Clinical Context

Osteoarthritis (OA) and chronic low back pain are 2 of the most common causes of chronic pain in clinical practice, and a study by Gore and colleagues, which appeared in the September 2012 issue of Pain Practice,^[1] evaluated the application of different forms of treatment for these conditions in the United States. Researchers evaluated claims data from 2008 in a large database to assess this issue.

More than 100,000 case patients provided treatment data for OA and chronic low back pain. The most frequently prescribed treatment was opioids, which were used in more than 70% of all patients. More than 50% of patients received nonsteroidal anti-inflammatory drugs (NSAIDs), and more than 30% were prescribed antidepressants. Benzodiazepines were prescribed to more than 20% of patients, and 15% received sedative hypnotics.

Relative to prescription medications, other forms of treatment were less commonly used. There were 20% of patients with OA or chronic low back pain who received physical therapy, and 34% of patients with back pain saw a chiropractor, compared with 11% of patients with OA.

Although NSAIDs are frequently prescribed to patients with OA, there is a paucity of data that compares their relative efficacy across different doses. The current network meta-analysis by Trelle and colleagues addresses this issue.

Study Synopsis and Perspective

Diclofenac 150 mg/day was the most effective NSAID for improving both pain and function in knee or hip OA, and acetaminophen (paracetamol) was least effective and should not be used in this setting, according to a new network meta-analysis published online March 17 in The Lancet.^[2]

"On the basis of the available data, we see no role for single-agent paracetamol for the treatment of patients with osteoarthritis irrespective of dose," write Sven Trelle, MD, University of Bern, Bern, Switzerland, and colleagues. "We provide sound evidence that diclofenac 150 mg/day is the most effective NSAID available at present, in terms of improving both pain and function. Nevertheless, in view of the safety profile of these drugs, physicians need to consider our results together with all known safety information when selecting the preparation and dose for individual patients."

This study reinforces data previously reported by other researchers^[3] that showed similar lack of efficacy for acetaminophen in this setting.

The new report, which includes substantially more patients than the previous network meta-analysis, is likely to increase calls for reconsideration of OA treatment guidelines that position the drug as a first-line treatment. Guidelines in which acetaminophen is recommended as first-line treatment include those from the American College of Physicians, the American Pain Society, the European League Against Rheumatism, the American College of Rheumatology, the Osteoarthritis Research Society International, and the UK National Institute for Health and Care Excellence.

Dr Trelle and colleagues used a network meta-analysis in an attempt to fill in some of the blanks left by previous NSAID efficacy studies, most of which reported only on NSAID pain relief vs placebo "and therefore are only of restricted use for clinical practice," the authors write. The network meta-analysis approach integrates data from all randomized controlled trials (RCTs) that compared different doses of NSAIDs either head to head or with placebo while respecting random selection, and thus allows for comparison among active agents.

Moreover, the authors used an extension of multivariable Bayesian random-effects models for mixed multiple treatment comparisons, which allowed for comparison of all available treatments across trials and accounted for multiple comparisons in trials with more than 2 treatment groups.

Studies eligible for inclusion were RCTs of patients with knee or hip OA that had at least 100 patients randomly assigned to each treatment group. Treatments examined included acetaminophen, rofecoxib, lumiracoxib, etoricoxib, diclofenac, celecoxib, naproxen, and ibuprofen.

The primary outcome was pain, which was assessed at 1 week, 2 weeks, 4 weeks, 6 weeks, 3 months, 6 months, and 12 months. The researchers defined minimal clinically significant difference as a decrease of 0.37 units, corresponding to a 9-mm difference on a 100-mm visual analog scale.

The secondary outcome was physical function, which was assessed at 1 week, 2 weeks, and at the end of treatment.

The authors initially identified 8973 manuscripts, of which 74 RCTs comprising 58,556 patients were ultimately included in the analysis.

Among approved maximal daily doses, diclofenac 150 mg/day and etoricoxib 60 mg/day were the most effective at reducing pain, both with 100% probability of reaching the minimal clinically important difference.

Four other treatments had a 95% probability of reaching the prespecified threshold for clinically significant impact: etoricoxib 30 mg/day and 90 mg/day, and rofecoxib 25 mg/day and 50 mg/day. The authors noted that etoricoxib is less available than diclofenac because it has marketing approval in fewer countries.

Five treatments were not superior to placebo using the data available: acetaminophen less than 2000 mg/day and 3000 mg/day, diclofenac 70 mg/day, naproxen 750 mg/day, and ibuprofen 1200 mg/day.

"The magnitude of treatment effect estimates varied greatly across different NSAIDs and doses," the authors write. "Whereas paracetamol had nearly a null effect on pain symptoms at various doses (effect size of -0.17, corresponding to 4 mm difference on a 100 mm visual analogue scale), diclofenac 150 mg/day had a moderate to large effect size of -0.57, corresponding to difference on a 100 mm visual analogue scale of 14 mm. This is 1.5 times the minimum clinically important difference for chronic pain of -0.37."

NSAID efficacy generally varied with dose, but acetaminophen was ineffective at all doses tested.

In contrast, diclofenac 150 mg/day was the most effective treatment for both pain and physical disability and was superior to the maximal doses of ibuprofen, naproxen, and celecoxib.

In a press statement, Dr Trelle said, "NSAIDs are usually only used to treat short-term episodes of pain in osteoarthritis because the side-effects are thought to outweigh the benefits when used longer term. Because of this, paracetamol is often prescribed to manage long-term pain instead of NSAIDs. However, our results suggest that paracetamol at any dose is not effective in managing pain in osteoarthritis, but that certain NSAIDs are effective and can be used intermittently without paracetamol."

In a linked Comment,^[4] Professor Nicholas Moore and colleagues from the Department of Pharmacology at the University of Bordeaux, Bordeaux, France, wrote that one limitation of the study was that widely used NSAIDs were not included in this meta-analysis. This is probably because no recent trials have been done of these drugs or because any recent trials that did assess them were too small.

Dr Moore and colleagues concluded, "The most remarkable result is that paracetamol does not seem to confer any demonstrable effect or benefit in osteoarthritis, at any dose. This finding is not entirely unexpected. Paracetamol has been on the market for as long as most of us remember. Its efficacy has never been properly established or quantified in chronic diseases, and is probably not as great as many would believe. Its safety is also questioned, not just in overdose."

The commentators suggested that patients might be experiencing needless pain and discomfort because of perceived risks associated with NSAID use and apparently nonexistent benefits of acetaminophen use.

Dr Moore and colleagues write, "Shorter-term intermittent use of NSAIDs with gastroprotection might also explain why upper gastrointestinal bleeding risks derived from full-dose long-term trials without gastroprotection are not found in patients in real-life settings."

Study coauthor Peter Jüni, MD, has received research grants from AstraZeneca, Biotronik, Biosensors International, Eli Lilly, and The Medicines Company and serves as an unpaid member of the steering group of trials funded by AstraZeneca, Biotronik, Biosensors, St Jude Medical, and The Medicines Company. Coauthor Simon Wandel, MD, is an employee of Novartis Pharma AG, Biometrics and Data Management, Novartis Oncology, and was previously an employee of and currently holds shares in Cogitars GmbH Switzerland. All other authors have disclosed no relevant financial relationships. Dr Moore has received grants for studies at the University of Bordeaux and personal fees for work related to NSAIDs and paracetamol from Boots, Reckitt Benckiser, Novartis, Pfizer, Roche, Rhône-Poulenc, Sanofi, and Helsinn. All of the other commentators have disclosed no relevant financial relationships.

Clinical Implications

• A previous study of a large commercial claims database by Gore and colleagues found that opioids were the most frequently prescribed treatment for OA and chronic low back pain.
• The researchers of the current network meta-analysis conclude that diclofenac 150 mg/day appears to be the most effective NSAID for OA.
• Implications for the Healthcare Team: The current study suggests that diclofenac may be slightly more effective compared with other NSAIDs for the treatment of OA. The healthcare team should also evaluate the safety of NSAID use, particularly among high-risk patients, in deciding on the best treatment for OA. These treatment decisions should be tailored to individual patients based on their symptoms and risk profiles.

CME Test