Citations

Abstract Background Handshake antimicrobial stewardship (HAS) is an effective stewardship approach characterized by regular, face-to-face interactions with antimicrobial prescribers and broad review of all active antimicrobials. Benefits of this approach include enhanced relationship building and opportunities for targeted education. We conducted a mixed methods study to evaluate the impact of once-weekly HAS rounds among Hospital Internal Medicine (HIM) teams in an academic medical center. Methods An infectious diseases (ID) physician and pharmacist conducted weekly table rounds with a HIM teaching team consisting of a hospitalist, advanced practice providers (APPs), APP fellow(s), and sometimes medical students. All patients on antimicrobials and without an active ID consult were reviewed and recommendations were provided. Thereafter, the stewardship team solicited ad hoc questions from other HIM teams via walking rounds in their offices. The number, type, and acceptance rate of recommendations were collected prospectively. A survey was sent to all HIM providers to obtain feedback on HAS rounds. Results HAS rounds were conducted weekly from February 7, 2024, to February 19, 2025, with a total of 52 sessions. Over the study period, 340 recommendations were provided, and 314 recommendations were implemented, yielding a recommendation implementation rate of 92.4%. The most common recommendation types were antimicrobial discontinuation (n=76, 22.4%), duration modification (n=63, 18.5%), and de-escalation (n=39, 11.5%). The survey was sent to 102 HIM providers and completed by 19 for a response rate of 18.6%. All respondents stated HAS added value or greatly added value to their practice. Antimicrobial selection (n=16, 84.2%), de-escalation (n=12, 63.2%), and pharmacy expertise (n=6, 31.6%) were areas in which providers saw the greatest benefit. Eighteen providers (94.7%) recommended expansion of HAS to non-HIM services. Conclusion Targeted, once-weekly HAS rounds resulted in high-yield interventions that were frequently accepted and perceived as highly valuable by HIM providers. Once-weekly or intermittent HAS practices might be considered in practice settings where resources or logistical constraints limit broader, hospital-wide handshake stewardship practices. Disclosures All Authors: No reported disclosures

Abstract Background Antimicrobial stewardship programs (ASP) commonly employ formulary restriction and prospective audit with feedback; however, frontline prescribers and other unit-based clinicians may enhance core ASP tactics as ASP extenders. We hypothesized that unit-based pharmacists act as ASP extenders and sought to characterize ASP interventions in their routine work.Top individual antimicrobial classes intervened on by unit-based pharmacistsUnit-based pharmacist antimicrobial intervention types Methods Retrospective, observational study of ASP interventions by unit-based pharmacists from July 2017 - April 2025 across the Mayo Clinic Enterprise. ASP interventions recorded using a standardized pharmacist documentation tool (Epic iVents [Software]) were collected retrospectively. Interventions on hospitalized, adult patients categorized as ‘ASP’ or linked with an antimicrobial were included; interventions by ASP dedicated staff were excluded. Secondary outcomes included: intervention subtypes and outcomes, location, and associated antimicrobial(s). Frequencies and percentages were used to summarize categorical data. Results A total of 104,055 interventions were included from 59,106 unique patients at 26 hospitals with the majority performed at the three destination sites. On average, 1,107 interventions were completed monthly by unit-based pharmacists. Frequently targeted antimicrobial classes were vancomycin and derivatives, penicillins, and 3rd/4th-generation cephalosporins (e.g., ceftriaxone and cefepime) (Figure 1); 20,111 (19.3%) interventions were on ≥ 1 antimicrobial. Of the 66,827 interventions with a documented subtype, ASP was the most common (77.8%), with the remainder classified as dose (17.3%), medication (3%), or duration changes (1%) (Figure 2). A minority (2.2%) were documented as conducted under pharmacist collaborative practice agreements. Documented prescriber intervention rejections were uncommon (6.6%). Reported outcomes, when available, included therapy optimization, cost savings and avoidance of adverse drug events. Conclusion Unit-based pharmacists acted frequently as ASP extenders. These findings support unit-based pharmacist and ASP collaborative efforts. Disclosures All Authors: No reported disclosures

Abstract Aim To assess the long‐term efficacy of vagus nerve stimulation (VNS) in Dravet syndrome and identify key factors influencing treatment response. Method We conducted a two‐center, retrospective cohort study of 15 patients (six females, nine males; median age at seizure onset 4 months [range 2–12 months]) with Dravet syndrome who underwent VNS implantation. Seizure frequency was evaluated over 12 months and up to 10 years after implantation. Linear mixed‐effects regression was used to analyze the impact of antiseizure medications (ASMs) and VNS parameters on seizure outcomes. Results A 50% or greater reduction in seizure frequency was observed in 93.3% (14 of 15) of patients at 12 months and 86.7% (13 of 15) of patients across the 10‐year follow‐up. This seizure reduction was independent of ASM use; the VNS duty cycle was a significant predictor of seizure reduction. Two individuals (13.3%) experienced minor adverse events. Interpretation VNS provides durable seizure reduction in Dravet syndrome, independent of ASM adjustments. These findings highlight the importance of early VNS initiation and parameter optimization in managing genetically defined epilepsy syndromes.

Abstract Background Hematopoietic stem cell transplant (HSCT) recipients are susceptible to infections with respiratory viruses resulting in higher morbidity and mortality. Parainfluenza virus (PIV) infection in the setting of leukopenia is associated with higher risk of lower respiratory tract infection and mortality. Intravenous immunoglobulin (IVIG) has been used as adjunctive treatment in viral respiratory infections, although data guiding its use is limited.Table 1:Patient CharacteristicsTable 2:Clinical Outcomes Methods We conducted a retrospective cohort study of adult HSCT recipients diagnosed with PIV infection at our institution from January 2014 to September 2024. Diagnosis of PIV infection was made via BioFire® Respiratory Panel multiplex polymerase chain reaction (PCR) or viral culture from nasopharyngeal or bronchoalveolar lavage. Data on clinical characteristics and outcomes were collected and reviewed.Table 3:Clinical Outcomes for Patients with IgG <767 mg/dL Results A total of 110 HSCT recipients were diagnosed with PIV infection by PCR or viral culture within the study period. Patient characteristics and clinical outcomes are summarized in Tables 1 and 2. PIV type 3 was the most common, followed by PIV type 4 and type 1. Eighteen patients (16.36%) had co-infections with another virus. A total of 16 patients received IVIG at time of PIV infection. The mean cumulative IVIG dose was 28.75 grams. Patients who received IVIG were more likely to have been hospitalized due to symptoms related to the viral infection (p=0.03) and had a lower immunoglobulin G level (p< 0.001). Overall, there was no significant difference in rates of mechanical ventilation (12.5% vs 7.45%, p=0.616), development of secondary fungal or bacterial infection (12.5% vs 12.7%, p=1), or 90-day mortality (18.75% vs 6.38%, p=0.123) among patients treated with or without IVIG, respectively. Similarly, in a subset analysis of patients with IgG level < 767 mg/dL, there were no significant difference in outcomes for patients treated with or without IVIG (Table 3). Conclusion There were no significant differences in clinical outcomes, including rates of mechanical ventilation, secondary infections, or 90-day mortality, between HSCT patients who received IVIG and those who did not for PIV infection. Larger studies are needed to gain insight into efficacy of IVIG as adjunctive treatment in PIV infections in this population. Disclosures All Authors: No reported disclosures

Abstract Background Lyme arthritis is a manifestation of disseminated infection with Borrelia spp., typically presenting as knee monoarthritis weeks to months after infection. Most patients respond to antibiotics, but some develop antibiotic-refractory Lyme arthritis (ARLA), characterized by proliferative or autoimmune synovitis despite clearance of active infection, with proposed risk factors including treatment delay, host genetics, and intra-articular steroid use prior to antibiotics. Table 1 Characteristics of cohort. IQR = interquartile range, ESR = erythrocyte sedimentation rate, CRP = C-reactive protein, WBC = white blood cell. Table 2 Comparison of characteristics and outcomes between ARLA and non-ARLA patients Methods We performed a retrospective chart review of adult patients with Lyme PCR-positive synovial fluid at our center between 01/2004 and 02/2025. ARLA was defined as persistent inflammatory arthritis despite ≥8 weeks of antibiotics and/or negative repeat PCR after antibiotics. Statistical analysis was done on R and STATA. Figure 1 Type of antibiotics administered during the first course (A) and the second course (B) Figure 2 Box plot comparing the median time from symptom onset and treatment initiation in patients who had complete symptomatic resolution with any duration of antibiotic therapy, and those that did not have symptomatic resolution with antibiotic therapy. Results Table 1 summarizes the characteristics of the study cohort. Definitive Lyme arthritis was diagnosed in 67 patients. Of these, 35 patients (52%) achieved complete symptom resolution after one course of antibiotics, and 45 patients (67%) achieved resolution after up to two courses. Doxycycline was used as first-line therapy in 97% of cases (Figure 1). Antibiotic-refractory Lyme arthritis (ARLA) developed in 20 patients (30%); of these, 7 required disease-modifying antirheumatic drugs (DMARDs), and 3 underwent synovectomy. ARLA was associated with incomplete response to a single antibiotic course and persistent joint effusion on post-treatment radiography (Table 2). Intra-articular steroid use prior to antibiotic therapy was associated with a threefold increased odds of ARLA, although this did not reach statistical significance. Patients who responded to antibiotics had a shorter median time to treatment (19 days) compared to non-responders (42 days) (Figure 2). Conclusion In this cohort, 30% of patients were ultimately diagnosed with antibiotic-refractory Lyme arthritis (ARLA), a rate notably higher than the previously reported 10%. Despite advancements in Lyme disease diagnostics, we observed significant delays in diagnosis and initiation of appropriate antibiotic therapy, as well as the use of intra-articular steroids prior to antibiotics. These factors may contribute to the development of ARLA. Disclosures All Authors: No reported disclosures

Abstract Background Staphylococci are the most frequent microorganisms in periprosthetic joint infection (PJI) and treatment remains challenging. Phage is being considered as a potential treatment. Here, the in vitro activity of phages SaMD07FSphi1or SaNSI1469phi1, alone and in combination with vancomycin against staphylococcal biofilms was evaluated. Figure 1 In vitro activity of phage SaMD07FSphi1 against S. aureus and S. epidermidis biofilms Figure 2 In vitro activity of phage SaNSI1469phi1 against S. aureus and S. epidermidis biofilms Methods Biofilms were grown at 37°C on 4 cm stainless steel K-wires in tryptic soy broth (TSB) with 106 cfu/ml of S. epidermidis or S. aureus (8 PJI isolates with phage activity against planktonically-grown bacteria). After 24 h, 3 K-wires were quantitatively cultured to define pre-treatment bacterial loads, with the remaining assigned to one of four treatment groups: TSB, phage (4 x 106 plaque forming units/ml), vancomycin (16 µg/ml) or phage/vancomycin. K-wires were incubated for 6 or 24 h at 37°C, rinsed, sonicated, and quantitatively cultured. Testing was performed in triplicate and on 3 separate days and mean log10 cfu/K-wire reduction factors (LRFs) versus TSB calculated. Results Phage SaMD07FSphi1 (Figure 1) plus vancomycin showed LRFs of 2.8 (p, 0.0003) and 2.3 (p, 0.0177) log10 cfu/K-wire at 24 h for S. aureus IDRL-6121 and S. aureus IDRL-8454, respectively, whereas phage alone showed was 3.1 (p, 0.0003) and 3.0 (p, 0.0003), respectively. All treatments were ≤1.0 log10 cfu/K-wire 24 h LRFs for S. aureus IDRL-6175, S. epidermidis IDRL-5966 and S. epidermidis IDRL-6108.This phage plus vancomycin showed LRFs of 0.6 log10 cfu/K-wire at 6 h for S. epidermidis IDRL-5966 and S. aureus IDRL-6175, with bacterial regrowth at 24 h. Phage SaNSI1469phi1 (Figure 2) plus vancomycin showed 24 h LRFs of 2.0 (p, 0.02) and 3.2 (p, 0.003) log10 cfu/K-wire for S. epidermidis IDRL-6955 and S. aureus IDRL-7299, respectively, whereas phage alone was 0.2 (p, 0.17) and 0.5 (p, 0.03) log10 cfu/K-wire, respectively. All treatments showed ≤1.0 log10 cfu/K-wire 24-hour LRFs log10 cfu/K-wire for S. epidermidis IDRL-10273. This phage alone showed 6 h LRFs of 1.0 and 1.3 for S. epidermidis IDRL-6955 and S. aureus IDRL-7299, with bacterial regrowth at 24 h. Conclusion Phages SaMD07FSphi1 and SaNSI1469phi1 alone and with vancomycin showed activity against staphylococcal biofilms formed by 4 isolates. Increased activity was observed at 6 h in some cases, with bacterial regrowth at 24 h. Disclosures Robin Patel, MD, a patent on Bordetella pertussis/parapertussis PCR issued, a patent on a device/method for sonication, a patent on PET imaging of bacterial infection: a patent on Bordetella pertussis/parapertussis PCR issued, a patent on a device/method for sonication, a patent on PET imaging of bacterial infection|MicuRx Pharmaceuticals and bioMérieux: Grant/Research Support|PhAST, Day Zero Diagnostics, DEEPULL DIAGNOSTICS, S.L., Nostics, HealthTrackRx, bioMérieux and CARB-X: Advisor/Consultant|Up-to-Date and the Infectious Diseases Board Review Course: Honoraria

Abstract Background A unified definition of prosthetic joint infection outcomes (PJI) is crucial for consistent data, study comparisons, and better care. This study systematically reviews how PJIs outcomes are defined, building on our prior research on native vertebral osteomyelitis and postoperative spinal infection.Figure 1Network plot illustrating the co-occurrence of outcome criteria in prosthetic joint infection studies. Edge thickness reflects the frequency of the co-occurrence of each criterion. Each plot highlights one criterion, namely (1) clinical signs, (2) imaging, (3) microbiology, (4) histopathology, (5) antibiotics-related, (6) surgical-related, (7) mortality. Methods We conducted a comprehensive search of the Cochrane, Embase, Medline, and Scopus databases for studies published after 2005 that clearly defined outcomes for PJI. Eligible studies included randomized controlled trials and cohort studies with a minimum of 10 patients diagnosed with PJI. Our methodology adhered closely to the PRISMA guidelines. Using a saturation approach, we extracted and grouped explicit definitions of PJI outcomes into thematic criteria. A network plot (Figure 1) was used to visualize how frequently these criteria appeared and how they co-occurred across the included studies. Results Out of 7,584 screened studies, 552 (7.3%) met the inclusion criteria. This pilot analysis included 49 randomly selected studies (8,162 patients), mostly retrospective (48/49, 98%), focusing on mixed joint replacements (25/49, 51%), two-stage exchanges (15/49, 31%), or mixed surgeries (24/49, 49%). Only 6/49 (12.2%) applied a society-endorsed definition, and 13/49 (26.5%) cited a reference for their outcome definitions. Overall, 32/49 (65.3%) explicitly defined "success," 35/49 (71.4%) defined "failure," and 16/49 (32.7%) defined both. Most studies (25/49, 51%) had follow-up periods over two years, and 14/49 (28.6%) included patient-reported outcomes, identifying 14 unique PROMs (most commonly UCLA, HHS, and KSS scores). Functional outcomes were assessed in 13/49 (26.5%) studies. We identified 16 distinct outcome criteria across seven categories: clinical signs, microbiology, imaging, histopathology, antibiotics-related, surgical-related, and mortality. Network analysis revealed strong links among clinical signs, microbiology, surgical factors, mortality, and imaging. Conclusion Our pilot study found major variation in PJI outcome definitions, with few standardized criteria. A consensus-driven approach is needed to strengthen evidence synthesis and advance patient care in PJI. Disclosures All Authors: No reported disclosures

Abstract Background Despite a total decrease in age-adjusted mortality rate (AAMR) for HIV across all demographic variables, a multitude of factors have accounted for this cumulative effect of reduction in AAMR, including HIV-related hospitalizations, as 59% of these deaths occur in a medical facility. This study’s main objective is to provide insights from the national in-patient database for the recent years regarding the trends of HIV-associated in-hospital mortality and their healthcare resource utilization.Figure 1:(A). shows age-distribution of HIV-related hospitalizations (n=289,095); (B). gender; (C). age-groups; (D). racial/ethnic groups; (E). Trends in age-adjusted mortality rates among gender; (F). Trends in age-adjusted mortality rates among racial/ethnic groups.(A). Histogram (age).(B). Pie-chart (frequency).(C). Distribution dot-plot (horizontal).(D). Bar graph (vertical).(E). Line graph (two-groups).(F). Line graph (four-groups).Figure 2:Trend analysis for crude mortality of in-hospital HIV mortality among gender (A), race (B); inflation-adjusted hospitalization cost from 2016-2022 among gender (C), and race (D); and length of hospital stay among gender (E) and race (F).The Cochrane-Armitage test is used for crude mortality rate trend analysis for the years 2016 up to 2022.The Jonckheere-Terpstra test is used for non-parametric distribution of trends in length of stay and hospitalization costs (inflation-adjusted). Methods The National Inpatient Sample (NIS) database was searched from 2016 to 2022 to identify patients with HIV (ICD-10 Codes: B20-B24). Hospital discharge weights for national representation were utilized. Baseline social and demographic characteristics along with hospital-level variables were analyzed subsequently using the odds ratio with multivariate regression analysis for mortality adjusted with age. Secondary outcomes included length of hospital stay and inflation-adjusted costs. For cost-based analysis, inflation adjustment was performed according to the inflation index for the Medical Care Consumer Price Index (CPI).Figure 3:A-Univariate analysis of sociodemographic variables associated with in-hospital mortality with HIV; B-Significant predictors of in-hospital mortality with HIV in patients with comorbidities; C- Top 20 listed primary causes of admission stratified by ICD-10 codes in HIV patients.Top 20 listed primary causes of hospitalization in HIV patients from the year 2016 to 2022:1. B20 = HIV-related conditions.2. A41.9 = Sepsis, unspecified organism.3. J18.9 = Pneumonia, unspecified organism.4. N17.9 = Acute kidney failure, unspecified.5. J44.1 = Chronic obstructive pulmonary disease with (acute) exacerbation.6. U07.1 = COVID-19 infection.7. Z51.11 = Encounter for antineoplastic chemotherapy.8. I13.0 = Hypertensive heart and chronic kidney disease (CKD) with heart failure and stage 1 through stage 4 CKD.9. I10.0 = Hypertensive heart disease with heart failure.10. I21.4 = Non-ST elevation (NSTEMI) myocardial infarction.11. A41.89 = Other specified sepsis.12. I13.2 = Hypertensive heart and chronic kidney disease (CKD) with heart failure and with stage 5 CKD, or end-stage renal disease.13. F25.0 = Schizoaffective disorder, bipolar type.14. F33.2 = Major depressive disorder, recurrent severe without psychotic features.15. J96.01 = Acute respiratory failure with hypoxia.16. F10.239 Alcohol dependence with withdrawal, unspecified.17. O98.72 = HIV disease complicating childbirth.18. F25.9 = Schizoaffective disorder, unspecified.19. I12.0 = Hypertensive chronic kidney disease (CKD) with stage 5 (CKD) or end-stage renal disease.20. K52.9 = Noninfective gastroenteritis and colitis, unspecified.Table 1:Baseline characteristics of the study population.ED: emergency; HMO: health maintenance organization; DRG: diagnosis related groups; LOS: length of stay; SNF: skilled nursing facility; AMA: against medical device; NCHS: National Center for Health Statistics. Results A total of 289,095 HIV-related hospitalizations were included, which represents a weighted size of 1,445,475 hospital discharges. The mean age was 50.24 years, with 68.3% males and 51.85% African Americans. A total of 2.59% in-hospital deaths occurred; AAMR was 7029 per 100,000 hospitalizations in males, which was higher than females (5644 per 100,000). AAMR was found to be highest in the Hispanic population, at 8184 per 100,000. Males had a 3% annual percent change from 2016 to 2022, while trends show an increase in deaths in other races, like Asians/Pacific Islanders/Natives. Most deaths occurred in the West region (OR=1.43), non-metropolitan areas (OR=1.40), and medium-large bed-size hospitals (OR=1.28). Conclusion HIV-related in-hospital mortality has posed a significant challenge across various demographic factors. Utilizing mortality data from large databases can serve as a valuable screening tool for targeted public health surveillance and allocation of resources to obtain further development of tailored health programs and interventions aimed particularly at allocation of resources in reducing the impact of HIV-related deaths. Disclosures All Authors: No reported disclosures

Abstract Background The increased frequency of spinal surgeries has led to a rise in postoperative spine infections (PSIs), now recognized as the third most common surgical complication. Currently, there is no standardized definition of PSI, complicating clinical management and research. This meta-epidemiological study aimed to develop clear diagnostic criteria to assist clinicians and researchers in identifying and reporting PSIs consistently. 2020 PRISMA flow diagram Sankey diagram showing the distribution of combination of criteria or definitions used by the authors. This refers to 21 articles that used a definition derived by the authors. Methods Following PRISMA guidelines, a comprehensive literature search was conducted by a medical librarian using Ovid MEDLINE (March 14, 2024). Studies were included if they reported on at least ten adult patients diagnosed with PSI and provided clear diagnostic definitions. The primary objective was to evaluate existing PSI diagnostic criteria and categorize them into thematic clusters. We constructed standardized diagnostic combinations using 13 predefined criteria derived from clinical practice guidelines and prior research (Table 1).Characteristics of the included studies*Total can be more than 100% since some variables can be present in multiple categories.This refers to 21 articles that used a definition derived by the authors. For articles #411 and #630 categorization was not possible as the definition provided only information about the depth of infection and not the criteria for diagnosis. Article #1148 was included as, defining “presumed aspetic” cases meant, by definition, also defining infectious cases.Abbreviations. The American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP), Centers for Disease Control and Prevention (CDC), The National Healthcare Safety Network (NHSN), Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA), Magnetic Resonance Imaging (MRI) Results From an initial 1,466 studies screened, 101 were included (Figure 1). Most studies were retrospective (91/101, 92.1%), and a majority (77/101, 76.2%) relied on established criteria such as those from the American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) or the CDC. Seven studies combined predefined criteria with additional detailed explanations. Among the remaining studies using independent definitions, clinical signs and symptoms were the most common criteria (13/24, 54.2%) (Table 1). Criteria were grouped into clinical, microbiological, and radiological categories, with most studies primarily using clinical criteria alone or in combination with microbiological or imaging findings (Figure 2). Conclusion There was no clear consensus in the literature on what defines a PSI. Our findings emphasize the necessity of establishing a standardized PSI definition that integrates clinical, microbiological, and radiological criteria. Implementing such a unified framework could enhance diagnostic accuracy, address existing limitations associated with hardware complications, wound drainage interpretations, and imaging ambiguities, and improve upon the current shortcomings of surveillance systems. Disclosures Jeremy D. Shaw, MD, Globus: Grant/Research Support|Purgo Scientific: Advisor/Consultant|Purgo Scientific: Stocks/Bonds (Private Company)

Abstract Background Cardiac implantable electronic device infection (CIEDI) is a serious complication, occurring in approximately 1% of implantations. In addition to antimicrobial therapy, complete device extraction is considered essential for definitive cure. However, this may not always be feasible due to a variety of factors. In select cases, long-term chronic antibiotic suppression (CAS) has been used, yet clinical characteristics and outcomes of patients who received CAS have scarcely been reported. Methods A descriptive study analyzing all adult (age > 18) patients with definite CIEDI who received CAS between 2016 and 2023 across the Mayo Clinic Enterprise (Arizona, Florida, Minnesota, and Upper Midwest) was conducted. Patients with either ventricular assist devices or fungal infection were excluded. Results A total of 321 patients were diagnosed with definite CIEDI during the study period. A total of 37 (8.7%) patients received CAS due to device retention. The median age was 77.5 [IQR 70.5, 87.5] with 32.4% female. Permanent pacemakers were involved in most (64.9%) infections. Approximately one-third (32.4%) of patients had a prosthetic valve. The median time from initial device implantation to CIEDI was 10.6 years [IQR 4.8, 18.1]. CIED-related infective endocarditis was present in 70.3% of cases and 67.6% had lead involvement. The most commonly isolated organisms were Staphylococcus aureus (29.7%), coagulase-negative staphylococci (27%), Enterococcus faecalis (13.5%) and Streptococci (13.5%). Other pathogens or culture-negative cases comprised 16.3%. The primary reason for device retention and CAS was poor surgical candidacy. Doxycycline was the most commonly used agent for CAS, accounting for 37.8% of cases. The average length of suppression was 469.0 days [IQR158.0, 1264.0]. The rate of relapse was 8.1% with all occurring within 90 days of diagnosis. One year mortality was 27.0%. Conclusion CAS may be a feasible option for patients with device retention, although a small proportion experience CIEDI relapse. Unexpectedly, the pathogen distribution among CAS recipients differed from that seen in a general CIED infection population, highlighting selection bias and the distinct clinical characteristics of patients managed with CAS. Disclosures Larry M. Baddour, MD, UpToDate, Inc.: Royalty payments (authorship duties).