Burn patients are particularly susceptible to staphylococcal infection.Reference Kaiser, Thompson, Malinoski, Lane and Cinat 1 Methicillin-resistant Staphylococcus aureus (MRSA) is associated with increased resource use and outbreaks, as well as potentially increased morbidity and mortality in this population.Reference Kaiser, Thompson, Malinoski, Lane and Cinat 1 , Reference Girerd-Genessay, Benet and Vanhems 2 Efforts to control MRSA in this vulnerable population include hand hygiene, contact isolation precautions, environmental hygiene, positive pressure rooms with high-efficiency particulate air filtration, and active surveillance cultures (ASCs) upon admission.Reference Kaiser, Thompson, Malinoski, Lane and Cinat 1 , Reference Branski, Al-Mousawi, Rivero, Jeschke, Sanford and Herndon 3 At the study institution, ASC for MRSA and placement in contact isolation of patients with positive cultures were implemented in the burn units in the 1980s.Reference Dansby, Purdue and Hunt 4 The practice was discontinued at the end of August 2014, following the results of a multicenter study in adult critically ill patients that demonstrated that it is possible to achieve control of MRSA without ASCs,Reference Huang, Septimus, Kleinman, Moody, Hickok and Avery 5 even though the study did not include critically ill burn patients, based on the consensus opinion of the infection prevention and control committee. In this study, we describe MRSA rates among hospitalized burn patients before and after this policy change was implemented.
METHODS
The setting was the regional burn center in Parkland Hospital, an 861-bed county tax–supported, tertiary care academic referral center. Patients admitted to the burn center between December 2011 and February 2016 were included in the study. The burn center has a 9-bed intensive care unit (BICU) that admits 28 patients per month on average, an 18-bed acute care unit (BACU) that admits 53 patients per month on average, and a hydrotherapy unit. Patients with burn injuries as well as overflow patients from the medical and surgical intensive care units (ICUs;<20% of total admissions on average) are admitted to the BICU. Notably, the hospital moved all inpatients, including those in the the burn center, to a new facility in August 2015. MRSA control measures include standard precautions and contact isolation precautions for MRSA colonization or infection. Active surveillance cultures (ASC) were obtained from the nares, axilla, groin, and the burn wound of each patient at admission. Cultures were done using routine bacterial culture method in the microbiology lab until October 2008 when the method was changed to use the MRSA Chromagar (Becton Dickinson, Franklin Lakes, NJ). Decolonization was left to the discretion of the treating physician.
All data were obtained from the Department of Infection Prevention and the microbiology laboratory at Parkland Hospital. We then determined the monthly colonization prevalence at admission and incidence of MRSA hospital-onset bloodstream infection (HO-BSI). ‘Hospital-onset’ was defined as time of culture>3 calendar days after admission. Colonization was defined as a positive surveillance culture obtained from any of the surveillance sites.
The ASC period was December 2011–August 2014, during which the ASC program was in place. The follow-up period was September 2014–February 2016, during which the ASC program was no longer in place. The HO-BSI rate was expressed as number of infections per 1,000 patient days, and the rates during the ASC period and the follow-up period were compared for statistically significant difference. The critical level of α was 0.05, and the tests were 2-tailed. This project was undertaken as a Quality Improvement Initiative at Parkland Health and Hospital System, and as such it was not formally supervised by the institutional review board.
RESULTS
During the 33 months from December 2011 to August 2014, in the BICU and BACU, 725 of 4,006 (557 of 2,665 patients in the BICU; 168 of 1,341 patients in the BACU) patients were screened at admission for an overall screening compliance of 18.1%. Of the 725 patients, 60 were found to have MRSA colonization (36 of 557 patients in the BICU; 24 of 168 patients in the BACU) for an admission prevalence of 8.3%. During the same period, the overall incidence of MRSA HO-BSI was 1.23 per 1,000 patient days. The MRSA HO-BSI incidences in the BICU and BACU were 2.59 and 0.31 per 1,000 patient days, respectively. In the 18-month follow-up period from September 2014 to February 2016, when active surveillance cultures were no longer employed as an MRSA control strategy, the overall incidence of MRSA HO-BSI was 1.28 per 1,000 patient days, with incidences of 2.67 and 0.46 per 1,000 patient days in the BICU and BACU, respectively. The difference in incidence during the 2 periods was not statistically significant (P=.91). The incidences in the old Parkland Hospital were 3.72 and 0.50 per 1,000 patient days in the BICU and BACU, respectively. The incidences in the new Parkland Hospital were 0.75 and 0.41 per 1,000 patient days in the BICU and BACU, respectively (P=.09 for BICU and .91 for BACU) (Table 1).
TABLE 1 Rates of Hospital-Onset MRSA Bloodstream Infections Before and After Discontinuation of Active Surveillance Cultures
NOTE. MRSA, methicillin-resistant Staphylococcus aureus; HO-BSI, hospital-onset bloodstream infection; ASC, active surveillance culture; BICU, burn intensive care unit; BACU, burn acute-care unit.
DISCUSSION
In our observational study, standard precautions alone were sufficient to keep the incidence of MRSA HO-BSI at a relatively low rate of 1.28 per 1,000 patient days. Our finding is consistent with previous publications noting the lack of data demonstrating a reduction in MRSA infections in burn patientsReference Branski, Al-Mousawi, Rivero, Jeschke, Sanford and Herndon 3 , Reference Dansby, Purdue and Hunt 4 , Reference Bahemia, Muganza, Moore, Sahid and Menezes 6 – Reference Reighard, Diekema, Wibbenmeyer, Ward and Herwaldt 8 using ASC, particularly in non-outbreak situations. The MRSA incidence did not change after the move to a new hospital building. Our incidence reported here is lower than the rates reported in the study by Johnson et al,Reference Johnson, Nygaard, Cohen, Fey and Wagner 9 which reported a reduction in the incidence of MRSA BSI from 7.45 to 2.4 per 1,000 patient days when a universal decolonization protocol was implemented in a burn unit. The admission prevalence of 8.3% in our burn center is similar to the prevalence of 9.3% reported by Kaiser et al.Reference Kaiser, Thompson, Malinoski, Lane and Cinat 1 Our admission surveillance compliance of 18.1% in the BICU is within the range of 5%–21% reported by a study in which 12 BICUs across the country were examined.Reference Huang, Rifas-Shiman and Warren 10 We conclude that active surveillance cultures and contact isolation of colonized patients may not be critical to reducing incidence of HO-BSI caused by MRSA.
ACKNOWLEDGMENTS
The authors would like to thank Karla Voy-Hatter in Infection Prevention, Linda Byrd in Microbiology, and Steven Wolf, MD, Mark Poteet, and Christine Owens-Lane in the Parkland Burn Center for their support of the study.
Financial support: No financial support was provided relevant to this article.
Potential conflicts of interest: All authors report no conflicts of interest relevant to this article.
