Unrecognized burden of methicillin-resistant Staphylococcus aureus and vancomycin-resistant Enterococcus carriage in the pediatric intensive care unit.
ABSTRACT Routinely, children's hospitals use data from clinical cultures to estimate the burden of methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus (VRE) carriage. In our pediatric intensive care unit, a program of weekly surveillance cultures uncovered an unrecognized reservoir of MRSA and VRE carriers. This weekly surveillance enabled more accurate estimates of the incidence rates of MRSA and VRE carriage and led to an increased number of isolation-days for patients.
- SourceAvailable from: Paul D Stamper[show abstract] [hide abstract]
ABSTRACT: We compared the BD GeneOhm methicillin-resistant Staphylococcus aureus (MRSA) PCR assay to culture with BBL CHROMagar MRSA for nasal surveillance among 602 arrestees from the Baltimore City Jail. The sensitivity and specificity were 88.5% and 91.0%, respectively, and after secondary analysis using enrichment broth, they were 89.0% and 91.7%, respectively. Twenty-three of 42 false-positive PCR lysates contained methicillin-susceptible S. aureus.Journal of clinical microbiology 03/2008; 46(2):743-6. · 4.16 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: A meta-analysis was performed to summarize the impact of methicillin-resistance on mortality in Staphylococcus aureus bacteremia. A search of the MEDLINE database for studies published during the period of 1 January 1980 through 31 December 2000 and a bibliographic review identified English-language studies of S. aureus bacteremia. Studies were included if they contained the numbers of and mortality rates for patients with methicillin-resistant S. aureus (MRSA) and methicillin-susceptible S. aureus (MSSA) bacteremia. Data were extracted on demographic characteristics of the patients, adjustment for severity and comorbid illness, source of bacteremia, and crude and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for in-hospital mortality. When the results were pooled with a random-effects model, a significant increase in mortality associated with MRSA bacteremia was evident (OR, 1.93; 95% CI, 1.54-2.42; P<.001); significant heterogeneity was present. We explored the reasons for heterogeneity by means of subgroup analyses. MRSA bacteremia is associated with significantly higher mortality rate than is MSSA bacteremia.Clinical Infectious Diseases 01/2003; 36(1):53-9. · 9.37 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Active screening for vancomycin-resistant enterococci (VRE) in rectal and stool specimens has been recommended to limit the spread of antimicrobial resistance within certain high-risk populations. Directly from 502 rectal swabs and stool specimens, we evaluated the diagnostic performance of the BD GeneOhm VanR assay (BD GeneOhm, San Diego, CA), a rapid real-time PCR test that detects the presence of vanA and/or vanB genes. The VanR assay was compared to culture consisting of both bile-esculin-azide agar with 6 mug/ml vancomycin (BEAV agar) (BD Diagnostics, Sparks, MD) and BEAV broth with 8 mug/ml vancomycin (Hardy Diagnostics, Santa Maria, CA). Enterococci were identified to the species level using standard biochemical tests and a Phoenix automated microbiology system (BD Diagnostics, Sparks, MD). The susceptibility of the enterococci to vancomycin and teicoplanin was determined using an Etest (AB Biodisk, Solna, Sweden). VRE were initially isolated from 147 cultures, and the VanR assay detected 142 of the 147 positive cultures for a sensitivity of 96.6%. The specificity was 87.0% (309/355) largely due to false positives seen with the vanB portion of the assay. The sensitivity when testing rectal swabs was 98.3%, and the sensitivity for stool samples was 95.4% (P = 0.643). The specificity of rectal swabs was comparable to that of the stool specimens (87.5% and 86.5%, respectively). When used only to detect VanA resistance, the VanR assay was 94.4% (136/144) sensitive and 96.4% (345/358) specific, with positive and negative predictive values of 91.3% and 97.7%, respectively. In summary, the BD GeneOhm VanR assay is a good screening test for VRE in our population of predominantly vanA-colonized patients. However, patient samples testing only vanB positive should be confirmed by another method for the presence of VRE.Journal of Clinical Microbiology 11/2007; 45(10):3360-5. · 4.07 Impact Factor
infection control and hospital epidemiologydecember 2008, vol. 29, no. 12
Unrecognized Burden of Methicillin-
Resistant Staphylococcus aureus and
Carriage in the Pediatric Intensive
Aaron M. Milstone, MD; Xiaoyan Song, PhD, MD, MSc;
Claire Beers, RN, MSN; Ivor Berkowitz, MD;
Karen C. Carroll, MD; Trish M. Perl, MD, MSc
Routinely, children’s hospitals use data from clinical cultures to es-
timate the burden of methicillin-resistant Staphylococcus aureus
(MRSA) and vancomycin-resistant Enterococcus (VRE) carriage. In
our pediatric intensive care unit, a program of weekly surveillance
cultures uncovered an unrecognized reservoir of MRSA and VRE
carriers. This weekly surveillance enabled more accurate estimates
of the incidence rates of MRSA and VRE carriage and led to an
increased number of isolation-days for patients.
Infect Control Hosp Epidemiol 2008; 29:1174-1176
Multidrug-resistant organisms (MDROs), including methi-
cillin-resistant Staphylococcus aureus (MRSA) and vancomy-
cin-resistant enterococci (VRE), cause approximately 70% of
healthcare-associated infections.1The transmission of these
MDROs can result in an increased incidence of healthcare-
associated infections that have increased financial costs, mor-
bidity, and mortality.2,3
Little is known about the epidemiology of endemic MRSA
and VRE colonization and transmission in the pediatric in-
tensive care unit (PICU). The burden of MRSA and VRE
infection or colonization is better characterized in adult pa-
tients in intensive care units, where surveillance cultures en-
able the early identification and isolation of carriers and may
Numerous states have implemented or are considering leg-
islation mandating the screening of hospitalized patients for
MRSA and other MDROs. Given the unproven benefit of
routine surveillance in a pediatric patient population, many
children’s hospitals do not screen high-risk children for
MDRO colonization. Instead, children’s hospitals often assess
MRSA and VRE prevalence and incidence rates by relying on
data from microbiologic cultures performed in the course of
clinical care, a practice less sensitive at detecting carriers.6
Our objectives were to assess whether findings of clinical
cultures reflect the burden of MRSA and VRE carriage in the
PICU and whether screening for MRSA and VRE would in-
crease the number of isolation-days for patients and help us
more accurately estimate incidence rates.
Setting and surveillance program.
Hopkins Hospital, a 26-bed tertiary care PICU, admits med-
ical and surgical patients, including patients who have re-
ceived a hematopoietic stem cell transplant or an organ trans-
plant. Beginning in 2006 as partof ahospitalqualityassurance
project, all patients in the PICU had nares and rectal culture
samples obtained weekly for the detection of MRSA and VRE,
respectively. Given our limited resources and the fact that
there were no standard practice guidelines in pediatrics, ad-
mission cultures were not performed. A computerized sur-
veillance system (Theradoc) was used to identify clinical mi-
crobiologic cultures of samples from any body site that grew
MRSA and VRE during the specified period.
Laboratory methods.Nasal swab samples were plated on
an MRSA chrome agar medium (BBL CHROMagar MRSA;
BD Diagnostics).7Rectal samples were plated on bile esculin
azide agar with 6 mg/mL of vancomycin (BD Diagnostics).
Colonies that were suspected of being MRSA were subcul-
tured onto 5% sheep blood agar plates with a 30-mg van-
comycin disk and confirmed as resistant to vancomycin by
use of the Etest (AB Biodisk).8
Definitions. Prevalent carriers were defined as those who
during their PICU stay had a surveillance or clinical culture
positive for MRSA or VRE. Newly identified carriers were a
subset of prevalent carriers who had no history of MRSA or
VRE colonization or infection at our institution. Incident
cases were those with no institutional history of MRSA or
VRE infection or colonization who had a surveillance or clin-
ical culture positive for MRSA or VREfromasampleobtained
at least 2 days after admission to the PICU.
Data analysis.Prevalence ratios, incidence rate ratios
(IRRs), and 95% confidence intervals (CIs) were calculated
using Stata, version 9.0 (Stata).
The PICU at the Johns
During the 12-month period of our study, 1,501 patients were
admitted to the PICU. The mean patient age was 6.5 years,
and 826 (55%) of the 1,501 patients were male. The mean
length of PICU stay was 4.3 days. Weekly surveillance culture
samples were obtained from 979 patients (65%), with both
nares and rectal culture samples being obtained from a total
of 930 patients (for a compliance rate of 95%). Of those 979
patients screened, 42 (4.3%) were colonized with MRSA, and
31 patients (3.2%) with VRE. The performance of weekly
surveillance cultures increased the detection of MRSA carriers
by 100% and VRE carriers by 350% (Table).
The prevalence of MRSA colonization was 4.1 cases per
1,000 patient-days at risk, on the basis of results of clinical
unrecognized burden of mrsa or vre carriage in the picu 1175
With Methicillin-Resistant Staphylococcus aureus (MRSA) and Van-
comycin-Resistant Enterococci (VRE) in a Pediatric Intensive Care
Unit During a 12-Month Period, as DeterminedbyClinicalCultures,
Weekly Surveillance Cultures, or a Combination of Both
Estimated Prevalence and Incidence Rates of Colonization
Newly identified carriers
Newly identified carriers
aSome overlap may exist because patients may have had a clinical culture
and a surveillance culture positive for MRSA or VRE.
bPer 1,000 patient-days at risk.
Data are no. (%) of total identified patients, unless otherwise
resistant Staphylococcus aureus (MRSA) and vancomycin-resistant
enterococci (VRE) newly identified by surveillance and clinical cul-
tures, plotted by the month during which they were identified.
The number of cases of colonization with methicillin-
cultures, compared with 8.6 cases per 1,000 patient-days at
risk, on the basis of results of both clinical cultures andweekly
surveillance cultures combined (prevalence ratio, 0.48 [95%
CI, 0.29–0.78]). The prevalence of VRE colonization was 1.1
cases per 1.000 patient-days at risk, on the basis of results of
clinical cultures, compared with 5.3 cases per 1,000 patient-
days at risk, on the basis of both clinical cultures and weekly
surveillance cultures combined (prevalence ratio, 0.21 [95%
CI, 0.08–0.48]). The results of clinical cultures alone under-
estimated the prevalence of MRSA by 52% and the prevalence
of VRE by 79%. Similarly, the results from clinical cultures
alone underestimated the MRSA and VRE incidence rates by
67% (IRR, 0.40 [95% CI, 0.14–1.0]) and 100% (IRR, 0 [95%
CI, 0–0.20]), respectively. In the Figure, the number of newly
identified cases of MRSA and VRE colonization were plotted
by the month during which they were identified, to illustrate
the proportion of unrecognized MRSA and VRE carriers de-
tected by surveillance.
Early identification of colonization accounted for 189 of the
1,163 days that patients harboring MRSA were placed under
contact precautions and 391 of the 1,243 days that patients
harboring VRE were placed under contact precautions. There-
fore, the use of surveillance cultures increasedthe totalnumber
of days patients harboring MRSA and patients harboring VRE
were in isolation by 19% and 46%, respectively.
The burden of MDRO colonization in children hospitalized
in the PICU and the potential benefit of impending public
policy that mandates MDRO screening in the PICU are un-
known. In our preliminary attempt to quantify the burden
of MDRO colonization in the PICU, we demonstrated that
relying on the results of clinical cultures alone underestimated
the prevalence of MRSA and VRE and their incidence rates.
The use of weekly surveillance cultures increased the number
of detections of MRSA and VRE carriers in the PICU by
100% and 350%, respectively.
Our findings agree with those of studies demonstrating that
screening of adult ICU patients identifies MRSA and VRE
colonized individuals weeks before clinical cultures can and
enables early isolation of patients.9,10In our study, the use of
weekly surveillance cultures increased the number of days
that patients harboring MRSA and patients harboring VRE
were appropriately isolated by 19% and 46%, respectively.
In the absence of surveillance, the incidence of VRE col-
onization appeared to be a fraction of the incidence of MRSA
colonization in our PICU. Using surveillance cultures, we
found that the incidence rates of MRSA andVREcolonization
were nearly identical. Although the number of clinical VRE
isolates was low, the transmission of VRE from patient to
patient is likely more common than is recognized. Because
few hospitalized children have their stool cultured, most VRE
carriers would go undetected without the use of surveillance
Whether the use of surveillance cultures will identify chil-
dren in the PICU who are carriers and thus reduce the fre-
quency of transmission and subsequent infections is un-
known. The healthcare-associated MRSA and VRE infection
rates were not statistically different during the 1-year inter-
vention period when weekly surveillance cultures were used
and during the 1-year preintervention period (data not
shown). Whether surveillance is sufficiently beneficial in im-
proving patient outcomes to justify its cost is an important
area for further research.
We report these results cautiously, because in the absence
of admission surveillance cultures, incident cases may have
been misclassified. As a result, our data likely underestimated
1176 infection control and hospital epidemiologydecember 2008, vol. 29, no. 12
the true burden of MRSA and VRE carriage in our PICU,
and likely overestimated the corresponding incidence rates.
Future studies should evaluate the added impact of admission
surveillance on isolation days and rates of MDRO transmis-
sion in the PICU, but our study highlighted the unrecognized
burden of MRSA and VRE carriage in the PICU.
As pressure mounts to increase the use of surveillance for
MDRO in the healthcare setting, further studies are crucial
to determine (1) the impact of MRSA and VRE infection or
colonization in hospitalized children; (2) whether the use of
routine surveillance improves the outcomes of children hos-
pitalized in the PICU; (3) whether the use of routine sur-
veillance for MDROs in a pediatric population is cost-effec-
tive; and (4) the impact of increased surveillance and isolation
of children dependent on others for their daily care. As we
develop infection prevention strategies and public policies to
target MRSA and VRE infection or colonization, we should
not neglect children, a high-risk group, and an unrecognized
reservoir of MRSA and VRE.
We thank K. Alex Shangraw, MSPH; Kathleen Speck, MPH; and the micro-
biology laboratory staff, the pediatric intensive care unit nursing staff, and
the hospital epidemiology and infection control group of Johns Hopkins
Hospital for their support of this study. We thank Dr. Timothy Townsend
for careful review of the manuscript.
Financial support.A.M.M. was supported by a Pediatric Infectious Dis-
eases Society Fellowship Award funded by an educational grant from
AstraZeneca Pharmaceuticals and a Johns Hopkins Clinical Research Career
Development grant (K12 RR023266). T.M.P. is supportedbyCDCgrantUR8/
Potential conflicts of interest. A.M.M. and T.M.P. have grant support from
Sage Products. T.M.P. has grant support from 3M, is on an advisory panel
for Theradoc and Replidyne, and has given talkssponsoredbyOrtho-McNeil.
K.C.C. has received research and speaker honoraria from BD Diagnostics,
BD-GeneOhm, and Cepheid Diagnostics. All other authors report no con-
flicts of interest relevant to this study.
From the Department of Pediatrics, Division of Pediatric Infectious Dis-
eases (A.M.M.), the Department of Medicine, Division of Infectious Diseases
(X.S., K.C.C., T.M.P.), the Department of Anesthesiology and Critical Care
Medicine (I.B.), and the Department of Pathology, Division of Medical Mi-
crobiology (K.C.C.), Johns Hopkins University School of Medicine, and the
Department of Hospital Epidemiology and Infection Control (A.M.M., X.S.,
T.M.P.) and the Children’s Center (C.B.), Johns HopkinsHospital,Baltimore,
Address reprint requests to Aaron M. Milstone, MD, Johns Hopkins Uni-
versity, Departments of Pediatric Infectious Diseases and Hospital Epide-
miology and Infection Control, 600 North Wolfe St., Osler 425, Baltimore,
MD 21287 (firstname.lastname@example.org).
Presented in part: 17th Annual Scientific Meeting of the Society of Health-
care Epidemiology of America; Baltimore, Maryland; April 2007 (Abstract
Received March 14, 2008; accepted July 11, 2008; electronically published
November 4, 2008.
? 2008 by The Society for Healthcare Epidemiology of America. All rights
reserved. 0899-823X/2008/2912-0013$15.00. DOI: 10.1086/592093
1. AntimicrobialResistance inHealthcareSettings.CentersforDiseaseCon-
trol and Prevention. Available at: http://www.cdc.gov/drugresistance/
healthcare/problem.htm. Accessed January 4, 2008.
2. Cosgrove SE, Sakoulas G, Perencevich EN, Schwaber MJ, Karchmer AW,
Carmeli Y. Comparison of mortality associated with methicillin-resistant
and methicillin-susceptible Staphylococcus aureus bacteremia: a meta-
analysis. Clin Infect Dis 2003; 36:53-59.
3. Patel R. Clinical impact of vancomycin-resistant enterococci. J Antimi-
crob Chemother 2003; 51(Suppl 3):iii13-21.
4. Huang SS, Rifas-Shiman SL, Pottinger JM, et al. Improving the assess-
ment of vancomycin-resistant enterococci by routine screening. J Infect
Dis 2007; 195:339-346.
5. Huang SS, Rifas-Shiman SL, Warren DK, et al. Improving methicillin-
resistant Staphylococcus aureus surveillance and reporting in intensive
care units. J Infect Dis 2007; 195:330-338.
6. Salgado CD, Farr BM. What proportion of hospital patients colonized
with methicillin-resistant Staphylococcus aureus are identified by clinical
microbiological cultures? Infect ControlHosp Epidemiol2006; 27:116-121.
7. Farley JE, Stamper PD, Ross T, Cai M, Speser S, Carroll KC. Comparison
of the BD GeneOhm methicillin-resistant Staphylococcus aureus (MRSA)
PCR assay to culture by use of BBL CHROMagar MRSA for detection
of MRSA in nasal surveillance cultures from an at-risk community pop-
ulation. J Clin Microbiol 2008; 46:743-746.
8. Stamper PD, Cai M, Lema C, Eskey K, Carroll KC. Comparison of the
BD GeneOhm VanR assay to culture for identification of vancomycin-
resistant enterococci in rectal and stool specimens. J Clin Microbiol
9. Zuckerman RA, Steele L, Venezia RA, Tobin EH. Undetected vanco-
mycin-resistant Enterococcus in surgical intensive careunitpatients.Infect
Control Hosp Epidemiol 1999; 20:685-686.
10. Warren DK, Guth RM, Coopersmith CM, Merz LR, Zack JE, Fraser VJ.
Epidemiology of methicillin-resistant Staphylococcus aureus colonization
in a surgical intensive care unit. Infect Control Hosp Epidemiol 2006; 27: