Abstract
Background:
Staphylococcus aureus is an important cause of cardiac surgical site infection. Based on studies in adults, nasal screening to detect S aureuscolonization is used to guide decolonization and selection of prophylactic antibiotics. In our Children’s Hospital, a sensitive polymerase chain reaction (PCR)-based assay is used to screen patients undergoing cardiac surgery for nasal colonization with methicillin-sensitive S aureus (MSSA) and methicillin-resistant S aureus (MRSA). Additionally for patients in diapers, cultures are used to detect MRSA colonization of the groin. The purpose of this study was to determine whether screening two anatomic locations results in a higher MRSA detection rate among children undergoing cardiac surgery.
Methods:
A retrospective chart review determined whether the frequency of bacterial colonization with MRSA differed by anatomic site. Records for 322 pediatric cardiac surgery procedures performed between January 2009 and June 2011 were reviewed. Both a nasal PCR and a second anatomic site culture were performed before 102 procedures.
Results:
The overall rate of colonization with MRSA and MSSA was 4.2% and 29.1%, respectively. Of the seven dually screened patients who tested positive for MRSA, two were identified solely via a groin test, four by nasal screening alone, and one by both the tests. Screening of only the nose would have failed to detect 28.6% of the MRSA cases.
Conclusion:
Preoperative detection of MRSA colonization may be enhanced by screening both the nose and a second anatomic site. The clinical utility of the extranasal MRSA culture was limited due to the long assay turnaround time.
Keywords
Introduction
Infection following open heart surgery results in significant morbidity, increases mortality, prolongs hospital stay, increases the need for antimicrobial therapy, decreases the quality of life, and is associated with higher costs. 1 Although measures are taken to prevent infection, major infections still occur following cardiac surgery. 2,3 Staphylococcus aureus, in both its methicillin-sensitive (MSSA) and methicillin-resistant forms (MRSA), is a leading cause of surgical site infection (SSI), accounting for up to half of all deep surgical wound infections. 2
Studies conducted in adult populations have demonstrated that nasal colonization with S aureusis a risk factor for the development of SSI following heart surgery. 2 To prevent such infections, hospitals typically screen the nose for the bacterium and attempt decolonization using intranasal application of mupirocin ointment alone or in combination with chlorhexidine gluconate baths. 4,5 In addition, patients are treated preoperatively with an appropriately timed prophylactic antibiotic, the selection of which is influenced by the detection of MRSA.
Among patients colonized with S aureus, the nares are not uniformly positive, as patients may be colonized at other anatomic sites. 1,6,7 That, pediatric community-acquired MRSA infection often presents as folliculitis or abscesses of the groin, buttocks, and thighs likely reflects an important role for local colonization. Whether S aureus colonization of the groin is also a risk factor for postoperative health care-associated infections (HAIs) has not been established; however, this is certainly a concern in children as the groin is often used for vascular access, and diapers must be changed frequently. Knowledge of S aureus colonization and sensitivity patterns at various anatomic sites would provide useful information to guide decolonization and the selection of prophylactic antibiotics. The purpose of this study was to determine whether screening both the nose and the groin for S aureus colonization improves the detection of MRSA among diapered children undergoing cardiac surgery.
Patients and Methods
Records of 286 pediatric patients (46% female) who underwent 322 cardiac surgery procedures at the Cleveland Clinic between January 2009 and June 2011 were reviewed. Preoperative screening for S aureus colonization was performed on 93% (266 of 286) of the patients before 92% (297 of 322) of the procedures. Nasal carriage was primarily determined using a commercial real-time polymerase chain reaction (PCR) assay that detects both S aureus and the mecA gene that confers resistance to methicillin (GeneOhm StaphSR; BD, Franklin Lakes, New Jersey). For children still in diapers, our protocol recommended an additional culture to detect MRSA colonization of the groin area, unless the umbilical cord was still attached in which case an umbilical culture was recommended. Cultures and PCR tests were obtained using a dry culturette, using separate swabs for each anatomic site. Screening for MRSA colonization is performed on admission and weekly by culture of the nares and umbilicus in our neonatal intensive care unit (NICU) only. Other than for selected surgical patients, screening for MRSA colonization is not performed routinely in our pediatric intensive care unit. Use of contact precautions for patients colonized with MRSA is limited to patients in the NICU.
Preoperative chlorhexidine baths were preformed routinely the night before and on the day of surgery, unless the child was less than aged two months in which case the bath was with soap and water. The patients who tested positive for S aureus, regardless of the anatomic site or whether MRSA or MSSA, were subsequently treated with conventional mupirocin ointment (98%) twice daily to the nares by swab and if aged two months or older with daily chlorhexidine baths for five days.
Fisher exact test was used to determine whether the frequency of bacterial colonization with MRSA differed by anatomic site (nose or groin) and to compare each individual test (nose or groin) to the gold standard test of screening both locations. The sensitivities and specificities of both the nose and the groin tests were also determined.
Results
Of the 295 procedures that included a screening test for S aureus, 81 (27.3%) were positive. In 265 (82.3%) procedures, screening was performed by a nasal PCR assay that detects S aureus(MSSA and MRSA) colonization. Of these, the respective number of patients colonized with MSSA and MRSA was 67 (25.3%) and nine (3.4%).
The 286 patients underwent 1 (n = 256), 2 (n = 22), 3 (n = 3), or 4 (n = 1) surgical procedures. A total of 20 patients were screened by nasal PCR before two or more surgical procedures. In 14 (70%) of these patients, both the initial and the subsequent swabs were negative. In two (10%) patients, both the initial and the subsequent screens were positive for MSSA. Three patients were negative on the initial screen but were positive for either MSSA (n = 1, 5%) or MRSA (n = 2, 10%) on the subsequent screen. One (5%) patient was positive for MSSA on the initial screen but was negative on the subsequent test.
Screening at two anatomic sites was recommended for patients in diapers. Screens were performed before 148 procedures performed in children under the age of three years, and 117 procedures performed in children aged three years or older (Table 1). In the population aged less than three years, 64.2% (n = 95) were screened at two anatomic sites. The nasal PCR test detected 32 MSSA (21.6%) and 5 MRSA (3.4%) carriers. The extranasal MRSA culture was positive in three (3.1%) patients, and detected two patients with MRSA not identified by nasal PCR, increasing the MRSA rate to 4.7%. The overall rate of colonization with MSSA or MRSA in children younger than three years was 25.7% (38 of 148). Among the 117 children aged three years or older, 6.0% (n = 7) were screened at 2 anatomic sites, all of which were negative. In these older children, 29.9% (n = 35) were colonized with MSSA and 3.4% (n = 4) were colonized with MRSA.
Staphylococcus aureusColonization in Pediatric Patients With Heart Disease Screened Preoperatively by Nasal PCR Alone or in Combination With a Second Anatomic Site MRSA Culture.
Abbreviations: PCR, polymerase chain reaction; MSSA, methicillin-sensitive S aureus; MRSA, methicillin-resistant S aureus.
Both a nasal PCR test and an MRSA culture were performed before 102 procedures at different anatomic sites. Before 94 procedures, screening was performed by both nasal PCR and either a groin (n = 85) or rectal culture (n = 9). Eight neonatal patients had preprocedure screening by nasal PCR and an umbilical swab. Reflecting our screening protocol, dually screened patients were younger than patients screened at one site. The rate of nasal MSSA colonization and median age differed among the screening populations: nares (32.5%; 1535 days), nares plus groin (15.3%; 163 days), nares plus rectal (11.1%; 134 days), and nares plus umbilical (0%; 5 days). The rate of nasal colonization with MSSA and MRSA, as detected by PCR alone in the dually screened patients, was 13.7% (n = 14) and 4.9% (n = 5). Of the 102 patients, 85 had a groin swab (three positive) for MRSA, 9 had a perianal swab (zero positive), and 8 had an umbilical swab (zero positive). Overall, 7 (6.9%) of the 102 dually screened patients tested positive for MRSA at least in one site. Of these patients, four were identified solely via the nose test and two were identified solely via the groin test; only one patient was positive for MRSA in both the sites. One patient was determined by PCR to carry MSSA in the nose but had MRSA detected by groin culture.
Using the Fisher exact test, we tested the hypothesis that screening of only one anatomic site was inferior to screening of both the nose and the groin/rectum (or umbilicus) with regard to MRSA detection. Screening of the groin/rectum (or umbilicus) by culture resulted in a 28.6% increase in the rate of MRSA detection compared to PCR screening of the nose alone (P = not significant). We assumed that the combined ability of the PCR and culture tests to detect MRSA colonization was the “gold standard,” and that the specificity of each screening test was 1. Using this approximation, the respective sensitivities of the nasal and groin tests, compared to performing both tests, were estimated to be 0.71 and 0.43.
Four (1.5%) deep mediastinal infections complicated the 265 procedures that included a nasal PCR. These infections occurred following an initial procedure in four different patients, who were 591 (negative nasal PCR and groin culture), 81 (positive nasal PCR for MSSA, treated with nasal mupirocin), 69 (negative nasal PCR and groin culture), and 7 (negative nasal PCR) days old at the time of surgery. Responsible organisms in order by age included Pseudomonas aeruginosa (n = 2), coagulase negative Staphylococcus, and MRSA. No patient who tested positive for MSSA, MRSA, or both and who was subsequently decolonized contracted a S aureusinfection. Additionally, no patients who were screened in both the nose and the groin contracted an S aureus infection. Nonetheless, the design of our study precludes us from stating that detection and decolonization reduce the rate of deep mediastinal infections due to S aureus. It is notable that the proportion of deep mediastinal infection due to MSSA and MRSA, spanning the time from the introduction of our protocol back to January 1, 2000, was 30.8% (4 of 13) and 7.7% (1 of 13).
A major difference between the PCR screen and the MRSA culture tests is the turnaround time. In our hospital, the laboratory runs the PCR test at 7:00 AM to ensure that the data are available for surgical procedures performed on that day. Culture-based tests, however, are not considered final for four days, so cultures performed two or fewer days before a procedure are unlikely to have results available in time to impact preoperative clinical decision making. The number of days between the test and the surgery for the 265 PCR swabs and the 102 nasal MRSA cultures was similar (PCR screen: mean 6.5 days, median 4 days, and mode 2 days; MRSA culture: mean 6.3 days, median 3 days, and mode 2 days). The PCR tests were performed one or more days ahead of time on 97% (n = 257) of the screened patients, indicating that data would have been available preoperatively for nearly all the patients. In contrast, the data from MRSA cultures may not have been as available for preoperative clinical decision making; indeed 18.6% (n = 21) of the cultures were obtained either on the day of surgery (n = 6) or on the day before surgery (n = 13), and only one of three children with a positive groin MRSA culture received vancomycin as perioperative prophylaxis (a child who was also MRSA positive by nasal PCR).
Comment
In children undergoing surgery for congenital heart disease, infection contributes significantly to morbidity and cost of care. To fully reduce HAIs in these children, quality and patient safety efforts need to address the problem of HAI in a pediatric-specific manner. In this small study, we explored whether there is added benefit in performing an extranasal MRSA culture in addition to a PCR test of the nose, which detects both MSSA and MRSA. In diapered children undergoing cardiac surgery, we find that preoperative detection of MRSA colonization may be enhanced by screening more than one anatomic site. Since MRSA is resistant to the standard antibiotics used for surgical prophylaxis, our observation that nasal screening alone fails to detect 28.6% of patients colonized with MRSA might justify screening children at two anatomic sites. The long turnaround time of the MRSA culture, when compared to the PCR screen, however, limits the utility of the culture-based test in prophylactic antibiotic decision making.
For nasal screening, we used a highly sensitive PCR-based assay that detected MSSA or MRSA colonization. The rate of MSSA colonization exceeded by fivefold the rate of MRSA colonization. Since our PCR assay is validated and cleared only for nasal screening, cultures were used to detect MRSA colonization of the groin. The MRSA cultures are less expensive than our PCR assay but have the disadvantage of longer assay turnaround time and of not detecting MSSA colonization. Whether screening children for groin colonization using a PCR-based approach would improve turnaround time and enhance our ability to detect colonization with MRSA and whether MSSA colonization of the groin is a modifiable risk factor for SSI in children remain unknown. Our data suggest that children may be colonized with MRSA at more than one anatomic site. Given the high overall rate of colonization with S aureus, and the need to decolonize patients with either MSSA or MRSA colonization, an important question is whether a universal nasal and skin decolonization protocol coupled with a rapid screening test for MRSA alone would be simpler and more cost-effective than our multisite search and destroy approach.
Footnotes
Authors’ Note
The research was approved by the Cleveland Clinic’s Institutional Review Board (IRB) under expedited review, as the study was determined to involve minimal risk, using existing data without recorded identifiers.
Acknowledgements
The nurses and nurse practitioners of the Cleveland Clinic Children’s Hospital are thanked for providing dedicated perioperative care to children with heart disease and for implementing protocols to screen for colonization with S aureus.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.