Doi:10.1016/s0735-6757(03)00220-

A Prospective, Randomized Pilot Evaluation of
Topical Triple Antibiotic Versus Mupirocin for
the Prevention of Uncomplicated Soft Tissue
Wound Infection
ROBERT HOOD, MD,* KENNETH M. SHERMOCK, PHARMD,† Little data exists comparing the safety and efficacy of triple antibiotic
itracin zinc, and polymixin B sulfate) or placebo petrolatum ointment (TAO) and mupirocin for prevention of uncomplicated soft
ointment, there was a lower rate of infection in the TAO tissue wound infections. The purpose of this investigation was to con-
group (4.5%) when compared with the control group duct a pilot study of the relative safety, efficacy, and cost effectiveness
(17.6% P ϭ .0034).5 Mupirocin ointment (pseudomonic of the 2 preparations. This was a randomized, prospective, interventional
study to determine the difference in infection rates of uncomplicated soft
acid A) has been shown to be highly effective against tissue wounds between subjects treated with TAO and mupirocin oint-
normal skin pathogens such as Staphylococcus and Strep- ment after standard wound care and suturing. Subjects were enrolled at
tococcus6 and has been used extensively to treat impetigo7-9 presentation to the ED if they met the study inclusion criteria and were
as well as Staphylococcus aureus-infected nipples.10 required to make one follow-up visit to the ED to determine the status of
There is little information comparing the relative safety their wound (infected vs. not infected). A total of 99 patients were en-
and efficacy of TAO and mupirocin when used as prophy- rolled and assessed at the follow-up visit. The groups had similar rates
of self-reported compliance with wound care and dressing changes.
laxis for uncomplicated soft tissue wounds. These results Patients in the mupirocin group had a greater rate of signs of infection
are important because there is a cost difference between the (12% vs. 6.1%), and infection (4% vs. 0%) compared with patients in the
available formulations and there have been concerns around TAO group, although neither difference achieved statistical significance.
sensitization with TAO. The purpose of this investigation There were no serious adverse effects in either group. This pilot study
was to conduct a pilot study of the relative safety and found a similar rate of wound infection and adverse events between TAO
and mupirocin ointments. Results should be confirmed in a larger equiv-
alency trial. (Am J Emerg Med 2004;22:1-3. 2004 Elsevier Inc. All rights
reserved.)
This study was approved by the Institutional Review Soft tissue wounds are a common ED complaint. Al- Board. The study was designed to enroll 120 patients with though many injuries do well with appropriate cleansing the expectation that there would be a 20% dropout rate. The and debridment, some become infected. Factors related to inclusion criteria were uncomplicated soft tissue wound wound infection include the mechanism of injury, the within the last 24 hours and willingness to return within 7 amount of tissue damage, the presence of contaminants, the days for reevaluation. There was no limitation on underly- location of the wound, comorbid medical conditions, and ing medical conditions. This study was intended as a pilot the age of the wound. Previous studies have found a varying study to determine the necessity and feasibility of a larger rate of wound infection after ED treatment from 4.5% to 6.3%.1-4 Topical antibiotic preparations are approved for The exclusion criteria included puncture wounds, under- use to prevent wound infection, although only limited in- lying fracture, use of antibiotics within the last 7 days, formation is available comparing the agents and their effec- known allergy to the study agents, wounds closed with Dermabond, wounds which, in the opinion of the treating In a prospective study comparing infection rates between physician, required use of oral or parenteral antibiotics, and wounds treated in a prophylactio manner with triple antibi- wounds that were found to be infected at the time of otic ointment (TAO; combination of neomycin sulfate, bac- presentation. Patients who did not completely comply withstudy procedures but for whom follow-up data were avail-able are included in the study. Initial wound management, From the *Departments of Emergency Medicine, Cleveland Clinic including the choice of glove type, cleansing solution, irri- Foundation and Metro Health Medical Center, and the †Department gation, and anesthetic, was at the discretion of the treating of Pharmacy, Cleveland Clinic Foundation, Cleveland, Ohio.
Received January 16, 2003; accepted January 16, 2003.
The study medication was prepared by the hospital phar- Supported by a grant from Pfizer Consumer Healthcare.
Address reprint requests to Charles L. Emerman, MD, Depart- macy and dispensed in a coded tube with all other identi- ment of Emergency Medicine, Cleveland Clinic Foundation, 9500 fying information removed. The patient, treating physician, Euclid Avenue, E-19, Cleveland, OH 44195. E-mail: [email protected] and study investigators were blinded to the identity of the Key Words: Infection, wound care, antibiotic therapy, prophylaxis.
study medication. After informed consent was obtained, all 2004 Elsevier Inc. All rights reserved.
0735-6757/04/2201-0001$30.00/0 patients enrolled in the study underwent standard wound care and repair procedures, followed by the initial applica- AMERICAN JOURNAL OF EMERGENCY MEDICINE ■ Volume 22, Number 1 ■ January 2004 Demographic and Wound Characteristic Data This study was conducted between August and December 2001 enrolling 120 patients. Twenty-one patients were lost to follow up, leaving 99 patients available for data analysis.
The average age of the patients was 24.7 Ϯ 17.8 years and 72.7% of the patients were male. A total of 42.4% of the patients were under the age of 18. Three (3%) of the patients had a history of diabetes. Forty-six patients had full-thick- ness wounds requiring repair. These characteristics were not significantly different between the 2 groups except that the mupirocin group was slightly older than the TAO group (27.7 y vs. 21.6 y, P ϭ .09). The patients reported similar pain scores before treatment (Table 1).
Eighty patients reported full compliance with study pro- cedures (Table 2). The groups had similar rates of self- reported compliance with wound care and dressing changes.
Nine patients had predetermined possible signs of infection as assessed by the study nurses, whereas 2 of these patients were diagnosed with a wound infection by the treating physician (Table 2), Mild erythema was noted in 8 of the 9 patients, whereas 1 patient had a small amount of purulent drainage. There was no significant difference in rate ofwound infection between the groups.
Abbreviation: SD, standard deviation.
One patient in the TAO group reported paraesthesia around the wound. The mean post-period pain score was tion of the randomly assigned topical ointment. They were higher in the mupirocin group compared with the TAO then discharged with a blinded supply of study medication group (0.44 vs. 0.15, P ϭ .07).
and written instructions for use. Subjects were instructed to Because no difference was detected between the drug apply the ointment to the wounds 3 times per day until the groups in infection rate, the cost-effectiveness analysis was reduced to a cost-minimization analysis (simple comparison The primary end point was the presence of infection of cost). Neither of the 2 infections were judged to requiretreatment. Therefore, the cost analysis is a comparison of within 7 days. This was determined by evaluating for fever, the cost of medication. The average wholesale prices erythema, edema, induration, swelling, warmth, exudate, (AWP) for a 22-g tube of bacitracin-neomycin-polymycin B adenopathy, and lymphangitis. The wounds were all graded (Bactroban) is $41.00 ($1.86/g) and for a 28-g tube of and evaluated by 1 of 2 trained research nurses with phy- mupirocin (Neosporin) is $6.95 ($0.25/g). Bactroban costs sician oversight. The wounds were graded as follows: grade $34.05 more than Neosporin and was not associated with a 0, no sign of infection; grade 1, simple stitch abscess; grade detectable difference in preventing infection or signs of 2, surrounding cellulitis Ͻ 1 cm; grade 3, accompanying lymphangitis and/or lymphadenitis; and grade 4, systemicsymptoms. The secondary outcomes were pain associated DISCUSSION
with the wound evaluated by having the patient’s rate their This pilot study found a similar rate of wound infection pain on a visual analog scale and a safety assessment and adverse events between the TAO and mupirocin-treated looking for adverse events associated with study medication groups for simple soft tissue wounds. The wound infection rate in this study overall was low compared with that The difference in infection rates between patients in the 2 reported historically. This could represent patient selection medication groups and other categorical variables were an- bias, good initial wound management, or the effectiveness alyzed with a Chi-squared test of proportions or Fisher exact of topical antibiotic treatment for soft tissue wounds.
test. Continuous variables were analyzed using a Studentt-test for normally distributed data and the Wilcoxon ranksum test for nonparametric data.
Compliance With Wound Care and Infection Rates A marginal cost-effectiveness analysis was conducted with costs and effectiveness calculated from the payer per- spective.11 Medication and follow-up care in the event of an infection were considered as costs. The unit of effectiveness was the rate of infection-free wounds. If the groups were not statistically different in terms of outcome, the cost analysis became a cost-minimization analysis, a simple comparison of average cost between the 2 groups. Sensitivity analysis was conducted by varying all cost estimates to determine the effect of such variation on the results of the cost-effectiveness analysis.
Abbreviation: SD, standard deviation.
HOOD ET AL ■ EVALUATION OF TOPICAL TRIPLE ANTIBIOTIC VS. MUPIROCIN FOR PREVENTION OF WOUND INFECTION Previous studies have evaluated the rate of wound infec- CONCLUSION
tion in ED patients. Generally, the rates are very low forwounds that are not grossly contaminated and have received This pilot study found a similar rate of wound infection, appropriate irrigation, cleansing, and debridement before pain attenuation, and adverse events between TAO and repair. This low inherent rate of infection in simple soft mupirocin ointment. If confirmed in a larger trial, this would tissue wounds means that studies designed to detect differ- suggest that the less expensive and over-the-counter TAO ences between treatments require very large sample sizes.
can be recommended for prophylaxis following uncompli- In a recent study, Langford and workers treated 177 minor soft tissue injuries using a topical antibiotic prepara-tion (cetrimide, bacitracin, and polymyxin B sulfate), pov-idone iodine 10% w/v, or hydroxypropyl methylcellulose REFERENCES
gel (control).12 Those treated with the TAO had an overall 1. Gosnold J: Infection rate of sutured wounds. Practitioner 1977; infection rate of 1.6% compared with the control group with 12.5% (P Ͻ .05), whereas there was no significant differ- 2. Hutton P, Jones B, Law D: Depot penicillin as prophylaxis in ence between the povidone iodine group and the control accidental wounds. Br J Surg 1978;65:549-550 3. Rutherford W, Spence R: Infection in wounds sutured in the There have been concerns about sensitization and allergic accident and emergency department. Ann Emerg Med 1980;9:350-352 reactions resulting from the use of topical antibiotic prepa- 4. Thirlby R, Blair A, Thal E: The value of prophylactic antibiotics rations. This has especially been true with preparations for simple lacerations. Surg Gynecol Obstet 1983;156:212-216 containing neomycin. A review by MacDonald and col- 5. Dire D, Coppola M, Dwyer D, Lorette J, Karr J: A prospective leagues discusses in detail cases of skin sensitization and evaluation of topical antibiotics for uncomplicated soft-tissue lac- systemic toxicity following use of this agent.13 Hearing loss has been documented following topical use of neomycin.14, 6. Ealls L, Mertz P, Piovanetti Y, Pekoe G: An evaluation of the safety and efficacy of topical antimicrobial therapy for primary skin 15 However, the prevalence of contact dermatitis might not infections. J Invest Dermatol 1984;82:404 be as high as commonly thought. Prystowsky et al. reported 7. Hogan P: Pediatric dermatology: impetigo. Aust Fam Physi- the prevalence of contact dermatitis to TAO among the general population was reported to be approximately 1%.16, 8. Bass J, Chan D, Creamer K, et al: Comparison of oral cepha- 17 Neomycin has been reported to be a key contributor to the lexin, topical mupirocin and topical bacitracin for treatment of im- TAO’s antimicrobial efficacy against Staphylococcus spe- petigo. Pediatr Infect Dis J 1997;16:708-710 9. Hudson I: The efficacy of intranasal mupirocin in the presen- tation of staphylococcal infections: a review of recent experience. J A case report by Zappi and Brancaccio described a 68- year-old patient who developed a documented case of al- 10. Livingstone V, Stringer L: The treatment of Staphylococcus lergic contact dermatitis after the use of mupirocin on a aureus infected nipples: a randomized study. Journal of Human surgical wound resulting from basal cell carcinoma re- moval.21 The incidence of rash reported for mupirocin is 11. Drummond MF, O’Brien B, Stoddart GL, Torrance GW: Meth- 1%.22 No serious adverse events were reported in either ods for the Economic Evaluation of Health Care Programmes, 2nded. Oxford: Oxford Medical Publications; 1997 12. Langford J, Artemi P, Berimoj S: Topical antimicrobial pro- From a cost-effectiveness perspective, mupirocin costs phylaxis in minor wounds. Ann Pharmacother 1997;31:559-563 more than 7 times TAO on a per gram basis and was not 13. MacDonald R, Beck M: Neomycin: a review with particular superior in preventing infections or associated with a lower reference to dermatological usage. Clin Exp Dermatol 1983;8:249- serious side effect rate. Unless these, or other, benefits are proven in favor of mupirocin in future studies, TAO is cost 14. Johnson C: Hearing loss following the application of topical effective for the prevention of infections in uncomplicated neomycin. J Burn Care Rehabil 1988;9:162-164 15. Kelly D, Nilo E, Berggren R: Deafness after topical neomycin wound irrigation. N Engl J Med 1969;280:1338-1339 There are several limitations of the current study. This 16. Prystowsky S, Allen A, Smith R, et al: Allergic contact hyper- study was not powered to detect differences in efficacy or sensitivity to nickel, neomycin, ethylenediamine, and benzocaine.
adverse events between the groups. By intention, this was a pilot study to help guide decisions regarding future study 17. Prystowsky S, Nonomura J, Smith R, et al: Allergic hypersen- designs. Patients were selected for this trial study based on sitivity to neomycin: relationship between patch test reactions and“use” tests. Arch Dermatol 1979;115:713-715 the availability of the study coordinators. Patients who 18. Booth J, Benrimoj S, Nimmo G: In vitro interactions of neo- arrived in the ED when the study coordinators were not mycin sulfate, bacitracin, and polymyxin B sulfate. Int J Dermatol available such as late at night could comprise a group with a different risk for wound infections. Although the study 19. Hendley J, Ashe K: Effect of topical antimicrobial treatment was blinded, the determination of wound infection is a on aerobic bacteria in the stratum corneum of human skin. Antimi-crob Agents Chemother 1991;35:627-631 clinical one and could not have been applied uniformly 20. Davis S, Cazzaniga A, Mertz P: Can OTC Antimicrobial Ban- Finally, patient compliance was self-reported. No attempt Wounds? American Academy of Dermatology Poster Exhibit; Feb 2, was made to verify the accuracy of these self-reports. A 21. Zappi E, Brancaccio R: Allergic contact dermatitis from mupi- larger study, to the extent that it can, should include design rocin ointment. J Am Acad Dermatol 1997;36:266 elements to mitigate the impact of these and other threats to 22. Bactroban Product Information. Philadelphia: SmithKline

Source: http://www.zjpharma.cn/zlzx/UploadFiles_8612/200703/20070310131207520.pdf