Topical Nonsteroidal Anti-Inflammatory Drugs in the Treatment of Osteoarthritis

REVIEW

Topical nonsteroidal anti-inflammatory drugs in the treatment of osteoarthritis

Zahi Touma, Lan Chen Osteoarthritis is the most common degenerative joint disease and is particularly common in & Thurayya Arayssi† the elderly. Topical nonsteroidal anti-inflammatory drugs (NSAIDs) may represent a rational †Author for correspondence treatment in this situation because of their favorable side-effect profile. The majority of American University of topical NSAIDs decrease pain and relieve functional disability caused by osteoarthritis over Beirut Medical Center, the short term, however, there are insufficient data for their long-term efficacy. Department of Internal Medicine, Division of Rheumatology, 3 Dag Osteoarthritis (OA) is the most common disease patients with OA of the knees [9]. Other interven- Hammarskjöld Plaza, of synovial joints, associated with locomotor tions include the use of assistive devices (cane or 8th floor, New York, pain, chronic disability and morbidity [1]. It is crutch), application of braces and patellar taping NY, 10017, USA estimated by the WHO to be the fourth most and/or orthotics, especially in patients with insta- Tel.: +1 961 135 0000 Ext: 5383; important cause of disability among women and bility of the knee and varus malalignment [8,10,11]. Fax: +1 961 136 5189; the eighth amongst men [2]. Its prevalence and Participation in patient self-management pro- [email protected] incidence are projected to rise as the elderly programs and personalized social support through portion of the population increases, which will telephone contact have also been shown to help have a significant impact on society [3]. The in decreasing pain and improving function in onset of the disease most commonly occurs patients with OA [10]. between the age of 50 and 60 years and fre- Acupuncture has been evaluated in rando- quently affects the hands, spine, knees and hips mized controlled trials and been shown to [4]. Involvement of the wrists, elbows, ankles and provide added benefit in patients with knee OA shoulders is uncommon [5]. The majority of per- with a favorable safety profile [12,13]. sons show radiographic evidence of OA by Topical traditional Chinese medicine 65 years of age, although most are asymptomatic (TTCMs) and other alternative remedies are still [6]. In the Framingham study, for example, 30% widely used, especially in Chinese and other eth- of patients aged over 60 years had radiological nic communities. When all the ‘modern’ prepa- evidence of OA of the knee, which was not rations have been exhausted, some ‘old remedies’ always symptomatic [4]. have been explored by patients [14]. Litt has OA was previously thought to be a normal offered a list of 73 alternative topical measures consequence of aging, thereby linked to the term that are often beneficial, mostly for dermatologic degenerative joint disease. It is now recognized disorders [15]. TTCMs have many varieties. A that OA results from a complex interplay of common misconception is that TTCM is a com- multiple factors, including joint integrity, genet- plicated, exotic art. Difficulties in understanding ics, metabolic, local inflammation, mechanical these measures are due first to more than one forces, previous injury of a joint, and cellular and ingredient in one formulation, second to the biochemical processes [7,8]. myriad of preparations available and third due to To date, there are no means to prevent the labeling, with brands such as Tiger Balm, occurrence of OA. Thus, treatment is directed 3-Snake Oil and Dragon Balm. First and fore- towards pain relief, improving function and most, it must be understood that brand names health-related quality of life [8]. A combination are just brand names. The names are symbolic of nonpharmacologic (Box 1), pharmacologic [14]. Tiger Balm and 3-Snake Oil do not contain (Table 1) and surgical (Box 2) treatment is usually any material from either of these animals. Keywords: diclofenac, needed and should be individualized according Indeed, TTCMs are mixtures containing many ibuprofen, ketoprofen, meta-analysis, osteoarthritis, to patient needs, presence of comorbidities and herbs, ranging from three to 20 different types. piroxicam, randomized drug–drug interactions. Mixtures are based on the concept of traditional controlled trials, topical Chinese medicine. The formulations use differ- nonsteroidal anti- inflammatory drugs Nonpharmacologic treatment ent ingredients to balance the body and to bal- Current treatment guidelines recommend the ance the opposite properties of different herbals. part of combination of weight loss and exercise pro- However, all these myriads of preparations can grams to reduce pain and improve function in be grouped into three classes, according to usage.

Box 1. Nonpharmacologic therapy for be soothing. They usually are not irritating to patients with osteoarthritis. the skin unless the patients are allergic to these ingredients. Other paste preparations are mix- • Patient education tures of various herbals with a petrolatum base. • Self-management programs Zingiber officinale rhizoma, Polygonum multiflo- • Telephone contact rum radix, Peonia lactiflora radix, rhizoma et • Weight loss (if overweight) radix notopterygii, myrrha and other herbals • Physical therapy and aerobic exercise are commonly used for rheumatic pain, a • Patellar taping • Corrective footwear, bracing, joint protection, bi-syndrome in TTCM. None of these have lateral-wedged insoles (for genu varum) been assessed in controlled trials available in • Assistive devices for activities of daily living english [14,17]. • Laser • Pulsed electromagnetic field, ultrasound, Pharmacologic treatment transcutaneous electrical nerve stimulation Paracetamol, up to 4 g/day, remains the most • Acupuncture commonly prescribed drug in patients with mild- • Nutrients, herbal remedies, vitamins/minerals to-moderate pain. A meta-analysis published in •Spa 2004 confirmed the efficacy of acetaminophen in relieving pain due to OA, which should be the In each class, the ingredients revolve around a first-line treatment, reserving nonsteroidal common theme, with only minor differences. anti-inflammatory drugs (NSAIDs) for patients The three classes are: who do not respond [18]. • Oils, ointments, pastes for aches and pains; Both classical NSAIDs and cyclo-oxygenase (Cox)-2 inhibitors are used commonly and are • Oils, ointments, pastes for orthopedic injuries; more effective than placebo and acetaminophen • Lotions and ointments for skin diseases. in reducing pain and functional disabilities in Here, only the first class will be briefly patients with OA of knees and hips [19,20]. This summarized [14,16]. superiority of NSAIDS over acetaminophen, The classic example of class 1 TTCMs used however, is modest and because of the high inter- for rheumatic pain is Tiger Balm [17]. It is an individual variability in patient response to both oil-based balm containing camphor, menthol of these drug classes, it is impossible to predict and one or more essential oils, such as cinna- the patient’s response to them [8,21]. mon oil, oil of clove, cassia oil, citronella oil, oil Opiate analgesic agents, including tramadol, of lavender or cajuput oil. These are compound can also be used and may be of benefit in together in a base oil or petrolatum. It can thus patients with severe pain resistant to NSAIDs or be appreciated that the formulation is meant to in those who have contraindication to treatment with other drugs [11]. Glucosamine and chondroitin are natural Table 1. Pharmacologic therapy of osteoarthritis. substances derived from animal products that Route Pharmacologic agent have acquired substantial popularity in the Oral therapy Acetaminophen treatment of OA [22]. The most important merit NSAIDs is their safety, although their mechanisms of Cyclooxygenase-2 action are unclear [11]. In a 2003 meta-analysis that included 15 randomized, double-blind, Nonacetylated salicylate placebo-controlled trials that assessed either glu- Tramadol cosamine or chondroitin on structure modifica- Opioids tion of knee or hip OA, glucosamine was found Glucosamine sulfate to have a positive effect in reducing joint space Chondroitin sulfate narrowing [23]. In a recently published rand- Intra-articular therapy Glucocorticoids omized controlled trial of glucosamine and Hyaluronan chondroitin, the drugs were not helpful in effec- Tidal irrigation tively reducing pain in patients with OA of the knees. Exploratory analyses, however, suggest Topical therapy NSAIDs that the combination may be effective in the Capsaicin subgroup of patients with moderate-to-severe NSAID: Nonsteroidal anti-inflammatory drug. knee pain [24].

164 Future Rheumatol. (2007) 2(2) futurefuture sciencescience groupgroup Topical nonsteroidal anti-inflammatory drugs in the treatment of osteoarthritis – REVIEW

Box 2. Surgical treatment for patients with osteoarthritis. hospital with GI problems for regular users of oral NSAIDs is 1.3–1.6% annually [30]. In the • Arthroscopic management USA, GI tract bleeding secondary to NSAIDs – Lavage accounts for 41,000 hospitalization and 3300 – Debridement deaths each year [31]. Additionally, approxi- – Abrasion arthroplasty mately 40% of hospital admissions with upper – Subchondral penetration procedures (drilling and microfracture) GI bleeding, and 40% of associated deaths in – Laser/thermal chondroplasty older people, are related to NSAID use [32]. •Osteotomy Other significant side effects of NSAIDs • Arthodesis (stiffening of a joint by operative means) include renal insufficiency, especially in • Total joint replacement/unicompartmental knee replacement patients with reduced renal perfusion, hyper- • Grafting and cell transplantation (autologous osteochondral transplantation) tension, leg edema and exacerbation of heart failure and an increased risk of cardiovascular Intra-articular treatment with hyaluran and events, especially with the use of the new Cox-2 hyalans has recently become a popular sympto- inhibitors [3,5,22,29,33]. matic therapy in OA of the knees and has been approved by the US FDA. However, data on Topical NSAIDs their efficacy are controversial [5,8]. The use of topical NSAIDs is a highly controver- Intra-articular injection of long-acting sial topic in the medical community. In Ger- corticosteroids are widely used in the manage- many, for example, it accounts for two-thirds of ment of patients with OA of the knees, and may the most frequently prescribed NSAIDs for both be particularly beneficial in patients demon- acute and chronic conditions, while in other strating signs of local inflammation with a joint parts of the world it is thought of as junk effusion [25]. Evidence supports short-term medicine and a marker of bad prescribing [34]. (2–4 weeks) improvement in symptoms of OA Topical NSAIDs, however, are attractive sub- of the knee after intra-articular corticosteroid stitutes to oral therapy in reducing the symptoms injection and, compared with placebo, they sig- of OA, with minimal adverse side effects, includ- nificantly relieve pain but do not improve func- ing peptic ulcer disease and GI hemorrhage [3,35]. tional impairment [26,27]. Moreover, long-term They provide the advantage of local, enhanced (2 years) treatment of knee OA with repeated drug delivery to affected tissues in combination intra-articular steroid injections at a frequency with a reduced systemic absorption of NSAID, not exceeding four times/year appears to be in addition to the lack of interactions and ease of effective when compared with saline injections, use [3,35,36]. with no evidence of a deleterious effect on the In the next section of this paper, we will joint structure [28]. The adverse effects of more answer common questions relating to these frequent injections or injections for a period agents, including their efficacy in OA and other exceeding 2 years is unknown. conditions and their safety profile.

Safety concerns Do topical NSAIDs penetrate into The typical patient with OA is an elderly per- the joint? son with multiple medical problems on several The therapeutic effect of topically applied for- medications who will require treatment for mulations is dependent on the ability of the weeks if not years, and thus is at high risk for active ingredients to penetrate into tissue layers toxicity. The elderly are especially vulnerable to beneath the application site, on the high interin- drug toxicity for many reasons, including diffi- dividual difference of skin penetration and any culties with treatment adherence, nutritional other materials that are used to enhance skin insufficiency, altered pharmacokinetics, end- diffusion [37]. organ responsiveness and the enhanced poten- The skin layers through which topical tial for drug–drug interactions arising from NSAIDs must be transported are the stratum polypharmacy for various comorbidities [29]. corneum, epidermis, basement membrane and The most common toxicity attributed to the the dermis. Thus, for optimal penetration and treatment of OA is the increased morbidity and absorption into deeper tissues and the systemic mortality from gastrointestinal (GI) events in circulation, the topical NSAIDs require both patients taking NSAIDs [10,20]. In the UK, for hydrophilic and hydrophobic proportional example, the attributable risk of going to a qualities [38].

futurefuture sciencescience groupgroup www.futuremedicine.com 165 REVIEW – Touma, Chen & Arayssi

In general, plasma concentrations achieved via creams are less effective in skin permeation than topical delivery are 1–10% of those achieved by gels or sprays, but newer formulations such as systemic delivery [39]. Topical administration microemulsions have better potential [44–47]. produces high concentrations in meniscus, Some essential oils and their terpene constitu- tendon sheath and cartilage, higher than that in ents (e.g., eucalyptus and peppermint), as well as plasma or synovial fluid (Figure 1) [38,40]. 10% ethanol and dimethyl sulfoxide are occasion- For instance, the topical application of ally used as vehicles to enhance or accelerate diclofenac solution in 248 patients with primary absorption [3,48–50]. These enhancers may also pro- OA of the knee resulted in a peak serum level of vide a cooling effect, have a local analgesic effect 12 ng/ml, up to 125-times lower than peak and/or induce muscle relaxing action [48]. plasma levels for an equivalent amount of oral Phonophoresis and iontophoresis are two physi- diclofenac [41]. cal modalities used to improve penetration of topi- In another study comparing oral ketoprofen cal medications transdermally. Although drug with topical ketoprofen in 100 patients with knee delivery may be enhanced with these two modali- disorders requiring arthroscopy, levels of ties, its usefulness in improving efficacy of the drug ketoprofen were higher in cartilage and meniscus in patients with OA needs to be studied [51–53]. after topical delivery as compared with the oral To date, there are no studies comparing the var- form. As expected, mean plasma level was ious delivery forms of topical NSAIDs with each significantly lower following the use of topical other in OA. However, patches may be preferred ketoprofen [42]. Similar findings were observed in by some patients because they offer a practical, a study comparing oral with topical ibuprofen [43]. easy-to-use treatment that allows delivery of a well-defined fixed amount of drug per application. Are some forms of delivery better than others? Is there a difference between various Several formulations of topical NSAIDs exist in topical NSAIDs? the market, including gels, creams, foams, oil, A series of NSAIDs were studied in vitro to gen- aerosol, sprays and patches [35]. In general, erate an index to predict topical efficiency. Indomethacin, ketorolac, ketoprofen and Figure 1. Comparison of median maximum concentrations of diclofenac exhibited acceptable efficiency for ketoprofen in joint tissue after topical and oral concentration. external use. However, for dermatological formulations of oxicams (piroxicam and tenoxicam), the use of penetration enhancers may be una- [54] Topical Cmax Oral Cmax voidable . In human subjects, alclofenac and ketoprofen, among several NSAIDs (bufexamac, indomethacin, flufenamic acid, ibuprofen, flurbi- Cartilage profen, ketoprofen and naproxen), have the greatest absorption rates through the skin [55]. Few studies compared the clinical efficacy of topical NSAIDs head-to-head in OA. Only one study compared topical ketoprofen gel with Meniscus diclofenac emulgel in patients with OA, and found no difference in clinical outcome and safety profile [56]. Synovial Head-to-head comparisons of topical NSAIDs tissue are more extensively evaluated in acute pain, although the results may not necessarily be extrap- olated to patients with OA. In a study of Synovial 1575 patients with acute soft-tissue injury, fluid diclofenac was equivalent to ketoprofen with an added cooling effect observed with ketoprofen. Interestingly, in the same study piroxicam gel was 0100 200 300 400 500 600 less effective than both ketoprofen and diclofenac Ketoprofen (ng/g) gel [57]. In another study of 384 patients with acute soft-tissue injuries, diclofenac showed Cmax: Peak concentration. higher efficacy than felbinac gel [40].

166 Future Rheumatol. (2007) 2(2) futurefuture sciencescience groupgroup Topical nonsteroidal anti-inflammatory drugs in the treatment of osteoarthritis – REVIEW

A recent meta-analysis of 26 randomized con- Topical diclofenac administered three- or four- trolled trials of topical NSAIDs, which included times/day was compared with oral diclofenac 2853 patients with acute pain, demonstrated that (50 mg three-times/day) in one study and with topical NSAIDs were significantly better than pla- oral ibuprofen (400 mg three-times/day) in cebo in 19 of 26 trials and that different NSAIDs another. In both studies, equivalent results were have different efficacy, while indomethacin was found between the topical and oral NSAIDs [3,70]. barely distinguished from placebo. Ketoprofen Piroxicam gel administrated three-times/day was found to be significantly better than all other was compared with oral ibuprofen (400 mg studied NSAIDs (ibuprofen, felbinac, piroxicam three-times/day) and with routine NSAID use, and indomethacin) [58]. Similar findings were and in both studies was equivalent to the oral found when topical indomethacin was compared form [71,72]. head-to-head with topical piroxicam [58]. Only one randomized controlled trial compared topical NSAIDs with each other; ketopro- Are topical NSAIDs effective alternatives fen gel (four-times/day) with diclofenac emulgel to oral treatments in OA? (four-times/day) in 85 patients with OA of the The findings of randomized controlled trials and knee followed over a period of 4 weeks. Both meta-analyses advocate the benefit of topical groups had improvement in their knee func- NSAIDs in the treatment of patients with symp- tions, knee score and pain, with no significant tomatic OA of the knees. In our search, we iden- difference between the groups at the end of the tified 18 randomized controlled trials evaluating study [56]. the effect of topical NSAIDs in OA (Table 2); In 1998, a systematic review assessed the effec- 12 compared topical NSAIDs with topical pla- tiveness and safety of topical NSAIDs in acute cebo, four compared topical with oral NSAIDs, and chronic pain conditions in 86 trials, involv- one compared topical NSAIDs with oral NSAIDs ing 10,160 patients. In chronic conditions such and topical placebo, and only one compared as OA and tendonitis, topical NSAIDs were sig- different topical NSAIDs with each other. nificantly better than placebo when given over In the 12 randomized controlled trials that 2weeks [30]. compared the use of topical NSAIDs versus top- A more recent second systematic meta-analy- ical placebo or vehicle-controlled placebo (VCP) sis by Mason and colleagues in 2004, which in patients with OA (15 OA knees, one OA of included 25 trials (14 of which examined gen- fingers and two different joints), there were eral musculoskeletal disorders and 11 examined three topical NSAID preparations: diclofenac OA: nine knees, one finger joints and one mixed (nine randomized controlled trials), ibuprofen sites), produced similar results to the previous (two randomized controlled trials) and eltenac meta-analysis [59]. In this meta-analysis, (two randomized controlled trials). Eltenac is a 14 double-blind, placebo-controlled trials of topical NSAID that is similar in structure to 1502 patients showed that topical NSAIDs were diclofenac, with enhanced skin permeability that significantly better than placebo in the treatment is not in use in humans anymore [49]. Of the 15 of chronic pain. Three trials with 764 patients studies conducted over 4 weeks or less, only comparing a topical with oral NSAID found no three extended beyond 6 weeks. Superiority of difference in efficacy. There were insufficient diclofenac and ibuprofen over placebo or VCP data, however, to allow comparisons of efficacy was demonstrated for the majority of defined between different NSAIDs (Figure 2) [59]. efficacy variables, including pain, functional rat- In 2004, Lin and colleagues published a meta- ings, patient global assessment and physicians’ analysis of the efficacy of topical NSAIDs in the global assessment [50,41,60–69]. treatment of OA. It included 13 randomized In the five randomized controlled trials that controlled trials, representing 1983 patients, and compared the use of topical NSAIDs versus oral found that topical NSAIDs were superior to NSAIDs, only three topical NSAID prepara- placebo in pain reduction and functional tions were studied: diclofenac (two randomized improvement, but only in the first 2 weeks of controlled trials), piroxocam (two randomized treatment. However, no benefit was observed controlled trials) and eltenac (one randomized from topical NSAIDs over placebo in weeks 3 controlled trial). Duration of the studies varied and 4 and the efficacy was not sustained beyond from 4 to 12 weeks (two trials over 3 weeks, 2 weeks [73]. Additionally, topical NSAIDs were two trials over 4 weeks and one trial over inferior to oral NSAIDs in the first week of treat- 12 weeks) [3,35,49,70–72]. ment and, as expected, caused more local side futurefuture sciencescience groupgroup www.futuremedicine.com 167 REVIEW – Touma, Chen & Arayssi [3] [60] [61] [50] [62] [41] [63] [64] [65] Ref. Worse X X X X X Better Outcome X Same weeks and 3days Length of study weeks 6 week 1 12 weeks weeks 2 OA knee 3 weeksOA knee X 12 OA knee 4 weeks X OA knee 1 day OA knee 4 weeks X Diagnosis OA knee OA knee OA knee OA knee hritis; daily;q.i.d: t.i.d:Three-timesdaily. Four-times 4 over the q.i.d gel g of diclofenac q.i.d gel knee vs placebo affected over the affected knee 1.3 q.i.d solution of diclofenac ml vs knee vehicle the affected over knee affected the over q.i.d solution 1.3 q.i.d solution of diclofenac ml vs knee vehicle the affected over over the q.i.d solution control solution knee vs placebo affected knee the affected over q.i.d in 24- a 1% plasters tissugel Flector knee affected one over trial 72-h and plasters over the other vs placebo knee affected 90 over the gel t.i.d mg of eltenac vs 27 knee affected gel eltenac mg of of vs 9 mg knee affected the over t.i.d the affected over t.i.d gel eltenac over the t.i.d vs gel knee placebo knee affected Treatment regimen Treatment 1.3 of ml over diclofenac solution q.i.d affectedthe knee vs vehicle control solution q.i.d over knee the affected 200 over t.i.d cream of mg ibuprofen cream knee vs placebo the affected knee t.i.d over the affected 1.55 t.i.d solution diclofenac ml of vs t.i.d capsules placebo with oral 50 and t.i.d orally mg diclofenac the over t.i.d solution placebo joint affected 180 mg DHEP plastersbid on the joint vs placebo plasters b.i.d affected sters; NSAID: Nonsteroidal anti-inflammatory drug; OA: Osteoart Diclofenac gel gel Diclofenac 1.16% solution Diclofenac 1.5% solution Diclofenac 1.5% Flector tissugel 1% vs placebo gel 0.1%, Eltenac gel 0.3%, eltenac gel 1% eltenac Formulation solution Diclofenac 1.5% cream Ibuprofen 5% solution Diclofenac vs 1.5% 50 mg capsules diclofenac DHEP containing 180 mg diclofenac Active drug therapy diclofenac Topical solution vs vehicle-control solution cream Ibuprofen Topical cream vs placebo diclofenac Topical vs oral diclofenac solution plasters diclofenac Topical plasters vs placebo placebo gel placebo solution vs vehicle-control solution solution vs vehicle-control solution bioadhesive epolamine plaster vs placebo gel placebo n 216 50 622 103 238 gel vs diclofenac Topical 326 diclofenac Topical 248 diclofenac Topical 20 diclofenac Topical 237 gel vs eltenac Topical b.i.d: Twice daily; DHEP: Diclofenac hydroxyethylpyrrolidineb.i.d: pla Twice Table 2. Topical nonsteroidal anti-inflammatory drugs trial drugs anti-inflammatory nonsteroidal 2. Topical . Table Study Niethard (2005) Baer (2005) Roth (2004) Trnavsky (2004) Bookman (2004) Tugwell (2004) Giamber- ardino (2004) Bruhlmann (2003) Ottillinger (2001)

168 Future Rheumatol. (2007) 2(2) futurefuture sciencescience groupgroup Topical nonsteroidal anti-inflammatory drugs in the treatment of osteoarthritis – REVIEW [66] [70] [67] [49] [56] [68] [71] [69] [72] Ref. Worse X X Better Outcome X X X Same 3 weeks 3 1 and day Length of study weeks 3 weeks 4 weeks 2 weeks 2 1 and day OA knee 1 week X OA knee 2 weeks X OA knee 4 weeks X OA of superficial joints OA knee 4 weeks X Diagnosis OA fingers OA knee OA of different joints OA knee hritis; daily;q.i.d: t.i.d:Three-timesdaily. Four-times 200 over t.i.d cream of mg ibuprofen cream knee vs placebo the affected knee t.i.d over the affected 2.5 the over t.i.d gel gm of diclofenac gel t.i.d vs joint placebo affected 1 hydroalcoholic gm of ketoprofen 1vs knee on the affected q.i.d gel g q.i.d emulgel of diclofenac 1 or for q.i.d t.i.d gel g of piroxicam 14 the dose of half days plus prestudy alone piroxicam by followed NSAID peripheral for 14 days over affected joints vs oral NSAID for 28 days 1 (over affected gel g of piroxicam tablet placebo one and tid knee) 400 vs t.i.d orally orally ibuprofen mg t.i.d gel placebo plus t.i.d Treatment regimen Treatment the over q.i.d emulgel Diclofenac joint andaffected placebo tablet orally t.i.d vs placebo emulgelt.i.d and 400 tablets mg ibuprofen orally t.i.d 3 one plus t.i.d gm of eltenac gel vs 50 b.i.d orally tablet placebo mg with b.i.d orally tablet diclofenac t.i.d gel vs placebo t.i.d gel placebo b.i.d orally tablet placebo one with 2 vs q.i.d gel gm of diclofenac q.i.d gel placebo on the bid DHEP plasters 180mg joint vs placebo plasters b.i.d affected sters; NSAID: Nonsteroidal anti-inflammatory drug; OA: Osteoart Ibuprofen cream cream Ibuprofen 5% lecithin Diclofenac 2% organogel Ketoprofen gel hydroalcoholic vs diclofenac emulgel not (concentration specified) gel vs Piroxicam NSAID oral 0.5% gel Piroxicam vs 400 mg ibuprofen Formulation emulgel Diclofenac not (concentration vs 400specified) mg tablets ibuprofen 1% vs gel Eltenac 50 mg diclofenac tablets gel Diclofenac sodium 3%/2.5% gel hyaluronate DHEP plasters 180containing mg diclofenac Active drug therapy diclofenac Topical vs oral ibuprofen emulgel eltenac gel vs oralTopical placebo and diclofenac gel gel vs diclofenac Topical gel vehicle-control plasters diclofenac Topical plasters vs placebo vs placebo cream vs placebo gel placebo emulgel diclofenac NSAID oral ibuprofen n 321 290 119 155 100 cream Ibuprofen Topical 74 gel vs diclofenac Topical 85 vs gel ketoprofen Topical 191 gel vs piroxicam Topical 235 vs oral piroxicam Topical Table 2. Topical nonsteroidal anti-inflammatory drugs trial (Cont.). trial drugs anti-inflammatory nonsteroidal 2. Topical Table Study Rovensky (2001) Zacher (2001) Grace (1999) Sandelin (1997) Waikakul (1997) Roth (1995) Browning (1994) Dreiser (1993) Dickson (1991) daily; DHEP: Diclofenac hydroxyethylpyrrolidineb.i.d: pla Twice

futurefuture sciencescience groupgroup www.futuremedicine.com 169 REVIEW – Touma, Chen & Arayssi

Figure 2. Comparison of NNT of topical analgesics in chronic American College of Rheumatology guidelines, musculoskeletal pain. where topical NSAIDs have been considered as adjunctive or as monotherapy [10,25].

What are the side effects of topical Capsaicin compared with oral NSAIDs? Skin reactions (erythema, rash and itching) are the most frequent of all adverse events (Table 3). Salicylates Adverse event rates, however, are not significantly different from those of placebo ingredients, as has been demonstrated in several trials [35]. NSAID Topical analgesics Topical Mason and colleagues, in their meta-analysis involving 18 placebo trials, described no sta- 0 5 10 15 20 25 30 35 tistically significant difference between topical NNT with 95% confidence interval NSAIDs and topical placebo in the number of

NNT: Number needed to treat; NSAID: Nonsteroidal anti-inflammatory drug. patients experiencing local adverse events (6%), Extracted from [59]. systemic adverse events (3%) or the number withdrawing due to an adverse event (1%) [59]. effects. The conclusions regarding long-term Similar findings were found in a meta-analysis of efficacy, however, were based on the studies of topical NSAIDs for acute musculoskeletal pain [58]. eltenac gel only versus placebo gel [73]. The incidence of GI adverse events, including More recently, topical diclofenac solution nausea, vomiting, diarrhea and dyspepsia, is low (Pennsaid®) in patients with OA of the knee dem- and occurs at a similar rate in patients receiving onstrated effectiveness in a systematic review and placebo and topical NSAIDs. No GI adverse meta-analysis of four randomized controlled trials events were considered severe, and no patient (1385 patients), with a mean trial duration of developed perforation, ulcer or bleeding in any 8.5 weeks [20]. of the studies. The rate of GI events in these Biswal and colleagues approached the ques- 18 randomized, controlled trials was lower than tion of intermediate-term (4–12 weeks) efficacy the percentage reported with oral NSAIDs and of topical NSAIDs for pain control in primary similar or higher to placebo/vehicle control. knee OA in another meta-analysis [74]. Four trials were included and the duration of studies varied Conclusion from 4 to 12 weeks. Four of them compared The current evidence supports the use of topical topical NSAID with placebo or VCP. Pooled NSAIDs in the treatment of OA of the knees. effect of topical NSAID (diclofenac or eltenac) These agents provide an attractive method of measured at 4 weeks or beyond was superior to treatment in the elderly because of their favora- placebo/vehicle in pain relief in knee OA; ble side-effect profile. Long-term trials, how- however, this may not hold true for all the ever, are needed to confirm their long-term preparations [74]. benefit and safety. Recently, Bjordal and colleagues, in a meta- analysis comparing seven pharmacotherapeutic Future perspective interventions (including topical NSAIDs) in Today, OA is regarded as a disease that affects the knee OA, argued that none of the trials of topical whole joint, including subchondral bone, peri- NSAIDs have demonstrated clinically relevant articular muscles, ligaments, capsule, sensory mean effects above the minimally perceptible nerve endings and synovium. Diagnostic tools, threshold of 9.7 mm after 2 weeks or more of chemical and radiographic, that will permit therapy [75]. However, because of the favorable earlier detection of the disease will enable better safety profile, they remain an alternative choice, understanding of this condition and possibly especially in the elderly population [75]. less disability. Owing to the plethora of positive literature on Future therapeutic modalities should go topical NSAIDs, the 2003 European League beyond the cartilage and address the treatment of Against Rheumatism guidelines for the medical the periarticular structures as well as the correc- management of OA of the hip and knee have tion of the mechanical abnormalities. Safety of given greater consideration to the role of topical therapy should remain a priority as the majority NSAID in OA compared with the 2000 of those affected are in their golden years.

170 Future Rheumatol. (2007) 2(2) futurefuture sciencescience groupgroup Topical nonsteroidal anti-inflammatory drugs in the treatment of osteoarthritis – REVIEW [3] [60] [61] [50] [62] [41] [64] [63] [65] [66] Ref.

‡ NSAID NSAID (n = /0) 0 NSAID (n = 24/14.6) NSAID (n = 8/9.6) Vehicle-control (n = 12/15) NSAID (n = 1/2) NSAID NSAID (n = 12/11.2) diclofenac Oral (n = 150/48) Gastorintestinal (no. patients /%patients) Placebo Placebo (n = 2/1.7) Placebo Placebo (n = 18/11.1) Placebo (n = 11/13.1) Placebo (n = 0/0) Placebo Placebo (n = 6/5.5) effects No side Topical diclofenac (n = 108/35) Not reported effects No side * NSAID (n = 4/3.4) NSAID (n = 81/49.4) NSAID (n = 62/73.8) Vehicle-control (n = 41/51.3) NSAID (n = 3/5.9) NSAID (n = 46/43) diclofenac Oral (n = 14/4.5) /% patients) incke’s edema. edema. incke’s Local skinreactions (no. patients ( n = 3 / 2 . 5 ) (n = 53/33) Placebo (n = 12/14.2) (n = 2/3.8) Placebo Placebo ( n = 3 1 / 2 8 . 4 ) effects side no Topical diclofenac (n = 157/51) Not reported effects No side Control 8.30% No side effects Not reported Not reported No side effects No dropout Placebo vehicle 4% Topical 0% placebo Topical 4% placebo Topical Placebo g, numbness, pruritis, paresthesia and Qu Not reported reported Not diclofenac Topical 3% Placebo diclofenac Topical 6% diclofenac Topical 2% Not reportedreported Not Not reportedreported Not Adverse reactions requiring dropout requiring Adverse reactions Active drug 8.40% effects No side Not reported Not reported effects No side diarrhea,constipation,dyspepsia. melena and , irritation, dry skin, dermatitis, burnin NSAID (n = 70/65) diclofenac Oral NSAID (n = 4/3.4) NSAID (n = 126/77) NSAID (n = 78/93) Vehicle-control (n = 58/72.5) NSAID (n = 4/7.8) Overall adverse events (no. patients/% patients) Placebo Placebo (n = 11/9) (n = 88/54.3) Placebo (n = 24/29) Placebo (n = 3/5.8) Placebo + NSAI (n = 16/6.8) Placebo Placebo (n = ) 49/45 effects No side Topical diclofenac Not reported effects No side Nausea, dry mouth, epigastric pain, vomiting, Table 3. Side effects of topical nonsteroidal anti-inflammatory drugs anti-inflammatory nonsteroidal topical of 3. Side effects Table . Study Niethard (2005) Baer (2005) Roth (2004) Placebo Trnavsky (2004) Bookman (2004) (2004) Tugwell Bruhlmann (2003) Giamberardino (2004) Ottillinger (2001) Rovensky (2001) ‡ *Application site reaction: erythema/redness NSAID: Nonsteroidal anti-inflammatory drug.

futurefuture sciencescience groupgroup www.futuremedicine.com 171 REVIEW – Touma, Chen & Arayssi [70] [67] [49] [56] [68] [71] [69] [72] Ref.

‡ Oral ibuprofen Oral ibuprofen (n = 8/5) Diclofenac (n = 11/13.4) Eltenac (n = 6/4.8) Topical diclofenac (n = 0/0) NSAID NSAID (n = 1/2.9) Oral NSAID (n = 0/0) Oral ibuprofen (n = 11/9.3) Gastorintestinal (no. patients /%patients) diclofenac diclofenac (n = 1/0.6) Placebo Placebo (n = 6/7.3) effects No side Placebo (n = 1/1.3) Placebo Placebo (n = 2/5.7) effects No side feldene Topical (n = 2/1.9) Topical piroxicam (n = 15/12.9) * Diclofenac Diclofenac (n = 1/1.2) Eltenac (n = 16/12.7) NSAID (n = 12/20.3) Topical diclofenac (n = 1/1.3) NSAID (n = 4/11.4) Oral NSAID (n = 0/0) ibuprofen Oral (n = 4/3.3) /% patients) incke’s edema. edema. incke’s Local skinreactions (no. patients (n = 5/6.1) Placebo (n = 26/43.3) Placebo (n = 3/3.9) Placebo Placebo (n = 7/20) effects No side feldene Topical (n = 1/0.9) Topical piroxicam (n = 3/2.5) Control 0% placebo Topical No dropout Oral NSAID (no dropout) ibuprofen Oral 5.90% Oral ibuprofen 10% ibuprofen Oral 0% diclofenac Oral Placebo placebo Topical 6.70% Topical placebo1.30% g, numbness, pruritis, paresthesia and Qu Topical diclofenac diclofenac Topical 3% eltenac Topical 3.20% diclofenac Topical 5% No dropout Topical Adverse reactions requiring dropout requiring Adverse reactions Active drug diclofenac Topical 2.90% No dropout feldene Topical 0.90% piroxicam Topical 7.70% diarrhea,constipation,dyspepsia. melena and , irritation, dry skin, dermatitis, burnin NSAID (n = 6/17.1) Oral NSAID (n = 1/1.2) Oral ibuprofen (n = 27/22.9) Oral ibuprofen (n = 9/6) Diclofenac (n = 20/24.4) Eltenac (n = 34/27.0) NSAID 12/20.3) = (n Topical diclofenac (n = 1/1.3) Overall adverse events (no. patients/% patients) diclofenac diclofenac (n = 4/2.4) Placebo (n = 13/15.90) (n = 26/43.3) (n = 4/5.2) Placebo Placebo (n = 9/25.7) effects No side feldene Topical (n = 3/2.8) Topical piroxicam (n = 31/26.5) Nausea, dry mouth, epigastric pain, vomiting, Table 3. Side effects of topical nonsteroidal anti-inflammatory drugs (Cont.). drugs anti-inflammatory nonsteroidal topical of 3. Side effects Table Study (2001)Zacher Topical (1999) Grace Sandelin (1997) Waikakul (1997) Roth (1995) Placebo Browning (1994) (1993)Dreiser Placebo (1991) Dickson *Application site reaction: erythema/redness ‡ NSAID: Nonsteroidal anti-inflammatory drug.

172 Future Rheumatol. (2007) 2(2) futurefuture sciencescience groupgroup Topical nonsteroidal anti-inflammatory drugs in the treatment of osteoarthritis – REVIEW

Executive summary Osteoarthritis • Osteoarthritis is the most common disease of synovial joints in the elderly. • Nonpharmacologic, pharmacologic and surgical treatment modalities are available that help to reduce pain and improve functional ability. • Current pharmacologic treatment, especially nonsteroidal anti-inflammatory drugs (NSAIDs), is associated with increased risk of morbidity and mortality in the most affected population, the elderly. Penetration of topical NSAIDs into the joint • Currently available topical NSAIDs have adequate penetration through the skin into the joint. • Levels of topical NSAIDs in plasma are significantly lower than levels achieved by oral route. • Levels of topical NSAIDs in cartilage and meniscus are higher with topical route than oral route. Efficacy of topical NSAIDs in osteoarthritis • Eighteen randomized controlled trials and several meta-analyses compared topical NSAIDs with either placebo, oral NSAIDs or with each other. • The majority of topical NSAIDs decrease pain due to osteoarthritis and improve functional ability over the short term. • There are insufficient data for long-term efficacy of these drugs. Safety • The most common side effects of topical NSAIDs are local skin reactions. • Gastrointestinal side effects are minimal, comparable with placebo and better than oral NSAIDs.

Bibliography 9. Messier SP, Loeser RF, Mitchell MN et al.: 16. Zizeng L, Shude J: Applied traditional Papers of special note have been highlighted as Exercise and weight loss in obese older Chinese medicine in rheumatisms. Beijing: either of interest (•) or of considerable interest (••) adults with knee osteoarthritis: a Peoples’ Health (1998). to readers. preliminary study. J. Am. Geriatr. Soc. 48(9), 17. Ng SK: Topical traditional Chinese 1. Hunter DJ, Felson DT: Osteoarthritis. 1062–1072 (2000). medicine. A report from Singapore. Arch. Br. Med. J. 332(7542), 639–642 10. Recommendations for the medical Dermatol. 134(11), 1395–1396 (1998). (2006). management of osteoarthritis of the hip and 18. Zhang W, Jones A, Doherty M: Does •• Excellent clinical review of osteoarthritis. knee: 2000 update. American College of paracetamol (acetaminophen) reduce the 2. Cooper C, Jordan KM: Topical NSAIDs in Rheumatology Subcommittee on pain of osteoarthritis? A meta-analysis of osteoarthritis. Br. Med. J. 329(7461), Osteoarthritis Guidelines. Arthritis Rheum. randomised controlled trials. Ann. Rheum. 304–305 (2004). 43(9), 1905–1915 (2000). Dis. 63(8), 901–907 (2004). 3. Tugwell PS, Wells GA, Shainhouse JZ: •• American College of Rheumatology 19. Bjordal JM, Ljunggren AE, Klovning A, Equivalence study of a topical diclofenac guidelines for the management of Slordal L: Non-steroidal anti-inflammatory solution (pennsaid) compared with oral osteoarthritis. drugs, including cyclo-oxygenase-2 diclofenac in symptomatic treatment of 11. Ge Z, Mu Y, Heng BC et al.: Osteoarthritis inhibitors, in osteoarthritic knee pain: meta- osteoarthritis of the knee: a randomized and therapy. Arthritis Rheum. 55(3), analysis of randomised placebo controlled controlled trial. J. Rheumatol. 31(10), 493–500 (2006). trials. Br. Med. J. 329(7478), 1317(2004). 2002–2012 (2004). • Comprehensive review of acupuncture in •• Extensive paper on the efficacy of topical 4. Bird HA: When are NSAIDs appropriate in osteoarthritis. nonsteroidal anti-inflammatory drugs osteoarthritis? Drugs Aging 12(2), 87–95 12. Scharf HP, Mansmann U, Streitberger K (NSAIDs) in knee osteoarthritis. (1998). et al.: Acupuncture and knee osteoarthritis: 20. Towheed TE, Maxwell L, Judd MG, 5. Sarzi-Puttini P, Cimmino MA, Scarpa R a three-armed randomized trial. Ann. Intern. Catton M, Hochberg MC, Wells G: et al.: Osteoarthritis: an overview of the Med. 145(1), 12–20 (2006). Acetaminophen for osteoarthritis. Cochrane disease and its treatment strategies. Semin. 13. Kwon YD, Pittler MH, Ernst E: Database Syst. Rev. (1), CD004257 (2006). Arthritis Rheum. 35(1 Suppl. 1), 1–10 Acupuncture for peripheral joint 21. Bannwarth B: Acetaminophen or NSAIDs (2005). osteoarthritis: a systematic review and meta- for the treatment of osteoarthritis. Best Pract. 6. Lawrence RC, Helmick CG, Arnett FC analysis. Rheumatology (Oxford), 45(11), Res. Clin. Rheumatol. 20(1), 117–129 et al.: Estimates of the prevalence of arthritis 1331–1337 (2006). (2006). and selected musculoskeletal disorders in the 14. Chen LX, Goldman J, Pullman-Mooar S. 22. Glass GG: Osteoarthritis. Clin. Fam. Pract. United States. Arthritis Rheum. 41(5), Therapy for Chronic Pain. In: The Textbook 7, 161–179 (2005). 778–799 (1998). of Fibromyalgia and Other Non-Neuropathic 23. Richy F, Bruyere O, Ethgen O, Cucherat M, 7. Creamer P, Hochberg MC: Osteoarthritis. Pain Syndromes. Wallace D, Clauw D (Eds): Henrotin Y, Reginster JY: Structural and Lancet 350(9076), 503–508 (1997). Lippincott Williams & Wilkins, PA, USA symptomatic efficacy of glucosamine and 8. Felson DT: Clinical practice. Osteoarthritis (2005). chondroitin in knee osteoarthritis: a of the knee. N. Engl. J. Med. 354(8), 15. Litt JZ: Alternative topical therapy. comprehensive meta-analysis. Arch. Intern. 841–848 (2006). Dermatol. Clin. 7(1), 43–52 (1989). Med. 163(13), 1514–1522 (2003).

futurefuture sciencescience groupgroup www.futuremedicine.com 173 REVIEW – Touma, Chen & Arayssi

24. Clegg DO, Reda DJ, Harris CL et al.: 34. Sift Carter R, Ebner D, Brenner D et al.: Use 46. Yang JH, Kim YI, Kim KM: Preparation and Glucosamine, chondroitin sulfate, and the of topical NSAIDs in patients receiving evaluation of aceclofenac microemulsion for two in combination for painful knee systemic NSAID treatment: a pharmacy-based transdermal delivery system. Arch. Pharm. Res. osteoarthritis. N. Engl. J. Med. 354(8), study in Germany. J. Clin. Epidemiol. 50(2), 25(4), 534–540 (2002). 795–808 (2006). 217–218 (1997). 47. Yuan Y, Li SM, Mo FK, Zhong DF: 25. Jordan KM, Arden NK, Doherty M et al.: 35. Mazieres B: Topical ketoprofen patch. Drugs Investigation of microemulsion system for EULAR Recommendations 2003: an R D 6(6), 337–344 (2005). transdermal delivery of meloxicam. Int. J. evidence based approach to the management 36. Lan Chen TA: Is there a role for topical Pharm. 321(1–2), 117–123 (2006). of knee osteoarthritis: Report of a Task nonsteroidal anti-inflammatory drugs in the 48. Rosenstein ED: Topical agents in the Force of the Standing Committee for treatment of tendonitis? Aging Health 2(1), treatment of rheumatic disorders. Rheum. International Clinical Studies Including 59–60 (2006). Dis. Clin. North Am. 25(4), 899–918, viii Therapeutic Trials (ESCISIT). Ann. Rheum. 37. Dehghanyar P, Mayer BX, Namiranian K, (1999). Dis. 62(12), 1145–1155 (2003). Mascher H, Muller M, Brunner M: Topical 49. Sandelin J, Harilainen A, Crone H, •• European League Against skin penetration of diclofenac after single- Hamberg P, Forsskahl B, Tamelander G: Rheumatism guidelines for management and multiple-dose application. Int. J. Clin. Local NSAID gel (eltenac) in the treatment of knee osteoarthritis. Pharmacol. Ther. 42(7), 353–359 of osteoarthritis of the knee. A double blind 26. Arroll B, Goodyear-Smith F, Corticosteroid (2004). study comparing eltenac with oral diclofenac injections for osteoarthritis of the knee: 38. Heyneman CA, Lawless-Liday C, Wall GC: and placebo gel. Scand J. Rheumatol. 26(4), meta-analysis. Br. Med. J. 328(7444), 869 Oral versus topical NSAIDs in rheumatic 287–292 (1997). (2004). diseases: a comparison. Drugs 60(3), 50. Roth SH, Shainhouse JZ: Efficacy and safety 27. Ravaud P, Moulinier L, Giraudeau B et al.: 555–574 (2000). of a topical diclofenac solution (pennsaid) in Effects of joint lavage and steroid injection 39. Vaile JH, Davis P: Topical NSAIDs for the treatment of primary osteoarthritis of in patients with osteoarthritis of the knee: musculoskeletal conditions. A review of the knee: a randomized, double-blind, results of a multicenter, randomized, the literature. Drugs 56(5), 783–799 (1998). vehicle-controlled clinical trial. Arch. Intern. controlled trial. Arthritis Rheum. 42(3), 40. Bouchier-Hayes TA, Rotman H, Med. 164(18), 2017–2023 (2004). 475–482 (1999). Darekar BS: Comparison of the efficacy and 51. Welch V, Brosseau L, Peterson J, Shea B, 28. Raynauld JP, Buckland-Wright C, Ward R tolerability of diclofenac gel (Voltarol Tugwell P, Wells G: Therapeutic ultrasound et al.: Safety and efficacy of long-term Emulgel) and felbinac gel (Traxam) in the for osteoarthritis of the knee. Cochrane intraarticular steroid injections in treatment of soft tissue injuries. Br. J. Clin. Database Syst. Rev. (3), CD003132 (2001). osteoarthritis of the knee: a randomized, Pract. 44(8), 319–320 (1990). 52. Speed CA: Therapeutic ultrasound in soft double-blind, placebo-controlled trial. 41. Bookman AA, Williams KS, Shainhouse JZ: tissue lesions. Rheumatology (Oxford) Arthritis Rheum. 48(2), 370–377 (2003). Effect of a topical diclofenac solution for 40(12), 1331–1336 (2001). 29. Tannenbaum H, Bombardier C, Davis P, relieving symptoms of primary osteoarthritis 53. Gam AN, Johannsen F: Ultrasound Russell AS; Third Canadian Consensus of the knee: a randomized controlled trial. therapy in musculoskeletal disorders: a Conference Group: An evidence-based CMAJ 171(4), 333–338 (2004). meta-analysis. Pain 63(1), 85–91 (1995). approach to prescribing nonsteroidal 42. Rolf C, Engstrom B, Beauchard C, Jacobs LD, 54. Cordero JA, Camacho M, Obach R, antiinflammatory drugs. Third Canadian Le Liboux A: Intra-articular absorption and Domenech J, Vila L: In vitro based index of Consensus Conference. J. Rheumatol. 33(1), distribution of ketoprofen after topical plaster topical anti-inflammatory activity to 140–157 (2006). application and oral intake in 100 patients compare a series of NSAIDs. Eur. J. Pharm. 30. Moore RA, Tramer MR, Carroll D, Wiffen undergoing knee arthroscopy. Rheumatology Biopharm. 51(2), 135–142 (2001). PJ, McQuay HJ: Quantitative systematic (Oxford) 38(6), 564–567 (1999). 55. Yano T, Nakagawa A, Tsuji M, Noda K: review of topically applied non-steroidal 43. Dominkus M, Nicolakis M, Kotz R, Skin permeability of various non-steroidal anti-inflammatory drugs. Br. Med. J. Wilkinson FE, Kaiser RR, Chlud K: anti-inflammatory drugs in man. Life Sci. 316(7128), 333–338 (1998). Comparison of tissue and plasma levels of 39(12), 1043–1050 (1986). • Quantitative systemic review suggesting ibuprofen after oral and topical 56. Waikakul S, Penkitti P, Soparat K, that local NSAIDs are effective in acute administration. Arzneimittelforschung Boonsanong W: Topical analgesics for knee and chronic musculoskeletal condition. 46(12), 1138–1143 (1996). arthrosis: a parallel study of ketoprofen gel 31. Griffin MR: Epidemiology of nonsteroidal 44. Subramanian N, Ghosal SK, Moulik SP: and diclofenac emulgel. J. Med. Assoc. Thai. anti-inflammatory drug-associated Enhanced in vitro percutaneous absorption 80(9), 593–597 (1997). gastrointestinal injury. Am. J. Med. 104(3A), and in vivo anti-inflammatory effect of a 57. Patel RK, Leswell PF: Comparison of S23–S29 (1998). selective cyclooxygenase inhibitor using ketoprofen, piroxicam, and diclofenac gels 32. Langman MJ: Ulcer complications microemulsion. Drug Dev. Ind. Pharm. in the treatment of acute soft-tissue injury in associated with anti-inflammatory drug use. 31(4–5), 405–416 (2005). general practice. General Practice Study What is the extent of the disease burden? 45. Paolino D, Ventura CA, Nistico S, Puglisi G, Group. Clin. Ther. 18(3), 497–507 Pharmacoepidemiol. Drug Saf. 10(1), 13–19 Fresta M: Lecithin microemulsions for the (1996). (2001). topical administration of ketoprofen: 58. Mason L, Moore RA, Edwards JE, Derry S, 33. Fitzgerald GA: Coxibs and cardiovascular percutaneous adsorption through human McQuay HJ: Topical NSAIDs for acute pain: disease. N. Engl. J. Med. 351(17), skin and in vivo human skin tolerability. a meta-analysis. BMC Fam. Pract. 5, 10 1709–1711 (2004). Int. J. Pharm. 244(1–2), 21–31 (2002). (2004).

174 Future Rheumatol. (2007) 2(2) futurefuture sciencescience groupgroup Topical nonsteroidal anti-inflammatory drugs in the treatment of osteoarthritis – REVIEW

59. Mason L, Moore RA, Edwards JE, Derry S, double-blind, placebo-controlled study on 74. Biswal S, Medhi B, Pandhi P: Long-term McQuay HJ: Topical NSAIDs for chronic the efficacy and safety of a 5% ibuprofen efficacy of topical nonsteroidal musculoskeletal pain: systematic review and cream. Drugs Exp. Clin. Res. 27(5–6), antiinflammatory drugs in knee meta-analysis. BMC Musculoskelet. Disord. 5, 209–221 (2001). osteoarthritis: metaanalysis of randomized 28 (2004). 67. Grace D, Rogers J, Skeith K, Anderson K: placebo controlled clinical trials. •• Summary of topical NSAID use in chronic Topical diclofenac versus placebo: a double J. Rheumatol. 33(9), 1841–1844 (2006). musculoskeletal conditions. blind, randomized clinical trial in patients •• Meta-analysis showing intermediate-term 60. Niethard FU, Gold MS, Solomon GS et al.: with osteoarthritis of the knee. efficacy of topical NSAIDs. Efficacy of topical diclofenac diethylamine gel J. Rheumatol. 26(12), 2659–2663 (1999). 75. Bjordal JM, Klovning A, Ljunggren AE, in osteoarthritis of the knee. J. Rheumatol. 68. Roth SH: A controlled clinical Slordal L: Short-term efficacy of 32(12), 2384–2392 (2005). investigation of 3% diclofenac/2.5% pharmacotherapeutic interventions in 61. Baer PA, Thomas LM, Shainhouse Z: sodium hyaluronate topical gel in the osteoarthritic knee pain: a meta-analysis of Treatment of osteoarthritis of the knee with a treatment of uncontrolled pain in chronic randomised placebo-controlled trials. Eur. J. topical diclofenac solution: a randomised oral NSAID users with osteoarthritis. Int. J. Pain 11(2), 125–138 (2007). controlled, 6-week trial Tissue React. 17(4), 129–132 (1995). •• Extensive review challenging the clinical [ISRCTN53366886]. BMC Musculoskelet. 69. Dreiser RL, Tisre-Camus M: DHEP efficacy of commonly used agents for Disord. 6, 44 (2005). plasters as a topical treatment of knee osteoarthritis of knees. 62. Trnavsky K, Fischer M, Vogtle-Junkert U, osteoarthritis – a double-blind placebo- Schreyger F: Efficacy and safety of 5% controlled study. Drugs Exp. Clin. Res. Affiliations ibuprofen cream treatment in knee 19(3), 117–123 (1993). • Zahi Touma, MD osteoarthritis. Results of a randomized, 70. Zacher J, Burger KJ, Farber L, Grave M, American University of Beirut Medical Center, double-blind, placebo-controlled study. Abberger H, Bertsch K: Topical diclofenac Department of Internal Medicine, Division of J. Rheumatol. 31(3), 565–572 (2004). versus oral ibuprofen: a double blind, Rheumatology, 3 Dag Hammarskjöld Plaza, 63. Giamberardino MA: Proof of the randomized clinical trial to demonstrate 8th floor, New York, NY, USA pharmacodynamic activity of Flector Plaster efficacy and tolerability in patients with Tel.: +1 961 387 4677; in healthy volunteers and patients suffering activated osteoarthritis of the finger joints Fax: +1 961 136 5189; from osteoarthritis of the knee. Presse Med. (Heberden and/or Bouchard arthritis). [email protected] (33), 355–359 (2004). Aktuelle Rheumatologie 26(1), 7–14 (2001). •Lan Chen, MD, PhD 64. Bruhlmann, P, Michel BA: Topical 71. Browning RC, Johson K: Reducing the University of Pennsylvania, diclofenac patch in patients with knee dose of oral NSAIDs by use of Feldene Gel: Department of Internal Medicine, osteoarthritis: a randomized, an open study in elderly patients with Division of Rheumatology, PA, USA double-blind, controlled clinical trial. osteoarthritis. Adv. Ther. 11(4), 198–207 Tel.: +1 215 662 4333; Clin. Exp. Rheumatol. 21(2), 193–198 (1994). Fax: +1 215 349 8900; (2003). 72. Dickson DJ: A double-blind evaluation of [email protected] 65. Ottillinger B, Gomor B, Michel BA, topical piroxicam gel with oral ibuprofen in • Thurayya Arayssi, MD Pavelka K, Beck W, Elsasser U: Efficacy and osteoarthritis of the knee. Curr. Ther. Res. American University of Beirut Medical Center, safety of eltenac gel in the treatment of knee Clin. Exp. 49, 199–207 (1991). Internal Medicine Department, osteoarthritis. Osteoarthr. Cartil. 9(3), 73. Lin J, Zhang W, Jones A, Doherty M: 3 Dag Hammarskjöld Plaza, 8th floor, New 273–280 (2001). Efficacy of topical non-steroidal York, NY, 10017, USA 66. Rovensky J, Micekova D, Gubzova Z et al.: anti-inflammatory drugs in the treatment Tel.: +1 961 135 0000 ext: 5383; Treatment of knee osteoarthritis with a of osteoarthritis: meta-analysis of Fax: +1 961 136 5189; topical non-steroidal antiinflammatory randomised controlled trials. Br. Med. J. [email protected] drug. Results of a randomized, 329(7461), 324 (2004).

futurefuture sciencescience groupgroup www.futuremedicine.com 175