EVIDENCE-BASED ANSWER
The use of topical triple-antibiotic ointments significantly decreases infection rates in minor contaminated wounds compared with a petrolatum control. Plain petrolatum ointment is equivalent to triple-antibiotic ointments for sterile wounds as a post-procedure wound dressing (strength of recommendation [SOR]: A, based on randomized controlled trials [RCTs]).
Mupirocin cream is as effective as oral cephalexin in the treatment of secondarily infected minor wounds and, because of better tolerability, is the treatment of choice for the prevention and treatment of Staphylococcus aureus and Staphylococcus pyogenes infections. Emerging resistance, including methicillin-resistant S aureus (MRSA), makes it prudent to check for clinical response in 24 to 48 hours. Major contaminated wounds requiring parenteral antibiotics do not appear to additionally benefit from topical antibiotics (SOR: A, based on RCTs).
Topical antibiotics may also aid in the healing of chronic wounds (SOR: B, based on a systematic review of low-quality RCTs), as does the application of honey (SOR: B, based on a systematic review of cohort studies).
CLINICAL COMMENTARY
It would be helpful to have objective criteria to properly classify skin wounds
Michael Mendoza, MD, MPH
Pritzker School of Medicine, University of Chicago
These results are encouraging, but they do not fully account for variability in the diagnosis of skin wounds or in the practical use of topical agents. The evaluation of skin wounds is inherently subjective. In order to properly apply these findings to my practice, it would be helpful to have more objective diagnostic criteria to properly classify skin wounds.
Furthermore, how patients use topical agents varies considerably. Patients apply topical agents differently, due to individual preference or perhaps inconsistent recommendations from their physician. Used improperly, topical agents may not provide the same potential for clinical improvement.
Evidence summary
Topical antibiotics for prophylaxis
Numerous studies support the prophylactic application of topical antibiotics to wounds that are clean. Topical bacitracin zinc (Bacitracin), a triple ointment of neomycin sulfate, bacitracin zinc, and polymyxin B sulfate (Neosporin), and silver sulfadiazine (Silvadene) were compared with petrolatum as a control in a well-conducted RCT of 426 patients with uncomplicated wounds seen at a military community hospital. Wound infection rates were 17.6% (19/108) for petrolatum, 5.5% (6/109) for Bacitracin (number needed to treat [NNT]=8), 4.5% (5/110) for Neosporin (NNT=8), and 12.1% (12/99) for Silvadene (NNT=18).1 Most (60%) of the infections were “stitch abscesses” and were treated with local care only. There was no difference in rates of more serious infections between groups. One patient (0.9%) developed a hypersensitivity reaction to Neosporin.
A clinical trial compared the efficacy of a cetrimide, bacitracin zinc, and polymyxin B sulfate gel (a combination not available in the US) with placebo and povidone-iodine cream in preventing infections in 177 minor wounds (cuts, grazes, scrapes, and scratches) among children. The antibiotic gel was found to be superior to placebo and equivalent to povidone-iodine, in that it reduced clinical infections from 12.5% to 1.6% (absolute risk reduction [ARR]=0.109; 95% confidence interval [CI], 0.011–0.207; NNT=11).2
A double-blind study of 59 patients found Neosporin superior to placebo ointment in the prevention of streptococcal pyoderma for children with minor wounds. Infection occurred in 47% of placebo-treated children compared with 15% treated with the triple-antibiotic ointment (NNT=32; P=.01).3
A small randomized prospective trial of 99 patients, who self-reported compliance with wound care and dressing changes, compared Neosporin with mupirocin (Bactroban) in preventing infections in uncomplicated soft tissue wounds. The study found no statistical difference in infection rates, and the authors recommend the more cost-effective Neosporin, as well as a larger trial to confirm the results.4
Another randomized controlled trial of 933 outpatients—with a total of 1249 wounds from sterile dermatologic surgeries—compared white petrolatum with bacitracin zinc ointment prophylaxis. The study found no statistically significant differences in post-procedure infection rates, though only 13 patients developed an infection (2% in petrolatum group vs. 0.9% in bacitracin zinc group; 95% CI for the difference, –0.4 to 2.7).5
Topical antibiotics for treatment
Topical antimicrobials are appealing for the treatment of secondarily infected wounds for the sake of convenience and because they may reduce the risk of adverse effects.
An open randomized trial with 48 volunteers compared the effects of Neosporin with several antiseptics (3% hydrogen peroxide, 1% povidone-iodine, 0.25% acetic acid, 0.5% sodium hydrochloride) and a wound protectant (Johnson & Johnson First Aid Cream without antimicrobial agent) on blister wounds (6 blisters per volunteer) intentionally contaminated with S aureus. Only Neosporin eliminated the infection after 2 applications (at 16 and 24 hours). Both the antibiotic ointment and the wound protectant led to faster wound healing by about 4 days compared with the antiseptics or no treatment.6
Another study with 2 parallel, identical RCTs of a total of 706 patients found mupirocin cream (Bactroban) to be equivalent to oral cephalexin in the treatment of secondarily infected minor wounds, such as small lacerations, abrasions, or sutured wounds. Clinical success (95.1% for mupirocin and 95.3% for cephalexin), bacteriologic success (96.9% for mupirocin and 98.9% for cephalexin), as well as the intention-to-treat success rate of 83% at follow-up were equivalent in the 2 groups.7