I have my fan club here. All right, so good afternoon. My name is Lexi Crawford. Again, I'm a third-year critical care fellow at Children's National Hospital in D.C. And I'm really excited to talk to you today about my research project, which is entitled the MRSA PCR Nasal Swab, a Tool for Antimicrobial Stewardship in Critically Ill Pediatric Patients. Let me make sure I know how to, okay. This is my About Me slide. Again, I have a specific interest in antimicrobial stewardship as well as medical education and quality improvement. And then myself and my co-authors that are listed here on this slide have nothing to disclose, unfortunately. So our background information, and actually, I apologize, there we go. So as intensivists, we are perpetually concerned about treating patients for presumed bacterial infection, including MRSA. Therefore, patients admitted to the PICU are frequently initiated on antibiotic regimens that include MRSA-active agents, even though MRSA infections are rare in the pediatric population, with an incidence of less than 2% amongst PICU patients in the published literature. In addition, there is a declining prevalence of methicillin resistance among staph aureus in the U.S., a trend that is reflective in our patient population, as the graph on this slide shows trends in methicillin resistance among staph aureus isolates over a seven-year period at Children's National. However, determining which patients warrant anti-MRSA coverage continues to represent a clinical dilemma, as failure to promptly recognize and treat invasive MRSA infection has significant mortality and morbidity outcomes. Therefore, hospitalized patients are often exposed to excessive or suboptimal use of antibiotics, such as vancomycin, as part of their empiric antibiotic regimen. MRSA colonization is an important risk factor for invasive MRSA infection. So the question arises is whether there is a way to supplement our current infectious workup to better guide antimicrobial selection, specifically pertaining to the use of anti-MRSA agents as part of empiric coverage. The MRSA PCR nasal swab is a noninvasive test for MRSA colonization and has been shown to have a high negative predictive value for the presence of invasive MRSA disease in the adult population. However, large studies in the pediatric population are lacking. So this led our team to identify an important area for investigation within the pediatric patient population. The lack of established protocols to quantify a PICU patient's risk of invasive MRSA disease at the outset of an infectious evaluation leads to the initiation and unnecessarily prolonged use of vancomycin, resulting in increased patient risk of iatrogenic kidney injury, development of resistant organisms, and subjection of frequent blood draws for drug level monitoring, as well as kidney function monitoring in this already vulnerable patient population. One of the goals of antimicrobial stewardship is the development of evidence-based, reliable, and low-cost tests, like the MRSA PCR, to help guide appropriate empirical antimicrobial selection, leading to reduction in inappropriate antibiotic courses and ultimately significant risk reduction. Therefore, our study aimed to determine the test characteristics of the MRSA PCR to predict the presence of invasive MRSA infection in the general PICU population. And we hypothesized that the MRSA PCR would have a high negative predictive value and that results in the MRSA PCR could be used to establish a patient's risk for invasive MRSA infection. We designed a retrospective cohort study that was performed in a 44-bed PICU in an academic children's hospital in a large metropolitan area, and cases were identified by querying our electronic medical record for patients who went MRSA PCR testing during a six-year period when it was standard for all of our ICU admissions to be screened with a MRSA PCR on admissions each seven days during their time in the ICU. 7,696 patients aged from birth to 19 years old who were admitted to the PICU between January 2013 and December 2019 with a screening MRSA PCR were first identified. And then we performed a structured chart review on only patients with infectious signs and symptoms based on the pediatric SIRS criteria and cultures that were obtained to evaluate for invasive infection within seven days of the MRSA PCR collection were included in our study, leading to a final study population of 3,139 patients. Importantly, the cutoff of seven days was chosen based on previous literature that suggests that MRSA colonization for inpatients can change within as few as seven days from initial testing. And we also included cultures that were obtained within 24 hours before the MRSA PCR collection to account for workup that was initiated within our emergency room, as well as on the acute care pediatric floor, with literature that supported that even if a patient had been initiated on an anti-staphylococcal antibiotic prior to the MRSA PCR collection, the decrease in sensitivity of the PCR within the first 24 to 48 hours is limited. We then performed a structured chart review and documented demographic, admission, MRSA PCR and clinical data for each patient encounter as listed here in this table. And then elements of past medical history, recent medical history and clinical data as listed in this table were collected for the 29 patients that were identified with invasive MRSA infection. The test characteristics of the MRSA PCR for detecting clinical proven MRSA invasive infection were then calculated and a subgroup analysis of patients with invasive MRSA infection was performed to determine a risk factor analysis to determine which risk factors occurred at highest rates among patients within this cohort. This table shows our patient and diagnostic demographics of our 3,139 unique patient encounters. Fifty-five percent of our patients were male and the median age was 2 years old. Ninety-six percent of our patients were admitted to the PICU on day 0 or 1 of admission to the hospital and 93% of our patients had cultures obtained within 24 hours before and 48 hours after the MRSA PCR collection. There was a 9% MRSA colonization rate amongst patients included in our study and we defined this as a positive MRSA PCR result. And then the date, time and result of all clinical cultures were recorded and 91% of our patients had either a blood or respiratory culture obtained as part of their infectious evaluation. Our results revealed that 25% of our patients had positive clinical cultures confirming some form of invasive bacterial infection. And of these patients, 16% had Staph aureus infection identified. 22% of our Staph aureus infections were identified as MRSA, a percent that was consistent with our hospital's antibiogram. And of the 29 patients with invasive MRSA infection, 83% had a positive MRSA PCR. There were 5 patients with invasive MRSA infection that had a negative MRSA PCR. Invasive MRSA infection was identified in only 0.9% of patients with an incidence rate of 1.6 per 1,000 patient days, confirming that this was a rare event within our patient population consistent with the literature. And a majority of our MRSA cultures were isolated from the respiratory tract. This is kind of our money slide. So this is our chart that shows the statistical analysis of all of our patients on the top column are patients that received treatment with vancomycin, so patients that clinicians might have been more concerned about possible MRSA infection, and then data that is broken up specifically based on culture site. So importantly for all patients, the positive predictive value of the MRSA PCR was determined to be low at only 8.4%. But the negative predictive value was high at 99.8% with a narrow 95% confidence interval. And the post-test probability of a negative PCR test was found to be very low at 0.2%. Analysis of patients treated with vancomycin that represented 44% of our patients in our study and analysis based on our specific culture site revealed similar results with the negative predictive values all greater than 98% and low post-test probabilities for a negative PCR. I then evaluated the risk factors for invasive MRSA infection in the PICU population with the goal to determine factors that may increase a patient's pre-test probability for invasive MRSA infection. And four risk factors listed here occurred at high rates amongst patients with confirmed MRSA infection, with almost all of these patients having at least one risk factor and more than half of the patients having three or more risk factors present. Our studies going through our strengths and limitations, our study's biggest strength is its size. This is the largest study within the pediatric population looking at the MRSA PCR and it complements the results in the adult population. Our limitations include the fact that this was a retrospective study that did preclude more rigorous control over such variables as timing of MRSA PCR collection as well as standardization of our culture workup. And we do acknowledge that our results might not be generalizable in all situations, specifically in patient populations with a higher rate of invasive MRSA infection. Our conclusions, we showed that the MRSA PCR has a poor positive predictive value but a very high negative predictive value and a very low post-test probability for a negative test for invasive MRSA infection in critically ill pediatric patients. The test characteristics of the MRSA PCR nasal swab make it a potentially powerful tool for antimicrobial stewardship in the pediatric patient population. And then finally, there are certain risk factors that occur at high rates among patients with confirmed MRSA infection. In terms of next steps, we're looking at creation of protocols to guide antimicrobial selection based on MRSA PCR results and concurrent risk factors that may lead to significant risk reduction in this vulnerable patient population. And next steps might include further evaluation of our identified risk factors with logistic regression analysis to determine if we can create a tool to guide antimicrobial selection. And then also a prospective multicenter study would allow these results to control for confounding variables including the MRSA swab collection timing as well as the standardization of culture workup to strengthen our evidence. And I'm happy to take any questions.

MRSA PCR nasal swabs

antimicrobial stewardship

invasive MRSA infection

negative predictive value

prospective multicenter study