Are NSAIDs more effective than acetaminophen in patients with osteoarthritis?

Journal of Family Practice, Oct, 2001 by David Blandino

Pincus TG, Koch GG, Sokka T, et al. A randomized, double-blind, crossover clinical trial of diclofenac plus misoprostol versus acetaminophen in patients with osteoarthritis of the hip or knee. Arth Rheum 2001; 44:1587-98.

* BACKGROUND Two smaller randomized controlled trials failed to show a statistically significant difference between acetaminophen and the nonsteroidal anti-inflammatory drugs (NSAIDs) ibuprofen and naproxen in the treatment of osteoarthritis (OA). However, survey data showing a benefit with NSAIDs and a meta-analysis suggesting their superiority led the researchers to conduct this larger clinical trial.

* POPULATION STUDIED A total of 227 patients were recruited in 12 ambulatory sites either directly or by advertising. The study was not conducted in a primary care setting. Eighty percent of the patients had already seen a rheumatologist before recruitment in the study. Patients were older than 40 years, had Kellgren-Lawrence radiographic scale grade 2 to 40A of the hip or knee, and a visual analog pain scale score of 30 mm or greater (range=0-100). Patients with severe comorbidities and hypersensitivity to the medications were excluded

* STUDY DESIGN AND VALIDITY This was a double-blind crossover trial with all patients receiving both therapies. The study consisted of 2 treatment periods of 6 weeks each, separated by a 3- to 7-day washout period. In period 1, half the patients took diclofenac 75 mg plus 200 [micro]g misoprostol twice daffy, and the other group of patients took acetaminophen 1000 mg 4 times daffy. Both groups took a placebo of the other medication. In period 2, the therapies for the 2 groups were reversed.

The study protocol had good internal validity with respect to randomization, blinding, and the use of an intention-to-treat analysis. All patients served as their own controls, which greatly enhances our confidence in the results. These results may be less relevant in primary care, since these patients were seen by rheumatologists and probably had more severe or less responsive disease than the typical primary care patient.

* OUTCOMES MEASURED There were 2 primary outcome measures. The first was the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) directed to the primary involved joint. The second was the Multidimensional' Health Assessment Questionnaire (MDHAQ) visual analog pain scale. These pain, stiffness, and functional status scoring systems are the best, although imperfect, measurement tools for OA. Since the scores are derived from patient self-reporting, they are reasonable surrogates for patient-oriented outcomes. Other outcomes measured included gastrointestinal distress, global patient status, a general bodily pain score, and the investigator's estimate of patient status.

* RESULTS For the first 6-week treatment period, the WOMAC index improved by 12.2 points (on a 100-point scale) for the diclofenac-treated patients and by 6.6 points for the acetaminophen-treated patients. For the second period, the improvement was 12.9 points and 2.1 points, again favoring diclofenac. Likewise, the MDHAQ pain scale improved more with diclofenac plus misoprostol in both treatment periods, 20.8 points (also a 100-point scale) compared with 13.1 for acetaminophen in period 1, and 24.6 points versus 0.4 points for acetaminophen in period 2.

Outcomes in the secondary measures showed similar benefit for diclofenac, except for the gastrointestinal distress score, which was worse in the diclofenac-treated patients despite the concomitant use of misoprostol. Overall, 57% of patients felt that the NSAID therapy was better than acetaminophen, but 20% felt that acetaminophen was better. Subgroup analyses indicated that most of these differences were more pronounced in the patients with moderate to severe OA and did not reach statistical significance for mild OA (defined as one third of all patients with the lowest scores based on their composite severity index).

RECOMMENDATIONS FOR CLINICAL PRACTICE

This well-designed trial found the NSAID diclofenac to be more effective titan acetaminophen in patients with moderate to severe arthritis. The 2 drugs provided similar pain relief in patients with mild symptoms. For now, patients with mild OA should still be offered acetaminophen based on its better side effect profile and its therapeutic equivalence. For certain patients with more severe symptoms, NSAIDS will be the better choice. Whether either of these agents should be offered before nonpharmacologic or nonsystemic therapy still has not been adequately studied. Perhaps more important, we can now more enthusiastically recommend NSAIDs for our OA patients who do not have contraindications and who have had an inadequate response to acetaminophen. It should be comforting for them and for us to know that there likely will be an added benefit.

David Blandino, MD
UPMC Shadyside
Pittsburgh, Pennsylvania
E-mail: Blandinoda@msx.upmc.edu