 to .0001. Pain relief was gradual, beginning at 6-8 weeks and sometimes continuing long after the end of the treatment period. On MRI there was a significant decrease in the volume of Modic-1 findings in the antibiotic group but no reduction in the placebo group. Side effects (mostly gastrointestinal) were reported by 65% in the antibiotic group and 23% in the placebo group. There were 13 dropouts in the antibiotic group (only 4 of these were due to side effects) and 5 in the placebo group (due to no shows and new disc herniations). There was a trend towards a dose-response relationship with double-dose antibiotics being more effective (not statistically significant).
Discussion
The authors point out that the improvements in this group of traditionally resistant chronic low back pain patients with this protocol were substantially greater than those described with other treatments. They felt the course of improvement was consistent with slow resolution of anaerobic infection in poorly vascularized tissue followed by gradual healing. They stressed that their findings applied only to this specific group of patients with disc herniation and Modic-1 changes on MRI, not to all patients with back pain. They also stressed that high-dose long-term antibiotics should not be prescribed without careful consideration because of the risks to individuals and to the community (antibiotic resistance).
Conclusion
This study is encouraging but far from definitive. It might justify antibiotics for patients with chronic low back pain and Modic-1 MRI changes following a herniated disc when they have failed to respond to all other treatment options. This was a well-designed study, carefully carried out, with a credible rationale, impressive results, and a cautious interpretation. This is how science should be done.
The authors cautioned:
We rely on our fellow colleagues to use clear evidence-based criteria and to avoid excessive antibiotic use.
Back pain is a frustrating problem, and patients who learn about these results may ask for a trial of antibiotics even if they don’t fit the criteria of the study. That would not be justified and might be expected to do more harm than good. Chronic low back pain is a mixed bag with various etiologies, and it must be stressed that this study addressed only one very limited sub-group of back pain patients. That said, it is a ray of hope for those patients, and I hope it will be confirmed.
This article was originally published in the Science-Based Medicine blog.
