Non-steroidal anti-inflammatory drugs, including cyclo-oxygenase-2 inhibitors, in osteoarthritic knee pain: meta-analysis of randomised plac
http://www.100md.com
《英国医生杂志》
1 Department of Public Health and Primary Health Care, University of Bergen, 5018 Bergen, Norway, 2 Department of Laboratory Medicine, Children's and Women's Health, Norwegian University of Science and Technology, 7489 Trondheim, Norway
Correspondence to: J M Bjordal jmb@hib.no
Abstract
Osteoarthritis of the knee is the most common type of osteoarthritis,1 the prevalence of which is rising in parallel with the increasing age of the population.2 The condition is associated with pain and inflammation of the joint capsule,3-5 impaired muscular stability,6 7 reduced range of motion,8 and functional disability.9 Treatment guidelines for knee osteoarthritis recommend pharmacological intervention, initially with paracetamol and subsequently with a non-steroidal anti-inflammatory drug (NSAID).10 In a recent UK survey, 15% of patients with osteoarthritis used paracetamol, whereas 50% reported regular use of NSAIDs. Of the latter, 32% were using traditional NSAIDs and 18% were using cyclo-oxygenase-2 inhibitors (coxibs).11 This widespread use is one explanation for the interest in tolerability and efficacy issues regarding these drugs.10 12 13 The recent introduction of coxibs seemed to promise a reduction in serious adverse events related to NSAIDs,13 14 but this remains controversial.15-18
Guidelines from the European League Against Rheumatism (EULAR) state that both pharmacological and non-pharmacological interventions are needed for optimal treatment of knee osteoarthritis.19 The various potentially effective pharmacological interventions at the clinicians' disposal19 highlight the need for information regarding treatment efficacy.
Meta-analyses can be used for reliable comparison of the efficacy of different interventions.20 Effect size measures the magnitude of a treatment effect independent of sample size.21 There is no current operational definition for what constitutes a sufficiently large effect size for a therapeutic intervention to be considered as useful, but a value of 0.2 is usually considered small, 0.5 moderate, and 0.8 large.22
A recent systematic review of therapeutic alternatives in knee osteoarthritis gives no effect sizes for paracetamol and an imprecise range (0.47-0.96), derived from a minority of available trials, for NSAIDs.19 Neither other reviewers nor the Cochrane library provide comprehensive and robust effect size data for the efficacy of either of these interventions in osteoarthritis of the knee.10 13 23-25 Calculations of effect size require data for mean change and standard deviation (SD). If not provided, these data can be obtained by indirect means from standard errors, P values, t values, and 95% confidence intervals when sample sizes are known. The lack of data on effect size is surprising because treatment with NSAIDs for knee osteoarthritis is established to the point of being a reference against which other interventions are often compared.
We carried out a meta-analysis of published randomised placebo controlled trials to estimate the analgesic efficacy of NSAIDs, including coxibs, in patients with knee osteoarthritis.
Methods
Included studies
We evaluated 268 randomised controlled trials of NSAIDs for knee osteoarthritis (fig 1). Of these, 4 did not provide pain scale data, 126 did not have a placebo control group, and 115 presented combined data for osteoarthritis in several sites with no separate data for the knee. This provided a final sample of 23 trials that satisfied the inclusion criteria.30-52 Of these, 16 were sponsored by the pharmaceutical industry,31-34 36 39 40 43 44 46-52 while three others did not state sponsorship but gave an address of a pharmaceutical company as the workplace of most of the authors.37 38 41 The final sample included 10 845 patients, of whom 7767 received NSAIDs and 3078 received placebo (table).
Fig 1 Selection of trials for inclusion in meta-analysis
Characteristics of trials of NSAIDs for pain relief in patients with knee osteoarthritis
Patients and possible selection bias
The included patients had a median age of 62.5 years, and three trials had an upper age limit of 75 years.35 45 49 There were more women (67.9%) than men, and the median duration of symptoms was 8.2 years. The weighted mean baseline pain of 64.1 mm on a visual analogue scale was calculated from all but three trials.37 51 52 Six reports provided data of body mass index with a median mean value of 31.2.32 35 38 43 44 48 Most trials excluded individuals with concomitant health disorders by a median of 14 exclusion criteria. All trials had a minimum limit for pain intensity or disease activity for inclusion, and they all used a pretreatment washout period of 3-14 days for previous pharmacotherapy. Thirteen trials used an additional criterion by requiring a predefined minimum flare of symptoms when NSAID treatment was discontinued in the pretreatment wash out period.31 33-35 39 40 43 44 46 47 50-52 Five of these trials reported the proportion of regular NSAIDs users, ranging from 66% to 100% (median 90.5%).31 34 35 40 47
Trial quality
The methodological quality was adequate or good (table). All trials were randomised and double blinded, but adequate randomisation procedure, concealed allocation to groups, and blinding procedure were described satisfactorily in only eight studies.33 34 36 39 43 44 46 47 All trials described dropouts and withdrawals well, but one trial did not perform intention to treat analyses.45
Presentation of data on outcome measures
Only four trials presented outcome data as the mean difference of change with SD between NSAIDs and placebo groups.34 42 46 48 Fourteen trials presented the mean difference of change for each group with P values, SE of mean, or 95% confidence intervals but not mean differences between groups.31 33 35-39 41 43 47 49-52 Four trials did not present mean change data but only before and after means and P values.30 32 44 45 Eleven trials presented data on overall pain on 100 mm visual analogue scales or on a categorical five point scale,30-32 37 38 41 42 47 50-52 while the 12 remaining trials presented either categorical or continuous data from the WOMAC subscale for pain. All 23 trials reported data on pain intensity, and 11 trials reported data on functional disability.
Effect size for reduction in functional disability and pain
We excluded from analysis six intervention groups (n = 867) in five trials because patients did not receive an adequate NSAID dose.34-36 39 43 As most trials with multiple time points showed rather stable results from 2-13 weeks, we pooled data. Tests for heterogeneity showed positive results for reduction in functional disability (Q = 39.9, P < 0.01) and reduction in pain (Q = 56.6, P < 0.01). For this reason, we decided to use a random effects model. Eleven trials with 7433 patients provided separate scores for reduction in functional disability,33 35 36 39-41 43 44 46-48 and their combined effect size was 0.29 (95% confidence interval 0.18 to 0.40). One trial reported long term effects on pain but found no significant difference between tiaprofenic acid and placebo at one, two, three, and four years after start of treatment.42 For short term effects (2-13 weeks) the pooled effect size of all included trials was 10.1 mm on visual analogue scale (7.4 to 12.8) or 15.6% better than placebo, and the effect size was 0.32 (0.24 to 0.39) (fig 2).
Fig 2 Effect size (pain) for all trials on NSAID use in knee osteoarthritis
Subgroup analyses
We carried out post hoc subgroup analyses as heterogeneity was evident and trial procedures differed in selection of patients, duration of treatment, and pain scales. For subgroups of trials that used short durations of treatment (< 6 weeks) or WOMAC subscale for pain, neither effect size (0.35 or 0.39) nor heterogeneity (Q = 48.3, P < 0.01, or Q = 43.0, P < 0.01) changed significantly from the results of the total material. For the subgroup of 10 trials (n = 4565) that did not require patients to have a minimum flare of symptoms after treatment with NSAIDs was stopped before the trial, trial results were homogeneous both for function (Q = 2.6, P = 0.275) and pain (Q = 10.0, P = 0.263).30 32 36-38 41 42 45 48 49 Three of these trials (n = 2928) provided data for reduction in functional disability36 41 48 and we calculated effect size by a fixed effects model to be 0.20 (0.09 to 0.30). For pain reduction, we a used fixed effect model to calculate a pooled effect size of 0.23 (0.16 to 0.31) or 5.9 mm on visual analogue scale (3.8 to 7.9) (fig 3).
Fig 3 Effect size (pain) for trials on NSAID use in knee osteoarthritis without identified selection bias
Discussion
Andrianakos A, Trontzas P, Christoyannis F, Dantis P, Voudouris C, Georgountzos A, et al. Prevalence of rheumatic diseases in Greece: a cross-sectional population based epidemiological study. The ESORDIG study. J Rheumatol 2003;30: 1589-601.
Felson DT, Lawrence RC, Dieppe PA, Hirsch R, Helmick CG, Jordan JM, et al. Osteoarthritis: new insights. Part 1: the disease and its risk factors. Ann Intern Med 2000;133: 635-46.
Vaatainen U, Lohmander LS, Thonar E, Hongisto T, Agren U, Ronkko S, et al. Markers of cartilage and synovial metabolism in joint fluid and serum of patients with chondromalacia of the patella. Osteoarthr Cartil 1998;6: 115-24.
Suenaga S, Abeyama K, Indo H, Shigeta K, Noikura T. Temporomandibular disorders: MR assessment of inflammatory changes in the posterior disk attachment during the menstrual cycle. J Comput Assist Tomogr 2001;25: 476-81.
Speldewinde GC, Bashford GM, Davidson IR. Diagnostic cervical zygapophyseal joint blocks for chronic cervical pain. Med J Aust 2001;174: 174-6.
Radebold A, Cholewicki J, Polzhofer GK, Greene HS. Impaired postural control of the lumbar spine is associated with delayed muscle response times in patients with chronic idiopathic low back pain. Spine 2001;26: 724-30.
Cowan SM, Bennell KL, Hodges PW, Crossley KM, McConnell J. Delayed onset of electromyographic activity of vastus medialis obliquus relative to vastus lateralis in subjects with patellofemoral pain syndrome. Arch Phys Med Rehabil 2001;82: 183-9.
Steultjens MP, Dekker J, van Baar ME, Oostendorp RA, Bijlsma JW. Range of joint motion and disability in patients with osteoarthritis of the knee or hip. Rheumatology (Oxford) 2000;39: 955-61.
Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and therapeutic criteria committee of the American Rheumatism Association. Arthritis Rheum 1986;29: 1039-49.
Wegman A, van der Windt D, van Tulder M, Stalman W, de Vries T. Nonsteroidal anti-inflammatory drugs or acetaminophen for osteoarthritis of the hip or knee? A systematic review of evidence and guidelines. J Rheumatol 2004;31: 344-54.
OA Nation survey. http://oanation.arthritiscare.org.ukfileadmin/oanation/downloads/OA_Nation_report.pdf (accessed 10 Sept 2004).
Tramer MR, Moore RA, Reynolds DJ, McQuay HJ. Quantitative estimation of rare adverse events which follow a biological progression: a new model applied to chronic NSAID use. Pain 2000;85: 169-82.
Deeks JJ, Smith LA, Bradley MD. Efficacy, tolerability, and upper gastrointestinal safety of celecoxib for treatment of osteoarthritis and rheumatoid arthritis: systematic review of randomised controlled trials. BMJ 2002;325: 619.
Zabinski RA, Burke TA, Johnson J, Lavoie F, Fitzsimon C, Tretiak R, et al. An economic model for determining the costs and consequences of using various treatment alternatives for the management of arthritis in Canada. Pharmacoeconomics 2001;19(suppl 1): 49-58.
Juni P, Rutjes AW, Dieppe PA. Are selective COX 2 inhibitors superior to traditional non-steroidal anti-inflammatory drugs? BMJ 2002;324: 1287-8.
Cerezo J, Hristov RL, Sansuan AJ, Rodriguez JJV. Outcome trials of COX-2 selective inhibitors: global safety evaluation does not promise benefits. Eur J Clin Pharmacol 2003;59: 169-75.
Wright JM. The double-edged sword of COX-2 selective NSAIDs. CMAJ 2002;167: 1131-7.
Sturkenboom MC, Romano F, Simon G, Correa-Leite ML, Villa M, Nicolosi A, et al. The iatrogenic costs of NSAID therapy: a population study. Arthritis Rheum 2002;47: 132-40.
Jordan KM, Arden NK, Doherty M, Bannwarth B, Bijlsma JW, Dieppe P, et al. EULAR recommendations 2003: an evidence based approach to the management of knee osteoarthritis: report of a task force of the standing committee for international clinical studies including therapeutic trials (ESCISIT). Ann Rheum Dis 2003;62: 1145-55.
Song F, Altman DG, Glenny AM, Deeks JJ. Validity of indirect comparison for estimating efficacy of competing interventions: empirical evidence from published meta-analyses. BMJ 2003;326: 472.
Kazis LE, Anderson JJ, Meenan RF. Effect sizes for interpreting changes in health status. Med Care 1989;27:suppl: S178-89.
Cohen J. Statistical power analysis for behavioural sciences. New York: Academic Press, 1977.
Eccles M, Freemantle N, Mason J. North of England evidence based guideline development project: summary guideline for non-steroidal anti-inflammatory drugs versus basic analgesia in treating the pain of degenerative arthritis. The north of England non-steroidal anti-inflammatory drug guideline development group. BMJ 1998;317: 526-30.
Watson MC, Brookes ST, Kirwan JR, Faulkner A Non-aspirin, non-steroidal anti-inflammatory drugs for treating osteoarthritis of the knee. Cochrane Database Syst Rev 2004;(3): CD000142.
Zhang W, Jones A, Doherty M. Does paracetamol (acetaminophen) reduce the pain of osteoarthritis? A meta-analysis of randomised controlled trials. Ann Rheum Dis 2004;63: 901-7.
Jadad AR, McQuay HJ. Meta-analyses to evaluate analgesic interventions: a systematic qualitative review of their methodology. J Clin Epidemiol 1996;49: 235-43.
Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of hip or knee. J Rheumatol 1988;15: 1833-40.
Roos EM, Klassbo M, Lohmander LS. WOMAC osteoarthritis index. Reliability, validity, and responsiveness in patients with arthroscopically assessed osteoarthritis. Western Ontario and MacMaster Universities. Scand J Rheumatol 1999;28: 210-5.
Fleiss J. The statistical basis of meta-analysis. Stat Methods Med Res 1993;2: 121-45.
Uzun H, Tuzun S, Ozaras N, Aydin S, Ozaras R, Dondurmaci S, et al. The effect of flurbiprofen and tiaprofenic acid on serum cytokine levels of patients with osteoarthrosis. Acta Orthop Scand 2001;72: 499-502.
McKenna F, Weaver A, Fiechtner JJ, Bello AE, Fort J. COX-2 Specific inhibitors in the management of osteoarthritis of the knee: A placebo-controlled, randomized double-blind study. J Clin Rheumatol 2001;7: 151-9.
Fleischmann RM, Flint K, Constantine G, Kolecki B. A double-masked comparison of Naprelan and nabumetone in osteoarthritis of the knee. Naprelan study group. Clin Ther 1997;19: 642-55.
Gibofsky A, Williams GW, McKenna F, Fort JG. Comparing the efficacy of cyclooxygenase 2-specific inhibitors in treating osteoarthritis: appropriate trial design considerations and results of a randomized, placebo-controlled trial. Arthritis Rheum 2003;48: 3102-11.
Gottesdiener K, Schnitzer T, Fisher C, Bockow B, Markenson J, Ko A, et al. Results of a randomized, dose-ranging trial of etoricoxib in patients with osteoarthritis. Rheumatology (Oxford) 2002;41: 1052-61.
Case JP, Baliunas AJ, Block JA. Lack of efficacy of acetaminophen in treating symptomatic knee osteoarthritis: a randomized, double-blind, placebo-controlled comparison trial with diclofenac sodium. Arch Intern Med 2003;163: 169-78.
Kivitz A, Eisen G, Zhao WW, Bevirt T, Recker DP. Randomized placebo-controlled trial comparing efficacy and safety of valdecoxib with naproxen in patients with osteoarthritis. J Fam Pract 2002;51: 530-7.
Dore R, Ballard I, Constantine G, McDonald P. Efficacy and safety of etodolac and naproxen in patients with osteoarthritis of the knee: a double-blind, placebo-controlled study. Clin Ther 1995;17: 656-66.
Schnitzer TJ, Ballard IM, Constantine G, McDonald P. Double-blind, placebo-controlled comparison of the safety and efficacy of orally administered etodolac and nabumetone in patients with active osteoarthritis of the knee. Clin Ther 1995;17: 602-12.
Bensen WG, Fiechtner JJ, McMillen JI, Zhao WW, Yu SS, Woods EM, et al. Treatment of osteoarthritis with celecoxib, a cyclooxygenase-2 inhibitor: a randomized controlled trial. Mayo Clin Proc 1999;74: 1095-105.
McKenna F, Borenstein D, Wendt H, Wallemark C, Lefkowith JB, Geis GS. Celecoxib versus diclofenac in the management of osteoarthritis of the knee. Scand J Rheumatol 2001;30: 11-8.
Lund B, Distel M, Bluhmki E. A double-blind, randomized, placebo-controlled study of efficacy and tolerance of meloxicam treatment in patients with osteoarthritis of the knee. Scand J Rheumatol 1998;27: 32-7.
Scott DL, Berry H, Capell H, Coppock J, Daymond T, Doyle DV, et al. The long-term effects of non-steroidal anti-inflammatory drugs in osteoarthritis of the knee: a randomized placebo-controlled trial. Rheumatology (Oxford) 2000;39: 1095-101.
Zhao SZ, McMillen JI, Markenson JA, Dedhiya SD, Zhao WW, Osterhaus JT, et al. Evaluation of the functional status aspects of health-related quality of life of patients with osteoarthritis treated with celecoxib. Pharmacotherapy 1999;19: 1269-78.
Williams GW, Hubbard RC, Yu SS, Zhao W, Geis GS. Comparison of once-daily and twice-daily administration of celecoxib for the treatment of osteoarthritis of the knee. Clin Ther 2001;23: 213-27.
Williams PI, Hosie J, Scott DL. Etodolac therapy for osteoarthritis: a double-blind, placebo-controlled trial. Curr Med Res Opin 1989;11: 463-70.
Ehrich EW, Bolognese JA, Watson DJ, Kong SX. Effect of rofecoxib therapy on measures of health-related quality of life in patients with osteoarthritis. Am J Manag Care 2001;7: 609-16.
Kivitz AJ, Greenwald MW, Cohen SB, Polis AB, Najarian DK, Dixon ME, et al. Efficacy and safety of Rofecoxib 12.5 mg versus Nabumetone 1,000 mg in patients with osteoarthritis of the knee: a randomized controlled trial. J Am Geriatr Soc 2004;52: 666-74.
Tannenbaum H, Berenbaum F, Reginster JY, Zacher J, Robinson J, Poor G, et al. Lumiracoxib is effective in the treatment of osteoarthritis of the knee: a 13-week, randomized, double-blind study versus placebo and celecoxib. Ann Rheum Dis 2004;63: 1419-26.
Lee P, Davis P, Prat A. The efficacy of diflunisal in osteoarthritis of the knee. A Canadian multicenter study. J Rheumatol 1985;12: 544-8.
Simon LS, Lanza FL, Lipsky PE, Hubbard RC, Talwalker S, Schwartz BD, et al. Preliminary study of the safety and efficacy of SC-58635, a novel cyclooxygenase 2 inhibitor: efficacy and safety in two placebo-controlled trials in osteoarthritis and rheumatoid arthritis, and studies of gastrointestinal and platelet effects. Arthritis Rheum 1998;41: 1591-602.
Weaver A, Rubin B, Caldwell J, McMahon FG, Lee D, Makarowski W, et al. Comparison of the efficacy and safety of oxaprozin and nabumetone in the treatment of patients with osteoarthritis of the knee. Clin Ther 1995;17: 735-45.
Makarowski W, Weaver A, Rubin B, Caldwell J, McMahon FG, Noveck RJ, et al. The efficacy, tolerability, and safety of 1200 mg/d of oxaprozin and 1500 mg/d of nabumetone in the treatment of patients with osteoarthritis of the knee. Clin Ther 1996;18: 114-24.
Dieppe PA, Ebrahim S, Martin RM, Juni P. Lessons from the withdrawal of rofecoxib. BMJ 2004;329: 867-8.
Oxford league table of analgesics in acute pain. www.jr2.ox.ac.uk/bandolier/booth/painpag/Acutrev/Analgesics/Leagtab.html (accessed 10 Sep 2004).
Barden J, Edwards JE, McQuay HJ, Moore RA. Single dose oral celecoxib for postoperative pain. Cochrane Database Syst Rev 2003;(2): CD004233.
Rowbotham MC. What is a "clinically meaningful" reduction in pain? Pain 2001;94: 131-2.
Angst F, Aeschlimann A, Michel BA, Stucki G. Minimal clinically important rehabilitation effects in patients with osteoarthritis of the lower extremities. J Rheumatol 2002;29: 131-8.
Ehrich EW, Davies GM, Watson DJ, Bolognese JA, Seidenberg BC, Bellamy N. Minimal perceptible clinical improvement with the Western Ontario and McMaster Universities osteoarthritis index questionnaire and global assessments in patients with osteoarthritis. J Rheumatol 2000;27: 2635-41.
Johnson AG. NSAIDs and increased blood pressure. What is the clinical significance? Drug Saf 1997;17: 277-89.
Bleumink GS, Feenstra J, Sturkenboom MC, Stricker BH. Nonsteroidal anti-inflammatory drugs and heart failure. Drugs 2003;63: 525-34.
Whelton A. Nephrotoxicity of nonsteroidal anti-inflammatory drugs: physiologic foundations and clinical implications. Am J Med 1999;106(5B): 13-24S.
Phelan KM, Mosholder AD, Lu S. Lithium interaction with the cyclooxygenase 2 inhibitors rofecoxib and celecoxib and other nonsteroidal anti-inflammatory drugs. J Clin Psychiatry 2003;64: 1328-34.(Jan Magnus Bjordal, resea)
Correspondence to: J M Bjordal jmb@hib.no
Abstract
Osteoarthritis of the knee is the most common type of osteoarthritis,1 the prevalence of which is rising in parallel with the increasing age of the population.2 The condition is associated with pain and inflammation of the joint capsule,3-5 impaired muscular stability,6 7 reduced range of motion,8 and functional disability.9 Treatment guidelines for knee osteoarthritis recommend pharmacological intervention, initially with paracetamol and subsequently with a non-steroidal anti-inflammatory drug (NSAID).10 In a recent UK survey, 15% of patients with osteoarthritis used paracetamol, whereas 50% reported regular use of NSAIDs. Of the latter, 32% were using traditional NSAIDs and 18% were using cyclo-oxygenase-2 inhibitors (coxibs).11 This widespread use is one explanation for the interest in tolerability and efficacy issues regarding these drugs.10 12 13 The recent introduction of coxibs seemed to promise a reduction in serious adverse events related to NSAIDs,13 14 but this remains controversial.15-18
Guidelines from the European League Against Rheumatism (EULAR) state that both pharmacological and non-pharmacological interventions are needed for optimal treatment of knee osteoarthritis.19 The various potentially effective pharmacological interventions at the clinicians' disposal19 highlight the need for information regarding treatment efficacy.
Meta-analyses can be used for reliable comparison of the efficacy of different interventions.20 Effect size measures the magnitude of a treatment effect independent of sample size.21 There is no current operational definition for what constitutes a sufficiently large effect size for a therapeutic intervention to be considered as useful, but a value of 0.2 is usually considered small, 0.5 moderate, and 0.8 large.22
A recent systematic review of therapeutic alternatives in knee osteoarthritis gives no effect sizes for paracetamol and an imprecise range (0.47-0.96), derived from a minority of available trials, for NSAIDs.19 Neither other reviewers nor the Cochrane library provide comprehensive and robust effect size data for the efficacy of either of these interventions in osteoarthritis of the knee.10 13 23-25 Calculations of effect size require data for mean change and standard deviation (SD). If not provided, these data can be obtained by indirect means from standard errors, P values, t values, and 95% confidence intervals when sample sizes are known. The lack of data on effect size is surprising because treatment with NSAIDs for knee osteoarthritis is established to the point of being a reference against which other interventions are often compared.
We carried out a meta-analysis of published randomised placebo controlled trials to estimate the analgesic efficacy of NSAIDs, including coxibs, in patients with knee osteoarthritis.
Methods
Included studies
We evaluated 268 randomised controlled trials of NSAIDs for knee osteoarthritis (fig 1). Of these, 4 did not provide pain scale data, 126 did not have a placebo control group, and 115 presented combined data for osteoarthritis in several sites with no separate data for the knee. This provided a final sample of 23 trials that satisfied the inclusion criteria.30-52 Of these, 16 were sponsored by the pharmaceutical industry,31-34 36 39 40 43 44 46-52 while three others did not state sponsorship but gave an address of a pharmaceutical company as the workplace of most of the authors.37 38 41 The final sample included 10 845 patients, of whom 7767 received NSAIDs and 3078 received placebo (table).
Fig 1 Selection of trials for inclusion in meta-analysis
Characteristics of trials of NSAIDs for pain relief in patients with knee osteoarthritis
Patients and possible selection bias
The included patients had a median age of 62.5 years, and three trials had an upper age limit of 75 years.35 45 49 There were more women (67.9%) than men, and the median duration of symptoms was 8.2 years. The weighted mean baseline pain of 64.1 mm on a visual analogue scale was calculated from all but three trials.37 51 52 Six reports provided data of body mass index with a median mean value of 31.2.32 35 38 43 44 48 Most trials excluded individuals with concomitant health disorders by a median of 14 exclusion criteria. All trials had a minimum limit for pain intensity or disease activity for inclusion, and they all used a pretreatment washout period of 3-14 days for previous pharmacotherapy. Thirteen trials used an additional criterion by requiring a predefined minimum flare of symptoms when NSAID treatment was discontinued in the pretreatment wash out period.31 33-35 39 40 43 44 46 47 50-52 Five of these trials reported the proportion of regular NSAIDs users, ranging from 66% to 100% (median 90.5%).31 34 35 40 47
Trial quality
The methodological quality was adequate or good (table). All trials were randomised and double blinded, but adequate randomisation procedure, concealed allocation to groups, and blinding procedure were described satisfactorily in only eight studies.33 34 36 39 43 44 46 47 All trials described dropouts and withdrawals well, but one trial did not perform intention to treat analyses.45
Presentation of data on outcome measures
Only four trials presented outcome data as the mean difference of change with SD between NSAIDs and placebo groups.34 42 46 48 Fourteen trials presented the mean difference of change for each group with P values, SE of mean, or 95% confidence intervals but not mean differences between groups.31 33 35-39 41 43 47 49-52 Four trials did not present mean change data but only before and after means and P values.30 32 44 45 Eleven trials presented data on overall pain on 100 mm visual analogue scales or on a categorical five point scale,30-32 37 38 41 42 47 50-52 while the 12 remaining trials presented either categorical or continuous data from the WOMAC subscale for pain. All 23 trials reported data on pain intensity, and 11 trials reported data on functional disability.
Effect size for reduction in functional disability and pain
We excluded from analysis six intervention groups (n = 867) in five trials because patients did not receive an adequate NSAID dose.34-36 39 43 As most trials with multiple time points showed rather stable results from 2-13 weeks, we pooled data. Tests for heterogeneity showed positive results for reduction in functional disability (Q = 39.9, P < 0.01) and reduction in pain (Q = 56.6, P < 0.01). For this reason, we decided to use a random effects model. Eleven trials with 7433 patients provided separate scores for reduction in functional disability,33 35 36 39-41 43 44 46-48 and their combined effect size was 0.29 (95% confidence interval 0.18 to 0.40). One trial reported long term effects on pain but found no significant difference between tiaprofenic acid and placebo at one, two, three, and four years after start of treatment.42 For short term effects (2-13 weeks) the pooled effect size of all included trials was 10.1 mm on visual analogue scale (7.4 to 12.8) or 15.6% better than placebo, and the effect size was 0.32 (0.24 to 0.39) (fig 2).
Fig 2 Effect size (pain) for all trials on NSAID use in knee osteoarthritis
Subgroup analyses
We carried out post hoc subgroup analyses as heterogeneity was evident and trial procedures differed in selection of patients, duration of treatment, and pain scales. For subgroups of trials that used short durations of treatment (< 6 weeks) or WOMAC subscale for pain, neither effect size (0.35 or 0.39) nor heterogeneity (Q = 48.3, P < 0.01, or Q = 43.0, P < 0.01) changed significantly from the results of the total material. For the subgroup of 10 trials (n = 4565) that did not require patients to have a minimum flare of symptoms after treatment with NSAIDs was stopped before the trial, trial results were homogeneous both for function (Q = 2.6, P = 0.275) and pain (Q = 10.0, P = 0.263).30 32 36-38 41 42 45 48 49 Three of these trials (n = 2928) provided data for reduction in functional disability36 41 48 and we calculated effect size by a fixed effects model to be 0.20 (0.09 to 0.30). For pain reduction, we a used fixed effect model to calculate a pooled effect size of 0.23 (0.16 to 0.31) or 5.9 mm on visual analogue scale (3.8 to 7.9) (fig 3).
Fig 3 Effect size (pain) for trials on NSAID use in knee osteoarthritis without identified selection bias
Discussion
Andrianakos A, Trontzas P, Christoyannis F, Dantis P, Voudouris C, Georgountzos A, et al. Prevalence of rheumatic diseases in Greece: a cross-sectional population based epidemiological study. The ESORDIG study. J Rheumatol 2003;30: 1589-601.
Felson DT, Lawrence RC, Dieppe PA, Hirsch R, Helmick CG, Jordan JM, et al. Osteoarthritis: new insights. Part 1: the disease and its risk factors. Ann Intern Med 2000;133: 635-46.
Vaatainen U, Lohmander LS, Thonar E, Hongisto T, Agren U, Ronkko S, et al. Markers of cartilage and synovial metabolism in joint fluid and serum of patients with chondromalacia of the patella. Osteoarthr Cartil 1998;6: 115-24.
Suenaga S, Abeyama K, Indo H, Shigeta K, Noikura T. Temporomandibular disorders: MR assessment of inflammatory changes in the posterior disk attachment during the menstrual cycle. J Comput Assist Tomogr 2001;25: 476-81.
Speldewinde GC, Bashford GM, Davidson IR. Diagnostic cervical zygapophyseal joint blocks for chronic cervical pain. Med J Aust 2001;174: 174-6.
Radebold A, Cholewicki J, Polzhofer GK, Greene HS. Impaired postural control of the lumbar spine is associated with delayed muscle response times in patients with chronic idiopathic low back pain. Spine 2001;26: 724-30.
Cowan SM, Bennell KL, Hodges PW, Crossley KM, McConnell J. Delayed onset of electromyographic activity of vastus medialis obliquus relative to vastus lateralis in subjects with patellofemoral pain syndrome. Arch Phys Med Rehabil 2001;82: 183-9.
Steultjens MP, Dekker J, van Baar ME, Oostendorp RA, Bijlsma JW. Range of joint motion and disability in patients with osteoarthritis of the knee or hip. Rheumatology (Oxford) 2000;39: 955-61.
Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and therapeutic criteria committee of the American Rheumatism Association. Arthritis Rheum 1986;29: 1039-49.
Wegman A, van der Windt D, van Tulder M, Stalman W, de Vries T. Nonsteroidal anti-inflammatory drugs or acetaminophen for osteoarthritis of the hip or knee? A systematic review of evidence and guidelines. J Rheumatol 2004;31: 344-54.
OA Nation survey. http://oanation.arthritiscare.org.ukfileadmin/oanation/downloads/OA_Nation_report.pdf (accessed 10 Sept 2004).
Tramer MR, Moore RA, Reynolds DJ, McQuay HJ. Quantitative estimation of rare adverse events which follow a biological progression: a new model applied to chronic NSAID use. Pain 2000;85: 169-82.
Deeks JJ, Smith LA, Bradley MD. Efficacy, tolerability, and upper gastrointestinal safety of celecoxib for treatment of osteoarthritis and rheumatoid arthritis: systematic review of randomised controlled trials. BMJ 2002;325: 619.
Zabinski RA, Burke TA, Johnson J, Lavoie F, Fitzsimon C, Tretiak R, et al. An economic model for determining the costs and consequences of using various treatment alternatives for the management of arthritis in Canada. Pharmacoeconomics 2001;19(suppl 1): 49-58.
Juni P, Rutjes AW, Dieppe PA. Are selective COX 2 inhibitors superior to traditional non-steroidal anti-inflammatory drugs? BMJ 2002;324: 1287-8.
Cerezo J, Hristov RL, Sansuan AJ, Rodriguez JJV. Outcome trials of COX-2 selective inhibitors: global safety evaluation does not promise benefits. Eur J Clin Pharmacol 2003;59: 169-75.
Wright JM. The double-edged sword of COX-2 selective NSAIDs. CMAJ 2002;167: 1131-7.
Sturkenboom MC, Romano F, Simon G, Correa-Leite ML, Villa M, Nicolosi A, et al. The iatrogenic costs of NSAID therapy: a population study. Arthritis Rheum 2002;47: 132-40.
Jordan KM, Arden NK, Doherty M, Bannwarth B, Bijlsma JW, Dieppe P, et al. EULAR recommendations 2003: an evidence based approach to the management of knee osteoarthritis: report of a task force of the standing committee for international clinical studies including therapeutic trials (ESCISIT). Ann Rheum Dis 2003;62: 1145-55.
Song F, Altman DG, Glenny AM, Deeks JJ. Validity of indirect comparison for estimating efficacy of competing interventions: empirical evidence from published meta-analyses. BMJ 2003;326: 472.
Kazis LE, Anderson JJ, Meenan RF. Effect sizes for interpreting changes in health status. Med Care 1989;27:suppl: S178-89.
Cohen J. Statistical power analysis for behavioural sciences. New York: Academic Press, 1977.
Eccles M, Freemantle N, Mason J. North of England evidence based guideline development project: summary guideline for non-steroidal anti-inflammatory drugs versus basic analgesia in treating the pain of degenerative arthritis. The north of England non-steroidal anti-inflammatory drug guideline development group. BMJ 1998;317: 526-30.
Watson MC, Brookes ST, Kirwan JR, Faulkner A Non-aspirin, non-steroidal anti-inflammatory drugs for treating osteoarthritis of the knee. Cochrane Database Syst Rev 2004;(3): CD000142.
Zhang W, Jones A, Doherty M. Does paracetamol (acetaminophen) reduce the pain of osteoarthritis? A meta-analysis of randomised controlled trials. Ann Rheum Dis 2004;63: 901-7.
Jadad AR, McQuay HJ. Meta-analyses to evaluate analgesic interventions: a systematic qualitative review of their methodology. J Clin Epidemiol 1996;49: 235-43.
Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of hip or knee. J Rheumatol 1988;15: 1833-40.
Roos EM, Klassbo M, Lohmander LS. WOMAC osteoarthritis index. Reliability, validity, and responsiveness in patients with arthroscopically assessed osteoarthritis. Western Ontario and MacMaster Universities. Scand J Rheumatol 1999;28: 210-5.
Fleiss J. The statistical basis of meta-analysis. Stat Methods Med Res 1993;2: 121-45.
Uzun H, Tuzun S, Ozaras N, Aydin S, Ozaras R, Dondurmaci S, et al. The effect of flurbiprofen and tiaprofenic acid on serum cytokine levels of patients with osteoarthrosis. Acta Orthop Scand 2001;72: 499-502.
McKenna F, Weaver A, Fiechtner JJ, Bello AE, Fort J. COX-2 Specific inhibitors in the management of osteoarthritis of the knee: A placebo-controlled, randomized double-blind study. J Clin Rheumatol 2001;7: 151-9.
Fleischmann RM, Flint K, Constantine G, Kolecki B. A double-masked comparison of Naprelan and nabumetone in osteoarthritis of the knee. Naprelan study group. Clin Ther 1997;19: 642-55.
Gibofsky A, Williams GW, McKenna F, Fort JG. Comparing the efficacy of cyclooxygenase 2-specific inhibitors in treating osteoarthritis: appropriate trial design considerations and results of a randomized, placebo-controlled trial. Arthritis Rheum 2003;48: 3102-11.
Gottesdiener K, Schnitzer T, Fisher C, Bockow B, Markenson J, Ko A, et al. Results of a randomized, dose-ranging trial of etoricoxib in patients with osteoarthritis. Rheumatology (Oxford) 2002;41: 1052-61.
Case JP, Baliunas AJ, Block JA. Lack of efficacy of acetaminophen in treating symptomatic knee osteoarthritis: a randomized, double-blind, placebo-controlled comparison trial with diclofenac sodium. Arch Intern Med 2003;163: 169-78.
Kivitz A, Eisen G, Zhao WW, Bevirt T, Recker DP. Randomized placebo-controlled trial comparing efficacy and safety of valdecoxib with naproxen in patients with osteoarthritis. J Fam Pract 2002;51: 530-7.
Dore R, Ballard I, Constantine G, McDonald P. Efficacy and safety of etodolac and naproxen in patients with osteoarthritis of the knee: a double-blind, placebo-controlled study. Clin Ther 1995;17: 656-66.
Schnitzer TJ, Ballard IM, Constantine G, McDonald P. Double-blind, placebo-controlled comparison of the safety and efficacy of orally administered etodolac and nabumetone in patients with active osteoarthritis of the knee. Clin Ther 1995;17: 602-12.
Bensen WG, Fiechtner JJ, McMillen JI, Zhao WW, Yu SS, Woods EM, et al. Treatment of osteoarthritis with celecoxib, a cyclooxygenase-2 inhibitor: a randomized controlled trial. Mayo Clin Proc 1999;74: 1095-105.
McKenna F, Borenstein D, Wendt H, Wallemark C, Lefkowith JB, Geis GS. Celecoxib versus diclofenac in the management of osteoarthritis of the knee. Scand J Rheumatol 2001;30: 11-8.
Lund B, Distel M, Bluhmki E. A double-blind, randomized, placebo-controlled study of efficacy and tolerance of meloxicam treatment in patients with osteoarthritis of the knee. Scand J Rheumatol 1998;27: 32-7.
Scott DL, Berry H, Capell H, Coppock J, Daymond T, Doyle DV, et al. The long-term effects of non-steroidal anti-inflammatory drugs in osteoarthritis of the knee: a randomized placebo-controlled trial. Rheumatology (Oxford) 2000;39: 1095-101.
Zhao SZ, McMillen JI, Markenson JA, Dedhiya SD, Zhao WW, Osterhaus JT, et al. Evaluation of the functional status aspects of health-related quality of life of patients with osteoarthritis treated with celecoxib. Pharmacotherapy 1999;19: 1269-78.
Williams GW, Hubbard RC, Yu SS, Zhao W, Geis GS. Comparison of once-daily and twice-daily administration of celecoxib for the treatment of osteoarthritis of the knee. Clin Ther 2001;23: 213-27.
Williams PI, Hosie J, Scott DL. Etodolac therapy for osteoarthritis: a double-blind, placebo-controlled trial. Curr Med Res Opin 1989;11: 463-70.
Ehrich EW, Bolognese JA, Watson DJ, Kong SX. Effect of rofecoxib therapy on measures of health-related quality of life in patients with osteoarthritis. Am J Manag Care 2001;7: 609-16.
Kivitz AJ, Greenwald MW, Cohen SB, Polis AB, Najarian DK, Dixon ME, et al. Efficacy and safety of Rofecoxib 12.5 mg versus Nabumetone 1,000 mg in patients with osteoarthritis of the knee: a randomized controlled trial. J Am Geriatr Soc 2004;52: 666-74.
Tannenbaum H, Berenbaum F, Reginster JY, Zacher J, Robinson J, Poor G, et al. Lumiracoxib is effective in the treatment of osteoarthritis of the knee: a 13-week, randomized, double-blind study versus placebo and celecoxib. Ann Rheum Dis 2004;63: 1419-26.
Lee P, Davis P, Prat A. The efficacy of diflunisal in osteoarthritis of the knee. A Canadian multicenter study. J Rheumatol 1985;12: 544-8.
Simon LS, Lanza FL, Lipsky PE, Hubbard RC, Talwalker S, Schwartz BD, et al. Preliminary study of the safety and efficacy of SC-58635, a novel cyclooxygenase 2 inhibitor: efficacy and safety in two placebo-controlled trials in osteoarthritis and rheumatoid arthritis, and studies of gastrointestinal and platelet effects. Arthritis Rheum 1998;41: 1591-602.
Weaver A, Rubin B, Caldwell J, McMahon FG, Lee D, Makarowski W, et al. Comparison of the efficacy and safety of oxaprozin and nabumetone in the treatment of patients with osteoarthritis of the knee. Clin Ther 1995;17: 735-45.
Makarowski W, Weaver A, Rubin B, Caldwell J, McMahon FG, Noveck RJ, et al. The efficacy, tolerability, and safety of 1200 mg/d of oxaprozin and 1500 mg/d of nabumetone in the treatment of patients with osteoarthritis of the knee. Clin Ther 1996;18: 114-24.
Dieppe PA, Ebrahim S, Martin RM, Juni P. Lessons from the withdrawal of rofecoxib. BMJ 2004;329: 867-8.
Oxford league table of analgesics in acute pain. www.jr2.ox.ac.uk/bandolier/booth/painpag/Acutrev/Analgesics/Leagtab.html (accessed 10 Sep 2004).
Barden J, Edwards JE, McQuay HJ, Moore RA. Single dose oral celecoxib for postoperative pain. Cochrane Database Syst Rev 2003;(2): CD004233.
Rowbotham MC. What is a "clinically meaningful" reduction in pain? Pain 2001;94: 131-2.
Angst F, Aeschlimann A, Michel BA, Stucki G. Minimal clinically important rehabilitation effects in patients with osteoarthritis of the lower extremities. J Rheumatol 2002;29: 131-8.
Ehrich EW, Davies GM, Watson DJ, Bolognese JA, Seidenberg BC, Bellamy N. Minimal perceptible clinical improvement with the Western Ontario and McMaster Universities osteoarthritis index questionnaire and global assessments in patients with osteoarthritis. J Rheumatol 2000;27: 2635-41.
Johnson AG. NSAIDs and increased blood pressure. What is the clinical significance? Drug Saf 1997;17: 277-89.
Bleumink GS, Feenstra J, Sturkenboom MC, Stricker BH. Nonsteroidal anti-inflammatory drugs and heart failure. Drugs 2003;63: 525-34.
Whelton A. Nephrotoxicity of nonsteroidal anti-inflammatory drugs: physiologic foundations and clinical implications. Am J Med 1999;106(5B): 13-24S.
Phelan KM, Mosholder AD, Lu S. Lithium interaction with the cyclooxygenase 2 inhibitors rofecoxib and celecoxib and other nonsteroidal anti-inflammatory drugs. J Clin Psychiatry 2003;64: 1328-34.(Jan Magnus Bjordal, resea)