Mupirocin. This topical antibiotic has activity against streptococci and staphylococci. Its primary use is for treatment of superficial skin infections, but in recent years it has been frequently prescribed for application to the anterior nares in patients with recurrent staphylococcal skin infections. Indeed, intranasal mupirocin(Drug information on mupirocin) is recommended for staphylococcal decolonization more frequently than any other single intervention.12
Intranasal mupirocin can eradicate nasal S aureus carriage, at least transiently.37-41 In one large study, 339 colonized health care workers were randomly assigned to receive either intranasal mupirocin ointment or a placebo ointment twice daily for 5 days.37,38 At the end of therapy, 13% of mupirocin recipients remained colonized, compared with 93% of placebo recipients. Mupirocin’s effectiveness is only temporary, however; recolonization is the rule. Multiple studies have confirmed that most mupirocintreated patients will again have S aureus recovered from intranasal cultures within 12 months of the end of treatment.37-41
Mupirocin is less active against MRSA than against methicillin-susceptible S aureus in vitro,42 and it is less than optimally effective for decolonization of MRSA carriers as well. In one study, 102 adult inpatients colonized with MRSA were randomly assigned to receive 5 days of intranasal therapy with mupirocin or placebo; all patients also bathed daily with chlorhexidine(Drug information on chlorhexidine). On day 26 after completion of therapy, MRSA was recovered from 1 or more sites in 75% of mupirocin recipients and in 82% of placebo recipients (P = .40). Moreover, 56% of mupirocin recipients had MRSA recovered from nasal cultures.43
A central assumption in the rationale for nasal decolonization therapy is that eradicating staphylococcal colonization will prevent recurrent skin infections. Extant evidence, however, contradicts this assumption. Staphylococcal infections still develop in patients who are successfully decolonized by mupirocin and at rates similar to those in patients who do not receive decolonization therapy.
An RCT of 127 nursing home residents colonized with S aureus found a nonsignificant reduction in infection rates in those who received intranasal mupirocin for 14 days compared with those who received placebo (5% vs 15%; P = .10), even though mupirocin recipients were more likely than controls to have negative cultures after therapy (88% vs 13%; P < .001). Furthermore, all 3 mupirocin recipients in whom staphylococcal infections developed had been successfully decolonized and remained free of nasal S aureus at the time of their infections.44
In another study, 134 soldiers colonized with MRSA were randomly assigned (in clusters, by training class) to receive intranasal mupirocin or placebo twice daily for 5 days. Over 16 weeks of observation, skin or soft tissue infections developed in 7.7% of placebo recipients and in 10.6% of mupirocin recipients (rate difference [RD], –2.9%; 95% confidence interval [CI], –7.5% to 1.7%). Moreover, decolonization of MRSA carriers did not significantly change the infection rate in close contacts. Of the colonized soldiers’ training classmates, infections developed in 4.3% (63 of 1459) of the classmates of those in the placebo group and in 3.5% (56 of 1607) of the classmates of those in the mupirocin group (RD, 0.8%; 95% CI, –1.0% to 2.7%).45
These data demonstrate that a single course of intranasal mupirocin is not effective for prevention of recurrent staphylococcal skin infections. Whether repeated courses of mupirocin can achieve this goal has not been as well studied.
In an RCT, S aureus–colonized adults with recurrent furunculosis received monthly courses of intranasal mupirocin (twice daily for the first 5 days of the month) or placebo for 1 year. Results favored the mupirocin group: 9 of 17 patients in this group had further skin infections, compared with 16 of 17 patients in the placebo group (P = .02).46 This study should be interpreted cautiously, however, because the authors did not perform an intent-to-treat analysis; instead they excluded 6 enrolled patients from the final analysis. Moreover, this study did not specifically evaluate patients colonized with MRSA, and it is doubtful that its results could be extrapolated to that population.
The clinical usefulness of mupirocin is also limited by the worldwide emergence of resistant MRSA strains.47-54 In 2000, a survey of 2159 clinical isolates of S aureus found mupirocin resistance (defined in this study as a minimal inhibitory concentration [MIC] of 16 μg/mL or greater) in 18% of MRSA isolates from Europe and in 14% of those from North America.47 Moreover, a 1999-2002 study in Saskatchewan found that more than 50% of 184 MRSA strains showed high-level mupirocin resistance (MIC, 256 μg/mL or greater).54
Limited data suggest that mupirocin might be useful for prevention of health care–associated infections in selected patient populations. 55,56 Therefore, indiscriminate use of intranasal mupirocin in patients with recurrent skin infections—for whom its effectiveness is unproved—might render it less useful for other, potentially more important indications.
Tea tree oil. This is an essential oil derived from the Australian plant Melaleuca alternifolia; it has broad antimicrobial activity.57 A variety of products containing tea tree oil are available, including soaps, shampoos, and creams, and their use has been advocated for such conditions as furunculosis, acne, superficial fungal infections, and herpes labialis.57,58
Tea tree oil is less active than mupirocin against MRSA in vitro,51 and it appears to be no better than mupirocin for MRSA decolonization. A study of 236 adults colonized with MRSA compared 5 days of a tea tree oil regimen (tea tree oil 5% body wash daily plus tea tree oil 10% cream intranasally 3 times daily) to a “standard” regimen (chlorhexidine 4% body wash daily plus mupirocin 2% ointment intranasally 3 times daily). Fourteen days after the completion of therapy, MRSA had been eradicated from 41% of those who received the tea tree oil regimen and from 49% of those who received the standard regimen (P = .29).58
Patients who wish to use “natural” products may find the idea of using tea tree oil appealing. Nonetheless, available data do not support the use of tea tree oil to prevent recurrent MRSA skin infections.
In theory, systemically administered antibiotics (such as TMP/SMX, rifampin, clindamycin(Drug information on clindamycin), or doxycycline(Drug information on doxycycline)) might eradicate colonization from mucosal sites beyond the reach of topical agents and, thus, either alone or in combination with topical and/or intranasal interventions, might eliminate the staphylococcal carrier state. In a review of MRSA decolonization strategies, Boyce59 listed 12 different oral antibiotics that have been used, alone or in combination, for this purpose. Nonetheless, a systematic review of published trials concluded that insufficient evidence exists to recommend systemic antibiotics for MRSA decolonization.60
Limited data suggest that prophylactic administration of systemic antibiotics might reduce the frequency of recurrent S aureus skin infections. In an RCT, new abscesses developed during the treatment period in 2 of 11 adults who received oral clindamycin (150 mg/d) for 3 months, compared with 7 of 11 patients who received placebo.61 Although this difference is of marginal statistical significance, interpretation of this study is hindered by its small sample size and its lack of an intent-to-treat analysis (3 enrolled patients were excluded from the efficacy analysis).
To date, no RCT has specifically evaluated the efficacy of systemic antibiotics for preventing recurrent skin infections caused by MRSA. Moreover, even if 1 or more drugs is found to be effective, the anticipated benefit of prolonged systemic therapy will still need to be weighed against the risks of toxicity and of increasing antimicrobial resistance.
A recent study evaluated a combined regimen of topical, intranasal, and systemic interventions for MRSA decolonization in 146 hospitalized adults. Patients were randomly assigned to receive treatment (2% chlorhexidine washes, 2% mupirocin ointment intranasally, and oral rifampin and doxycycline for 7 days) or no treatment. Three months after completing therapy, 74% of treated patients had culture results that were negative for MRSA, compared with 32% of those not treated (P < .0001). However, at an 8- month follow-up, only 54% of treated patients still had cultures that were negative for MRSA.62
At present, there is no evidence that combinations of topical, intranasal, and oral antimicrobials are effective in preventing MRSA skin infections. On the other hand, combined regimens expose patients to both costs and potential side effects. Such regimens should probably be avoided, at least until the efficacy of one of them is confirmed.