Mupirocin has been widely used as a topical antimicrobial agent for treatment of skin infections and for eradicating Staphylococcus aureus nasal carriage in patients and healthcare personnel.Reference Patel, Gorwitz and Jernigan ¹ In the United States, mupirocin resistance among methicillin-resistant Staphylococcus aureus (MRSA) isolates have emerged in adult populations, especially among patients from long-term care facilities and patients with skin and soft-tissue infections.Reference McDanel, Murphy and Diekema ^{2 ,} Reference Fritz, Hogan and Camins ³ MRSA has become endemic in some neonatal intensive care units (NICUs).Reference Popoola, Budd and Wittig ⁴ Neonates that acquire MRSA in the NICU have a high risk of developing a MRSA infection.Reference Popoola, Budd and Wittig ^{4 ,} Reference Huang, Chou and Su ⁵ Some NICUs treat MRSA-colonized neonates with intranasal mupirocin to prevent MRSA transmission and infections. Data are limited on the prevalence of mupirocin resistance among MRSA in US NICUs in the setting of selective mupirocin exposure. Thus, we evaluated the prevalence of mupirocin resistance among unit-acquired MRSA isolates from a NICU with a long-standing active surveillance culture and targeted decolonization program.

The Johns Hopkins Hospital (JHH) is a tertiary-care hospital with an imbedded 200-bed Children’s Center that houses a 45-bed, level IIIC NICU. The NICU’s MRSA surveillance program has been previously described in detail and includes weekly nares cultures of all neonates and admission nares cultures of neonates admitted from home or from an outside hospital.Reference Popoola, Budd and Wittig ^{4 ,} Reference Popoola, Carroll, Ross, Reich, Perl and Milstone ⁶ MRSA-colonized neonates receive intranasal mupirocin twice daily for 5 days and some neonates receive 2 baths with 2% chlorhexidine gluconate-impregnated cloths. We retrospectively identified a cohort of neonates with positive MRSA cultures who were in the NICU between April 1, 2007, and August 31, 2013. Unit-acquired MRSA was defined as a MRSA isolate obtained from a neonate who was born at JHH or who was not born at JHH and had a surveillance or clinical culture obtained ≥3 days after NICU admission that grew MRSA. The first available MRSA isolates from neonates were tested for mupirocin susceptibility using Etest (bioMerieux SA, Marcy l’Etaile, France). Minimum inhibitory concentrations (MICs) were determined using standard Clinical and Laboratory Standards Institute (CLSI) methodology. ⁷ Mupirocin susceptibility was defined as an MIC ≤4 μg/mL, low-level mupirocin resistance (LLMR) was defined as MIC between 8 and 64 μg/mL, and high-level mupirocin resistance (HLMR) was defined as MIC ≥512 μg/mL.Reference Patel, Gorwitz and Jernigan ^{1 , 7} We performed pulsed-field gel electrophoresis (PFGE) on mupirocin-resistant isolates as previously described and considered isolates to be related if their patterns had 3 or fewer band differences.Reference Popoola, Budd and Wittig ^{4 ,} Reference Popoola, Carroll, Ross, Reich, Perl and Milstone ⁶ This study was approved by the Johns Hopkins University Institutional Review Board.

During the study period, 4,030 neonates were admitted to the NICU; 101 neonates (2.5%) had a culture that grew MRSA, and 20 of those 101 (0.5%) developed a MRSA infection. A total of 87 MRSA isolates (86.1%) were acquired in the hospital unit; 71 neonates with MRSA (70.1%) received mupirocin as part of a decolonization protocol. A total of 84 MRSA isolates were available for mupirocin susceptibility testing, of which 73 (86.9%) were unit-acquired. Only 3 isolates (3.6%) were LLMR strains (Table 1), and 2 isolates (2.4%) were unit-acquired MRSA strains. One neonate had an LLMR isolate cultured from an abdominal wound 6 weeks into the NICU hospitalization. This patient was treated with mupirocin and intravenous antibiotic and had resolution of clinical infection. The median mupirocin MIC of all isolates was 0.19 μg/mL. The 3 isolates with LLMR were different PFGE types, including USA300, USA100, and a unique strain (Table 1).

TABLE 1

Clinical Characteristics and Outcomes of Neonates with Mupirocin-Resistant MRSA Isolates

NOTE. MRSA, methicillin-resistant Staphylococcus aureus; MIC, minimum inhibitory concentration; PFGE, pulsed-field gel electrophoresis.

^a Of 83 available MRSA isolates, 3 isolates (3.6%) had low-level mupirocin resistance strains.

^b Isolates from cases 2 and 3 were unit-acquired MRSA strains. The mean (SD) time to positive culture from hospitalization for unit-acquired mupirocin-susceptible MRSA strains was 34.1 (35.8) days.

^c Patient received intranasal mupirocin and intravenous antibiotic for MRSA abdominal wound infection.

Our study found a low prevalence of mupirocin resistance among unit-acquired MRSA isolates in a NICU. Only LLMR isolates were identified. LLMR in S. aureus is mediated via the mutation of the native isoleucyl RNA synthetase gene (ileS), and the clinical significance is unclear.Reference Patel, Gorwitz and Jernigan ^{1 –} Reference Popoola, Budd and Wittig ⁴ In contrast, HLMR in S. aureus is mediated by a plasmid-encoded mupA gene and has been associated with decolonization failure.Reference Patel, Gorwitz and Jernigan ^{1 –} Reference Popoola, Budd and Wittig ⁴ In the context of targeted decolonization for MRSA colonized neonates, there was no evidence of HLMR or selection of a predominant clonal strain. Our findings are consistent with previous studies reporting successful programs that included mupirocin prophylaxis to prevent MRSA transmission and infections in high-risk neonates without emergence of mupirocin resistance.Reference Delaney, Wang and Melish ^{8 ,} Reference Hitomi, Kubota and Mori ⁹ Delaney et alReference Delaney, Wang and Melish ⁸ did not detect mupirocin resistance in 19 MRSA isolates collected during a 5-year campaign of universal mupirocin prophylaxis. Hiromi et alReference Hitomi, Kubota and Mori ⁹ found that the median mupirocin MIC slightly increased from 0.25 to 0.75 μg/mL after 1 month of universal mupirocin prophylaxis use without HLMR emerging. Previous studies in other populations have detected dramatic increases in mupirocin resistance after broad use.Reference Miller, Dascal, Portnoy and Mendelson ¹⁰ Surveillance is needed to monitor LLMR and HLMR as mupirocin use increases in many hospital settings including NICUs, adult ICUs, and preoperative patients.

Some limitations of our study should be considered. Among the isolates, 17 were not available for testing, but these were not systematically excluded and should not bias the results. We have reported only the first MRSA isolate from each patient. Serial isolates were not available from patients to identify possible selection of mupirocin resistance within a patient exposed to mupirocin decolonization. If patients did develop mupirocin-resistant strains after treatment, we likely would have detected mupirocin resistance in other unit-acquired strains.

In conclusion, a targeted MRSA decolonization program in neonates over a 7-year period did not lead to the emergence of mupirocin resistance in unit-acquired MRSA isolates. Mupirocin decolonization in neonates coupled with active surveillance cultures is one strategy to prevent MRSA transmission and infections in the NICU. Mupirocin susceptibility testing of the MRSA isolates should be considered in hospitals systematically decolonizing patients to detect selection of mupirocin-resistant strains.