Results from this study showed that the mean one-time point MRSA carriage rate of patients receiving hemodialysis in northern Taiwan was 3.8%, and the rate was up to 6.9% for those receiving two-time point surveys. For patients receiving hemodialysis, the nasal MRSA carriage rate was different, with a wide range, in different countries and regions, and also different with the different number of samplings. In previous reports from Taiwan, the average one-time point MRSA carriage rate among patients receiving hemodialysis was 2.4% (of 509 patients) in a report from southern Taiwan , and the rate was up to 9.5% of 306 patients with two serial surveys in another study from northern Taiwan . Both rates were comparable to those in the present study.
For one-time point survey, the rate of nasal MRSA carriage (3.8%) among the patients receiving hemodialysis in the present study was also similar to that of adult patients visiting emergency room (3.8% of 502 patients)  and that of adults for health examination (3.8% of 3098 adults) reported from Taiwan . Since most of the patients receiving hemodialysis came from community settings and both out-patient hemodialysis clinics were located in independent spaces of both hospitals, these patients were not frequently exposed to the patients with at-risk for MRSA acquisition, such as patients in intensive care units (ICU).
In the present study, we also found that no significant difference for nasal S. aureus, either MSSA or MRSA, carriage rate was found between patients treated at the medical center and the local hospital. Though the patients receiving hemodialysis may be associated with an increased risk of MSSA or MRSA colonization [20, 25, 26], no additional significant risk factor for S. aureus was identified among patients receiving hemodialysis in the present study. No significant difference was found, either, between the isolates from both hospitals in terms of antibiotics susceptibilities and molecular characteristics. These results suggest that the characteristics of the patients as well as the environments of out-patient hemodialysis at a local hospital and a medical center in northern Taiwan were similar.
Molecular characterization of all MRSA isolates in the present study showed that 70% of the isolates shared a common characteristics of endemic CA-MRSA clones (ST 59 or its single locus variant, 338) in Taiwan . The remaining 6 isolates carried type IV or V SCCmec, also suggesting community strains . The scenario of community strains being transmitted to healthcare facilities was indicated. The patient with the isolate of ST 30 had a travel history to other Asian countries, where the strains of ST 30 prevailed in the community; whether the patient acquired the isolate abroad needs further studies.
Scanty studies assessed the effect of MRSA decolonization for patients receiving hemodialysis and indicated that these patients might benefit from decolonization, though repeated courses of treatment are needed and the effects are modest . In the present study, only one of 9 MRSA colonizers without successful elimination was persistent colonizer; in contrast, without elimination procedures, nearly two-thirds of 28 MSSA colonizations were persistent. Furthermore, with successful decolonization for most colonizers, no significant difference was found between the patients with and without MRSA colonization for subsequent MRSA infection, though the size of the patient’s number was small. These findings support that patients receiving hemodialysis with MRSA colonization may be successfully decolonized with intranasal mupirocin treatment plus chlorhexidine bath and might benefit from the reduction of subsequent MRSA infection to a rate comparable to that among those without MRSA colonization. The issue whether decolonization of MRSA may reduce subsequent infection in this population needs a large scale randomized control study.
There are several limitations for the present study. First, less than 60%, lower than expected, of the patients receiving hemodialysis in CGMH participated in this study, which reduced the size of case number for evaluation and indirectly affected the analysis of statistic significance. Second, from each study subject, samplings for MRSA detection were obtained only from one site (nares) once or twice, so some MRSA colonizing patients might be undetected . Third, for each patient receiving de-colonization therapy, only one follow-up sampling was obtained one week after treatment, which might be inadequate for the proof of successful decolonization [30, 31]. It is an issue that how many follow-up samplings and how long the observation duration are adequate for the proof of successful decolonization.