A short opioid course for acute back and neck pain did not beat placebo

In the only large placebo-controlled trial of a short opioid course for acute low back and neck pain, the drug did not reduce pain more than placebo — and at six weeks the opioid group's mean pain score was slightly higher, not lower [s1]. For chronic back and osteoarthritis pain, a separate twelve-month randomised trial found opioids no better than non-opioid drugs on function, and worse on pain intensity [s2]. Neither result says pain should go untreated. Both say the drug class that has been reached for by default was not doing what it was assumed to do.

OPAL

OPAL was a triple-blinded, placebo-controlled randomised trial recruiting adults presenting to one of 157 primary care or emergency department sites in Sydney, Australia, with twelve weeks or less of low back or neck pain, or both, of at least moderate severity [s1]. Participants were randomised 1:1 to guideline-recommended care plus an opioid — oxycodone–naloxone, up to 20 mg of oxycodone per day orally — or to guideline-recommended care plus an identical placebo, for up to six weeks [s1]. Recruitment ran from 29 February 2016 to 10 March 2022 and enrolled 347 participants, 174 to the opioid group and 173 to placebo; 170 of 346 (49%) were female and 176 (51%) male [s1].

The primary outcome was pain severity at six weeks on the Brief Pain Inventory severity subscale, a 10-point scale. Mean pain was 2.78 (SE 0.20) in the opioid group and 2.25 (0.19) in the placebo group, an adjusted mean difference of 0.53 (95% CI −0.00 to 1.07, P = 0.051) [s1]. That difference points the wrong way for the drug. It does not reach conventional significance, and 0.53 points on a 10-point scale would not be clinically meaningful in either direction — but the trial certainly did not show a benefit.

Attrition was slightly higher in the opioid arm: 33 of 174 (19%) discontinued by week 6, against 25 of 172 (15%) on placebo, leaving 151 and 159 participants in the primary analysis [s1]. Overall adverse events did not differ significantly — 61 of 174 (35%) in the opioid group and 51 of 172 (30%) on placebo, P = 0.30 — but opioid-related events were more common, with constipation reported by 13 of 174 (7.5%) against six of 173 (3.5%) [s1].

The authors' conclusion is unambiguous: opioids should not be recommended for acute non-specific low back pain or neck pain, and the finding calls for a change in their frequent use for these conditions [s1].

SPACE, for chronic pain

The chronic-pain question was tested pragmatically. SPACE recruited patients from Veterans Affairs primary care clinics between June 2013 and December 2015 who had moderate to severe chronic back pain or hip or knee osteoarthritis pain despite analgesic use [s2]. Of 265 patients enrolled, 25 withdrew before randomisation and 240 were randomised [s2]. Both arms followed a treat-to-target strategy with three medication steps: the opioid arm began with immediate-release morphine, oxycodone or hydrocodone with acetaminophen; the non-opioid arm began with acetaminophen or an NSAID [s2]. Mean age was 58.3 years and 32 participants (13.0%) were women; 234 of 240 (97.5%) completed the trial [s2].

Over twelve months, pain-related function did not differ between groups (overall P = .58): mean 12-month Brief Pain Inventory interference was 3.4 in the opioid group and 3.3 in the non-opioid group, a difference of 0.1 (95% CI −0.5 to 0.7) [s2]. Pain intensity was significantly better in the non-opioid group (overall P = .03), at 4.0 versus 3.5 on the BPI severity scale, a difference of 0.5 (0.0 to 1.0) [s2]. Medication-related adverse symptoms were significantly more common in the opioid group (P = .03), at a mean of 1.8 against 0.9 at twelve months [s2].

Acute injury outside the back

A systematic review of opioid versus non-opioid drugs for acute traumatic pain assembled 14 studies — 12 randomised trials and two pseudo-randomised trials — covering 2,347 patients, most of them in emergency departments, with a low-to-moderate risk of bias [s3]. The studies were heterogeneous in both drug and outcome, with only two homogeneous enough to compare directly, and a minimum clinically important difference was evaluated in eight of them [s3]. The review's conclusion is that non-opioids can be considered an alternative to opioids for short-term management of acute musculoskeletal injury, and it notes that intravenous ketamine may cause more adverse events than other routes of administration [s3].

What these trials do not show

They do not show that opioids are ineffective analgesics. They show that in specific conditions — acute non-specific back and neck pain, chronic back and osteoarthritis pain — the average patient did no better on them than on the comparator over the trial period. Cancer pain, post-surgical pain, palliative care and severe acute injury were not the populations studied, and nothing here speaks to them.

They also do not argue for leaving pain untreated. Under-treatment is a harm with its own literature, and the alternative arms in these trials were not "nothing": OPAL's placebo group received guideline-recommended care [s1], and SPACE's non-opioid arm received a structured, escalating medication strategy aimed at the same target [s2]. The comparison being made is between two ways of treating pain, not between treating and ignoring it.

The American College of Physicians guideline positions opioids accordingly: an option only for patients who have not responded to the alternatives, only where the potential benefits outweigh the risks for that individual, and only after a discussion of the known risks and realistic benefits — a weak recommendation on moderate-quality evidence [s4].

What to watch

OPAL took six years to recruit 347 patients for a six-week course of a widely prescribed drug, which is itself a signal about how hard placebo-controlled opioid trials are to run [s1]. Whether anyone attempts a larger one, and whether prescribing in the emergency and primary care settings OPAL sampled actually shifts, are the two things worth following.

This article is informational and is not medical advice. Doses named here are trial parameters, not guidance. Decisions about analgesia belong with a reader and their clinician.

Sources

• Opioid analgesia for acute low back pain and neck pain (the OPAL trial): a randomised placebo-controlled trial — The Lancet, 2023-06-28
• Effect of Opioid vs Nonopioid Medications on Pain-Related Function in Patients With Chronic Back Pain or Hip or Knee Osteoarthritis Pain: The SPACE Randomized Clinical Trial — JAMA, 2018-03-06
• Pain management in acute musculoskeletal injury: Effect of opioid vs nonopioid medications — World Journal of Orthopedics, 2024-09-18
• Noninvasive Treatments for Acute, Subacute, and Chronic Low Back Pain: A Clinical Practice Guideline From the American College of Physicians — Annals of Internal Medicine, 2017-02-13