Objective: The objective of this study was to analyze the frequency and rates of community respiratory virus infections detected in NIH Clinical Center (NIHCC) patients from January 2015 through March 2021, comparing the trends before and during the COVID-19 pandemic. Methods: We conducted a retrospective study comparing frequency and rates of community respiratory viruses detected in NIHCC patients from January 2015 through March 2021. Test results from nasopharyngeal swabs/washes, bronchoalveolar lavages, and bronchial washes were included in this study. Results from viral challenge studies and repeat positives were excluded. A quantitative data analysis was completed using cross tabulations; comparisons were done using mixed models, applying Dunnett’s correction for multiplicity. Results: Frequency of all respiratory pathogens declined from an annual range of 0.88-1.97% from January 2015 through March 2020 to 0.29% between April 2020 and March 2021. Individual viral pathogens declined sharply in frequency during the same timeframe, with zero cases of influenza A/B or parainfluenza and one case of RSV. Rhino/enterovirus detection continued, but with a substantially lower frequency of 4.27% between April 2020 and March 2021, compared with an annual range of 8.65-18.28% from January 215 through March 2020. Discussion: The decrease in viral respiratory infections detected in NIHCC patients during the pandemic was likely due to the layered COVID-19 prevention and mitigation measures implemented in the community and the hospital. Hospitals should consider continued use of nonpharmaceutical interventions in the future to prevent nosocomial transmission of respiratory viruses during times of high community viral load.

Background During the first year of the COVID-19 pandemic, nonpharmaceutical interventions had a broad impact on viral transmission apart from SARS-CoV-2. The NIH Clinical Center has used the BioFire FilmArray multiplex PCR respiratory pathogen panel (RPP) for evaluation of upper respiratory symptoms since 2014. Beginning in 3/20, respiratory samples from symptomatic patients were tested by SARS-CoV-2 PCR and the RPP. We performed a retrospective study comparing frequency and rates of community respiratory viruses detected by RPP from 1/14 through 3/21. Methods Results of RPPs from nasopharyngeal swabs/washes, bronchoalveolar lavages, and bronchial washes were included. Results from viral challenge studies were excluded. Charts were reviewed to determine whether repeat positives for the same virus within 12 months represented new infections; repeats from the same infection were excluded. A quantitative data analysis was completed using cross tabulations; comparisons were done using mixed models, applying Dunnett’s correction for multiplicity. Results A total of 3,329 patients underwent 8,122 RPPs from 1/14 through 3/21. Frequency of all respiratory pathogens declined from an annual range of 0.88-1.97% from 1/14-3/20 to 0.29% in 4/20-3/21 (p < 0.001). Individual viral pathogens declined sharply in frequency during the pandemic, with zero cases of influenza A/B, parainfluenza, or metapneumovirus detected from 4/20-3/21. One case each of adenovirus, RSV, CoV OC43, and CoV HKU1 were detected in 4/20-3/21. Rhino/enterovirus detection continued, but with a substantially lower frequency of 4.27% in 4/20-3/21, compared with an annual range of 8.65-18.28% from 1/14-3/20 (p < 0.001). Frequency of detection of all respiratory pathogens tested using the Biofire FilmArray multiplex PCR respiratory pathogen panel from January 2014 through March 2021. The frequency of pathogen detection from April 2020 through March 2021 declined substantially in comparison with previous years. Frequency of detection of influenza A, influenza B, rhinovirus/enterovirus, parainfluenza (1, 2, 3, 4), and respiratory syncytial virus from January 2014 through March 2021. The frequency of detection of these pathogens declined sharply starting in April 2020. Conclusion During the pandemic, the burden of viral respiratory infections detected among patients at the NIH Clinical Center improved considerably. This reprieve was likely thanks to the layered COVID-19 prevention and mitigation measures implemented in the community and the hospital: masking, distancing, symptom screening, isolation and testing symptomatic persons. As COVID-19 vaccination allows relaxation of masking, community transmission of respiratory viruses will likely resume; continued mask-wearing in the hospital may provide an enduring benefit by preventing nosocomial transmission. Disclosures Tara N. Palmore, MD, Nothing to disclose

Background: Transmission of carbapenemase-producing organisms (CPO) threatens patient safety in healthcare facilities. As a result of a 2011 outbreak of blaKPC+ Klebsiella pneumoniae, the NIH Clinical Center (NIHCC) has prioritized early detection and isolation of CPO carriers, using point-prevalence surveys and targeted high-risk ward surveillance since 2011 and admission surveillance since 2013. We describe our experience over 6 years of admission surveillance. Methods: The NIHCC is a 200-bed research hospital that provides care for a highly immunocompromised patient population. From September 2013 to September 2019, perirectal swabs were ordered automatically for all patients on admission to nonbehavioral health wards. Swabs were ordered twice weekly for ICU patients, weekly in other high-risk wards, and monthly for hospital-wide point prevalence (excluding behavioral health). Patients hospitalized in the United States in the previous week or abroad in the previous 6 months were considered high risk for carriage and isolated pending results from 2 swabs. Most swabs (n = 37,526) were cultured onto HardyCHROM CRE. If gram-negative bacilli (GNB) were present, a molecular screen for carbapenemases was performed on a sweep of cultured material (day 1) pending organism isolation. GNB were identified by MALDI-TOF MS. Prior to June 2019, isolates were screened by blaKPC/blaNDM PCR. Starting in June 2019, Enterobacteriaceae and Pseudomonas aeruginosa were screened using the phenotypic modified carbapenem inactivation method (mCIM), reflexing to the GeneXpert CARBA-R molecular assay if positive; other GNB were tested directly with CARBA-R. Selected GNB underwent susceptibility testing (Sensititre). Whole-genome sequencing was used to assess relatedness among CPO isolates. Swabs from high-risk patients were tested directly by blaKPC PCR (n = 699) until August 2019 (most in parallel with culture) and thereafter by CARBA-R (n = 13). Results: Among 54,188 orders for perirectal swabs, 38,238 were collected from 14,497 patients (compliance 71%). Among 33 CPO-colonized patients identified from September 2013 through September 2019, 15 were identified on admission, 6 were identified in point-prevalence surveys, 8 were identified from high-risk ward surveillance, and 4 were identified from clinical cultures. Sequencing demonstrated no relatedness among CPO isolates. Although only 1.4% of patients sampled on admission were colonized with CPO, those meeting high-risk criteria were 21 times as likely to be colonized. Conclusion: Admission surveillance for CPO identified a low rate of colonization, but it detected nearly half of known CPO-colonized NIHCC patients over the past 6 years. Modest compliance with swab collection leaves room for improvement and likely results in missed instances of colonization. Although we cannot determine its effectiveness, we view our strategy as one of several key safety measures for our highly vulnerable patient population.Funding: NoneDisclosures: None

Background: The NIH Clinical Center implemented multiple testing protocols to facilitate early detection and isolation of SARS-CoV-2 infected patients and rooming-in family members (RIFMs). Beginning in February 2020, all symptomatic patients were tested; in March 2020, all patients were tested prior to aerosol-generating procedures (AGPs); and in May 2020, all patients and RIFMs were tested on admission. We sought to determine the value of SARS-CoV-2 testing practices in our hospital. Methods: Respiratory specimens collected March 2020 through June 2021 tested for SARS-CoV-2 by RT-PCR were reviewed, and corresponding patient clinical and demographic variables were collected. Repeated tests from SARS-CoV-2–positive persons were excluded from the data. Results associated with multiple testing indications were assigned the highest priority reason based on a predetermined hierarchy. Data were analyzed using the χ2 test and logistic regression. Results: Of 12,706 results from 5,704 patients, primary testing reasons were pre-AGP (n = 5,387, 43.0%), admission (n = 2,733; 21.8%), and symptomatic testing (n = 2,701; 21.6%). Overall, 159 tests (1.25%) were positive for SARS-CoV-2. Asymptomatic patients tested on admission were 1.8 times more likely to be positive than outpatients tested for any reason (P = .003) and 4.2 times more likely than asymptomatic inpatients tested prior to AGP (P = .003). Within asymptomatic pre-AGP testing, there was no difference between inpatients (0.46%) and outpatients (0.65%). Hispanic patients were 1.9 times more likely to be positive. (p Conclusions: At a hospital with a geographically broad referral base, admissions COVID-19 testing was far more fruitful than pre-AGP testing of inpatients. Pre-AGP used the most testing resources yet had the lowest yield. Admissions testing remains beneficial regardless of community transmission rates, while testing prior to AGP could be pared back when community rates of COVID-19 are low and redeployed when community rates rise. Conclusions: Our findings that Hispanic persons had higher risk and that transplant patients had lower risk of testing positive suggests differences in the extent to which each subgroup may have been able to shelter from COVID-19 in the community during this earlier phase of the pandemic. Keeping immunocompromised patients safe from COVID-19 while they undergo longitudinal care involves layered precautions in the hospital and in the community that must evolve in response to evidence and epidemiological trends.Funding: NoneDisclosures: None