Background Many studies reporting the excess cost of healthcare-associated infections (HAIs) have used data from positive cultures found in electronic medical records. These data are often limited in a number of ways including whether the positive culture truly represents an actual infection and not just colonization, when the onset of the infection occurred, and whether the infection was device related. Our objective was to generate estimates of the excess cost of HAIs in one facility, the Central Texas Veterans Health Care System (CTVHCS), using highly reliable information extracted from chart review regarding the infection. Methods Our study included data from all inpatient admissions between 1/1/2014 and 12/31/2019 at CTVHCS. HAIs were identified via chart review and were classified based on the infecting organism, specimen location, and whether the infection was device related using 2017 National Health and Safety Network definitions. To minimize time-dependent bias in our estimates of the excess cost of HAIs, we matched each patient with an HAI on day t of their inpatient stay to up to 4 patients who had not had an HAI up until day t of their inpatient stay. We used multivariable generalized estimating equations models to compare the inpatient costs prior to discharge between patients with and without HAIs. Finally, we used multivariable 2-part models to estimate the impact of HAIs on post-discharge readmission costs. Results Our analysis cohort consisted of 425 patients with HAIs who were matched to 1,645 patients without HAIs. HAIs were associated with $29,412 (95% CI:$18,064-$40,759) excess pre-discharge costs, 46.3% increase in the odds of a post-discharge readmission, and $16,049 excess post-discharge costs (see Table 1). The excess pre-discharge costs for central line, catheter-associated urinary tract, and surgical site infections (SSIs) were $96,655 (95% CI:$61,557-$131,754), $34,026 (95% CI:$9,562-$58,491), and $20,014 (95% CI:$5,324-$34,703), respectively. As seen in Table 2, the excess readmission cost for SSIs was $28,222 (95% CI:$9,523-$46,920). Table 1Multivariable regression model results - excess pre-discharge costs associated with healthcare-associated infectionsTable 2Multivariable regression model results - excess post-discharge costs associated with healthcare-associated infections Conclusion We found that HAIs significantly increased both pre- and post-discharge inpatient costs. These costs could potentially be saved if interventions to prevent HAIs could be successfully implemented. Disclosures Piyali Chatterjee, PhD, AHRQ Grant # 1R03HS027667-01: Grant/Research Support|AHRQ Grant # 1R03HS027667-01: Central Texas Veterans Health Care System Chetan Jinadatha, MD, MPH, AHRQ R01 Grant-5R01HS025598: Grant/Research Support|EOS Surfaces: Copper Coupons and materials for testing.

Background Carbapenem-resistant Acinetobacter baumannii (CRAb) is increasing due to widespread use of antibiotics. Multidrug resistant (MDR) CRAb is a major threat to public health as treatment options are limited. The objective of this study is to elucidate the molecular epidemiology of circulating antibiotic resistance genes causing MDR CRAb infections by using a combination of whole-genome Multi-Locus Sequence Typing (wgMLST) and antibiotic susceptibility phenotyping. Table 1.Molecular characterization of MDR CRAb isolates*The numbers indicate % of sequence identity match for each Beta-lactamase gene. Methods Bacterial isolates were derived from cultures taken from subjects 48 hours following admission as part of routine clinical care for patients between 2017-2020. Isolates were obtained from 16 hospital units (both ICU and non-ICU) across two hospitals in the Detroit area. Whole Genome Sequencing (WGS) was performed using Illumina MiniSeq or Nextseq. WgMLST analysis was performed using BioNumerics software v7.6. ResFinder software was used for analysis of antibiotic resistance genes. Isolates underwent antibiotic susceptibility testing using a broth microdilution method (VITEK2) and Clinical & Laboratory Standards Institute (CLSI) minimum inhibitory concentration (MIC) cut offs. Results Out of the 95 total isolates, 51(54%) were CRAb isolates and of the CRAb isolates, 21(41%) were MDR CRAb. WgMLST identified that majority of the circulating MDR CRAb isolates belonged to ST2Pas (ST195Ox and ST208Ox) based on CDC definitions (Table 1). MDR CRAb isolates were resistant to 3 different classes of antibiotics including aminoglycosides, fluroquinolones and β-lactams. β-lactamase genes present include (blaADC-25, blaOXA-23, blaOXA-66 and blaTEM1D) for both ST195Ox and ST208Ox and (blaADC-25, blaOXA-23 and blaOXA-223) for ST406Pas (ST310Ox). Among the patients with MDR CRAb infections, most were males with respiratory infections in a non-ICU setting. Conclusion The study demonstrated a high proportion of isolates belonged to ST2 Pas carrying multiple beta-lactamase genes including blaOXA-23 gene. ST406Pas might be an emerging lineage carrying the blaOXA-23 gene. In addition to stringent infection control measures, continuous surveillance is recommended in limiting the spread of MDR CRAb isolates in the healthcare settings. Disclosures Chetan Jinadatha, MD, MPH, AHRQ R01 Grant-5R01HS025598: Grant/Research Support|EOS Surfaces: Copper Coupons and materials for testing Keith S. Kaye, MD, MPH, Allecra: Advisor/Consultant|GlaxoSmithKline plc.: Receiving symposia honoraria|GlaxoSmithKline plc.: GlaxoSmithKline plc.-sponsored study 212502|Merck: Advisor/Consultant|qpex: Advisor/Consultant|Shionogi: Grant/Research Support|Spero: Advisor/Consultant Piyali Chatterjee, PhD, AHRQ Grant # 1R03HS027667-01: Grant/Research Support|AHRQ Grant # 1R03HS027667-01: Central Texas Veterans Health Care System.

Background Enterococcus faecium (E. faecium) is a common hospital-associated infection (HAI) that can lead to increased costs, morbidity, and mortality. Pulsed Xenon Ultraviolet light (PX-UV) has been shown to reduce bacterial bioburden levels on surfaces. This study aims to assess the effect of the addition of PX-UV to terminal cleaning on the clonal recovery of E. faecium sequence types (STs) using Whole Genome Sequencing (WGS) on patient isolates. Methods During 2017 to 2020 a prospective, randomized, double-blinded, sham-controlled, interventional, crossover trial in 2 separate Detroit hospitals (H1 and H2) compared HAI counts after the addition of either PX-UV or a non-UV sham device to terminal cleaning methods. The trial consisted of a total of 16 units randomized to have either the treatment of PX-UV (Group Q) or the sham control device with no UV (Group W) for 12 months. A washout period (Group R) of 6 months followed and the trial concluded with a 12-month crossover of treatments. A total of 60 E. faecium samples were collected, then WGS was performed by the Illumina Nextseq 550 instrument. de novo assembly was preformed using the SPAdes program. Whole Genome Multilocus Sequence Type (wgMLST) analysis was performed by BioNumerics (v7.6) to construct minimum spanning tree (MST). Results A total of 7 STs were obtained across the 2 hospitals (H1 and H2). ST117, ST17, and ST80 all had more than 10 total recovered isolates, ST117 being the most frequent with 24 isolates. Less than 3 isolates were recovered for all other STs. For all STs, Group Q (PX-UV) had 14, Group R (washout period) had 22, and Group W (non-UV sham) had 24. ST18 was only found in Group Q. ST412, ST584, and ST736 were not found in Group W. The data shows that the intervention PX-UV group had a reduction of clonal recovery by 10 STs as compared to the sham UV group. Table 1:Total number of each sequence type (ST) in H1 and H2 Conclusion The overall reductions in the number of isolates in the real UV units was driven by reductions in the ST117 a predominant strain in a hospital environment reported previously in Detroit and ST80. There were negligible differences in recovery of other ST between the groups. The reduction in clonal recovery of E. faecium isolates in Group Q as compared to Group R might be related to use of PX-UV following terminal room cleaning. Disclosures Piyali Chatterjee, PhD, AHRQ Grant # 1R03HS027667-01: Grant/Research Support|AHRQ Grant # 1R03HS027667-01: Central Texas Veterans Health Care System Keith S. Kaye, MD, MPH, Allecra: Advisor/Consultant|GlaxoSmithKline plc.: Receiving symposia honoraria|GlaxoSmithKline plc.: GlaxoSmithKline plc.-sponsored study 212502|Merck: Advisor/Consultant|qpex: Advisor/Consultant|Shionogi: Grant/Research Support|Spero: Advisor/Consultant Chetan Jinadatha, MD, MPH, AHRQ R01 Grant-5R01HS025598: Grant/Research Support|EOS Surfaces: Copper Coupons and materials for testing.

Background Healthcare-associated infections (HAIs) can be caused by some multidrug-resistant Escherichia coli, most commonly due to production of an extended spectrum beta-lactamase (ESBL), and lead to increased morbidity and mortality. Pulsed Xenon Ultraviolet light (PX-UV), in combination with terminal cleaning, has been shown to improve disinfection and has the potential to lower HAIs by reducing the horizontal spread of infections in hospitals. Here, we assess the effect of PX-UV on the clonal recovery pattern of several E. coli sequence types (STs) using Whole Genome Sequencing (WGS). Methods A prospective, sham-controlled, double-blinded, interventional, crossover trial was conducted to compare standard terminal cleaning with PX-UV (intervention, Group Q) and standard terminal cleaning with sham UV (control, Group W) in 2 Detroit hospitals from 2017 to 2020. Treatments lasted 12 months before crossover, with a 6-month washout period in between (Group R) during which UV was not used. A total of 67 E. coli samples were collected. WGS of the isolates was performed using the Nextseq 550 (Illumina). After de novo assembly, BioNumerics calculation engine (v7.6) was used to complete Whole Genome Multilocus Sequence Type (wgMLST) analysis, assembly free and assembly-based call, and construction of minimum spanning tree (MST). Results The total number of different STs found for the intervention UV device group (Q) and the sham UV device group (W) was 6 while the washout group (R) was 5. Out of the 9 total STs obtained, the most common was ST131 (Table 1). In Group Q, 11 of ST131 were found; 15 were found in Group W. During the washout period (R) 17 of ST131 were found. All other STs had 3 or less circulating clusters. After UV treatment ST1193, ST399, and ST7394 were not recovered. Fig. 1:Minimum spanning tree (MST) for E. coliTable 1:Total number of each sequence type (ST) per treatment group Conclusion During UV intervention 3 different STs (ST1193, ST399, ST7394) were not recovered, but there were negligible changes to the frequency of recovery of the other 6 STs as compared to sham UV treatment. ST131 was the dominant E. coli ST found in Detroit, which is consistent with previously published data stating ST131 as the predominant strain. While PX-UV has previously demonstrated effectiveness on decreasing bioburden, our data does not indicate any remarkable change in clonality and prevalence of E. coli STs after PX-UV use. Disclosures Piyali Chatterjee, PhD, AHRQ Grant # 1R03HS027667-01: Grant/Research Support|AHRQ Grant # 1R03HS027667-01: Central Texas Veterans Health Care System Keith S. Kaye, MD, MPH, Allecra: Advisor/Consultant|GlaxoSmithKline plc.: Receiving symposia honoraria|GlaxoSmithKline plc.: GlaxoSmithKline plc.-sponsored study 212502|Merck: Advisor/Consultant|qpex: Advisor/Consultant|Shionogi: Grant/Research Support|Spero: Advisor/Consultant Chetan Jinadatha, MD, MPH, AHRQ R01 Grant-5R01HS025598: Grant/Research Support|EOS Surfaces: Copper Coupons and materials for testing.

Background Clostridioides difficile (C. diff) is one of the most common causes of healthcare-associated infections (HAIs), leading to increased hospital stay, cost, and mortality. Elimination of C. diff spores is difficult as they are resistant to kill by common hospital grade disinfectants. Copper impregnated self-sanitizing surfaces (SSSCu) provide continuous reduction of vegetative pathogens, potentially lowering the risk of infections, but their efficacy on C. diff spores has not been previously evaluated. Methods Control (no copper) and copper coupons containing 20% copper-oxide were inoculated with varying C. diff spore loads ranging from 105 to 107 spores, prepared using environmental protection agency protocol, with or without 5% fetal bovine serum (FBS) soil load. After 4 hours of contact time, the C. diff spores were recovered and tested for viability. The efficacy of copper (log10 kill) was estimated using a Bayesian linear multilevel model. Results After 4 hours, the copper coupons, at mean initial spore load (6.67 log10) and no soil load, had a mean 1.21 (95% uncertainty interval: 1.13 - 1.31) log10 reduction compared to control coupons. With soil load, the treatment effect of copper decreased by 0.50 (95% uncertainty interval: 0.37 - 0.64) log10. Also, for each additional standard deviation (SD) increase in initial spore load there was a 0.59 (95% uncertainty interval: 0.47 - 0.72) log10 decrease in the treatment effect of copper. The soil load in combination with increasing initial spore load further decreased the treatment effect of copper by an additional 0.19 (95% uncertainty interval: 0.01 - 0.38) log10 for each SD increase in initial spore load. Conclusion Copper coupons can substantially reduce C. diff spores after 4 hours by 50%-97% depending on the initial spore concentration and presence or absence of organic material. Higher initial spore loads or excess organic material may prevent the spores from coming in contact with the copper-impregnated surface thus decreasing the kill efficacy. Continuous sporicidal effect of copper-impregnated surfaces might prevent transmission of spores and help reduce HAIs. Further studies are needed to evaluate the efficacy on C. diff HAI rates. Disclosures Piyali Chatterjee, PhD, AHRQ Grant # 1R03HS027667-01: Grant/Research Support|AHRQ Grant # 1R03HS027667-01: Central Texas Veterans Health Care System Chetan Jinadatha, MD, MPH, AHRQ R01 Grant-5R01HS025598: Grant/Research Support|EOS Surfaces: Copper Coupons and materials for testing.