Topical Agents for the Management of Musculoskeletal Pain Steven P

342 Journal of Pain and Symptom Management Vol. 33 No. 3 March 2007

Special Article Topical Agents for the Management of Musculoskeletal Pain Steven P. Stanos, DO Chronic Pain Care Center, Rehabilitation Institute of Chicago, and Department of Physical Medicine and Rehabilitation, Feinberg School of Medicine, Northwestern University, Chicago, Illinois, USA

Abstract The recent recognition of the magnitude of cardiovascular risk of both nonselective nonsteroidal anti-inflammatory drugs and COX-2 selective inhibitors, in addition to the persistent concerns about the use of opioids, has brought increased attention to nonsystemic, topical analgesics. These agents have a favorable safety profile and there is increasing evidence indicating their efficacy for a variety of pain disorders. The use of topical analgesics in the treatment of the most prevalent musculoskeletal pain syndromes is described, with a focus on mechanisms for drug delivery and clinical trials data. J Pain Symptom Manage 2007;33:342e355. Ó 2007 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.

Key Words Counterirritants, local anesthetics, musculoskeletal pain, topical analgesics

Introduction may provide a safe and effective therapeutic approach for some musculoskeletal pains. Musculoskeletal pain encompasses a wide spectrum of disorders from simple ligamen- tous injuries (e.g., ankle sprains) to intra- articular disorders (e.g., osteoarthritis and rheumatoid arthritis), muscle pain syndromes Epidemiology (e.g., myofascial pain and ﬁbromyalgia), and Low back pain (LBP) and osteoarthritis are various spine-related neck and low back condi- the most common musculoskeletal disorders. tions (e.g., disc degeneration, disc herniation, LBP creates a signiﬁcant burden on society facet arthropathies, and spondylosis). These due to its high prevalence (17.6% within the 1 disorders are highly prevalent and exact U.S. work force) and consequent economic 2 a huge toll in terms of individual quality of impact (estimated to exceed $50 billion). life and cost to society. Topical analgesics LBP is the most reported musculoskeletal pain3 and as many as 84% of adults will experience LBP in their lifetime.4 When chronic, the complex pathophysiology, often involving both musculoskeletal and neuropathic compo- Address reprint requests to: Steven P. Stanos, DO, Reha- nents, may cause the management of LBP to bilitation Institute of Chicago, Chronic Pain Care be notoriously challenging. Center, 1030 North Clark Street, Suite 320, Chicago, IL 60610, USA. E-mail: [email protected] Osteoarthritis affects more than 20 million Americans between the ages of 25 years and Accepted for publication: November 28, 2006. 75 years5 and is predicted to become the

Ó 2007 U.S. Cancer Pain Relief Committee 0885-3924/07/$esee front matter Published by Elsevier Inc. All rights reserved. doi:10.1016/j.jpainsymman.2006.11.005 Vol. 33 No. 3 March 2007 Topical Agents for Musculoskeletal Pain 343

fourth leading cause of disability by the year Topical Drug Delivery 2020, worldwide.3 Over the course of the dis- Controversy regarding postmarketing ad- ease progression, joints undergo degeneration verse event (AE) data and a re-examination of articular cartilage, osteophyte formation, of coxib and traditional NSAID safety studies and subchondral bone sclerosis.5 Nerve fibers ignited intense debate within both popular localize densely within bone, and chronic and medical media starting from the end of pain is thought to arise from sensitized noci- 12 2004 and still ongoing currently. Certainly, ceptors that become exposed due to the ero- the debates brought the issue of pharmaco- sion of articular cartilage.6 therapy safety to the forefront, and conse- Numerous other musculoskeletal pains are quently may have positively affected the prevalent. Ankle sprain alone has an incidence prescribing habits of clinicians. Hence, health of 52.7 per 10,000, according to a study con- care providers and the pharmaceutical indus- ducted in the United Kingdom.7 Bone pain try are currently reassessing the potential anal- may result from metastatic cancer, orthopedic gesic and additional safety advantages of one surgery, and traumatic injuriesdpotentially af- of the oldest routes of delivery, topical applica- fecting many people. Although the majority of 13 tion. Potentially, topical agents can achieve these injuries are benign and self-limiting, a similar efficacy to oral formulations without many patients may experience intermittent or the associated systemic side effects. Some evi- more persistent pain and/or loss of function, dence has shown that topically delivered both of which adversely affect a patient’s gen- agents can accumulate to therapeutic concen- eral quality of life.8 trations within the local tissues to which they have been applied while maintaining low serum levels. Potentially, lowering the systemic levels of medications reduces the associated risk of organ or tissue toxicity; a more com- The Biology of Pain: Pathologic plete list of the benefits and limitations of der- 14e32 vs. Physiologic Pain mal and transdermal delivery systems are listed in Table 1. In a healthy individual, pain may be defined Topical agents comprise a growing part of as a complex sensory experience associated the over-the-counter analgesic market, as well with actual or potential tissue damage.9 Injury as a smaller evolving niche market of com- in the periphery induces keratinocytes and pounding pharmacies in the United States blood vessels in the dermis to release excit- and Europe. Topical medications for pain ac- atory factors, such as prostaglandins, substance counted for 6.1% of the U.S. analgesic market P (SP), and calcitonin gene-related peptide, in 2000.33 Yet, an email survey in 2002 con- which bind to receptors on nociceptive fibers ducted by the American Society of Regional and cause depolarization. These unspecial- Anesthesia and Pain Medicine indicated that ized, peripheral nerve fibersdC and Ad poly- only 27% of clinicians prescribe compounded modal nociceptorsdcan be stimulated by topical analgesics, despite the perception that noxious thermal, chemical, and mechanical in- 43 3% of treated patients respond favorably puts. They transmit signals from the periphery to topical agents, with an average of 47 3% via the dorsal horn to higher cerebral struc- pain relief and few side effects.34 tures.10 Generally, the intensity, localization, and timing of the initiating stimuli are re- flected in the level of the neuronal signal.11 Topical vs. Transdermal In contrast, with inflamed tissue, an external The terms ‘‘topical’’ and ‘‘transdermal’’ are stimulus is not required to generate signal sometimes used interchangeably, although im- transduction and transmission to the dorsal portant differences need to be considered. horn. This hypersensitivity, seen in diseases Both delivery methods must transverse the like osteoarthritis, can be inhibited by a num- stratum corneumdthe major barrier to de- ber of different pharmacologic agents, such livering treatment.6 Transdermal methods as nonsteroidal anti-inflammatory drugs deliver medication through percutaneous ab- (NSAIDs), opioids, and cannabinoids.11 sorption, with the goal of achieving 344 Stanos Vol. 33 No. 3 March 2007

Table 1 shelters the live epidermis (Fig. 1).6,14 Once Benefits and Limitations of Analgesia past this relatively impermeable barrier, anal- by Cutaneous Delivery gesics may access the unmyelinated C-fibers Benefits Limitations and encounter the predominant keratinocytes First pass Diffusion across and melanocytes of the epidermal layer. Below metabolism and other the stratum corneum the epidermis, the dermal layer also contains variables associated with only occurs for the gastrointestinal tract molecules <500 Da.25 nociceptive fibers, along with fibroblasts, con- (such as pH and gastric Topical agents nective tissue, blood vessels, hair follicles, emptying time) are must have both e and glands. Collectively, the nerves present in avoided.15 17 aqueous and lipid Reduced side effects, and solubility.26 the epidermis and dermis are referred to as 6 the minimization of drug Both intra- and the cutaneous nociceptors. concentration peaks and interindividual Animal models have suggested that the vari- troughs in the blood.17,18 variability in the Ease of dose termination permeability ability in transcutaneous absorption rates be- in the event of untoward of skin, as well as tween topical agents is likely derived from side effects. differences between their ability to negotiate the superficial Delivery can be sustained healthy and diseased 35 and controlled over a skin, causes variable skin. Ideally, a topical agent will have a low 14 prolonged period.19,20 efficacy.27,28 molecular weight (<500 Da) and have both Direct access to the Skin enzymes can cause hydrophobic features in order to transverse target site.13,14 metabolism before Convenient and painless cutaneous absorption, the stratum corneum and hydrophilic charac- administration.15,16 reducing the potency teristics to penetrate the predominantly aque- Improved patient of the drug.29 ous epidermis.35,36 Furthermore, ex vivo acceptance and adherence Localized skin e to therapy.21 23 irritation, such studies of human skin tissue indicate that non- Ease of use may reduce as erythema, can be physiological pHs of the vehicle can reduce e overall health treatment common.30 32 the skin penetration of an agent, while occlu- costs.24 Provides a viable sion of the skin surface can have a beneficial solution for treatment impact on drug delivery.36 when oral dosing Several delivery agents have been engi- is not feasible (i.e., in unconscious or neered to increase the bioavailability of topi- nauseated patients.)17 cally delivered analgesics. For example, Adapted from Ref. 14. lecithin organogels are biocompatible jelly- like phases, composed primarily of hydrated therapeutic systemic levels of active drug com- phospholipids and organic liquid, which have parable to oral medications. Transdermal been used to deliver NSAIDs and other treat- pharmacotherapies can be, and often are, ad- ments for muscle spasm, cancer, and brux- ministered distal to the site of injury (i.e., sus- ism.37 Also, lecithin, pluronic gel, and tained release nicotine and clonidine patches isopropyl palmitate (called pluronic lecithin and long-acting fentanyl delivery systems), organogel-PLO) have been combined to cre- and typically deliver therapy over an extended ate a colloidal gel used to topically deliver period of time after a slow onset of action. In NSAIDs.38 Additionally, polyethylene glycol contrast, topical agents use cutaneous delivery and limonene39 have been used as topical pen- to specifically target the site of application. etration enhancers, increasing the absorption The sites of action for topical agents are the rate of NSAIDs by up to 75-fold.36 Dating soft tissues and peripheral nerves underlying back as far as a century ago, the delivery of top- the site of application.13 Serum levels generally ical therapeutics was ameliorated by nonphar- remain relatively low, and consequently, sys- macologic methods, such as iontophoresis.14 temic side effects or drug-drug interactions This method uses a low-level current to en- are more unlikely.13 hance the permeability of a topically applied The vehicle in which the active ingredi- agent. Finally, patch preparations and ‘‘plas- ent(s) are delivered can affect the skin pene- ters’’ incorporate adhesive cloth disposable sys- tration depth and absorption rate into the tems and may offer additional benefit to more epidermis.13 The penetration of topical modal- traditional topical creams, gels, or solutions, ities is limited by the stratum corneumda given their potential to deliver medication at dense layer of flattened keratinocytes that a continuous rate.40 Vol. 33 No. 3 March 2007 Topical Agents for Musculoskeletal Pain 345

Meissner’s Hair shaft corpuscle Stratum corneum Stratum Dermal papillae lucidum Pore Stratum Free nerve granulosum ending Local Anesthetics Sebaceous Stratum (oil) gland spinosum Stratum Arrector pili basale muscle Sensory Papillary layer nerve fiber Reticular layer Hair follicle Hair root Systemic Opioids Antidepressants Artery Hypodermis

Vein Eccrine sweat gland Deeper tissue Pacinian corpuscle Nonsteroidal Adipose tissue anti-inflammatory Root hair plexus agents (NSAIDs)

Fig. 1. Anatomy and physiology of the skin with the potential targets or sites of action of selected analgesics. Adapted from Ref. 14.

Although analgesics have been the most excitatory amino acids, as well as modulation thoroughly studied in controlled trials of all of G-protein mediated signal transduction.13 types of agents delivered topically, anesthetics A large variety of topical formulations of and counterirritants can be administered topi- NSAIDs are available commercially, including cally as well. In fact, some of the oldest topical the agents listed in Table 3.35 medicines may have been counterirritantsd Pharmacokinetic data suggest that topically mild or moderate noxious agents that suppress applied NSAIDs can result in enhanced local the perception of pain.13 Anesthetic agents have evolved considerably for the treatment of musculoskeletal pain, too. Some examples Table 2 of commercially available products and Examples of Topical Agents for the Treatment pharmacy-compounded agents are listed in of Musculoskeletal Pain and Prescribing Table 2. Considerations Pharmacy- Commercially Compounded Available Products Preparations

Topical Treatment Options: Clinical Examples Capsaicin Alpha-2 agonists 2.5% lidocaine/ Anticonvulsants Trial Evidence 2.5% prilocaine Local anesthetics Nonsteroidal Anti-Inﬂammatory Drugs Lidocaine NMDA antagonists patch 5% NSAIDs Topical NSAIDs have been more widely used Doxepin Opioids and studied in Europe than in the United TCAs States. Research primarily conducted in Considerations Consistency of Potential variability Europe has suggested several potential peri- preparation of preparation pheral mechanisms of analgesic activity, in- Established Lack of controlled cluding the inhibition of prostaglandin safety and efﬁcacy trials synthesis, the lipoxygenase pathway, and (FDA approval) 346 Stanos Vol. 33 No. 3 March 2007

Table 3 plasters over a five-day period, or a 50 mg Topical NSAIDs 42 oral dose. The median Cmax values for topical Formulation Active Ingredient and oral administration were 568.9 ng/g and Ointment Indomethacin 85.7 ng/g in the cartilage, respectively (a 6.8- Cream Diclofenac, ibuprofen, benzydamine, fold difference). In contrast, the plasma values salicylic acid were 18.7 ng/ml for topically administered ke- Spray Indomethacin Patch/plaster Diclofenac, flurbiprofen toprofen and 2,595.3 ng/ml for the oral route. Gel Piroxicam, diclofenac, felbinac, Overall, when applied topically, the ketopro- ketoprofen, indomethacin, fen levels were 30-fold greater in the cartilage ibuprofen, salicylic acid, eltenac 42 Drops Ketorolac, flurbiprofen, suprofen, than in the plasma. diclofenac Relative to any other topically administered Foam Ketoprofen, felbinac drug, the largest amount of clinical evidence Reprinted with permission from Ref. 35. has been accumulated for NSAIDs.13 A meta- analysis by Moore et al. considered 86 random- concentrations without significant toxic sys- ized, placebo-controlled trials of NSAIDs for temic levels. Heyneman et al. summarized treatment of pain conditions; 10,160 patients both single- and multiple-dose NSAID absorp- were included.43 The effectiveness measure tion studies.36 Collectively, the studies indi- nearest to one week following the start of treat- cated that following topical administration of ment for acute conditions, such as soft tissue NSAIDs, peak plasma levels were less than trauma, sprains, and strains, and the two- 10% of the concentrations obtained from week measure for chronic pain conditions, oral dosing (range, 0.2%e8.0%). As the total such as osteoarthritis and tendonitis, were systemic absorption from topical application used in the meta-analysis. Overall, the number is only 3%e5% of the oral route, systemic tox- needed to treat (NNT) was 3.9 for acute pain icity from topical NSAIDs are correspondingly conditions, more specifically: 2.6 for ketopro- rare. Also, the length of time before Cmax is fen, 3.0 for felbinac, 3.5 for ibuprofen, and achieved following topical application ranged 4.2 for piroxicam (benzydamine and indo- from 2.2 to 23 hours, approximately 10 times methacin had efficacies comparable to pla- longer than the time required for the equiva- cebo). For chronic pain conditions, the NNT lent oral dose. Topical NSAIDs achieve was 3.1 and ranged between 2.7 and 3.8. Fur- steady-state generally within 2e5 days of re- thermore, for all pain conditions studied, top- peated application.36 ical NSAIDs rarely induced local skin reactions Penetration studies also indicate that topi- (3.6%) and more rarely had any adverse sys- cally applied NSAIDs reach therapeutic con- temic effects (<0.5%).43 Similarly, a smaller centrations below the site of application.36 A meta-analysis of 26 double-blind, placebo- two-way crossover design assessed the levels of controlled trials by Mason et al., found that subcutaneous absorption and muscle absorp- topical NSAID treatment was safe and effective tion of 800 mg of oral ibuprofen or 16 g of for acute pain, following one week of 5% ibuprofen gel administered to the thigh.41 application.44 Microdialysis probes inserted 25e30 mm into Another meta-analysis of randomized, con- the muscle found average values of 63.5 trolled trials by Lin et al. assessed the efficacy 90.3 ng h mL1 and 213.4 117.2 ng h mL1 of topical NSAIDs relative to placebo or oral of ibuprofen for the topical and oral routes, re- formulations for the treatment of osteoarthri- spectively. The ibuprofen concentrations in tis.45 Compared to placebo, a positive treat- the dermis were 22.5-fold greater when deliv- ment effect was observed only during Weeks ered topically; the mean values of ibuprofen 1 and 2 of treatment; reported pooled treat- in the subcutaneous tissue were 731.2 ment effect sizes for pain relief were 0.41 605.0 ng h mL1 and 176.6 122.9 ng h mL1 (95% confidence interval [CI], 0.16e0.66) for the topical and oral routes, respectively.41 and 0.40 (95% CI, 0.15e0.65) for each week, Another study of 100 patients considered respectively. In contrast, even in the first the ketoprofen concentrations in the intra- week, topical NSAIDs were found inferior to articular tissues following a single application oral versions for improving pain and function, of a 30 mg plaster, multiple applications of and topical agents induced more local side Vol. 33 No. 3 March 2007 Topical Agents for Musculoskeletal Pain 347

effects as well. The authors concluded that no required to reach pain resolution at the site trial evidence has supported the benefit of of injury was significantly shorter for the study NSAIDs over placebo for treating osteoarthritis group (P < 0.0001). The study and control after two weeks of application, and even sug- groups experienced a similar frequency of gested that the current practice guidelines on AEs, most commonly local cutaneous reac- osteoarthritis that advocate the use of topical tions.52 Finally, a placebo-controlled study of e NSAIDs46 48 should be revised.45 However, patients with pain due to ankle sprain assessed two recent, randomized, controlled trials of the analgesia achieved by the application of a topical diclofenac solution for the treatment a 100 mg topical ketoprofen patch over of pain from knee osteoarthritis reported ben- a two-week period.8 There were significantly efit following four and 12 weeks of applica- fewer observations of spontaneous pain for tion.49,50 Measured by Western Ontario and the study group compared to the control McMaster Universities Osteoarthritis Index group during all visits (Days 3e4, Day 7 1, scores, the study group had a 42.9% decrease and Day 14 2). Most notably, there was in pain relative to baseline following four a 49.9 20.2 mm (73%) decrease in pain weeks, and a 45.7% decrease after 12 weeks, from baseline at Day 7 1 for patients given compared to a respective 26.9% and 33.3% de- the ketoprofen patch, compared to a crease in pain by the vehicle-control groups. 37.6 24.3 mm (57%) decrease among the Measurements of physical function, stiffness, patients given a placebo patch. The intergroup and pain on walking indicated similar benefit difference in pain relief was significant of the topical diclofenac over placebo in both (P ¼ 0.0007), but the difference in AEs was studies. Likely due to the skin penetration en- not. Thirty-one percent of the study group and hancer, dimethyl sulfoxide, 30 of the 84 pa- 24% of the control group experienced AEs.8 tients in the four-week study group and 68 of 164 patients in the 12-week study group re- Nitroglycerin (NTG) ported skin irritation (typically dryness), lead- Topically applied NTG has been studied for ing to the discontinuation of five patients in localized treatment of musculoskeletal pain. each of the treatment groups.49,50 NTG may be converted to an anti-inflammatory Various novel NSAID patches or plasters agentdnitric oxide (NO)dthat is released en- have been developed, conferring a more con- dogenously by activated macrophages.53 Treat- stant delivery of a standardized dose of analge- ment-generated NO may modulate the sic (although individual skin variability affects inflammatory process, as well as produce the actual amount absorbed). A study by Galer analgesia through mechanisms directed at noci- et al. used a multicenter, randomized, parallel ceptors, similar to the activity of cholinergic design to assess the efficacy of a topical drugs.53 For example, acetylcholine is a cholin- diclofenac patch for treatment of pain from ergic agent that can induce analgesia by stimu- sports-related soft tissue injuries, such as lating the release of NO, leading to an sprains, strains, or contusions.51 A diclofenac elevation of 3050 cyclic guanidine monophos- epolamine 1.3% or placebo patch was applied phate concentration in nociceptors. twice daily on 222 patients for two weeks. The A randomized, double-blind, placebo- study group achieved statistically significant controlled trial evaluated a 5 mg NTG patch pain intensity differences from placebo during for relief of shoulder pain when applied daily clinic visits on Day 3 (P ¼ 0.036) and Day 14 over a three-day period. While no change in (P # 0.044) following treatment initiation, pain intensity was observed in the control but not on Day 7. Forty percent of participants group, after 48 hours the NTG group reported given placebo reported AEs, while 34% of a significant decrease in pain intensity from study group patients reported side effects.51 baseline, on a 0e10 analog scale (2 0.3; A similar study comparing a diclofenac patch P < 0.003). Two of 10 patients in the treatment to placebo for patients with traumatic blunt group experienced headache.53 Paoloni et al. soft tissue injury found that the analgesic con- examined the use of topical NO for the treat- ferred a significantly beneficial effect over pla- ment of chronic extensor tendinosis in a ran- cebo (P < 0.0001), as measured by tenderness domized, placebo-controlled trial. Compared produced by pressure.52 Also, the time to the placebo group, patients in the NO 348 Stanos Vol. 33 No. 3 March 2007

group reported significantly reduced elbow patch 5% for 100 patients with knee osteoar- pain while active at two weeks (P < 0.05), re- thritis.61 All four composite measures were duced extensor tenderness at six and 12 weeks significantly improved (P < 0.001) following (P < 0.05), and an increase in wrist extensor two weeks of treatment. No treatment-related strength at 24 weeks (P < 0.05).54 AEs were reported in the group given lidocaine patch 5% monotherapy.61 Local Anesthetics Finally, an open-label study of 137 patients Topical anesthetics likely provide pain relief with knee osteoarthritis considered the use of by reducing ectopic discharges from superficial the lidocaine patch 5% as add-on therapy to somatic nerves.13 A number of mechanisms that a stable analgesic regimen (e.g., acetamino- contribute to peripheral sensitization may be phen, NSAIDs, COX-2 inhibitors, tramadol, decreased by topical anesthetics.55 Experiments opioids).62 Following two weeks of treatment that have used complete Freund’s adjuvant to once daily with the lidocaine patch 5%, aver- induce inflammation have shown a correlation age pain intensity scores decreased by 29% between the upregulation of tetrodotoxin- (4.2 2.2) compared to baseline scores resistant, voltage-gated sodium channels and (5.9 1.5; P < 0.001). Also, Brief Pain Inven- an inflammatory state.56,57 As sodium channels tory measurements of pain interference on play a fundamental role in the excitability of quality of life decreased from 37.2 13.7 at neurons, alterations in their expression levels baseline to 23.5 15.1 at Week 2 (P < 0.001). or function can lead to the neuronal hyperexcit- Fourteen patients experienced treatment- ability observed in many types of chronic pain related side effects that included headache conditions.58 Additionally, pharmacotherapies (n ¼ 4), dermatitis (n ¼ 3), and taste distur- which block sodium channels have been shown bance (n ¼ 2). Although there were no serious to provide effective analgesia for the manage- AEs reported, five patients discontinued ment of some types of chronic pain.58 treatment.62 The lidocaine patch 5% likely reduces the Open-label trial evidence has suggested that ectopic transmission of pain signals to the dor- the lidocaine patch 5% may be beneficial for sal horn of the spinal cord by inhibiting abnor- patients experiencing LBP as well. A six-week, mal sodium channels in dermal nociceptors nonrandomized trial conducted by Gimbel residing in the area of the localized pain.59 et al. considered the patch’s effectiveness for The current basic mechanistic understanding acute/subacute (<3 months, n ¼ 21), short- of osteoarthritis suggests that pain is the term chronic (3e12 months, n ¼ 33), or result of a chronic inflammatory state in in- long-term chronic (>12 months, n ¼ 77) tra-articular joints. Blocking upregulated LBP.63 Compared to baseline, significant re- sodium channels in inflamed tissues from in- ductions in pain intensity were reported by pa- ducing aberrant transmissions may provide tients within all three categories, at both two analgesia.60 A two-week pilot study tested the and six weeks (P # 0.007). Overall, 25 patients ability of the lidocaine patch 5% to provide an- reported treatment-related AEs; 20 of those algesia for 20 patients with knee osteoarthritis. AEs were mild-to-moderate in severity.63 A sim- At the completion of the trial, statistically sig- ilarly designed study of 71 patients also found nificant improvements were observed for all significant improvement measured by the Neu- Western Ontario and McMaster Universities ropathic Pain Scale composite at Weeks 2 and subscale scores of pain, as well as the compos- 6(P < 0.001).1 Eleven patients reported der- ite index (P < 0.01). Also, 84% of the patients mal side effects, while five patients reported reported the achievement of moderate- systemic side effects (dizziness and nausea).1 to-complete global pain relief, while 90% of the patients had no observed erythema. How- Topical Counterirritants ever, due to the lack of controls and small sam- Counterirritants, such as capsaicin, cam- ple size, further randomized, controlled trials phor, menthol, and garlic, are a category of were recommended.59 analgesics that excite and subsequently desen- Another open-label study used the Neuro- sitize nociceptive sensory neurons.64 Although pathic Pain Scale to assess the analgesia many of the members of this group have had achieved by the application of the lidocaine a long history of common medical use, it was Vol. 33 No. 3 March 2007 Topical Agents for Musculoskeletal Pain 349

not until recently that their molecular mecha- applied topically, capsaicin has a biphasic phar- nisms of action were elucidated. All of these macologic effect on sensory neurons. The first pungent plant derivatives act on the recently phase is characterized by the excitation and ac- elucidated transient receptor potential (TRP) tivation of TRPV1, while the second phase pro- superfamily, a group of structurally similar duces analgesia from the depletion of ion channels, such as TRPV1 (also called vanil- pronociceptive transmitters, including SP.70 loid receptor subtype 1), TRPV3, TRPM8, and There is significant pain, burning, itching, TRPA1.65 The superfamily is activated by and cutaneous vasodilatation experienced capsaicin, camphor, menthol, and garlic, upon initial application due to excitation and respectively64 (Fig. 2). These thermosensitive sensitization of cutaneous C- and A-fiber noci- receptors detect a wide range of temperatures ceptors; however, the repeated application of ranging from noxious heat to extreme cold, as capsaicin leads to the persistent desensitiza- well as other stimuli, including acidic pH, tion of nociceptors during Phase 2.71 lipids, changes in extracellular osmolarity A systematic review of topical capsaicin for and/or pressure, and depletion of intracellu- the relief of musculoskeletal pain pooled the lar Ca2þ stores (Fig. 3).6,66 These proteins are results of three double-blind, placebo- expressed in primary sensory neurons as well controlled trials, summing up to 368 patients as other tissues. Upon the activation of TRP re- in total.72 After four weeks of treatment with ceptors, the release of calcitonin gene-related capsaicin 0.025% or plaster, the mean re- peptide, SP, and other inflammatory neuro- sponse rate (the percentage of patients with transmitters are induced; hence, producing at least 50% pain relief) was 38% (range, local irritation and inflammation.67 This can 34%e42%), while the placebo response rate lead to two types of desensitization: 1) acute was 25% (range, 17%e37%). The NNT was or ‘‘pharmacologic’’ desensitization character- 8.1 and approximately one-third of the pa- ized by a diminished response during a constant tients experienced local, treatment-related agonist application, and 2) over a longer pe- AEs.72 An older meta-analysis also reported riod, tachyphylaxis or ‘‘functional’’ desensitiza- that capsaicin cream provided better pain re- tion characterized by a reduction in response lief for osteoarthritis than placebo (odds ra- after many stimulations.64 tio ¼ 4.36; 95% CI ¼ 2.77-6.88).73 However, As early as the 19th century, the selective ef- products with low concentrations of capsaicin fects of capsaicin on sensory nerve fibers were require multiple applications to provoke the recognized.6 The spicy ingredient in chili pep- desensitization of nerves,68 and often contam- pers has been used to relieve neuropathic inate the patient’s personal surroundings pain, uremic pruritus, and bladder overactivity, (bed, contact lenses, etc.).6 Those hindrances, as well as for providing analgesia.68 Currently, along with the side effect of burning and pain nonprescription creams, lotions, and patches upon application, reduce patient adherence containing 0.025%e0.075% capsaicin by and concomitantly, the efficacy of the treat- weight for treatment of musculoskeletal and ment.6,68 Yet due to the reversible desensitiza- neuropathic pains are available.69 When tion of nociceptive C-fibers and lack of

Fig. 2. TRP family receptors and their thermosensitivity (adapted from Ref. 69). 350 Stanos Vol. 33 No. 3 March 2007

O + H O N H

HO Acidosis Endogenous/exogenous Temperature pH<6 agonists: >43ºC anandamide , 12- hydroxyeicosatetraenoic acid, N-oleoyldopamine, capsaicin, resiniferatoxin, Ca2+ olvanil, DA-5016, SDZ 249- 665, civamide, and Na+ camphor Brain: burning pain sensation

Cell membrane

Sensory neuron

Peripheral nerve

Action potential

2+ Intracellular Ca signal and depolarization Spinal cord

Fig. 3. Acute activation of the TRPV1 receptor leads to local intracellular calcium signals and the sensation of heat or burning pain. TRPV-transient receptor potential vanilloid. Reprinted with permission from Ref. 6. systemic side effects, topical capsaicin was rec- The analgesic capacity of the compounds in ommended by the European League Against garlic (allicin and diallyl disulfide) have not Rheumatism (EULAR) in 2003 for the treat- been well-explored, although they have been ment of knee osteoarthritis.47 often used to treat hypertension, high blood The potent analog of capsaicin, resinifera- cholesterol, and thrombosis.67 Allicin and toxin (RTX), is an extract from the resin of diallyl disulfide both bear a structural resem- the Euphorbia cactus.6 RTX induces a slower de- blance to the components in mustard plants polarization and influx of Ca2þ into C-fibers74 which induce pain and inflammationdisothio- and induces less pain upon application.68 De- cyanates. Likewise, recent research has indi- spite some promising clinical trial results for cated that mustard and garlic depolarize treating detrusor hyperreflexia and bladder sensory neurons specifically through activation hyperreflexia,68 RTX has not been approved of the TRPA1 receptor to elicit inflammatory for use within the United States. pain.67,75 Camphor is derived from the wood of the In contrast, mentholdthe component that camphor laurel tree (Cinnamomum camphora) confers mint smell and flavor to the Mentha and historically, the sweet-smelling compound speciesdis often included in eutectic formula- has had many medicinal uses, including as a de- tions of local anesthetic agents.76 Anecdotally, congestant, cough suppressant, and antipruritic menthol induces tingling and cooling sensa- agent.64 Over-the-counter camphor-containing tions when applied topically. Menthol gener- balms have also been used to provide analgesia. ates analgesia through its Ca2þ channel Recent research has implicated three receptors blocking activity. In addition to binding in camphor’s mechanism of action, TRPV3, the TRPM8,65,69 menthol binds k-opioid receptors, capsaicin receptordTRPV1, and the garlic and possibly engenders analgesia through receptordTRPA1.64 both mechanisms.76 Animal studies have Vol. 33 No. 3 March 2007 Topical Agents for Musculoskeletal Pain 351

demonstrated that menthol can confer analge- recommended by practice guidelines in both sia following both noxious thermal and chem- the United States and the United Kingdom ical stimuli.76 Furthermore, similar to other for management of acute LBP.80 A random- terpenes, menthol is an effective topical per- ized, single-blind clinical trial compared meation enhancer for water-soluble drugs, a group of 113 patients treated with a 40C such as the antidepressant, imipramine.77 heat wrap for 8 hours per day to patients given Another type of counterirritant is topical ru- one of the two most common oral, nonpre- befacients containing salicylates. Although, it scription drugs in the United States: acetamino- has been postulated that analgesia is conferred phen (4,000 mg/ by a mode different than that of NSAIDs, as day; n ¼ 113) or ibuprofen (1,200 mg/day; yet, salicylates have an unidentified mecha- n ¼ 106).80 Significantly better mean pain relief nism of action.78 Despite this, salicylates are was obtained by the heat-wrap therapy group often found in many topical preparations. on Day 1, and then again by Days 3e4(extended Randomized, clinical trial data describing the mean pain relief ¼ 2.61, compared to 1.68 for efficacy of salicylates for pain relief have been ibuprofen, P ¼ 0.0001 and 1.95 for acetamino- systematically reviewed by Mason et al.78 Only phen, P ¼ 0.0009). Furthermore, lateral trunk three double-blind, placebo-controlled trials mobility and disability significantly improved have been published that consider the use of for the heat-wrap therapy group relative to the topical salicylates for the treatment of acute comparators.80 Another randomized, controlled musculoskeletal pain. The study groups ex- trial also found better functional outcomes for hibited significantly better pain reductions patients treated with low-level heat wraps than than placebo (relative benefit ¼ 3.6, 95% CI, for the control group.81 Combining heat wraps 2.4e5.6; NNT ¼ 2.1, range, 1.7e2.8). Al- with directional, preference-based exercise fur- though the long-term efficacy data and the ther reduced disability scores after a week of reporting of AEs were poor for chronic muscu- treatment to nearly half those of either therapy loskeletal pain, the information from six dou- alone.81 ble blind, placebo-controlled trials indicated For thousands of years, medicinal leeches a relative benefit over control of 1.5 (range, have been used for the treatment of various 1.3e1.9; NNT 5.3, range, 3.6e10.2).78 diseases, and more recently have offered Other counterirritants often found in over- a novel option for the treatment of venous the-counter analgesic preparations include congestion in plastic surgery82,83 and in marsh tea, peppermint oil, and poison ivy. A chronic musculoskeletal disorders.84 Leeches, randomized, controlled trial explored the use Hirudo medicinalis, purportedly have biologi- of a topical homeopathic gel that included cally active, anti-inflammatory and anticoagu- the ingredients, Symphytum officinale (comfrey), lant mediators within their saliva.85 In Rhus toxicodendron (poison ivy), and Ledum pal- a recent study, 24 patients with osteoarthritis ustre (marsh tea) for the management of knee of the knee were locally administered four to osteoarthritis pain.79 The 86 patients that com- six leeches for approximately 70 minutes, pleted the trial in the homeopathy group and compared with 27 patients treated with reported a reduction in pain of 16.5 mm by a 28-day regimen of topical diclofenac gel visual analogue scale, compared to an (300 mg, twice a day).84 The mean reduction 8.1 mm pain reduction by the 86 patients in in pain observed within the leech therapy the piroxicam group. The difference between group was 53.5 13.7 at baseline and the treatment groups was 8.4 mm (95% CI, 19.3 12.2 at Day 7. Compared with the diclo- 0.8e15.9). Twelve patients within the homeop- fenac group (51.5 16.8 at baseline and athy group and 16 patients within the NSAID 42.4 19.27 at Day 7), there was a significant group experienced an AE, causing five and between-group difference (23.9; CI, 32.8 nine patients, respectively, to withdraw from to 15.1; P < 0.001). However, 17 of the 24 pa- the study.79 tients administered leeches experienced local itching, while none of the patients given topi- Other Agents cal NSAIDs reported pruritus.84 Heat wraps are a popular therapy for back Although the evidence-based literature indi- pain, and the treatment has been cates that systemic opioids are effective at 352 Stanos Vol. 33 No. 3 March 2007

providing analgesia for a variety of painful con- of topical delivery systems, as new ones are de- ditions, topical preparations of opioids have veloped and others optimized, will be impor- not been well-explored, with the exception of tant to providing continuing best practice for some studies of relief from painful skin ul- patients requiring diverse, multimodal options cers.86,87 Potentially, the local application of for managing their pain. opioids could minimize systemic side effects, such as sedation, respiratory depression, and 88 nausea. A 2005 study followed three patients References administered topical morphine in PLO gel for relief of chronic arthritis pain.89 A satisfactory 1. Galer BS, Gammaitoni AR, Oleka N, Jensen MP, Argoff CE. Use of the lidocaine patch 5% in reduc- level of pain control was reported, although ing intensity of various pain qualities reported by likely due to the systemic absorption of the patients with low-back pain. Curr Med Res Opin 89 opioid, as indicated by urinalysis results. Fur- 2004;20(Suppl 2):S5eS12. thermore, an animal study indicated potential 2. Luo X, Pietrobon R, Sun SX, Liu GG, Hey L. synergy between topical lidocaine and topical Estimates and patterns of direct health care opioids.88 Possibly, combining different drug expenditures among individuals with back pain in classes to create a topical preparation that tar- the United States. Spine 2004;29(1):79e86. gets multiple peripheral sites may enhance the 3. Woolf AD, Pfleger B. Burden of major musculo- analgesia obtained. Preliminary results in an skeletal conditions. Bull World Health Organ 2003; animal model suggest that combining topical 81(9):646e656. Epub 2003, Nov 2014. morphine with topical cannabinoid may im- 4. Walker BF. The prevalence of low back pain: prove the antinociceptive effects that are a systematic review of the literature from 1966 to e achieved, while reducing central nervous sys- 1998. J Spinal Disord 2000;13(3):205 217. tem exposure to opioid.90 Other precursory 5. Moskowitz RW, Holderbaum D. Clinical and lab- results suggest that topical cannabinoid prepa- oratory findings in osteoarthritis. In: Koopman WJ, rations may buffer emerging pain signals at pe- ed. Arthritis and allied conditions, 14th ed. Phila- delphia: Lippincott Williams & Wilkins, 2001: ripheral sites via CB1 receptors, while avoiding 2216e2243. the dysphoric side effects and abuse potential of central-acting cannabinoid agents.91 This 6. Bley KR. Recent developments in transient receptor potential vanilloid receptor 1 agonist-based promising research could lead to the develop- therapies. Expert Opin Investig Drugs 2004; ment of another agent in the armamentarium 13(11):1445e1456. of topical analgesics. 7. Bridgman SA, Clement D, Downing A, et al. Population based epidemiology of ankle sprains at- tending accident and emergency units in the West Midlands of England, and a survey of UK practice Conclusion for severe ankle sprains. Emerg Med J 2003;20(6): e Recent research has significantly evolved our 508 510. understanding of the diverse mechanisms by 8. Mazieres B, Rouanet S, Velicy J, Scarsi C, which topical preparations induce analgesia. Reiner V. Topical ketoprofen patch (100 mg) for the treatment of ankle sprain: a randomized, Many topical agents have significantly less sys- double-blind, placebo-controlled study. Am J Sports temic AEs than oral pharmacotherapies, while Med 2005;33(4):515e523. providing effective pain relief. In particular, 9. Basbaum AI, Jessell TM. The perception of topical NSAIDs have demonstrated benefit pain. In: Kandel ER, Schwartz JH, Jessell TM, eds. for treating pain from acute and chronic mus- Principles of neural science, 4th ed. New York: culoskeletal conditions, while topical lidocaine The McGraw-Hill Companies, 2000: 472e479. has provided effective peripheral analgesia for 10. Millan MJ. The induction of pain: an integrative localized pain associated with joint and low review. Prog Neurobiol 1999;57(1):1e164. back ailments. Also, clinicians should become 11. Kidd BL, Urban LA. Mechanisms of inflamma- familiar with over-the-counter topical prepara- tory pain. Br J Anaesth 2001;87(1):3e11. tions for analgesia, as they are often used and 12. Grosser T, Fries S, FitzGerald GA. Biological ba- can contain ingredients with analgesic mecha- sis for the cardiovascular consequences of COX-2 in- nisms supported by evidence in the literature. hibition: therapeutic challenges and opportunities. Furthermore, maintaining an understanding J Clin Invest 2006;116(1):4e15. Vol. 33 No. 3 March 2007 Topical Agents for Musculoskeletal Pain 353

13. Galer BS. Topical medications. In: Loeser JD, 27. Southwell D, Barry BW, Woodford R. Variations Butler SH, Chapman CR, Turk DC, eds. Bonica’s in permeability of human skin within and between management of pain, 3rd ed. Philadelphia: Lippin- specimens. Int J Pharm 1984;18:299e309. e cott Williams & Wilkins, 2001: 1736 1742. 28. Larsen RH, Nielsen F, Sorensen JA, Nielsen JB. 14. Brown MB, Martin GP, Jones SA, Akomeah FK. Dermal penetration of fentanyl: inter- and intraindi- Dermal and transdermal drug delivery systems: cur- vidual variations. Pharmacol Toxicol 2003;93(5): rent and future prospects. Drug Deliv 2006;13(3): 244e248. e 175 187. 29. Steinstrasser I, Merkle HP. Dermal metabolism 15. Cleary GW. Transdermal delivery systems; a med- of topically applied drugs: pathways and models re- ical rationale. In: Shah VP, Maibach HI, eds. Topical considered. Pharm Acta Helv 1995;70(1):3e24. drug bioavailability, bioequivalence, and penetra- 30. Hogan DJ, Maibach HI. Adverse dermatologic e tion. New York: Plenum Press, 1993: 17 68. reactions to transdermal drug delivery systems. 16. Henzl MR, Loomba PK. Transdermal delivery of J Am Acad Dermatol 1990;22(5 Pt 1):811e814. sex steroids for hormone replacement therapy and 31. Carmichael AJ. Skin sensitivity and transdermal contraception. A review of principles and practice. drug delivery. A review of the problem. Drug Saf e J Reprod Med 2003;48(7):525 540. 1994;10(2):151e159. 17. Kornick CA, Santiago-Palma J, Moryl N, 32. Murphy M, Carmichael AJ. Transdermal drug Payne R, Obbens EA. Benefit-risk assessment of delivery systems and skin sensitivity reactions. Inci- transdermal fentanyl for the treatment of chronic dence and management. Am J Clin Dermatol 2000; pain. Drug Saf 2003;26(13):951e973. 1(6):361e368. 18. Cramer MP, Saks SR. Translating safety, efficacy 33. Datamonitor. United Statesdanalgesics. New and compliance into economic value for controlled York, NY. Reference code 72e751. Available from. release dosage forms. Pharmacoeconomics 1994; www.datamonitor.com. Accessed January 28, 2007. 5(6):482e504. 34. Ness TJ, Jones L, Smith H. Use of compounded 19. Varvel JR, Shafer SL, Hwang SS, Coen PA, topical analgesicsdresults of an Internet survey. Stanski DR. Absorption characteristics of transder- Reg Anesth Pain Med 2002;27(3):309e312. mally administered fentanyl. Anesthesiology 1989; 35. Vaile JH, Davis P. Topical NSAIDs for musculo- 70(6):928e934. skeletal conditions. A review of the literature. Drugs 20. Yang SI, Park HY, Lee SH, et al. Transdermal 1998;56(5):783e799. eperisone elicits more potent and longer-lasting 36. Heyneman CA, Lawless-Liday C, Wall GC. Oral muscle relaxation than oral eperisone. Pharmacol- e versus topical NSAIDs in rheumatic diseases: a ogy 2004;71(3):150 156. comparison. Drugs 2000;60(3):555e574. 21. Payne R, Mathias SD, Pasta DJ, et al. Quality of 37. Kumar R, Katare OP. Lecithin organogels as life and cancer pain: satisfaction and side effects a potential phospholipid-structured system for topi- with transdermal fentanyl versus oral morphine. e cal drug delivery: a review. AAPS PharmSciTech J Clin Oncol 1998;16(4):1588 1593. 2005;6(2):E298eE310. 22. Jarupanich T, Lamlertkittikul S, Chandeying V. 38. Franckum J, Ramsay D, Das NG, Das SK. Efficacy, safety and acceptability of a seven-day, Pluronic lecithin organogel for local delivery of transdermal estradiol patch for estrogen replace- anti-inflammatory drugs. Int J Pharmaceutical ment therapy. J Med Assoc Thai 2003;86(9): Compounding 2004;8(2):101e105. 836e845. 39. Yamane MA, Williams AC, Barry BW. Terpene 23. Archer DF, Cullins V, Creasy GW, Fisher AC. penetration enhancers in propylene glycol/water The impact of improved compliance with a weekly co-solvent systems: effectiveness and mechanism of contraceptive transdermal system (Ortho Evra) on action. J Pharm Pharmacol 1995;47(12A):978e989. contraceptive efficacy. Contraception 2004;69(3): 189e195. 40. Assandri A, Canali S, Giachetti C. Local tolera- bility and pharmacokinetic profile of a new 24. Frei A, Andersen S, Hole P, Jensen NH. A one transdermal delivery system, diclofenac hydroxye- year health economic model comparing transder- thylpyrrolidine plaster. Drugs Exp Clin Res 1993; mal fentanyl with sustained-release morphine in 19(3):89e95. the treatment of chronic noncancer pain. J Pain Palliat Care Pharmacother 2003;17(2):5e26. 41. Tegeder I, Muth-Selbach U, Lotsch J, et al. Ap- plication of microdialysis for the determination of 25. Bos JD, Meinardi MM. The 500 Dalton rule for muscle and subcutaneous tissue concentrations af- the skin penetration of chemical compounds and ter oral and topical ibuprofen administration. Clin drugs. Exp Dermatol 2000;9(3):165e169. Pharmacol Ther 1999;65(4):357e368. 26. Yano T, Nakagawa A, Tsuji M, Noda K. Skin per- 42. Rolf C, Engstrom B, Beauchard C, Jacobs LD, meability of various non-steroidal anti-inflammatory Le Liboux A. Intra-articular absorption and distribu- drugs in man. Life Sci 1986;39(12):1043e1050. tion of ketoprofen after topical plaster application 354 Stanos Vol. 33 No. 3 March 2007

and oral intake in 100 patients undergoing knee treatment of chronic extensor tendinosis at the arthroscopy. Rheumatology (Oxford) 1999;38(6): elbow: a randomized, double-blinded, placebo- 564e567. controlled clinical trial. Am J Sports Med 2003; 31(6):915e920. 43. Moore RA, Tramer MR, Carroll D, Wiffen PJ, McQuay HJ. Quantitative systematic review of topi- 55. Jensen TS, Gottrup H, Kasch H, et al. Has basic cally applied non-steroidal anti-inflammatory drugs. research contributed to chronic pain treatment? Br Med J 1998;316(7128):333e338. Acta Anaesthesiol Scand 2001;45(9):1128e1135. 44. Mason L, Moore RA, Edwards JE, Derry S, 56. Gould HJ 3rd, England JD, Liu ZP, Levinson SR. McQuay HJ. Topical NSAIDs for acute pain: a meta- Rapid sodium channel augmentation in response to analysis. BMC Fam Pract 2004;5:10. inflammation induced by complete Freund’s adju- 45. Lin J, Zhang W, Jones A, Doherty M. Efficacy of vant. Brain Res 1998;802(1e2):69e74. topical non-steroidal anti-inflammatory drugs in the treatment of osteoarthritis: meta-analysis of rando- 57. Gould HJ 3rd, Gould TN, Paul D, et al. Develop- mised controlled trials. Br Med J 2004;329(7461): ment of inflammatory hypersensitivity and augmen- 324. tation of sodium channels in rat dorsal root ganglia. Brain Res 1999;824(2):296e299. 46. Scott DL, Shipley M, Dawson A, et al. The clinical management of rheumatoid arthritis and osteo- 58. Lai J, Porreca F, Hunter JC, Gold MS. Voltage-- arthritis: strategies for improving clinical gated sodium channels and hyperalgesia. Annu e effectiveness. Br J Rheumatol 1998;37(5):546e554. Rev Pharmacol Toxicol 2004;44:371 397. 47. Jordan KM, Arden NK, Doherty M, et al. 59. Galer BS, Sheldon E, Patel N, et al. Topical lido- EULAR recommendations 2003: an evidence based caine patch 5% may target a novel underlying pain approach to the management of knee osteoarthritis. mechanism in osteoarthritis. Curr Med Res Opin Report of a Task Force of the Standing Committee 2004;20(9):1455e1458. for International Clinical Studies Including Thera- peutic Trials (ESCISIT). Ann Rheum Dis 2003; 60. Amir R, Argoff CE, Bennett GJ, et al. The role of 62(12):1145e1155. sodium channels in chronic inflammatory and neuropathic pain. J Pain 2006;7(5 Suppl 3):S1eS29. 48. American College of Rheumatology Subcom- mittee on Osteoarthritis Guidelines. Recommenda- 61. Gammaitoni AR, Galer BS, Onawola R, tions for the medical management of Jensen MP, Argoff CE. Lidocaine patch 5% and its osteoarthritis of the hip and knee: 2000 update. positive impact on pain qualities in osteoarthritis: Arthritis Rheum 2000;43(9):1905e1915. results of a pilot 2-week, open-label study using the Neuropathic Pain Scale. Curr Med Res Opin 49. Bookman AA, Williams KS, Shainhouse JZ. Ef- 2004;20(Suppl 2):S13eS19. fect of a topical diclofenac solution for relieving symptoms of primary osteoarthritis of the knee: 62. Burch F, Codding C, Patel N, Sheldon E. Lido- a randomized controlled trial. CMAJ 2004;171(4): caine patch 5% improves pain, stiffness, and physi- 333e338. cal function in osteoarthritis pain patients. A prospective, multicenter, open-label effectiveness 50. Roth SH, Shainhouse JZ. Efficacy and safety of e a topical diclofenac solution (pennsaid) in the treat- trial. Osteoarthr Cartil 2004;12(3):253 255. ment of primary osteoarthritis of the knee: a ran- 63. Gimbel J, Linn R, Hale M, Nicholson B. Lido- domized, double-blind, vehicle-controlled clinical caine patch treatment in patients with low back e trial. Arch Intern Med 2004;164(18):2017 2023. pain: results of an open-label, nonrandomized pilot e 51. Galer BS, Rowbotham M, Perander J, Devers A, study. Am J Ther 2005;12(4):311 319. Friedman E. Topical diclofenac patch relieves mi- 64. Xu H, Blair NT, Clapham DE. Camphor activates nor sports injury pain: results of a multicenter con- and strongly desensitizes the transient receptor trolled clinical trial. J Pain Symptom Manage 2000; e potential vanilloid subtype 1 channel in a vanilloid- 19(4):287 294. independent mechanism. J Neurosci 2005;25(39): 52. Predel HG, Koll R, Pabst H, et al. Diclofenac 8924e8937. patch for topical treatment of acute impact injuries: 65. Tominaga M, Caterina MJ. Thermosensation a randomised, double blind, placebo controlled, e multicentre study. Br J Sports Med 2004;38(3): and pain. J Neurobiol 2004;61(1):3 12. e 318 323. 66. Gunthorpe MJ, Benham CD, Randall A, 53. Berrazueta JR, Losada A, Poveda J, et al. Suc- Davis JB. The diversity in the vanilloid (TRPV) re- cessful treatment of shoulder pain syndrome due ceptor family of ion channels. Trends Pharmacol to supraspinatus tendinitis with transdermal nitro- Sci 2002;23(4):183e191. glycerin. A double blind study. Pain 1996;66(1): 67. Bautista DM, Movahed P, Hinman A, et al. Pun- 63e67. gent products from garlic activate the sensory ion 54. Paoloni JA, Appleyard RC, Nelson J, channel TRPA1. Proc Natl Acad Sci USA 2005; Murrell GA. Topical nitric oxide application in the 102(34):12248e12252. Vol. 33 No. 3 March 2007 Topical Agents for Musculoskeletal Pain 355

68. Szallasi A. Vanilloid (capsaicin) receptors in 80. Nadler SF, Steiner DJ, Erasala GN, et al. Contin- health and disease. Am J Clin Pathol 2002;118(1): uous low-level heat wrap therapy provides more effi- 110e121. cacy than ibuprofen and acetaminophen for acute e 69. McKemy DD, Neuhausser WM, Julius D. Identi- low back pain. Spine 2002;27(10):1012 1017. fication of a cold receptor reveals a general role for 81. Mayer JM, Ralph L, Look M, et al. Treating TRP channels in thermosensation. Nature 2002; acute low back pain with continuous low-level heat 416(6876):52e58. wrap therapy and/or exercise: a randomized con- e 70. Szallasi A, Blumberg PM. Vanilloid (capsaicin) trolled trial. Spine J 2005;5(4):395 403. receptors and mechanisms. Pharmacol Rev 1999; 82. Daane S, Zamora S, Rockwell WB. Clinical use 51(2):159e212. of leeches in reconstructive surgery. Am J Orthop e 71. Carpenter SE, Lynn B. Vascular and sensory re- 1997;26(8):528 532. sponses of human skin to mild injury after topical 83. Gideroglu K, Yildirim S, Akan M, Akoz T. Imme- treatment with capsaicin. Br J Pharmacol 1981; diate use of medicinal leeches to salvage venous 73(3):755e758. congested reverse pedicled neurocutaneous flaps. Scand J Plast Reconstr Surg Hand Surg 2003; 72. Mason L, Moore RA, Derry S, Edwards JE, e McQuay HJ. Systematic review of topical capsaicin 37(5):277 282. for the treatment of chronic pain. Br Med J 2004; 84. Michalsen A, Klotz S, Ludtke R, et al. Effective- 328(7446):991. ness of leech therapy in osteoarthritis of the knee: a randomized, controlled trial. Ann Intern Med 73. Zhang WY, Li Wan Po A. The effectiveness of e topically applied capsaicin. A meta-analysis. Eur J 2003;139(9):724 730. Clin Pharmacol 1994;46(6):517e522. 85. Markwardt F. Pharmacology of hirudin: one 74. Robbins W. Clinical applications of capsaici- hundred years after the first report of the anticoag- noids. Clin J Pain 2000;16(2 Suppl):S86eS89. ulant agent in medicinal leeches. Biomed Biochim Acta 1985;44(7e8):1007e1013. 75. Bautista DM, Jordt SE, Nikai T, et al. TRPA1 mediates the inflammatory actions of environmen- 86. Porzio G, Aielli F, Verna L, et al. Topical morphine in the treatment of painful ulcers. J Pain tal irritants and proalgesic agents. Cell 2006; e 124(6):1269e1282. Symptom Manage 2005;30(4):304 305. 76. Galeotti N, Di Cesare Mannelli L, Mazzanti G, 87. Gallagher RE, Arndt DR, Hunt KL. Analgesic effects of topical methadone: a report of four cases. Bartolini A, Ghelardini C. Menthol: a natural anal- e gesic compound. Neurosci Lett 2002;322(3): Clin J Pain 2005;21(2):190 192. 145e148. 88. Kolesnikov YA, Chereshnev I, Pasternak GW. 77. Jain AK, Thomas NS, Panchagnula R. Transder- Analgesic synergy between topical lidocaine and topical opioids. J Pharmacol Exp Ther 2000; mal drug delivery of imipramine hydrochloride. I. e Effect of terpenes. J Control Release 2002; 295(2):546 551. 79(1e3):93e101. 89. Wilken M, Ineck JR, Rule AM. Chronic arthritis 78. Mason L, Moore RA, Edwards JE, et al. System- pain management with topical morphine: case se- ries. J Pain Palliat Care Pharmacother 2005;19(4): atic review of efficacy of topical rubefacients con- e taining salicylates for the treatment of acute and 39 44. chronic pain. Br Med J 2004;328(7446):995. 90. Yesilyurt O, Dogrul A, Gul H, et al. Topical cannabinoid enhances topical morphine antinocicep- 79. van Haselen RA, Fisher PA. A randomized e e controlled trial comparing topical piroxicam gel tion. Pain 2003;105(1 2):303 308. with a homeopathic gel in osteoarthritis of the 91. Dogrul A, Gul H, Akar A, et al. Topical cannabi- knee. Rheumatology (Oxford) 2000;39(7): noid antinociception: synergy with spinal sites. Pain 714e719. 2003;105(1e2):11e16.