(Last Updated On: January 12, 2018)

Naproxen with Cyclobenzaprine, Oxycodone/Acetaminophen, or Placebo for Treating Acute Low Back Pain: A Randomized Clinical Trial.

Friedman, BW,  et al, JAMA 2015. [paper]

Why I chose this article

I chose this topic because while it may not be the most exciting question it is an important one. This is a question that I seem to deal with every day in both the emergency department (ED) and also in clinic. Acute low back pain is frustrating to treat and I often feel like I am doing my patient’s a disservice prescribing narcotics for them to go home with. I wanted to look at some of the data and see what the most successful treatment strategies are.


It is estimated that roughly 2.7 million visits (2.4% of total visits) occur in US EDs with the complaint of low back pain. Treatment is often ineffective as many patients continue to have significant pain weeks to months after presentation to the ED. There are many different treatment strategies utilized today, usually starting with NSAIDs as there has been shown proven benefit. Muscle relaxants are often used along with opiates, however there is little evidence to show benefit with narcotic usage.

Research Question

Among patients with non-radicular and non-traumatic acute low back pain who present to the ED is there an improvement in pain or functional outcomes with 1-week follow up with adding cyclobenzaprine or Oxycodone/Acetaminophen to a treatment of Naproxen compared to Placebo with Naproxen?

Study Design

This was a randomized, single center, double blinded, placebo controlled study that consisted of three different treatment groups. 2,588 patients with low back pain were assessed for eligibility with 323 eventually chosen to be randomized to 3 different arms.


  • age 21-64
  • Back pain defined as originating between lower border of scapula and the upper gluteal folds.
  • score greater than 5 on the Roland-Morris Disability Questionnaire* (RMDQ).

*The RMDQ is a questionnaire that is used to measure low back pain and functional impairment. It is 24 questions and scored from 0-24 with 24 meaning maximum impairment.


  • Radicular pain defined as pain radiating below gluteal folds.
  • Trauma to back within previous month.
  • Pain lasting longer than 2 weeks.
  • Recent history of greater than 1 low back pain episode per month.
  • Pregnant or lactating.
  • Unavailable for follow up.
  • Chronic opioid use currently or in past.
  • Allergy or contraindication to treatment medications.


All patients in the study received twenty 500-mg Naproxen tablets to be taken as one tablet every 12 hours. Then all patients were given 60 tablets of either placebo (107 patients), cyclobenzaprine (108 patients), or oxycodone/acetaminophen (108 Patients). Patients were instructed to take 1-2 tablets every 8 hours for 10 days. Every patient was also given a 10 minute educational session about how to manage low back pain.


  • Primary Outcome: Improvement (>5 considered significant) on RMDQ between ED discharge and the 7-day telephone follow-up.
  • Secondary Outcome: Measures of pain or functional impairment at 3 months


Analysis was performed using an intention-to-treat model. The primary outcome was analyzed by performing comparisons between the change in the RMDQ at ED discharge and after 1 week from discharge. A clinically significant change in the RMDQ was 5.0. Adjustments were made to account for those lost to follow-up.

  • In the Naproxen + placebo group at 1 week of follow up the RMDQ improved by an average of 9.8
  • Naproxen + Cyclobenzaprine RMDQ improved by 10.1
  • Naproxen + Oxycodone/acetaminophen RMDQ improved by 11.1

When comparing the differences between the RMDQs of the different treatment arms there was no statistical difference between treatment groups at 7 days or 3 months. Oxycodone/acetaminophen users were more likely to report mild pain levels compared to placebo (NNT of 6) but there was no statistical difference between the RMDQs. NNH was greatest with Oxycodone/Acetaminophen at 5.3.


When reading this study a few things stood out. There is a significant amount of exclusion criteria, so this data cannot be generalized to treat all non-radicular, non-traumatic lower back pain. The study also only took place at a single urban ED and so the results should probably only be applied to EDs that have similar populations. Also, most patient used Naproxen regularly but fewer participants took the other medications as prescribed.  However, for those in the acute setting who don’t have chronic back pain and are not on chronic opioids this article gives decent evidence that Naproxen alone is a fine and reasonable treatment.

Often, in my limited clinical practice, patients are sent home with all three medications used in this study. After reading this study I am going to push to send my patients home with just Naproxen if they are similar to the subjects in this study. This could help decrease the amount of narcotics that we prescribe from the ED and is also a better utilization of resources.

Justin Kosirog is a PGY 3 EM/IM resident