Print Return

In people with chronic low back pain, selective nonsteroidal anti-inflammatory drugs reduce pain

Enthoven WT, Roelofs PD, Deyo RA, et al. Non-steroidal anti-inflammatory drugs for chronic low back pain. Cochrane Database Syst Rev. 2016 Feb 10;2:CD012087.​​​​​​​

Review questions

In people with chronic (long-term) low back pain, do nonsteroidal anti-inflammatory drugs (NSAIDs) improve pain? If so, which NSAIDs work best?


Chronic low back pain is a common problem and cause of disability.

Treatment of low back pain is difficult and often not very effective. NSAIDs are often used to treat chronic back pain. There are 2 categories of NSAIDs: selective COX-2 NSAIDs and nonselective NSAIDs.

How the review was done

The researchers did a systematic review, searching for studies available up to June, 2015.

They found 13 randomized controlled trials with 4,807 people. The average age ranged from 42 to 55 years.

The key features of the trials were:

  • people had low back pain, with no known cause, for at least 12 weeks;
  • treatment was NSAIDs (naproxen, piroxicam patch, etoricoxib, valdecoxib, ibuprofen, diclofenac, or diflunisal);
  • treatment was compared with controls, which included placebo, acetaminophen, other NSAIDs, tramadol, pregabalin, or exercise therapy; and
  • outcomes were measured at 9 days to 4 months after treatment.

What the researchers found

Compared with placebo:

  • overall, NSAIDs reduced pain, but the effect was small;
  • selective COX-2 NSAIDs (etoricoxib and valdecoxib) reduced pain;
  • nonselective NSAIDs (diflunisal, naproxen, and piroxicam) did not reduce pain;
  • NSAIDs reduced disability, but the difference was small; and
  • NSAIDs did not increase side effects.

Compared with acetaminophen, NSAIDs did not reduce pain intensity or increase side effects (1 RCT of 30 people).

Compared with tramadol, NSAIDs increased global improvement and reduced side effects after 6 weeks (1 RCT of 1,583 people).

Compared with pregabalin, NSAIDs did not reduce pain intensity or increase side effects after 4 weeks (1 RCT of 72 people).


In people with chronic low back pain, low-quality evidence suggests that selective COX-2 NSAIDs reduce pain, but nonselective NSAIDs do not.

Effects of nonsteroidal anti-inflammatory drugs (NSAIDs) compared with placebo at up to 4 months in people with chronic low back pain*

OutcomesType of NSAIDNumber of trials (people)Effects of NSAIDs compared with placebo
Pain intensityAll6 (1,354)Small reduction in pain: 3.3 points on a 100-point scale (but could be as few as 1.3 points or as many as 5.3 points)


4 (847)No difference between groups.

Selective COX-22 (507)Reduced pain: 9.1 points on a 100-point scale (but could be as few as 4.7 points or as many as 13.6 points)
DisabilityAll4 (1,161)Small reduction in disability: 0.85 points on a 24-point scale (but could be as few as 0.40 points to 1.30 points)
Side effectsAll6 (1,354)No difference between groups.

*All evidence was of low quality.

This Evidence Summary was originally prepared for the McMaster Optimal Aging Portal.

Published: Monday, August 14, 2017

Please note that the information contained herein is not to be interpreted as an alternative to medical advice from a professional healthcare provider. If you have any questions about any medical matter, you should consult your professional healthcare providers, and should never delay seeking medical advice, disregard medical advice or discontinue medication based on information provided here.