How Well Do Opioids Stack Up for Musculoskeletal Pain in the ED?

For musculoskeletal pain in the emergency department (ED), opioids weren’t clearly better than other analgesics across studies in a systematic review and meta-analysis.

When evaluated at a median of around 2 hours of follow-up, opioids were not statistically or clinically more effective against pain than nonsteroidal anti-inflammatory drugs (NSAIDs; mean difference [MD] -0.1 on a 100-point visual analog scale, 95% CI -2.5 to 2.3) or systemic anesthetics like ketamine (MD -2.1, 95% CI -8.5 to 4.3).

Opioids appeared inferior to local anesthetics (MD 17.3, 95% CI 1.5-33.1), albeit with “very low certainty of evidence,” reported Caitlin M.P. Jones, PhD, of the University of Sydney, Australia, and colleagues in the Annals of Internal Medicine.

Opioids were statistically more effective in reducing pain compared with acetaminophen (paracetamol; MD -6.7, 95% CI -11.9 to -1.5) and better than placebo (MD -6.3, 95% CI -10.5 to -2.2), but neither difference was deemed clinically significant.

While the goal of opioid analgesic use in the ED is rapid control pain and avoiding unnecessary hospitalization or delay in discharge, Jones’ group noted, “The ED setting is the point of initiation of opioids for many people who may consequently experience opioid-related harms, such as dependence, overdose, and death.”

Because of such risks, ED clinical guidelines have recommended that opioids be prescribed or administered at the lowest effective dose to minimize the risk of harm, although the CDC recently proposed an update to opioid prescribing guidelines that would ease off on dose and duration limits. However, guidelines have not addressed differences in the expected benefits or harms of opioids compared with non-opioid analgesics.

Prior systematic reviews found opioids did not show superiority over acetaminophen for controlling musculoskeletal pain and were not superior to NSAIDs for acute pain after excluding lower back pain. Others have found a small benefit for chronic lower back pain with opioids compared to placebo, but not in an ED setting.

For this study, Jones and colleagues examined data from 42 randomized trials with a total of 6,128 adults who presented to the ED with musculoskeletal pain up to February 7, 2022. Trials of fracture, including patients awaiting surgery or conservative treatment, were allowed, but not those evaluating post-operative pain. Most of the trials assessed pain medication prescribed to adults in the ED (95%), with the remainder at ED discharge (5%). While 14 of the trials were in bony injury, soft tissue injuries accounted for three, spinal pain for four, and mixed presentations for 17.

Of the opioids assessed, morphine was most common (25 studies or groups), followed by oxycodone (six), codeine (five), and three each for fentanyl, hydrocodone, or tramadol. Two studies assessed meperidine (Demerol), and one looked at piritramide/tilidine (not FDA-approved). Notably, 27 evaluated opioids as a single ingredient, while 13 looked at a combination of an opioid with a non-opioid using different routes of administration.

The most common control comparators were systemic anesthetics, including ketamine (18 groups) and NSAIDs (15 groups). Next most common were acetaminophen (eight groups) and local anesthetics (six groups), followed by placebo (three groups) and muscle relaxants (one group).

No clinical or statistical difference was seen between opioids and non-opioids for pain in the short term up to 0.5 hours or at nearly 12 hours of follow-up. Control group patients were discharged faster than opioid patients (between 16-76 minutes faster), although looked at in only three studies.

Serious events were rare, without any significant differences among groups. Three opioid patients experienced hypoventilation and 10 others had non-specified serious events. Among ketorolac and lidocaine/epinephrine controls, two experienced non-specified events and one had total body numbness.

Although opioids had more adverse events overall than the other pain management medications (risk difference 7-21%), the evidence was very uncertain, the authors cautioned.

The authors acknowledged limitations to the data, including unexplained heterogeneity and a short follow-up. Some outcomes were based on low to very low GRADE ratings. Few trials investigated serious injuries and many delivered only a single dose, rather than multiple doses as usually required in clinical practice.