Multisystem inflammatory syndrome in Indian adolescents associated with SARS-CoV-2 infection: a case report

Background Adolescents with coronavirus disease 2019 (COVID-19) associated multisystem inflammatory syndrome (MIS) can present with shock and myocardial injury and mimic Kawasaki disease. Case presentation We describe 4 previously well adolescents (age 13–14 years), presenting with clinical features of MIS in children (MIS-C). All patients had nearly similar clinical presentation. Hematological investigations revealed elevated inflammatory markers, anemia, thrombocytopenia, and decreased neutrophil:lymphocyte ratio. All patients were negative on real-time polymerase chain reaction against severe acute respiratory syndrome coronavirus 2, but had elevated immunoglobulin G titers. Two patients had atypical Kawasaki disease. Three patients had severe disease with hypotensive shock and required intensive care with fluids and inotropes. Two patients required non-invasive respiratory support for dyspnea and one patient had biventricular dysfunction. All received empiric antibiotics, low-molecular weight heparin, steroids, and intravenous immunoglobulin. One patient succumbed, while others recovered well. Conclusions MIS-C may be a late presentation in adolescent with COVID-19. Individualized treatment with empiric antibiotics, immunomodulation, and thromboprophylaxis can result in significantly better outcome.

325. Empiric Antibiotics for COVID-19 and the Utility of Procalcitonin

Background Bacterial coinfection in COVID-19 is infrequent, yet empiric antibiotic use is common. The objectives of this study were to investigate the effect of empiric antibiotics on time to resolution of COVID-19 pneumonia, elucidate the impact of COVID-19 on procalcitonin levels, and determine the incidence of respiratory bacterial coinfection. Methods This was a retrospective study of adult patients hospitalized with COVID-19 between June 1, 2020 and September 30, 2020. Patients were included if they had at least one procalcitonin level. They were excluded if admitted to an intensive care unit within 24 hours of presentation or received antibiotics for an indication besides pneumonia. Patients were stratified into 4 groups based on procalcitonin level and receipt of antibiotics. The primary outcome was time to clinical resolution of pneumonia. A key secondary outcome was incidence of confirmed respiratory bacterial coinfection. Results A total of 199 patients were included. Patients with a procalcitonin greater than 0.25 ng/mL who received antibiotics had a longer median time to clinical resolution of pneumonia, 8 days (95% CI, 4 to 11 days) vs. 3 or 4 days in other groups (P< 0.001). Additionally, this same group required greater baseline oxygen supplementation, had more comorbidities, and increased mortality compared to all other groups. Median time to clinical resolution of pneumonia was also longer in patients who received antibiotics compared to those who did not (5 vs. 4 days, P=0.017) and in those with a procalcitonin greater than 0.25 ng/mL compared to those with PCT less than or equal to 0.25 ng/mL (7 vs. 4 days, P< 0.001). Renal dysfunction was more prevalent in patients with an elevated procalcitonin (45% vs. 17.5%). The overall incidence of confirmed respiratory bacterial coinfection was 1.5%. Conclusion Irrespective of procalcitonin level, empiric antibiotics were not associated with a shorter time to resolution of COVID-19 pneumonia in non-critically ill patients. Elevated procalcitonin is likely a reflection of the severity of COVID-19 disease and baseline renal function rather than bacterial infection. Additionally, the overall incidence of confirmed bacterial coinfection in non-critically ill patients hospitalized with COVID-19 was low. Disclosures Kiya D. Mohadjer, PharmD, BCPS, BCIDP, Eli Lilly and Company (Shareholder)Gilead Sciences (Shareholder)

1124. Evaluation of the Use and Impact of the BioFire FilmArray® Respiratory Panel on Diagnosis and Treatment of Pedidatric Respiratory Infections: A Quality Improvement Project

Background The BioFire FilmArray® Respiratory Panel is a respiratory pathogen PCR that is used to detect the presence of 20 different infectious organisms that cause respiratory illness. Young children are prone to viral upper respiratory illness and represent an age group most likely to receive an antibiotic. As rapid diagnostics evolve, utilizing these tools within antimicrobial stewardship programs to improve utilization of antimicrobials is ideal. The purpose of this study was to assess how the BioFire FilmArray® Respiratory Panel is being used in practice and optimize its use in a large, free-standing, academic children’s hospital. Methods Retrospective chart review evaluating all patients (inpatient and outpatient) that received the BioFire FilmArray® Respiratory Panel from December 1, to December 10, 2019. Patients were evaluated based on where the panel was administered, results of the panel, results of other cultures, utilization of antibiotics, and overall hospital course. Data was collected from the electronic medical record and entered into a REDCap database and then analyzed descriptively. Results 151 patients were included with an average age of 2.6 years with 78 (51.7%) being < 1 year of age. 105 (70%) were administered in either the clinic or ED. In the < 1 year group, 29 (37%) received antibiotic therapy, with 20 having positive viral panels and 11 had positive bacterial cultures. In the ≥ 1 year of age group, 38 (52%) received antibiotic therapy, with 28 having positive viral panels and 9 had positive bacterial cultures. In the outpatient group, 33/105 (31%) were given empiric antibiotics of which 66% had positive viral panels. In the inpatient group, 28/46 (61%) were given empiric antibiotics of which 68% had positive viral panels. Conclusion The BioFire FilmArray® Respiratory Panel was found to be primarily utilized in the young child and outpatient/ED setting. With approximately 67% of children who received empiric antibiotics having a positive viral panel, and the majority of these not having positive bacterial cultures, work can be done to decrease the initiation of empiric antibiotics or earlier discontinuation. Further studies are needed in order to determine the optimal strategy for using the viral panel to de-escalate and escalate antimicrobial therapy in practice. Disclosures All Authors: No reported disclosures

343. RURAL-COVID-19 Trial: Retrospective Analysis of COVID-19 Coinfections in Hospitalized Urban and Rural Adults

Background The impact of COVID-19 in rural communities has been well described. However, little is known regarding differences in coinfections among COVID-19 patients in rural vs. urban settings. Our primary objective is to evaluate community acquired coinfection (CACo) rates (< 72 hrs from admission) and healthcare-associated infection (HAI) rates ( > 72 hrs from admission) in these populations. Secondary objectives include use of empiric antibiotics, pathogen prevalence, and patient outcomes. Methods Retrospective analysis of the first 255 adult patients admitted to a tertiary medical center with symptomatic COVID-19 and confirmed by PCR. Rural and urban categories were determined using patient address and county census data. Isolated pathogens were individually evaluated and considered coinfections if the patient met predetermined criteria. Predetermined Coinfection Criteria Results The rates of CACo for rural (n = 90) and urban (n = 165) residents were 11.1% and 13.3%, respectively. Non-respiratory coinfections, such as bloodstream and urinary tract infections, were more common in urban residents; however, empiric antibiotics were started in 75.1% of all subjects. Methicillin susceptible staphylococcus aureus and Escherichia coli were the most common pathogens isolated on admission in both populations. HAI rates were 13.3% in the rural residents vs 13.9% in the urban residents with Methicillin resistant staphylococcus aureus as the most common respiratory pathogen, although Pseudomonas aeruginosa was the most prevalent overall pathogen. There was no significant difference in hospital length of stay or 30-day all-cause mortality among both populations. Patient Outcomes Among Rural and Urban Populations Conclusion There was no significant difference in the rate of CACo or HAI among rural or urban populations. Despite the high rate of antibiotic use to empirically cover community acquired respiratory infections at the start of the pandemic, only 1.9% of the subjects had a possible or proven respiratory coinfection on admission. Despite prior research showing worse outcomes for rural populations with COVID-19, our data demonstrates that coinfection rates and patient outcomes were similar among these populations when receiving medical care at an academic hospital. Disclosures All Authors: No reported disclosures

1133. Opportunities for Antibiotic Discontinuation and De-escalation after Discharge from the Emergency Department in Pediatric Patients with UTI

Background In children, urinary tract infection (UTI) represents one of the most common indications for antibiotics. While previous data has demonstrated high rates of misdiagnosis and inconsistencies with empiric antibiotics, the impact and opportunities for antibiotic reduction once final culture and susceptibility data are available, particularly in pediatric patients seen in the emergency department (ED), is unknown. Methods This was a retrospective study conducted over a period of 18-months, which included subjects less than 18 years of age who were discharged from the ED with a diagnosis of UTI. Episodes in which urine cultures were negative or grew only mixed urogenital flora were considered possible for discontinuation. De-escalation was considered possible in episodes in which identified bacteria were susceptible to more narrow spectrum agents than the prescribed empiric antibiotic. Rates of discontinuation and de-escalation were calculated as proportions, and excess days of therapy were described. Subjects whose empiric antibiotics were active against isolated bacteria were compared to those with bacteria resistant to empiric therapy. Results A total of 87 episodes of UTI were identified. Pathogenic bacteria were isolated in 51 (59%) of the 78 episodes in which urine cultures were sent, most commonly Escherichia coli (84%). Empiric antibiotic therapy and duration varied and were active against isolated bacteria in 39 (76%) of the episodes. Subjects whose antibiotics were inactive were more likely to be Hispanic and receive cephalexin [Table 3]. Antibiotics were discontinued in 3 of the 27 possible episodes (11%), resulting in 127 extra antibiotic days, median of 6 (IQR=10 days) days per episode. In 20 episodes there was an opportunity for de-escalation, but it was never attempted, leading to 131 extra days of broad-spectrum antibiotics (median 7.5 days, IQR=3). Table 1. Microbiology and resistance profile of bacteria isolated from urine cultures Table 2. Empiric antibiotic regimens, including type of antibiotic and duration Table 3. Comparison of episodes in which empiric antibiotics were active against isolated bacteria versus those in which empiric antibiotics were inactive Conclusion Antibiotics are rarely adjusted after discharge from the ED. Lack of adjustment results in unnecessary total and broad-spectrum antibiotic exposures. Initiatives designed to improve antibiotic use post-discharge could result in significant decreases in unnecessary antibiotics, and ultimately reduced rates of antibiotic resistance. Disclosures All Authors: No reported disclosures

165. Decreased Antimicrobial Consumption and Decreased Rates of Multi-drug Resistant Organisms Following Onset of the COVID-19 Pandemic: Experience from an Australian Tertiary Hospital

Background Current guidelines recommend empiric antibiotics be used only for severe cases of coronavirus disease 2019 (COVID-19) or in cases where there is high clinical suspicion for bacterial co-infection. Level of adherence to guideline-recommended prescribing is unknown and high rates of antimicrobial prescribing may lead to increased development of resistance. Methods We reviewed antimicrobial prescribing patterns for patients with COVID-19 managed at The Alfred Hospital in Melbourne, Australia in 2020. Adherence to World Health Organization (WHO) guideline-based prescribing was assessed by manual review of case notes. Monthly hospital-wide antibacterial consumption April-Dec 2020 (post-pandemic period) was compared to Jan 2019-Mar 2020 (pre-pandemic period), measured as days of therapy (DOT) per 1000 patient-days. Rates of multi-drug resistant organisms (MRO) (including MRSA, VRE, CPE, ESBL) were compared between months in 2019 and 2020 after pandemic onset (April 2020) and expressed as isolates per 1000 patient-days. Results 147 patients were managed for COVID-19 in 2020 at our centre. 101 patients required hospital admission and 58 (39%) were classified as either severe or critical in severity. 80 (54%) patients received empiric antimicrobial treatment, including 78/101 (77%) of hospital inpatients and 24/26 (92%) of ICU-admitted patients. 59 (73%) of antimicrobial prescriptions were adherent to WHO guidelines. Monthly antibacterial consumption was significantly lower post-pandemic than in the pre-pandemic period (mean 853 vs 902 DOT/1000 patient-days, P=0.0065). Antimicrobial use patterns varied, with significant decreases in commonly used antibiotics such as ceftriaxone, piperacillin-tazobactam, azithromycin and ciprofloxacin but no change in vancomycin or meropenem (Figure 1). There was a mean decrease of 0.77 MRO isolates/1000 patient-days (P=0.026) when each month in 2020 was compared with the corresponding month in 2019 (Figure 2). Antibacterial consumption in 2019 and 2020 by month, expressed as days of therapy/1000 patient-days. Rates of isolated multi-drug resistant organisms in 2019 and 2020 by month, expressed as isolates/1000 patient-days. Conclusion A high proportion of admitted patients with COVID-19 received empiric antibiotics. In spite of this, we observed a significant reduction in total antimicrobial consumption and reduced rates of MRO isolation in the post-pandemic period. Disclosures All Authors: No reported disclosures

1366. Deep Neck Infections: Bacteriology and Antimicrobial Susceptibilities, Akron Children’s Experience

Background Deep neck infections (DNI’s) are uncommon (~45,000 US cases annually) but, potentially serious. Published data regarding bacteriology and antibiotic usage for DNI’s in children is limited. In addition, geographic variation in the incidence of pathogens and their antimicrobial susceptibility limits generalization of treatment guidance. Reviewing our practice at Akron childrens we noted considerable variation in the choice of empiric antibiotics (ampicillin-sulbactam vs piperacillin-tazobactam vs Ceftriaxone and Clindamycin/vancomycin/linezolid). Admission unit (floors vs intensive care) and service (hospitalist vs infectious diseases) were some important determinants that influenced choice of empiric antibiotics. This retrospective study aimed to review local data and come up with standard guidance for empiric therapy. Summary of the predominant bacterial isolates. Methods We reviewed records of 125 patients who underwent surgical drainage of DNI’s from 1/2015 – 12/2019. In addition to demographic data we gathered information on bacterial isolates and their susceptibilities. Chart review was performed for patients with staphylococcus aureus, to look for any unique presenting features. Results Up on reviewing the data- peritonsillar abscesses were common in older children (Median age 11 years). As expected, retropharyngeal and parapharyngeal infections were common in younger ones (< 5 years). Group A streptococcus remained the most common aerobic isolate followed by Hemophilus influenzae/parainfluenzae. MRSA was detected in ~7 % of all cultures (see enclosed table). Notably, none of the MRSA isolates were clindamycin resistant. However, MSSA resistance to clindamycin was about 20%. No clinical characters predicted isolation of S. aureus. Anaerobic infections (polymicrobial) were overwhelmingly common across all abscess types. Conclusion Based on our review, Ampicillin-Sulbactam is a good empiric choice antibiotic for deep neck infections in our institution. Ceftriaxone with clindamycin is another option. Clindamycin monotherapy seems to be inadequate. Staph aureus and especially MRSA, were only isolated in a small percentage of cases. Unless a patient is ill appearing, vancomycin use seems unnecessary. Clinical presentation was not helpful to suspect infection with Staph aureus. Disclosures All Authors: No reported disclosures

174. Development of a Machine Learning Prediction Model to Select Empirical Antibiotics in Patients with Clinically Suspected Urinary Tract Infection using Urine Culture Data

Background Increasing antimicrobial resistance and the emergence of superbugs are problems globally. Inappropriate empiric antibiotic use would be a reason to cause antibiotic resistance. However, it has been a challenge to prescribe empiric antibiotics as it is difficult to identify the causative organism beforehand. In this study, we aimed to develop a prediction model to estimate the risk of antibiotics resistance using urine culture tests. Methods The study population included adult patients who had at least one of the results from a urine culture test and antibiotic susceptibility tests (from ampicillin, ceftriaxone, ciprofloxacin, gentamicin, levofloxacin, nitrofurantoin, tetracycline, trimethoprim/sulfamethoxazole) on admission to Ajou University Medical Center. Outcomes were defined as a resistant or intermediate susceptibility. Candidate predictors were diagnosis, prescription, visit, laboratory, procedures of the study population. We split data to 75:25 for training and test. Lasso logistic regression (LLR), extreme gradient boosting machine (XGB), Random Forest (RF) were used as model algorithms. The models were evaluated by an area under the curve of receiver operator characteristics curve (AUROC), precision-recall curve (AUPRC), and its calibration. All codes are available in https://github.com/ABMI/AbxBetterChoice Results Total 33 covariates were selected for final prediction models. The RF showed the highest AUROC in the ceftriaxone and tetracycline models (0.823, 0.626, respectively). The XGB presented the highest AUROC for ciprofloxacin and nitrofurantoin (0.731, 0.706, respectively). The AUROC of RF and the XGB were the same in an ampicillin model (0.633). For gentamicin, levofloxacin, and trimethoprim/sulfamethoxazole, the AUROC of LLR was the highest (0.838, 0.831, 0.615, respectively). Among the models, the AUROC was the highest in the gentamicin model regardless of algorithms. All calibrations of the models were acceptable. Conclusion We developed prediction models with competing performances of discrimination and calibration. It would contribute to the proper selection of empiric antibiotics susceptible to those causative pathogens in hospitalized patients with a clinically suspected urinary tract infection. Disclosures All Authors: No reported disclosures