Authors: Atlas SJ., Ann Intern Med 2017 Feb 14;
The American College of Physicians has evaluated a broad array of treatment options.
Sponsoring Organization: American College of Physicians (ACP)
Target Audience: All clinicians
Target Population: Adults with acute (<4 weeks), subacute (4–12 weeks), or chronic (>12 weeks) back pain in primary care settings
The ACP has developed a guideline that updates its 2007 recommendations on noninvasive approaches to treating patients with low back pain (NEJM JW Gen Med Nov 15 2007 and Ann Intern Med 2007; 147:478). Randomized trials and systematic reviews were included in the analysis. Reviewers evaluated outcomes including pain, function, health-related quality of life, work disability and return to work, global improvement, and patient satisfaction. The guideline does not include topical therapies or epidural injections.
- Pharmacologic therapies for acute or subacute low back pain:
- Nonsteroidal anti-inflammatory drugs (NSAIDs) yielded a small improvement in pain and slightly better function, compared with placebo. No significant differences in efficacy were found between traditional NSAIDs and cyclooxygenase (COX)-2–selective NSAIDs.
- Skeletal muscle relaxants (SMRs) improved short-term pain; inconsistent results were found on pain reduction with NSAIDs plus SMRs versus NSAIDs alone.
- Systemic corticosteroids yielded no significant improvement in pain or function.
- Evidence was insufficient to determine the effectiveness of antidepressants, benzodiazepines, antiseizure medications, or opioids.
- Pharmacologic therapies for chronic low back pain:
- NSAIDs were associated with small-to-moderate improvements in pain, and no-to-small improvements in function, relative to placebo. No significant differences in efficacy were found among traditional NSAIDs; no data were available on COX-2–selective NSAIDs.
- Strong opioids were associated with small short-term improvements in pain and function, compared with placebo. Tramadol resulted in moderate short-term pain relief and small improvements in function.
- Evidence was insufficient to evaluate efficacy of SMRs.
- The benzodiazepine tetrazepam improved pain at 5 to 7 days and improved function at 10 to 14 days.
- No difference in pain was seen with tricyclic antidepressants (TCAs) or selective serotonin reuptake inhibitors relative to placebo. Duloxetine was associated with small improvements in pain and function.
- Investigators found insufficient evidence to evaluate the effects of acetaminophen, systemic corticosteroids, or antiseizure medications.
- Nonpharmacologic therapies for acute or subacute low back pain:
- No difference was found between exercise and usual care for pain or function.
- Acupuncture yielded a small reduction in pain and slightly increased likelihood of overall improvement relative to NSAIDs.
- Massage moderately improved short-term pain. Spinal manipulation did not show clear improvement in pain; however, spinal manipulation slightly improved function.
- Heat wraps moderately improved pain.
- Lumbar supports were not beneficial.
- Evidence was insufficient to determine the efficacy of other therapies, including transcutaneous electrical nerve stimulation (TENS), Pilates, tai chi, yoga, or multidisciplinary rehabilitation.
- Nonpharmacologic therapies for chronic low back pain:
- Exercise resulted in small improvements in pain and function.
- Tai chi and yoga resulted in moderate pain improvement; Pilates resulted in small or no improvement.
- Acupuncture yielded moderate reduction in pain.
- Rehabilitation resulted in moderate reductions in pain and disability.
- Massage yielded small improvements in pain.
- Spinal manipulation, ultrasound, and TENS did not yield significant benefits.
- For patients with acute or subacute low back pain, use superficial heat, massage, acupuncture, or spinal manipulation. Add an NSAID or SMR, if needed. (Strong recommendation)
- For chronic low back pain, select a nonpharmacologic therapy. (Strong recommendation)
- For chronic low back pain with inadequate response to nonpharmacologic therapies, NSAIDs should be first-line therapy, with tramadol or duloxetine as second-line therapy. Opioids should be considered only if first- or second-line pharmacologic therapies have failed and after discussing the benefits and harms with the patient. (Weak recommendation)
A significant change from the 2007 guideline is dropping acetaminophen and TCAs from recommended medications. The deletion of acetaminophen is based on a study in which researchers evaluated pain at 3 weeks; no improvement in pain was seen relative to placebo. However, head-to-head studies failed to demonstrate a difference in efficacy between NSAIDs and acetaminophen.
Most patients do not require any specific intervention for low back pain. For those who do, many available options are not very effective — most interventions had only a small benefit on pain or function. Given how little we have to offer, one can summarize these recommendations as: Do no harm (and avoid opioids).