ABSTRACT
Background. Increased antibiotic prescribing rates during the early phases of the COVID-19 pandemic have been widely reported. We previously reported that while both antibiotic days of therapy (DOT) and total days present (DP) declined in the first 5 months of 2020 at Veterans Affairs (VA) acute care facilities nationwide relative to the comparable period in 2019, antibiotic DOT per 1000 DP increased by 11.3%, largely reversing declines in VA antimicrobial utilization from 2015 - 2019. We now evaluate whether these changes in antibiotic use persisted throughout the COVID-19 pandemic. Methods. Data on antibacterial use, patient days present, and COVID-19 care for acute inpatient care units in 108 VA level 1 and 2 facilities were extracted through the VA Informatics and Computing Infrastructure;level 3 facilities which provide limited acute inpatient services were excluded. DOT per 1000 DP were calculated and stratified by CDC-defined antibiotic classes. Results. From 1/2020 to 2/2021, care for 34,096 COVID-19 patients accounted for 13% of all acute inpatient days of care in the VA. Following the onset of COVID-19 pandemic, monthly total acute care antibiotic use increased from 533 DOT/1000 DP in 1/2020 to a peak of 583 DOT/1000 DP in 4/2020;during that month COVID-19 patients accounted for 13% of all DP (Figure). In subsequent months, total antibiotic use declined such that for the full year the change of antibiotic use from 2019 to 2020 (a decrease of 18 DOT/1000 DP) was similar to the rate of decline from 2015 to 2019 (mean decrease of 13 DOT/1000 DP;Table). The decreased DOT/1000 DP from 5/2020 to 2/2021 occurred even as the percentage of all DP due to COVID-19 peaked at 14 - 24% from 11/2020 to 2/2021. Conclusion. Although rates of antibiotic use increased within the VA during the early phases of the COVID-19 pandemic, rates subsequently decreased to below previous baseline levels even as the proportion of COVID-19 DP spiked between 11/2020 and 02/2021. Although the degree to which the initial increase in antibiotic use is attributable to concerns of bacterial superinfection versus changes in case-mix (e.g., decreased elective admission) remains to be assessed, these data support the continued effectiveness of antimicrobial stewardship programs in the VA.
ABSTRACT
Background. The data on CAPA in the U.S. are limited to date and clinical characteristics unique to this phenomenon have not been widely reported. Methods. This retrospective observational study was conducted at multiple VA hospitals across southern California and Arizona. CAPA cases were identified in inpatients with laboratory-confirmed COVID-19 based on microbiologic or serologic evidence of aspergillosis and pulmonary abnormalities on imaging, and were classified according to ECMM/ISHAM consensus definitions. Characteristics of interest included immunosuppressive/modulatory agents used prior to onset of CAPA, COVID-19 disease course, length of hospitalization, and mortality. Results. Seventeen patients with probable (18%) or possible (82%) CAPA were identified from April 2020 to March 2021. Values below reported as medians. All patients were male and 13 (76%) were white, with age 74 years and BMI 26 kg/m2. Baseline comorbidities included diabetes mellitus (47%), cardiovascular disease (65%), and pulmonary disease (71%). Evidence of aspergillosis was mostly based on respiratory culture, with mainly A. fumigatus (75%). Systemic corticosteroids were used in 14 patients, with a total dose of 400 mg prednisone equivalents starting 10 days prior to Aspergillus detection. Patients also received tocilizumab (18%), leflunomide (6%), tacrolimus (6%), mycophenolate (6%), and investigational agent LSALT or placebo (6%);2 patients (12%) did not receive any immunosuppression/modulation. Length of hospitalization for COVID-19 was 22 days. Death occurred in 12 patients (71%), including all patients with probable CAPA, at 34 days after COVID-19 diagnosis and 16 days after CAPA diagnosis. Eight patients (47%) were treated for aspergillosis;mortality did not appear to differ with treatment (75% vs. 67%). Conclusion. This case series reports high mortality among patients with CAPA;the primary contributor to this outcome is unclear. Frequency of lower respiratory tract sampling in patients with COVID-19 may have limited diagnosis of CAPA. Interestingly, inpatient respiratory cultures with Aspergillus spp. increased compared to previous years. Future work will attempt to identify risk factors for CAPA and attributable mortality via comparison to inpatients with COVID-19 without CAPA.
ABSTRACT
Background. Bamlanivimab and casirivimab/imdevimab were the first monoclonal antibodies (mAb) developed against SARS-CoV-2 and proved beneficial early in the course of infection. However, real-world administration of these therapies presents logistical challenges. We present our experience implementing mAb treatment at a large VA Medical Center and review the efficacy of therapy in preventing hospitalization from COVID-19 in a closed healthcare system. Methods. All positive outpatient COVID tests performed at VA Greater Los Angeles Healthcare System (GLA) were reviewed by the Emergency Medicine (EM) and Infectious Diseases (ID) Sections for mAb eligibility beginning 12/2/2020. Due to limited supply, treatment was prioritized for patients at highest risk of developing severe disease, as determined by EM/ID with input from a machine learning ensemble risk estimation model produced by VA National Artificial Intelligence Institute (Figure 1). If a patient declined or did not reply, treatment was offered to the next patient on a ranked eligibility list. Those who declined or were eligible but not treated were included in the analysis. Patients were excluded if they were hospitalized before treatment was offered. We collected data on age, comorbidities, date of diagnosis, and admission at 30 days after diagnosis. A multivariate log binomial regression was performed to determine the relative risk of admission within 30 days of diagnosis for those who received mAb therapy as compared to those who did not, adjusting for age and comorbidity. All analysis was done in R (version 4.0.5). Results. 139 patients met inclusion criteria. 45 (32%) received mAb therapy, 48 (35%) declined mAb therapy, and the remaining 46 (33%) either did not respond or were not offered mAb therapy. Hospitalizations following diagnosis in each group are illustrated in Figure 2. There was a trend towards reduced absolute and relative risk of hospitalization (Table 1). There were no anaphylactic events in patients who received mAb therapy. Conclusion. At our facility, a system for rapid identification of candidates and a coordinated distribution plan was essential in ensuring timely administration of mAb therapy to eligible patients. Administration of mAb showed a trend towards decreased risk of hospitalization due to SARS-CoV-2.
ABSTRACT
Background: The kinetics of antibody responses to SARS-CoV-2 infection are not fully understood. We analyzed IgG responses to the SARS-CoV-2 Spike protein receptor binding domain (RBD) in COVID-19 patients admitted to VA Greater Los Angeles (VAGLA) and correlated with clinical outcomes. Methods: Serially admitted patients from March 20-May 10, 2020 with at least one available residual serum specimen were included in this analysis. Serum samples selected for analysis included first, last, and intermediaries spaced ≥ 5 days apart, as available. Anti-RBD IgG was detected with an enzyme immunoassay (EIA) using recombinant RBD protein. Serum from an uninfected individual collected April 2019 was used as control. The average optical density of the control in triplicate plus 3 standard deviations was considered the threshold positive/negative value. The highest dilution above the threshold value was considered the IgG titer. Clinical groups were defined as asymptomatic, moderate/severe (no ICU) or critical (mechanical ventilation, cytokine storm and/or death). Results: Of the 43 consecutive patients admitted to VAGLA with COVID-19 in this analysis, 40 developed detectable RBD IgG responses with maximum inverse titers (MIT) ranging 100-819,200, geometric mean 12,152. Five patients remained asymptomatic but had positive EIAs with median MIT 3200 (IQR 800-3200). Twenty-five had moderate-severe illness with median MIT 25600 (IQR 6400-102400). Ten patients with critical disease had median MIT 38400 (IQR 8800-51200). The median time to positive IgG was 10 days for asymptomatic (IQR 10,10), 4 days for moderate-severe (IQR 3,15), and 7 days for critical (IQR 3.5,14.5). The figure depicts RBD IgG titers over time after onset of symptoms. Asymptomatic patients had a more gradual rate of increase and lower peak titers, while critical patients had the fastest rate of rise and the highest peak titers. Of the 21 patients with samples > 30 days after symptom onset (range 31-67 days), there was no evidence for decrease in anti-RBD IgG. Kinetics of IgG to SARS-CoV-2 receptor binding domain by clinical severity Conclusion: Following infection with SARS-CoV-2, disease severity correlates with both the rate of increase and peak in antibody titers. Anti-RBD IgG titers did not decrease over the observation period.
ABSTRACT
Background: The VA initiated an antimicrobial stewardship program in 2011, which includes participation in the Center for Disease Control (CDC) Antimicrobial Use Option, educational webinars, training programs for antimicrobial stewards, required staffing & reporting, and quality improvement initiatives, that has led to ongoing decreases in antimicrobial therapy nationwide. With the onset of the COVID-19 pandemic, however, there are several factors that may contribute increases in antimicrobial use (increased presentations of lower respiratory tract infection, concern for bacterial co-infection with SARS-CoV-2, etc.). We sought to compare patterns of antibacterial use in the VA from January - May 2020 with corresponding time periods in prior years. Methods: Data on antibacterial use from 2015 - 2020 were extracted from the VA Corporate Data Warehouse for acute inpatient care units in 84 VA facilities (facilities which provide limited acute inpatient services were excluded). To control for seasonal effects, only data from January to May for each year were included in the analysis. Days of therapy (DOT) per 1000 days-present (DP) were calculated and stratified by CDC-defined antibiotic classes. Results: From 2015 - 2019, total antibiotic use from January to May decreased by a mean of 9.1 DOT/1000 DP per year. In contrast, from 2019 to 2020, antibiotic use over the same months increased by 26.4 DOT/1000 DP (Table). Increases were observed in all drug classes except for a decrease in narrow spectrum ß-lactam antibiotics. Total antibiotic DOT in 2020 increased by 27.9 and 7.3 DOT/1000 DP in facilities in the highest and lowest terciles of use in 2019 (Figure). Conclusion: We observed a broad increase in antibacterial use during the initial surge of COVID-19 cases in VA facilities that abruptly reversed steady reductions in use over the prior 4 years. The degree to which this increase reflects potentially appropriate use in the setting of increased patient vulnerability and provider uncertainty, inappropriately decreased provider thresholds for initiating or continuing therapy, or stresses on the structure and staffing of antimicrobial stewardship programs requires further study.
ABSTRACT
Background: Despite numerous outbreaks, antibody responses to SARS-CoV-2 in residents of skilled nursing facilities (SNF) are not well described. We reviewed serological test results in a cohort of SNF residents who had been repetitively screened for SARS-CoV-2 infection by nasopharyngeal swab PCR. Methods: In late March 2019, we identified symptomatic SARS-CoV-2 PCR positive residents at a SNF. In response, all remaining SNF patients were serially screened, and all SARS-CoV-2 PCR positive patients were transferred to the acute care hospital or cohorted in a separate COVID Recovery Unit (CRU) in the SNF. In early June, all SNF residents (SARS-CoV-2 PCR positive and negative) underwent serologic testing for SARS-CoV-2 Spike (S1/S2) IgG (DiaSorin). DiaSorin IgG-positive results for patients that were SARS-CoV-2 PCR-negative were reflexed to nucleocapsid IgG (Abbott). Antibody testing occurred a median of 69 days (63-70 IQR) after PCR positivity. Results: Nineteen SARS-CoV-2 PCR positive residents were identified from the outbreak and an additional 9 were transferred from the acute care hospital to the CRU;1 died and 1 received convalescent plasma leaving 26 SARS-CoV-2 PCR positive residents, including 6 who were asymptomatic, that were eligible for serologic testing. Twenty-four of the 26 were positive for IgG by the DiaSorin assay;one seronegative resident was one of the asymptomatic residents. There were an additional 121 residents in the SNF whose SARS-CoV-2 PCR was negative at least once. Among these 121 SNF residents with negative SARS-CoV-2 RT-PCR, all but two were seronegative by the Diasorin assay. The two seropositive residents had no nucleocapsid antibodies when reflex tested by the Abbott assay. Conclusion: In a limited sample of SNF residents with SARS-CoV-2 PCR positivity, the sensitivity of the Diasorin assay was 92% (24/26) and the specificity was 98% (119/121). None of the residents with negative SARS-CoV-2 PCR had confirmed positive antibody results using reflex testing (DiaSorin/Abbott). Despite high risk exposure in congregate living facilities, we found no evidence of additional SARS-CoV-2 exposure, reinforcing the importance of serial surveillance SARS-CoV-2 testing and early cohorting in SNF settings. (Table Presented).