Is Universal Screening of Children for Nasal Methicillin-Resistant Staphylococcus aureus Colonization Necessary on Hospital Admission?

Abstract

Methicillin-resistant Staphylococcus aureus (MRSA) infections have increased significantly over the last decade. Increased prevalence and concern for nosocomial spread has led many states to mandate screening for nasal MRSA carriage. In December 2007, our institution implemented universal surveillance of all admitted patients. Nasal MRSA colonization rates and the role of universal surveillance in prevention and control of MRSA transmission in pediatric wards is not well characterized. The objectives of our study were to determine the incidence of nasal MRSA colonization, the concordance between nasal MRSA colonization and culture-proven MRSA disease at admission, and the impact of universal surveillance on MRSA transmission among children admitted to our pediatric ward. Loyola University Medical Center is an academic center with a pediatric hospital within the larger hospital. Of the 123 pediatric beds, 34 are ward beds. All children admitted from December 1, 2007, through June 30, 2010, were screened for nasal MRSA colonization, using the Cepheid GeneXpert system (Cepheid), and they were identified through the Epic electronic medical record system (Epic Systems). Contact isolation precautions were instituted if a polymerase chain reaction test or a clinical culture was positive. Microbiology laboratory data were used to identify patients with a positive nasal screen or MRSA cultures. Data on hospitalization within the previous 12 months, presence of an indwelling device, and site of the positive MRSA culture were collected. Recent antibiotic use and family history of MRSA disease were not analyzed because these data could not be reliably ascertained from charts. Antimicrobial susceptibilities of MRSA clinical isolates against oxacillin, gentamicin, cefazolin, rifampin, tetracycline, clindamycin, erythromycin, trimethoprim-sulfamethoxazole, and vancomycin were recorded. Isolates were considered to be community-associated MRSA (CA-MRSA) if they were resistant only to b-lactams erythromycin. Resistance to other antibiotics tested was considered to indicate healthcareassociated MRSA (HA-MRSA). Infection control data were reviewed to determine nosocomial MRSA cases in the pediatric ward. Four thousand and eighty-four patients were admitted to the pediatric ward, and 3,934 were screened for nasal MRSA colonization during the study. Of these, 234 patients (5.7%) were colonized, and 99 had culture-proven MRSA disease (2.4%). MRSA nasal colonization and disease rates were uniform throughout the study period. MRSA colonization and disease were highest among young children; 8.4% of infants less than 1 year old were colonized versus 5.2% of children 1 or more years old ( ). MRSA disease was more P p .004 prevalent in children 5 years old and under, 3.6% versus 1% in those more than 5 years old ( ). The incidence of P ! .002 nasal colonization markedly decreased in older age groups (Table 1). Most pediatric patients with MRSA colonization or disease had no identifiable risk factors; 9% had an indwelling catheter, and 39% were hospitalized in the year before. Most positive MRSA cultures (89%) were from skin and soft tissue sites, and 84% were CA-MRSA. Bacteremia was noted in 5 patients, 3 of whom had concurrent osteomyelitis. Only 54% of patients admitted with MRSA disease had a positive nasal screen. The use of MRSA nasal colonization to predict MRSA disease on admission had a sensitivity of 54% and a specificity of 96%. The positive predictive value was 24%, and the negative predictive value was 99%. Nosocomial MRSA infections were infrequent in our pediatric ward. There were 2 cases in the 2 years before universal surveillance and 1 during universal surveillance. The overall cost of universal screening of pediatric ward patients for the institute for the study period was $204,200, at $50 per test. Nasal MRSA colonization rates vary by geographic area and have increased over the last decade from less than 1% to 10%. Nasal MRSA colonization rate in admitted children was 5.7%, which is similar to the rate in our adult population (∼7%). The majority of colonized or infected children had no risk factors for HA-MRSA (!60%). Skin and soft tissue infections, mainly due to CA-MRSA, comprised the bulk of MRSA disease. Concerns for nosocomial spread of CA-MRSA have led many states to mandate screening of hospitalized patients and to observe contact precautions of identified carriers. While absence of MRSA nasal colonization is highly predictive of absence of disease, we found poor concordance between