INTRODUCTION
There are 20-24 million prescriptions (predominantly oral) for non-steroidal anti-inflammatory drugs (NSAIDs) in the United Kingdom each year, 5% of all NHS prescriptions. (Moore et al 1998). Moreover, £30million pounds are spent every year on prescribing topical NSAIDs in England. (NPC 1997).
Topical non-steroidal anti-inflammatory drugs (NSAIDs) are promoted on the basis that they diffuse directly into the target tissue. The advantages of this are to benefit from the local delivery without the side effects that oral NSAIDs can produce. (MeReC 1997). However, recent research has shown that this may not be the case.
Recent studies, suggest that contrary to promotional claims, topical NSAIDs enter synovial joints via the systemic circulation and not directly into the target tissue. (NPC 1997). Indeed, some smaller clinical trails have shown equal amounts of topical NSAID concentrations in both knees after application, even though only a single knee was treated. However, Anon (1991) reports that the plasma concentrations of this are likely to be substantially lower than oral NSAID therapy. However, there is some dispute, as some studies have identified that topical NSAIDs do act directly into skin, muscle and subcutaneous fat rather than systemically. (Grahame 1995)
DO NSAIDs WORK?
Following an injury, inflammation occurs at the site of the injury. Prostoglandins and leukotrienes are produced in response to the inflammation. Prostoglandins i.e PGE-2 cause blood vessels to expand whilst leukotrienes i.e LTB-4 cause white blood cells to infiltrate the traumatised and inflamed area. PGE-2 encourages the white blood cells to release enzymes resulting in further pain and inflammation.
Topical NSAIDs act to inhibit the inflammatory process by preventing synthesis of both PGE-2 and LTB-4, thus reducing both inflammation and pain at the site of injury. (NHS Journal 2002)
Many trials have been reported into the efficacy of topical NSAIDs. However, the reliability and validity of some of these studies are questionable as are the interpretation of the results.
Several trials have compared topical NSAIDs with placebo, but all have limitations. A marked placebo effect has been estimated in 40%-60% of 20 randomised controlled trials involving topical NSAIDs. (MeReC 1997). This in effect is a high percentage, which could reflect the benefit of the rubbing action itself, or indeed the power of the practitioner. (NPC 1997). However, the trials also suggest that an average 70% of patients respond to active topical NSAIDs.
Unfortunately, most of these studies used low subject numbers, poorly designed methodologies, selective randomisation, short-term trials and subjective rating scales to measure the efficacy of the topical NSAID. These obviously need to be taken into consideration before reaching conclusions.
THE EVIDENCE
There are limited reliable published papers on the effectiveness of topical NSAIDs on soft tissue injuries.
Topical Ketoprofen 2.5% gel was compared to placebo in 56 patients with unspecified soft tissue injuries over seven days. 83% of the patients responded well to the topical NSAID compared with 51% in the placebo group. However, these results were not statistically significant. (Airaksinen et al 1993).
Topical Piroxcam 0.5% gel produced an overall improvement compared to placebo in a two-week period on 246 patients. (Kageyama 1987). However, the report also suggests a large placebo effect in the study.
A similar study on Piroxicam 0.5% gel in 200 patients found no significant difference in its effectiveness compared to placebo. (Russell 1991)
Topical Ibuprofen 5% cream was not significantly more effective than placebo in 80 patients with sports injuries. Both groups of patients improved rapidly, but it was noted that physicians did favour the Ibuprofen group. There was no discussion as to the reason. (Ramesh 1983)
Topical Felbinac 3% gel has been shown to be significantly more effective than placebo in 231 patients with soft tissue injuries. 73% responded well in the Felbinac group, and 41% in the placebo group. (MeRec 1997)
A comprehensive review by Moore et al (1998) reviewed the effectiveness of selected topical NSAIDs in 86 trials involving 10,160 patients with acute and chronic soft tissue trauma. Utilising and reviewing a series of randomised and placebo controlled studies, they concluded topical NSAIDs to be effective in the treatment of soft tissue injuries compared to placebo. They found Ketoprofen (number needed to treat 2.6), Felbinac (3.0), Ibuprofen (3.5), and Piroxicam (4.2) had significant efficacy. However, Duerden et al (1998) questions the methodology of this review suggesting fundamental flaws such as low subject numbers and the use of short-term studies were used.
Some authors have also questioned the way we compare the use of topical NSAIDs against placebo, suggesting that a more valid comparison would be to compare a topical NSAID with another topical NSAID. (Duerden 1998). Clinically, this would make more sense and be more valid in research. The other consideration to take into account is publication bias, and noticeably, most research is conducted or funded by the pharmaceutical companies themselves.
Reliable evidence on the efficacy of topical NSAIDs for the use in Osteoarthritic/chronic degenerative joints also remains limited. Only a few research papers were available, the majority were not placebo controlled or double blinded and most had fundamental flaws in the methodology.
Overall, the quality of evidence comparing topical NSAIDs to placebo is poor, and more reliable research is needed to confirm their effectiveness. Similarly, the high level of prescribing of topical NSAIDs is not justified by the available evidence according to MeReC Bulletin (1997). The bulletin further recommends the use of interventions such as rest, ice, compression and elevation with simple analgesics such as paracetamol should be considered first. They even suggest taken oral Ibuprofen (if tolerated) or oral rubafacient should be considered, as it is more cost effective
ARE TOPICAL NSAIDs SAFE?
Overwhelming evidence supports the safety of topical NSAIDs in small doses. Very few side effects have been reported in literature with local skin reactions being the most common, approx 1.5%. (MeRec 1997).
Other side effects include bronchospasm in asthmatics if applied in large amounts (Anon 1989), 3 cases of renal failure have been reported (NPC 1997) and in 21 of 21,923 (0.1%), GI adverse reactions were noted with Felbinac 3% gel over a two-week trial. (Newbury et al 1992). However, these risks are minimal in small to moderate doses. However, caution should be noted in patients with predisposing GI pathology, or history of Asthma.
Precautions also need to be taken into consideration. These include application on broken skin, mucous membranes and nursing mothers because of the systemic effect of topical NSAIDs.
Other precautions to note include the application of topical NSAIDs in those patients taking oral NSAIDs. This is a fundamental point, as Scott et al (1995) found in a study that it is not uncommon for patients to take topical NSAIDs supplementary to oral NSAIDs for long-term use.
CONCLUSION
Evidence on the reliability of topical NSAIDs remains limited. Research based evidence does however show that topical NSAIDs are slightly more effective than their placebo counterparts, even though placebo has been shown to work well.
Caution does need to be taken into consideration when advising on such applications, especially when patients are taking oral NSAIDs at the same time. Indeed, the high level of prescribing and advising of topical NSAIDs, may not justify the available evidence.
More reliable research needs to be published to identify the short and long term benefits of topical NSAIDs, even though topical NSAIDs are safer than their oral counterparts.
REFERENCES
Airaksinen O, Venalainen J, Pietilainen T, (1993) Ketoprofen 2.5% gel versus placebo gel in the treatment of acute soft tissue injuries, International Journal of clinical Pharmacology Therapy and Toxicology, 31, 561-564
Anon, (1991) Topical non-steroidal anti-inflammatory drugs, MeReC Bulletin, 2, 21-24
Duerden M, Barton S, Johnstone E, MacLean K, Underhill J, Walley T, (1998) Topical NSAIDs are better than placebo, British Medical Journal, 317, 67
Grahame R, (1995) Transdermal non-steroidal anti-inflammatory agents, British Journal in Clinical Practice, 49, 33-36
Kageyama T, (1987) A double-blinded placebo controlled multi-centre study of piroxicam 0.5% gel osteoarthritis of the knee, European journal of rheumotology Inflammation, 8, 114-115
National Prescribing Centre (NPC) (1997), Topical non-steroidal anti-inflammatory drugs, Circular release, 8
MeReC Bulletin, (1997) Topical non-steroidal anti-inflammatory drugs, 8, 8, 29-32
Moore R, Tramer M, Carroll D, Wiffen P, McQuay H, (1998) Quantitive systematic review of topically applied non-steroidal anti-inflammatory drugs, British Medical Journal, 316, 333-338
Newbury R, Shuttleworth P, Rapier C, (1992) A multi centre post marketing surveillance study to evaluate the safety and efficacy of Felbinac 3% gel in the treatment of musculoskeletal disorders in general practice, European Journal of clinical Res, 3, 139-150
NHS Journal (2002) An excellent choice for the treatment of mild arthritis and thematic conditions, NHS Journal for healthcare professionals, May 2002.
Ramesh N, Steuber U, (1983) Dolgit cream in accident and sports related injuries in medical practice, results of a double-blinded study, Therapiewolche, 33, 4563-4570
Russell A, (1991) Prioxicam 0.5% topical gel compared top placebo in the treatment of acute soft tissue injuries: a double blinded study comparing the efficacy and safety, Clinical Invest Medicine, 14, 35-42
Scott E, Briggs A, Henning C, (1995) Patient survey on topical non-steroidal anti-inflammatory drugs, The Pharmaceutical Journal, 35
