Your browser does not support the NLM PubReader view. Go to this page to see a list of supported browsers or return to the regular view. |
Agency for Healthcare Research and Quality (US), Rockville (MD)
Chou R, McDonagh MS, Nakamoto E, et al. Analgesics for Osteoarthritis: An Update of the 2006 Comparative Effectiveness Review [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2011 Oct. (Comparative Effectiveness Reviews, No. 38.) Appendix J, Evidence Tables: Topical NSAIDs.
| Author Year | Eligibility Criteria | Demographics (age, Gender, Race) | Study Design/Type | Interventions (Drug, Dose, Duration) | Allowed Other Medications/Interventions | Other Population Characteristics (Diagnosis, etc.) | Number Screened/Eligible/Enrolled | Number Withdrawn/Lost to fu/Analyzed | Results | Adverse Events Assessment: Pre-Specified, Active or Passive Ascertainment, Assessed the Severity of Adverse Events? | Adverse Events Reported | Total Withdrawals; Withdrawals due to Adverse Events | Run-in/Washout | Class Naïve Patients Only | Notes |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Dickson, 1991298 | Male and female patients between 18 and 86 years old with well-documented, mild osteoarthritis of the knee | Mean age: 63 years (range 21–86 years) Female: 66% Race: NR | RCT |
| Paracetamol up to 4 mg allowed during washout and throughout trial; no significant difference between groups | Baseline overall pain during day: 5.0 vs. 5.0 | NR/NR/235 (117 topical piroxicam, 118 oral ibuprofen) | 39/3/196 (101 topical piroxicam, 95 oral ibuprofen) | Topical gel piroxicam vs. oral ibuprofen, at 4 weeks Overall pain during day (median, 1–9 scale): 3.0 vs. 2.0, p=0.56 Overall pain during night (median, 1–9 scale): 3.0 vs.3.0, p=0.54 Ability to perform specified activity (median, 1–9 scale): 5.0 vs. 5.0, p=0.33 Rescue analgesic use: 69% vs. 62% | Prespecified: No (general question) Active or passive ascertainment: Active Assessment of severity: Yes | Topical gel piroxicam (n=117) vs. oral ibuprofen (n=118) Any adverse event judged to be definitely or possibly related to study treatment: 26% vs. 23% Upper GI events: 10% vs. 8.5% Other GI events: 2.6% vs. 0.8% CNS events: 6.0% vs. 6.8% Rash events: 0.8% vs. 0.8% Dependent edema: 0% vs. 6.8% Local effects: 1.7% vs. 0.8% | Topical gel piroxicam vs. oral ibuprofen Total withdrawals: 14% vs. 19% Withdrawal due to adverse events: 7.7% vs. 9.9% Withdrawal due to upper GI events: 5.1% vs. 3.4% Withdrawal due to other GI events: 0.9% vs. 0% Withdrawal due to CNS events: 1.7% vs. 2.5% Withdrawal due to rash: 0% vs. 0.8% | 7-day washout free of anti-inflammatory medication | No | |
| Kosuwon, 2010312 | Ambulatory males, or non-pregnant females, between 40 and 80 years of age diagnosed with OA ≥ 6 months prior to screening according to ACR and confirmed by radiolographic evidence (grade 2 or 3 on Kellgren ascale). Pain in the knee < 80 mm (100 mm VAS scale) and a baseline minimum joint space width in the medial and lateral compartments of the index knee of > 1.5 and > 2.5 mm, respectively. | Age Range: 44–82 (Mean= 61 years) 100% Female Race: NR | RCT, cross-over design | Capsicum tincture 45.50 g (equivalent to capsaicin 0.0125%) per 100 g of Capsika gel®. Subjects applied 2 inches of gel topically 3 times per day for 12 weeks. | Acetaminophen for pain—500 mg three times a day (or every 4–6 hours). | Baseline (Mean +/−) VAS: 6.40 +/− 1.64 WOMAC : 51.65 +/− 13.30 | NR/NR/100 | 1/0/99 |
| Unclear if active or passive ascertainment and did not assess severity. | Burning sensation: 66 episodes (17%) in placebo group vs. 272 episodes (67%) in the capsaicin group during 4 week study period. 34 patients (34%) in placebo vs. 57 patients (57.58%) had a burning sensation during the capsaicin period (p < 0.05). | 0/1 | Washout was 4 weeks (middle of placebo or treatement weeks) | No | |
| Rother, 2007299 | Minimum of 6 months’ history of osteoarthritis with 2 of 3 criteria: 1) morning stiffness < 30 minutes/duration, crepitus on motion and age ≥ 40 years; 2) pain rating as ≥3 on a 5 point Likert scale; 3) oral NASIDs at least 3 days per week in the past 3 months or >25 of the past 30 days AND meeting of three osteoarthritis flare criteria | Mean age: 63 years (range NR) Female: 79% Race: NR | RCT |
| 2,000 mg paracetamol per day for 3 days any week except 48 hours before study visit | Baseline WOMAC pain score (mean, 0 to 100): 55 vs. 56 Baseline WOMAC stiffness score (mean, 0 to 100): 49 vs. 51 Baseline WOMAC physical function score (mean, 0 to 100): 54 vs. 55 Baseline patient global assessment of osteoarthritis (mean, 0 to 4): 3.9 vs. 3.9 | 499/NR/397 (138 topical ketoprofen, 132 oral celecoxib) | Topical ketoprofen and oral celecoxib arms only 48/1/270 (138 topical ketoprofen, 132 oral celecoxib) | Topical ketoprofen + IDEA-033 vs. oral celecoxib, at 6 weeks WOMAC pain score (mean change from baseline, 0 to 100 scale): −19 vs. −21, p not reported WOMAC physical function score (mean change from baseline, 0 to 100 scale): −16 vs. −18, p not reported Patient global assessment excellent (poor, fair, good, or excellent): 12% vs. 11% Patient global assessment good or excellent: 46% vs. 39% Withdrawal due to lack of efficacy: 0.7% vs. 2.3% | Prespecified: Unclear Active or passive ascertainment: Active Assessment of severity: No | Topical ketoprofen + IDEA-033 (n=138) vs. oral celecoxib (n=132) Any GI event: 9.4% vs. 14% Upper abdominal pain: 1.4% vs. 3.0% Dyspepsia: 0.7% vs. 3.0% Nausea: 1.4% vs. 2.3% Musculoskeletal and connective tissue disorders: 8.7% vs. 14% Respiratory, thoracic and mediastinal: 12% vs. 11% Allergic dermatitis: 1.4% vs. 0.8% Erythema: 21% vs. 14% | Topical ketoprofen + IDEA-033 vs. oral celecoxib Total withdrawals: 18% vs. 17% Withdrawal due to adverse events: 17% vs. 14% | NR/NR | No | |
| Sandelin, 1997300 | Male and female outpatient patients with radiologically confirmed OA including osteophytes of one or both knees and with pain symptoms for most days of the prior month where analgesics was needed | Mean age: 61 years (range NR) Female: 66% Race: NR | RCT |
| NR | Bilateral OA: 53% vs. 51% Baseline pain (mean, 0 to 100 VAS): 48 vs. 52 Baseline Lequesne index score (mean, 0 to 24): 9.5 vs. 10 | NR/NR/290 (number randomized in each group unclear) | 9/0/281 (124 topical eltenac, 89 oral diclofenac) | Topical eltenac vs. oral diclofenac, average at 2–4 weeks Overall pain (mean, 0–100 VAS): 31 vs. 30 Lequesne Index (mean, 0–24 scale): 6.9 vs. 7.3 Physician rated effect “good” (none, slight, moderate, or good): 18% vs. 30% | Prespecified: Unclear Active or passive ascertainment: Unclear Assessment of severity: No | Topical eltenac (n=126) vs. oral diclofenac (n=82) Any adverse events: 27% vs. 24% Any GI event: 4.8% vs. 13% CNS events: 9.5% vs. 7.3% Local skin reactions: 13% vs. 1.2% Other: 5.6% vs. 4.9% | Topical eltenac vs. oral diclofenac Total withdrawals: Not reported Withdrawal due to adverse events: 5% vs. 1.2% | NR/NR | No | |
| Simon, 2009301 | Male and non-pregnant women aged 40–85 with primary OA of the knee based on a) standard radiological criteria from a recent examination within 3 months; b) pain with regular use of pain meds; c) a flare of pain and a minimum Likert pain score of 8 at baseline | Mean age: 62 years (range NR) Female: 65% Non-white: 22% | RCT |
| Acetaminophen (up to four, 325 mg per day), except 3 days before efficacy assessment Glucosamine, chondroitin, anti-depressants or proton pump inhibitor, or low dose (≤325 mg/day) aspirin allowed | Bilateral OA: 99% vs. 99% Baseline WOMAC pain score (mean, 0 to 20): 13 vs. 13 Baseline WOMAC physical function score (mean, 0 to 68): 42 vs. 42 Baseline WOMAC stiffness score (mean, 0 to 8): 5.1 vs. 5.2 | 1396 (overall)/NR/775 (154 to topical diclofenac, 151 to oral diclofenac) | Topical and oral diclofenac arms only 95/4/305 (154 topical diclofenac, 151 oral diclofenac) | Topical diclofenac vs. oral dicofenac, at 12 weeks WOMAC pain score (mean change from baseline, 0–20): −6.0 vs. −6.4, p=0.43 WOMAC physical function score (mean change from baseline, 0 to 68): −16 vs. −18, p=0.32 WOMAC stiffness score (mean change from baseline, 0 to 8): −1.9 vs. −2.1, p=0.60 Patient overall health assessment score (mean change from baseline, 0 to 4): −0.95 vs. −0.88, p=0.96 Patient global assessment of the study knee (mean change from baseline, 0 to 4): −1.4 vs. −1.4, p=0.44 Withdrawal due to lack of efficacy: 10% vs. 3.3% | Pre-specified: Unclear Active or passive ascertainment: Active Assessment of severity: No | Topical diclofenac (n=154) vs. oral diclofenac (n=151) Any adverse event: 62% vs. 62% Any GI event: 6.5% vs. 24% Abdominal pain: 3.2% vs. 7.3% Dyspepsia: 2.6% vs. 4.0% Nausea: 0% vs. 2.0% Dry skin at application site: 18% vs. 2.6% Contact dermatitis at application site: 2.6% vs. 0.7% Rash: 2.6% vs. 0% Headache: 18% vs. 17% Back pain: 10% vs. 7.3% Arthralgia: 9.1% vs. 7.9% | Topical diclofenac vs. oral diclofenac Total withdrawals: 33% vs. 29% Withdrawal due to adverse events: 10% vs. 13% | NR/NR | No | Has topical diclofenac + oral diclofenac group |
| Tugwell, 2004303 | Men and nonpregnant women 40 to 85 years old, with symptomatic primary OA of the knee and recent (<3 months) x ray showing osteoarthritis (confirmed by radiologist) | Mean age: 64 years (range NR) Female: 57% Non-white: 6% | RCT |
| Aspirin up to 325 mg/day for cardiovascular prophylaxis (use comparable in groups 14% topical and 15% oral) | Mean OA duration: NR Total x-ray score (mean, maximum 27): 6.4 vs. 6.2 Baseline WOMAC pain score (mean, 0 to 500): 288 vs. 289 Baseline WOMAC physical function score (mean, 0 to 1700): 979 vs. 983 WOMAC stiffness score (mean, 0 to 200): 123 vs. 124 | 1057/NR/622 (311 topical diclofenac, 311 oral diclofenac) | 145/10/604 (303 topical diclofenac, 301 oral diclofenac) | Topical vs. oral diclofenac, at 12 weeks WOMAC pain score (mean change from baseline, 0–500 scale): −118 vs. −134; difference 16 (−3.4 to 36.1), p=0.10 WOMAC physical function score (mean change from baseline, 0–1700 scale): −348 vs. −438; difference 90 (24 to 156), p=0.008 WOMAC stiffness score (mean change from baseline, 0–200 scale): −45 vs. −52; p=0.14 Pain on walking (mean change from baseline, 0 to 100 scale [based on 1st item of the WOMAC pain subscale): −25 vs. −24; difference 1.7 (−2.9 to 6.4), p NS Patient global assessment (mean change from baseline, 0–100 scale): −27 vs. −32; difference 4.5 (−0.5 to 9.6), p=0.08 Number of responders (OMERACT criteria, >=50% improvement in pain or function that was >=20 mm on a 100 mm VAS, or >=20% improvement in at least two of pain, function, or patient global assessment that was >=10 mm on a 100 mm VAS): 66% vs. 70%, p=0.37 Withdrawal due to lack of efficacy: 9.0% vs. 3.2% | Prespecified: Unclear Active or passive ascertainment: Unclear Assessment of severity: Yes | Topical diclofenac (n=311) vs. oral diclofenac (n=311) Any GI events: 35% vs. 48%, p=0.0006 Abdominal pain: 12% vs. 22%, p=0.0008 Diarrhea: 9% vs. 17%, p=0.001 Dyspepsia: 15% vs. 26%, p=0.001 Flatulence: 10% vs. 17%, p=0.009 Melena: 1% vs. 2%, NS Nausea: 25% vs. 41%, p=0.4 Dry skin: 27% vs. 1%; p<0.0001 Rash: 12% vs. 2%, p<0.0001 Vesiculobullous rash: 5% vs. 0%, p<0.0001 Asthma: 0.6% vs. 3%, p=0.02 Dizziness: 0.6% vs. 4%, p=0.002 Dyspnea: 0% vs. 2%, p=0.01 | Topical diclofenac vs. oral diclofenac Total withdrawals: 41% vs. 37% Withdrawal due to adverse events: 21% vs. 25% | NR/washout 3–10 days | No | |
| Underwood, 200749 (TOIB study) | Literate men and women ≥ 50 years of age with troublesome pain around the knee most days for at least 1 month with knee pain more than three months out of preceding year; consultation with or treatment prescribed by GP for knee pain in the last 3 years. | Mean age: 64 years (range 50–89 years) Female: 56% Non-white: 1% | RCT |
| Not specified | Met ACR criteria for OA: 97% vs. 98% Baseline WOMAC pain score (mean, 0 to 100): 19 vs. 22 Baseline WOMAC stiffness score (mean, 0 to 100): 25 vs. 26 Baseline WOMAC physical function score (mean, 0 to 100): 23 vs. 18 Baseline WOMAC global assessment (mean, 0 to 100): 18 vs. 22 | Number assessed and eligible for RCT unclear/282 randomized (138 to advice for topical NSAID, 144 to advice for oral NSAID) | 18 at 3 months, 34 at 1 year/NR/264 at 3 months, 248 at 1 year | Advice to use a topical NSAID vs. advice to use an oral NSAID, at 3 months, 1 year, 2 years, and end of study (last value carried forward or 2 years); positive scores favor oral WOMAC pain score (difference in change from baseline, 0 to 100): −2 (−6 to 2), 1 (−4 to 6), 6 (0 to 12), 5 (0 to 9) WOMAC stiffness score (difference in change from baseline, 0 to 100): −3 (−8 to 2), 0 (−6 to 5), −1 (−8 to 6), −2 (−7 to 4) WOMAC physical function score (difference in change from baseline, 0 to 100): −2 (−5 to 2), 3 (−2 to 7), 5 (−1 to 10), 3 (−2 to 7) WOMAC global assessment (mean difference in change from baseline, 0 to 100: −2 (−5 to 2), 2 (−2 to 6), 4 (−1 to 10), 3 (−1 to 7) | Prespecified: Yes Active of passive ascertainment: Unclear Assessment of severity: Yes | Advice to use topical NSAID (n=136) vs. advice to use oral NSAID (n=140) Deaths by 24 months: 0% vs. 0% Gastric bleeding by 24 months: 0% vs. 0% Emergency hospital admission (any reason) by 24 months: 7% vs. 4% (difference 3.1%, −2.5 to 8.6%) Cardiovascular hospital admission by 24 months: 2.9% vs. 3.5% Defined GI adverse event (dyspepsia, laboratory evidence of anemia) by or at 12 months: 42% vs. 40% (difference 2.5%, −9 to 14%) New diagnosis of heart failure at 12 months: 1% vs. 0% Increase in systolic blood pressure >=20 mm Hg at 12 months: 13% vs. 11% Peak expiratory flow reduced by 15% or more at 12 months: 8% vs. 18%; difference −10% (−19 to −1%) Minor GI events: 42% vs. 40% Minor renovascular events: 16% vs. 15% Minor respiratory events: 7% vs. 17% Any minor adverse event: 56% vs. 56% | Advice to use topical NSAID vs. advice to use oral NSAID Missing follow-up data: 12% vs. 12% at 12 months; 42% vs. 36% at 24 months Withdrawal due to adverse events: Not reported | NR/NR | No | Comprehensive data available, also has patient preference data of oral vs. topical as well as cost-effectiveness analyses |
| Tiso, 2010302 | Subjects from a pain management practice who were ≥ 50 years old and ≥ 3 months of knee pain | Mean age 58 years Female: 89% | RCT |
| Not specified | Pain duration >12 months: 95% Chronic Grade Pain:
| 30/22/20 | 0/1/19 | -- | Prespecified: Yes Active of passive ascertainment: Unclear Assessment of severity: Yes | NR | NR | NR/2 days | No |
ACR = American College of Rheumatology; bid = twice daily; CNS = central nervous system; DMSO = dimethyl sulfoxide; GI = gastrointestinal; GP = general practitioner; IDEA = drug name for epicutaneous ketoprofen in transfersome; NR = not reported; NSAID = nonsteroidal anti-inflammatory drug; OA = osteoarthritis; OMERACT = Outcomes Measures in Arthritis Clinical Trials; QR = quality result; RCT = randomized controlled trial; tid = three times daily; TOIB = Topical or Oral Ibuprofen study; VAS = visual analogue scale; WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index
| Author Year | Aims | Time Period Covered | Eligibility Criteria | Number of Patients | Characteristics of Identified Articles: Study Designs | Characteristics of Identified Articles: Populations | Characteristics of Identified Articles: Interventions | Main Results | Subgroups | Comments |
|---|---|---|---|---|---|---|---|---|---|---|
| Lin 2004306 | To access the efficacy of topical NSAIDS in the treatment of osteoarthritis | MEDLINE, Embase, CINAHL, Scientific Search Index and Cochrane Library, and conference abstracts 1966 to 10/31/2003 | RCTs comparing topical NSAIDs with placebo OR oral NSAIDs Studies included those with clinical or radiographical (cross checked by 2 radiologists) evidence of osteoarthritis | n=1983 | 13 RCTs: double blinded crossovers, double blinded parallel | Patients with diagnosis of radiographical evidence of osteoarthritis |
|
| Efficacy: pain reduction, topical NSAIDs were superior to placebo in the first two weeks of treatment; topical NSAIDs were less effective than oral NSAIDs numerically at any week and statistically in the first week | Adverse events (A) Rate Ratio: 1.02 (0.62 to 1.68); (C) Rate ratio: 0.99 (.77 to 1.27) Topical NSAIDs had no more side effects than placebo. Compared with oral NSAIDs, fewer patient taking topical NSAIDs had any adverse events, withdrawals due to side effects and GI side effects, but significantly more patients had local side effects such as rash, itch and burning. |
| Mason 2004305 | To access the efficacy of topical NSAIDS in relieving pain | MEDLINE, Embase, Pre Medline, Cochrane Library and references supplied by pharmaceutical companies 1966 to April 2003 | Double blinded RCTs in which treatments were given to adult patients with moderate to severe chronic pain resulting from musculoskeletal or other painful disorders | n=1,502 (efficacy) n=2,302 (trials with adverse events) | 14 efficacy trials 18 placebo controlled trials | Generally, patients were over 40 years of age with predominantly musculoskeletal disorder and with baseline pain of moderate to severe intensity | Pennsaid vs. Placebo (3 trials) WOMAC
|
| Efficacy: for 4 or 5 patients with chronic pain treated with topical NSAID, one would benefit who would not have with placebo 95% CI Osteoarthritis of the knee with topical NSAIDs: 2.02 (1.57, 2.60) Topical NSAIDs vs. placebo for chronic pain 1.87 (1.61, 2.17) | RR (95% CI) Local adverse events: 1.0 (0.7 to 1.5) Systematic events: 1.7 (0.96 to 2/85) Withdrawal due to adverse events 0.9 (0.4 to 2.1) |
| Mason 2004313 (capsaicin) | To determine the efficacy and safety for topically applied capsaicin for chronic pain from neuropathic or musculoskeletal disorder | MEDLINE, Cochrane Library, Embase, and PubMed up to April 2003 | 16 trials | n=1556 | RCT | Patients aged 20 to 95 years with 11 trials of a baseline pain of moderate to severe and 7 allowed concomitant drugs | Capsaicin vs. placebo
| Relative benefit (95% CI)
| Topical capsaicin is better than placebo for the treatment of chronic pain. Local adverse events are common. | Local 3.6 (2.6 to 5.0) Withdrawals 4.0 (2.3 to 6.8) |
| Mason 200430 (rubefacients) | To determine the efficacy and safety of topical rubefacients containing salicylates in acute and chronic pain | MEDLINE, Cochrane Library, Embase, and PubMed up to March 2003 | 3 trials, acute conditions 5 trials, chronic conditions | n=862 | Randomized placebo controlled | Patients age ranged from 14 to 86 years and treatments contained salicylate as the primary ingredient | Topical vs. placebo
| Relative benefit (95% CI)
| Efficacious in acute pain and moderately to poorly effective in chronic arthritic and rheumatic pain. Longest trial lasted 28 days most lasted 14 days | Acute pain local: 1.1 (0.4 to 3.5) |
| Towheed 2006180 | To access the efficacy of topical diclofenac in patients with osteoarthritis of the knee | MEDLINE (1966 to February 2nd, 2005), Embase, CSDR, ACP Journal Club, DARE, CCTR | 4 trials, Pennsaid vs. VCP vs. placebo; 2 trials Pennsaid vs. VCP; Pennsaid vs. oral diclofenac | n=1412 (randomized subjects) n=666 (Pennsaid) n=746 randomized to comparator groups n= 970 completed trials | 4 RCTs | Mean trial duration was 8.5 weeks, all patients had osteoarthritis of the knee and in 3 trials specified radiographic criteria used by investigators to establish OA diagnosis | Pennsaid vs. VCP (3 trials)
| RR 95% CI
| Pennsaid was of equivalent efficacy as oral dicofenac in WOMAC outcomes and was significantly better tolerated than oral diclofenac |
|
ACP = American College of Physicians; CI = confidence interval; GI = gastrointestinal; NSAID = nonsteroidal anti-inflammatory drug; OA = osteoarthritis; OR = odds ratio; RCT = randomized controlled trial; RR = relative risk; VCP = vehicle control placebo; WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index.