Comparison of Oral Cephalexin, Topical Mupirocin and Topical Bacitracin for Treatment of Impetigo

During the past decade penicillin-resistant Staphylococcus aureus emerged as the predominant cause of impetigo and during this period numerous studies have been conducted to determine what constitutes optimal treatment for this common pediatric infection.1 Drugs that are not inactivated by penicillin beta-lactamases including the penicillinase-resistent penicillins, the cephalosporins and macrolides all appear to be logical candidates and have been studied extensively. Erythromycin evolved as a popular and effective form of oral treatment for this infection during the past decade from the prospective of effectiveness, acceptance, tolerance and comparative cost. This drug has been the standard with which other drugs were compared in a number of studies.2-8 In a previous prospective double blind study we compared oral penicillin V, erythromycin and cephalexin for treatment of impetigo.9 We found that S. aureus was the most common cause of impetigo in our study population, that cephalexin was the most effective treatment, that erythromycin was nearly equally effective and that penicillin V was inadequate for treatment of impetigo. The observation that S. aureus is the most common cause of impetigo worldwide has since become appreciated.10, 11 Moreover the observation that resistance of this organism to erythromycin has become an increasing problem in many areas of the world has caused concern that this mode of treatment may not be optimal. Studies published in the early 1980s reported a 10% resistance rate12 whereas some published in the late 1980s and 1990s have reported resistance rates of 24%,13 28%,14 35%,15 50%6 and 52%.16 Topical mupirocin has been available for the past decade and its use for treatment of impetigo has been extensively studied. It has been shown to be superior to placebo17 and topical Neosporin® (polymyxin B and neomycin)18 and equal to or superior to oral erythromycin in a number of blinded tests,2-8 but no study has compared it with oral cephalexin. Topical bacitracin has been available for treatment of skin infections since the late 1940s.19 Although its effectiveness is time-honored, no report could be found evaluating its effectiveness for treatment of impetigo in a blinded placebo-controlled study, and no study has compared its efficacy with that of oral cephalexin or topical mupirocin. This prospective double blind placebo-controlled study was performed to compare the effectiveness of oral cephalexin, topical mupirocin and topical bacitracin for treatment of impetigo in children. Methods. This study is a continuation of our previous study with the same design except for the treatment regimens evaluated.9 The study design, study population, baseline information, exclusionary criteria, culture technique and compliance measurements are detailed in the previous report. Antimicrobial treatment regimens. After the initial clinic evaluation participants were referred to the pharmacy where a clinical pharmacist assigned them by a table of random numbers to one of three treatment groups: Group 1, cephalexin monohydrate suspension in a dosage of 50 mg/kg/day (maximum, 500 mg/dose) in three divided doses plus 30 g of a placebo topical ointment (petrolatum plus glycerin) to be applied to affected areas three times daily; Group 2, mupirocin 2%, ointment 30 g, to be applied topically to affected areas three times daily plus an oral liquid placebo matched to oral cephalexin to be given in a dosage comparable with that of cephalexin in three divided daily doses; or Group 3, bacitracin 500 units/g of ointment 30 g, to be applied topically to affected areas three times daily plus an oral liquid placebo matched to oral cephalexin to be given in a dosage comparable with that of cephalexin in three divided daily doses. Treatment in all three groups was for 10 days. The clinician, patients and their parents were not aware of which of the three treatment regimens they were assigned. Patients were instructed to wash their skin lesions daily with soap and water but no other medications were to be used other than those provided. Initial assessment and follow-up. Baseline clinical information was recorded including the character of impetigo lesions (bullous, pustular or honey-crusted) and the total area involved in square centimeters. Patients were reevaluated at 3 to 5 and 8 to 10 days after beginning therapy by one of the treating clinicians blinded to the treatment regimens. Response to treatment was assessed by determining lesion character and size. At the 3- to 5-day visit patients were classified as "treatment failure" if there was no improvement in initial lesions and/or new lesions were observed. They were classified as "improved" if clearing of initial lesions and no new lesions was observed. Patients with persistent lesions at the 8- to 10-day visit were judged "treatment failures" and retreated with cephalexin for 10 days. Those with almost complete clearing of their lesions (scant residual with no signs of active infection) and no new lesions were judged "improved" and those with complete clearing of their lesions were judged "cured." Statistical analysis. The difference in treatment failure among the three treatment groups and sex distribution was tested for significance using Fisher's exact test. The level of significance was set at P < 0.05. Two-tailed P values were calculated. Analysis of variance for continuous data (age, duration and size of lesion(s) and compliance rates) was performed by a nonparametric test (Kruskal-Wallis one-way analysis of variance on ranks).20 Results. Thirty-two patients were entered into the study. Three patients (2 mupirocin-treated and 1 bacitracin-treated) did not return for either follow-up visit and 3 mupirocin-treated patients did not return for the second follow-up visit. These patients were dropped from the study leaving 26 (mean age, 3.8 ± 0.55 SE years) evaluable patients: 10 cephalexin (4.9 ± 1.3); 7 mupirocin (3.0 ± 0.6); and 9 bacitracin (3.2 ± 0.6). The three treatment groups were comparable in sex and age distribution as well as duration, size and type of lesions. Table 1 shows clinical data on the patients in each treatment group as well as outcome. No treatment failure occurred with mupirocin or cephalexin. Bacitracin was less effective. Most patients treated with bacitracin were treatment failures. Significant differences among the treatment groups are: cephalexin > bacitracin (P < 0.005, power 0.94); mupirocin > bacitracin (P = 0.01, power 0.88); and cephalexin = mupirocin (P = 0.999).TABLE 1: Comparison of the three treatment groups S. aureus was cultured from all of 22 patients who had cultures of their lesions when entered into the study. Of these, group A beta-hemolytic Streptococcus in combination with S. aureus were cultured in 3 (14%). Of 21 antibiotic susceptibility tests performed on S. aureus isolates, 20 (95%) showed resistance to penicillin, 4 (19%) to erythromycin and none to cephalexin (as determined by cephalothin disks). Sixteen study patients submitted one or more urine samples for antibiotic activity testing. All of 6 cephalexin-treated patients, 1 of 5 mupirocin-treated patients and none of 5 bacitracin-treated patients tested positive. Compliance (percent consumed) was measured on 17 medication bottles (7 cephalexin, 5 mupirocin, 5 bacitracin). Calculations were made to determine the percent taken in relation to the amount that should have been taken at the time of follow-up. The mean percentages of liquid consumed for cephalexin, mupirocin and bacitracin were 80, 97 and 75%, respectively (P > 0.05). Examination of the jars of mupirocin, bacitracin and placebo ointment at the 8- to 10-day follow-up visit revealed >25% of the contents of all jars missing consistent with use as prescribed. Discussion. Although the numbers in this study are small the differences are large; therefore the power associated with the comparison against bacitracin is large. Most patients treated with bacitracin were treatment failures. This mode of treatment can no longer be recommended. Our findings show that topical mupirocin is comparable with oral cephalexin treatment in effectiveness for treatment of impetigo. Because there were no treatment failures in any of the patients in these two treatment groups, both regimens can be recommended equally. Topical mupirocin may best be utilized for treatment of patients with only a few lesions involving a small area whereas oral cephalexin may be more appropriate for treatment of patients with more extensive involvement, or some patients may prefer one of these forms of treatment over the other. Combining the results of our previous similarly designed study9 with this study we have now treated 35 impetigo-infected children with oral cephalexin followed by blinded evaluation and comparison with other treatment regimens, and there were no treatment failures with cephalexin. On the basis of efficacy, taste, acceptance, compliance and relative cost, cephalexin would appear to be the drug of choice for oral antibiotic treatment of children with impetigo. Acknowledgments. We gratefully acknowledge the assistance of John R. Claybaugh, Ph.D., for the statistical analysis of the data, and Phyllis McFadden, MT (ASCP), for her assistance in the laboratory. We also thank the late Leki Ratanamaneichatara for his support of this study. James W. Bass, M.D., M.P.H. Debora S. Chan, B.S., F.A.S.H.P. Kevin M. Creamer, M.D. Mark W. Thompson, M.D. Francis J. Malone, M.D. Theresa M. Becker, D.O. Stephen N. Marks, M.D.