Background: Acute-care hospitals in the United States are required to submit 6 healthcare-associated infection (HAI) metrics to the CMS for reporting and performance purposes prior to payment. We examined the association between HAI rate trends and hospital-onset bloodstream infection (HO-BSI) rate trends across a large, multihospital health system. Methods: HO-BSI events were identified across 52 hospitals attributable to Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, Klebsiella pneumoniae, or Candida spp using the NHSN Lab ID event definition of ≥day 4 of admission. We compared the performance from January 2016 to March 2019 for HO-BSI and the 6 NHSN-defined HAIs: central-line–associated bloodstream infection (CLABSI), catheter-associated urinary tract infection (CAUTI), Clostridioides difficile, methicillin-resistant Staphylococcus aureus (MRSA) bacteremia, abdominal hysterectomy surgical site infections (SSIs), and colon SSI. We calculated 2 “infection composite scores” to account for the 6 HAIs based on all observed or predicted events (score 1) and an average of the 6 HAI standardized infection ratios (SIRs; score 2). We normalized both measures to 1 for a 12-months rolling baseline. We evaluated the HO-BSI rate change over time and compared it to the change in the infection score over the same period. We compared the change in the 12-month rolling rates of the 2 HAI scores and the HO-BSI rate. Results: During the 39-month period, 3,288 HO-BSI events occurred over 9,775,118 patient days. The source of HO-BSI events included S. aureus (33.5%), P. aeruginosa (10.2%), E. coli (19.7%), K. pneumoniae (13.8%), and Candida spp (22.8%). HO-BSI event rates decreased by 17.3% from 12-month rolling baseline to last 12 months (3.70 vs 3.06 per 10,000 patient days). Similarly, 7,648 HAI events were observed, with the source of events being Clostridioides difficile (57.0%), CAUTI (15.1%), CLABSI (12.8%), MRSA (7.0%), colon SSI (6.4%), and abdominal hysterectomy SSI (1.7%). The 2 HAI scores and the HO-BSI rate all showed a notable decrease from the 2016 baseline period (Fig. 1). The reductions in the HAI scores were both strongly correlated with the reduction in the HO-BSI rate, with the HAI score 1 having a stronger correlation (r = 0.949; P < .001) than was observed for HAI score 2 (r = 0.867; P < .001). Conclusions: Utilization of a HO-BSI measure may prove useful as a correlated but distinct marker of infection prevention improvement or trends. HO-BSI could be useful as an objective electronically obtainable measure to assist in the evaluation of performance within and across facilities.

Funding: None

Disclosures: None

Background: Adverse events following immunization (AEFI) surveillance largely depends on the ability of the healthcare worker (HCW) to timely detect and report cases using the correct reporting tools through an appropriate system. AEFI surveillance is carried out regularly during both routine immunization services and supplemental immunization activities in the state. Objective: We assessed knowledge of adverse events following immunization reporting tools and system among primary HCWs in Jigawa state, northwestern Nigeria. Method: A descriptive cross-sectional design was used for this study. A multistage sampling technique was used to select 290 HCWs that had spent at least 6 months in immunization units of primary healthcare centers of Jigawa state. Data were collected using pretested self-administered structured questionnaire with open and closed ended questions and were analyzed using IBM SPSS version 20 software. All statistical tests were 2-tailed with P < .05 as the statistical significance level. Results: Most of the primary HCWs (93.2%) had AEFI reporting forms in their health facilities, and 68.9% said that the AEFI reporting form could be obtained from a focal or contact person in the health facility. Up to 96.4% of the primary HCWs were aware of how to report AEFI. Also, ~76.6% of primary HCWs knew the correct AEFI reporting flow, but only 15.8% knew that only serious AEFIs are reported. Furthermore, ~78.8% and 19.4% of HCWs mentioned telephone and filling forms as some of the appropriate methods of AEFI notification, respectively. Conclusions: Most primary HCWs had reporting forms in their health facilities and were aware of how to report an AEFI. Most of the respondents knew the correct AEFI reporting flow. The state in collaboration with local government authorities should provide quality training on AEFI reporting and reporting system.

Funding: None

Disclosures: None

Background: Carbapenemase-producing Enterobacterales (CPE) have become an increasingly common cause of hospital-acquired infections while their reservoirs within the clinical setting remain poorly understood. Outbreaks have been linked to hospital sinks, which have been shown to harbor and, under certain conditions, disperse CPE to surrounding surfaces. Hospital and laboratory studies have proposed that Gram-negative organisms, including CPE, can migrate through plumbing biofilms, leading to widespread contamination of the drainage system. Methods: To assess the prevalence of CPE in hospital sinks, drain swabs and waste trap water samples were taken from 10 sinks in 10 hospitals. Hospitals were in different regions of England; 4 had reported recent cases of CPE infection. To investigate spread and dispersal of CPE, waste traps from a single hospital were installed in a laboratory model sink system. Built to simulate a clinical setting, the model incorporated 12 sinks, 6 of which were connected through a common waste pipe. All 12 taps were automatically flushed. Drainage was automatically controlled. Nutrients were provided daily to maintain the bacterial populations, which were regularly sampled to monitor their composition. At 3 weeks after installation, the waste traps were subjected to a drainage backflow event. Waste trap water populations continued to be monitored, and when transfer between sinks was suspected, isolates were characterized and compared using whole-genome sequencing. Results: Between January and June 2019, 200 samples were taken from 103 sinks. In total, 24 (23%) sinks (in 8 hospitals) harbored CRE; of which 10 (in 5 hospitals) harbored at least 1 CPE. Immediately after a backflow event in the laboratory model system, 2 KPC-producing E. cloacae were recovered from a waste trap in which CPE had not been previously detected. The isolates were identified as ST501 and ST31 and were genetically indistinguishable from those colonizing sinks elsewhere in the system. Following intersink transfer, KPC-producing E. cloacae ST501 successfully integrated into the microbiome of the recipient sink and was detected in the waste trap water at least 6 months after the backflow event. At 2 and 3 months after the backflow, other intersink transfers involving Escherichia coli and KPC-producing E. cloacae were also observed. Conclusions: Sink waste traps and drains are a reservoir for CPE in hospitals. Once established, CPE contamination might not be confined to a single sink and could spread through wastewater plumbing. Hospitals frequently report drainage problems, which could cause or facilitate CPE transfer between sinks and could lead to long-term establishment.

Funding: None

Disclosures: None

Background: Accurate identification of Clostridioides difficile infections (CDIs) from electronic data sources is important for surveillance. We evaluated how frequently laboratory findings were supported by diagnostic coding and treatment data in the electronic health record. Methods: We analyzed a retrospective cohort of patients in the Veterans’ Affairs Health System from 2006 through 2016. A CDI event was defined as a positive laboratory test for C. difficile toxin or toxin genes in the inpatient, outpatient, or long-term care setting with no prior positive test in the preceding 14 days. Events were classified as incident (no CDI in the prior 56 days), or recurrent (CDI in the prior 56 days) and were evaluated for evidence of clinical diagnosis based on International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) and ICD-10-CM codes and at least 1 dose of an anti-CDI agent (intravenous or oral metronidazole, fidaxomicin, or oral vancomycin). We further assessed the possibility of treatment without testing by quantifying positive laboratory tests and diagnostic codes among inpatients receiving an anti-CDI agent. A course of anti-CDI therapy was defined as continuous treatment with the same drug. Results: Among 119,063 incident and recurrent CDI events, 70,114 (58.9%) had a diagnosis code and 15,850 (13.3%) had no accompanying treatment. The proportion of patients with ICD codes was highest among patients treated with fidaxomicin (82.6% of 906) or oral vancomycin (74.3% of 30,777) and was lower among patients receiving metronidazole (63.3% of 103,231) and those without treatment (29.9% of 15,850). The proportion of events with ICD codes and treatment was similar between incident and recurrent episodes. During the study period, there were ~470,000 inpatient courses of metronidazole, fidaxomicin, and oral vancomycin. Table 1 shows the presence of ICD codes and positive laboratory tests by anti-CDI agents. Among 51,100 courses of oral vancomycin, 51% had an ICD code and 44% had a positive test for C. difficile within 7 days of treatment initiation. Among 1,013 courses of fidaxomicin, 79% had an ICD code and 56% had a positive laboratory test. Conclusions: In this large cohort, there was evidence of substantial CDI treatment without confirmatory C. difficile testing and, to a lesser extent, some positive tests without accompanying treatment or coding. A combination of data sources may be needed to more accurately identify CDI from electronic health records for surveillance purposes.

Funding: None

Disclosures: None

Background: A sterile processing service (SPS) technician was found to inadequately clean fiberoptic endoscope channels during reprocessing prior to high-level disinfection. Channels were only brushed once and 7 of 30 audited scopes had measurable bioburden. Consistent with VA policy, a retrospective investigation, along with public disclosure, was performed. Methods: A potentially exposed case (PEC) was defined as any patient who had flexible fiberoptic endoscopy between April 19, 2015, and June 23, 2017, when the identified SPS technician worked in the endoscope reprocessing station. Using the internal log of the automated high–level disinfection equipment (Medivators/Cantel, Minneapolis, MN), device serial numbers were matched to patients in endoscopy suite procedure logs. Additionally, the VA Corporate Data Warehouse (CDW) was queried for CPT and International Classification of Disease, Ninth Revision (ICD-9) and ICD-10 procedure codes to verify identified cases and to search for other PECs. All PECs were notified by telephone and mail, and serologic testing for human immunodeficiency virus (HIV-1), hepatitis C virus (HCV), and hepatitis B virus (HBV) was offered. Results were compared to prior bloodborne pathogen (BBP) testing results extracted from the CDW. Facility microbiology laboratory records of positive cultures/microscopy for enteric pathogens also were compared to the list of PECs; no active testing was performed. Results: Of the 565 PECs, 552 (98%) were successfully contacted. 8 declined testing or preferred non-VA testing, and 22 died before testing could be initiated. Repeat testing at 6 months was requested for PECs who had initial testing performed <6 months after exposure; 32 refused additional tests or did not respond to additional requests. In total, 522 PECs (92%) had testing performed for 1 or more BBPs: (1) 521 were anti–HIV negative with 1 previously known positive; (2) 481 were anti–HCV negative—43 were previously known positive and 1 PEC with an undetectable HCV viral load was newly identified; (3) 461 were negative for both HBV core or surface antibodies and surface antigen—32 were previously known positive and 17 were newly positive for one or both antibody tests with negative HBV surface antigen. Of 17 newly identified positive PECs, 16 had undetectable HBV DNA; 1 died prior to HBV DNA testing. Conclusions: There was no evidence of transmission of BBPs in this cohort of PECs who had procedures with potentially improperly cleaned fiberoptic endoscopes. Although not all patients completed all retrospective BBP testing, <10% were missing all or some tests. Local passive surveillance did not indicate enteric pathogen transmission. Additional education regarding and monitoring of reprocessing procedures have been instituted.

Funding: None

Disclosures: None

Background: Annually, an estimated 1.03 million leptospirosis cases lead to 2.9 million disability adjusted life years. A cluster of fever cases was reported in Keerakadu village, Kollihills block in Namakkal district of Tamilnadu state, India, on April 28, 2017. We investigated to control the outbreak. Methods: We did a cross-sectional survey between April 29 and May 1. We defined a case of fever as any resident of Keerakadu village with fever for >2 days, with or without headache or myalgia, between April 15 and May 1, 2017. We conducted active surveillance. We reviewed medical records. We collected the line list from nearby health centers. We computed proportions to calculate the attack rate. We collected 11 serum samples and tested for dengue, scrub typhus, hepatitis A and leptospirosis by IgM ELISA method. We did a Widal slide agglutination test. We conducted an environmental survey to identify water sources. We performed a dengue larval survey. We collected 5 water samples: 1 from unprotected well, 1 from overhead tank and 3 from the houses of residents. We tested for fecal coliforms in the district public health laboratory. Results: The population of Keeradu village was 540. We identified 11 cases, for an attack rate of 2% (11 of 540). The hospitalization rate of cases was 81% (9 of 11). Median age was 45 years (range, 23–65). Of 11 samples, 3 were positive for leptospirosis; all were negative for dengue, scrub typhus, hepatitis A, and typhoid. The single water source for the whole village was an open, unprotected well. This well supplied water every day to the community, both for drinking purpose and domestic use. No breeding of dengue larva was observed. All the 5 water samples tested positive for fecal coliforms. Water was not chlorinated regularly. All patients were isolated and treated in the primary health center. Prophylactic antibiotics were given to the whole community. Conclusions: There was a leptospirosis outbreak in Keerakadu village, probably due to contaminated water from unprotected well. There were no cases after May 1, 2017. We recommended that the community chlorinate the water regularly and protect the well. We also recommend continued surveillance and a rodent survey.

Funding: None

Disclosures: None

Background: The Department of Veterans’ Affairs (VA) operates 146 hospitals providing healthcare to >6 million veterans annually, including dental care to qualified veterans. Although bloodborne pathogen transmission after dental procedures is rare, little is known of risk when there are breaches. A standardized approach to performing lookback investigations after dental infection control breaches could better quantify these risks. We reviewed dental lookback investigations from the past decade conducted by our VA office for lessons learned to improve processes. Methods: Three VA hospitals had dental infection control breaches during 1992–2016. Facility A had dental instruments that were not cleaned according to the manufacturer’s recommendations, and dentists at facilities B and C failed to adhere to proper infection control standards. Exposed veterans who underwent dental procedures were notified of possible exposure and were offered testing for human immunodeficiency (HIV-1), hepatitis B virus (HBV), and hepatitis C virus (HCV). Prior clinical testing was also reviewed. Newly identified positive results were compared to known positives prior to exposure to determine strain relatedness when sufficient plasma viral load was present for viral sequence comparison. Results: There were 2,939 patients with potential exposures in these dental investigations: 2,667 were tested for HBV, 2,642 were tested for HCV and 2,599 were tested for HIV-1. No evidence of viral transmission was found based on genetic sequence comparison of positive cases, but relatively few samples were available for this testing. Lessons Learned: Each facility faced different challenges with their investigation; however, several key processes were identified. (1) Early engagement by our office with local facility leadership and lookback teams resulted in more efficient investigation and testing processes. (2) To improve standardization, a lookback manual detailing of investigation procedures was created in 2009 and was updated subsequently. The contents of this manual include identifying and notifying patients; providing services to veterans responding to notifications; laboratory testing algorithms; disclosure and documentation of test results and clinical follow-up; and epidemiologic investigation of patients with newly identified infection. (3) Prompt patient notification and obtaining adequate samples for initial and follow-up pathogen genetic testing is critical. Determination of genetic linkages was greatly limited because specimens were unavailable for supplemental testing. (4) Ethics and legal counsel staff are key partners in providing guidance on appropriate disclosure procedures and documentation. (5) Designating a single mechanism for reporting results ensures consistent communication among stakeholders. (6) Education of dental staff on importance of following the manufacturers’ cleaning recommendations, not using outside equipment and reporting instances of concern promptly can help prevent future infection control breaches.

Funding: None

Disclosures: None

Background: In 2017, the IDSA and the SHEA released updated Clostridium difficile practice guidelines. Implementing institutionally accepted criteria for identifying clinically appropriate patients for testing was endorsed. When utilizing NAAT as the sole laboratory testing methodology, testing clinically symptomatic patients is important to reduce inappropriate treatment of C. difficile colonization. C. difficile rates at a regional community health system were higher than expected, and patient case reviews identified inappropriate patient testing as an issue. Therefore, the infection prevention team sought to optimize and standardize protocols surrounding appropriate patient selection for C. difficile testing. Methods: Current recommendations were evaluated, and processes formulated to implement an innovative process to support our clinicians in identifying clinically appropriate patients to test for C. difficile infection. The electronic decision support is summarized as a bundled approach with 4 best practice alerts that incorporate algorithms that warn providers of potentially inappropriate testing scenarios. These alerts include the following criteria: (1) a laxative having been administered within 48 hours of attempted order, (2) a negative test resulted within 7 days, (3) a positive test resulted within 14 days, and (4) identification of patients at high risk for C. difficile infection (based on recent long-term care facility exposure, recent inpatient hospital visits, recent antimicrobial therapy). Outcomes of our acute-care hospitals were monitored by real-time evaluation of each hospital’s quarterly C. difficile LabID standardized infection ratio (SIR) as defined by the NHSN. For statistical analyses, the cumulative second and third quarters of 2018 (before the intervention) were compared to the cumulative second and third quarters of 2019 to account for seasonality of C. difficile infections. Results: Utilizing the NHSN statistical calculator to compare 2 SIRs, there was a statistically significant decrease (P = .0026) in the largest hospital’s C. difficile LabID SIR when comparing representative preintervention cumulative quarters to the postintervention cumulative quarters. Although the other hospitals did not see a statistically significant decrease in their C. difficile LabID SIR, a clinically significant decrease was appreciated for 2 of our hospitals. Conclusions: Electronic health record–based decision support helps clinicians identify clinically appropriate patients to test by NAAT alone for C. difficile infection. By limiting the number of patients tested without clinical signs or symptoms of infection and/or after receiving laxatives, hospitals more accurately capture their true C. difficile rates and maximize reimbursement based on this measure within the CMS Safety of Care Measure.

Funding: None

Disclosures: None

Background: New York City is a gateway for emerging pathogens and global threats. In 2013, faculty from Montefiore Medical Center and Memorial Sloan Kettering developed a free half-day workshop for postgraduate trainees in antimicrobial stewardship (AS), infection prevention (IP), hospital epidemiology, and public health. This annual workshop, sponsored by the Infectious Diseases Society of New York (IDSNY), incorporates case studies and expert panel discussions on timely topics such as Ebola, Candida auris, Clostridiodes difficile, measles, nosocomial influenza, drug shortages, and AS/IP “big data.” Methods: From 2013 through 2017, the workshop involved 10–15 interactive AS/IP cases with audience response questions and panel discussions. In 2018–2019, based on feedback, the format was revised to emphasize breakout sessions in which participants actively practiced AS/IP tools, (eg, medication utilization evaluations, epidemiologic curves, and performance improvement devices). Examples of 2018–2019 cases are shown in Figure 1. A pre- and postseminar paper survey was conducted yearly to understand baseline training in AS/IP, desire for future AS/IP careers, and self-reported effectiveness of the workshop. Results: Initially, the primary audience was NYC ID fellows. From 2018 onward, we opened enrollment to pharmacy residents. Approximately 45 NYC ID fellows were eligible for the course each year. Results from 2013 to 2016 surveys were reported previously (Fig. 2). There were 32 attendees in 2018, 42 in 2019. The survey response rate was 88% in 2018 and 95% in 2019, with 68 (92%) total participants. Most participants had received previous training in IP (82%) and AS (94%) (Fig. 3). Most participants reported that the program was a good supplement to their ID training (98%) and that case studies were an effective means of learning IP (100%) and AS (98%). Furthermore, 92% stated they would like additional AS/IP training, and many since 2013 have requested a full-day course. Self-reported interest in future involvement in AS/IP increased after the workshop: IP, 68%–83% (P =.04) and AS, 88%–91% (P = .61). Conclusions: Most trainees reported satisfaction with the workshop and case-study learning method; interest in future AS/IP careers increased after the seminar. We intend to explore Funding: to expand to a full-day program for all NYC postgraduate trainees and AS/IP junior faculty. As such, we hope to obtain the endorsement of professional societies such as SHEA. This workshop could address a crucial educational gap in AS/IP postgraduate training and help sustain our future workforce.

Funding: None

Disclosures: None

Background: Carbapenemase-producing carbapenem-resistant Enterobacteriaceae (CP-CRE) pose a serious public health threat. The CDC guidelines for combating CP-CRE include a recommendation to screen selected high-risk patients. Objective: We describe a program to identify and screen patients at risk for CP-CRE. Setting: An academic, tertiary-care center with 1,297 licensed beds and 62,071 admissions per year. Methods: A report was created in the electronic medical record (EMR) to identify adult patients admitted in the previous 24 hours from countries and states with known CP-CRE transmission based on address and ZIP code. Patients with a known travel history outside the United States were also reviewed, but these data were inaccurate. Initially, a physician from Infection Prevention and Control (IPAC) placed orders for CP-CRE screening of these patients. Subsequently, a nursing protocol was developed to facilitate infection preventionists placing orders for CP-CRE screening earlier in the eligible patient’s hospital stay. An electronic communication is sent via the EMR alerting the patient care team to the order, the rationale for the order, and links to a tool kit with resources to help answer patient questions. A single perirectal swab is obtained by the nurse caring for the patient and is tested for Klebsiella pneumoniae carbapenemase (KPC), New Delhi metallo-β-lactamase (NDM), oxacillinase-48 (OXA-48), and Verona integron-encoded metallo-β-lactamase (VIM) by polymerase chain reaction (PCR). Results: From May 2018 to November 2019, 688 patients were screened for CP-CRE using the case-finding report and the nursing protocol. Overall, 9 patients with CP-CRE were identified: 1 KPC, 2 NDM, and 5 OXA-48, and 1 patient was identified to have both NDM and OXA-48. The yield of 1.3% from this screening is higher than that reported previously in the literature. Use of the nursing protocol has enabled IPAC to complete timely CP-CRE surveillance and prevent transmission to other patients. We are currently using a similar process to identify and screen persons at risk for the emerging infection Candida auris.Conclusions: The EMR can be leveraged for early identification and screening of patients with epidemiologically significant pathogens. Protocols within the EMR can be effectively replicated and modified to respond to emerging infections and changing surveillance guidelines.

Funding: None

Disclosures: Consulting fee-—Merck (Priya Sampathkumar)

Background: The identification of risk factors for infections in surgical patients with lower-limb fractures and blood transfusions has increased in recent years. Surgical site infections (SSIs) increase hospitalization, care costs, and patient suffering. Correction surgery for lower-limb fractures and blood transfusion is quite common between surgical procedures. The aim of this study was to describe the relationship between blood transfusion and SSI in patients undergoing orthopedic surgery on lower limbs. Methods: We conducted a prospective cohort study to identify risk factors for SSI in blood transfused patients undergoing fracture repair in lower-limb surgeries between February 2017 and May 2019 in 2 reference tertiary-care hospitals in Belo Horizonte, a city of 3 million people in Brazil. Data regarding patient characteristics, surgical procedures, blood transfusions, and surgical infections were collected. Patient characterization was performed by calculating the absolute and relative frequencies of categorical variables and calculating mean, median, minimum, maximum, standard deviation, and coefficient of variation for quantitative variables. The incidence of surgical site infection, the risk of postoperative hospital death, and the total length of hospital stay were calculated by point estimates and 95% confidence intervals identified by statistical tests of bilateral hypotheses, considering the level of significance of 5%. A multivariate analysis (logistic regression) was performed to identify SSI risk factors. Results: Patients who had an indication for blood transfusion (n = 38) but who did not receive blood (n = 4) had significantly lower hemoglobin, comparing discharge with admission, than the group who received blood. Intraoperative transfusion was a risk factor for SSI (OR, 4.7) (Fig. 1). Among the 205 patients with no indication for transfusion, 98 received blood even without the indication: there was no difference in hemoglobin outcome when discharge and admission were compared, and the 98 patients were exposed to unnecessary risk. Regarding restrictive versus liberal transfusion strategies, there were differences in the variables, age (P = .000), duration of surgery (P = .003), number of comorbidities (P = .000), body mass index (BMI) (P = .027), previous hemoglobin (P = .000), and high hemoglobin (P = .000), considering the transfusion practice employed (Fig. 2). Conclusions: The indications for and definition of protocols and careful evaluation of blood transfusion are critical to avoid infectious complications in orthopedic patients with lower-limb fractures.

Funding: None

Disclosures: None

Background: In the state of Wisconsin, 3%–4% of bats submitted for rabies testing are positive. Inpatient bat encounters at 2 affiliated healthcare facilities at nearly the same time were brought to the attention of the infection prevention and control (IPC) team. The first bat was captured in a patient room and was submitted for testing. Postexposure prophylaxis (PEP) was initiated for 1 patient before the bat testing results came back negative. The second bat was found in a transplant unit hallway and was released before we could request testing. We observed significant variations in responses, including decision to administer PEP and submission of bats for rabies testing. The IPC team developed a protocol to minimize unnecessary PEP, to prevent nosocomial rabies infection from bat exposure, and to limit associated panic. Methods: A systematic literature review of multiple databases was performed. A search of nonscientific articles using Google was also performed to assess unpublished inpatient bat encounters. A workgroup was established including IPC staff, physicians, and facilities management. The county animal services department and the state public health department veterinarian were consulted to aid in development of a protocol. Results: Literature review yielded a single report of a bat discovered in a neonatal intensive care unit (NICU). A lack of protocol resulted in PEP administration to 7 neonates without observed exposure after the bat was released instead of being submitted for testing. Of the first 100 articles retrieved via Google search of “bat in hospital,” 9 pertained to nosocomial discovery of bats in 5 different states over the past 7 years. Encounters included infestations requiring unit shutdowns and PEP administration. One tertiary-care referral center reported 10 encounters per year but did not elaborate on associated procedures. The county animal services staff assisted in training maintenance and engineering services (MES) personnel on how to secure bats for testing and helped develop a “bat kit” with protective gear and equipment to do so safely. In the new protocol, an inpatient bat encounter prompts personnel to capture the bat and begin an investigation into known or potential occult exposure. Known or potential exposures merit submission of the bat for rabies testing, the results of which guide PEP recommendations. All encounters are investigated for point of entry or roost. Conclusions: Inpatient bat encounters are not uncommon. Encounters should prompt systematic assessment for exposures and an investigation of the root cause. Following a protocol may limit unnecessary PEP administration, prevent nosocomial transmission of rabies from bat to patient, and attenuate associated anxiety.

Funding: None

Disclosures: None

Background: Linezolid an oxazolidinone drug available in both parenteral and oral formulations has emerged as a novel alternative to vancomycin and other second-generation drugs for the treatment of infections from gram-positive cocci. Clinical isolates of linezolid-resistant staphylococci and enterococci were reported. Since then, linezolid-resistant strains have become an increasing problem worldwide. The most frequently reported mechanisms of linezolid resistance include the mutation in 23S ribosomal nucleic acid and presence of cfr gene. Methicillin-resistant coagulase-negative staphylococci (MR-CoNS) and vancomycin-resistant Enterococcus (VRE) have become a worrisome clinical problem. Objective: Therefore, we aimed to determine the distribution of linezolid-resistant strains in an inpatient setting of a tertiary-care hospital in India and to evaluate the resistance mechanisms among these isolates. In addition, the clonal diversity of the isolates was determined by pulsed-field gel electrophoresis (PFGE). Methods: The distribution, clonal diversity, and resistance mechanism of linezolid resistant-Staphylococcus haemolyticus (LRSH) strains were determined. The isolates were identified by MALDI-TOF. The mechanism of resistance was determined by sequence analysis of the domain V of 23SrRNA and screening for cfr gene. Clonal relatedness was defined by PFGE. Results: In total, 13 LRSH isolates were recovered from pus specimens. The 13 LRSH strains isolated had an linezolid MIC of 256 g/mL. Sequencing results revealed G2576T mutations in 7 (53.8%), G2447U in 4 (30.7%) and C2534U in 1 (7.6%) isolate of S. haemolyticus. One isolate of S. haemolyticus showed 2 simultaneous mutations (G2576T and G2447U) in the domain V region of 23Sr RNA gene. PFGE of the LR-SH isolates revealed the presence of 11 clones. Of the 11 clones, clones I and II had 2 isolates each. Isolates of clone I exhibited a band pattern identical with the previous isolates of LRSH isolated from the orthopedic unit. Similarly, isolates of clone II also shared the same band pattern with the previous LRSH isolates from the dermatology unit of our center. Conclusions: This study highlights the importance of continuous monitoring of vigilance of linezolid resistance in staphylococci. Rationalizing the use of linezolid and implementing methods to control the spread of hospital clones is of paramount importance to prevent further dissemination of these strains.

Funding: None

Disclosures: None

Background:Clostridioides difficile infection (CDI) following colorectal surgery can lead to significant adverse outcomes. Although previous studies have identified risk factors for CDI, their relative importance for predicting complications remains unclear. Objective: We sought to use machine-learning algorithms to accurately determine which perioperative risk factors are most predictive of adverse outcomes after CDI. Methods: The National Surgical Quality Improvement Project (NSQIP) database was used to identify all patients who developed CDI after a colorectal operation in 2016 (N = 14,392). We excluded patients without CDI and patients <18 years of age. Any missing data were replaced with multivariate singular value decomposition imputation. We collected data on patient demographics, comorbidities, preoperative laboratory values, operative details, and outcomes, including infectious, cardiovascular, hematologic, renal, and pulmonary complications, unplanned returns to the operating room (RTOR), non–home discharge, readmission, and mortality. Data were univariably assessed for significant association with outcomes. If an input variable significantly correlated with ≥5 outcomes, it was included in our machine-learning models. We utilized bootstrap aggregation with random forests to improve prediction accuracy. We then calculated each input variable’s importance to the model outcome (VIP). The VIPs of each variable were averaged to yield an overall impact. Each model’s accuracy was determined by the area under the receiver operator curve (AUROC). Results: There were 841 patients in our cohort (median age 66 years (IQR, 55–75.8), 482 (57%) were women, and the mean American Society of Anesthesiologists [ASA] class score was 2.9 (SD, ±0.7). Of all colorectal surgeries, 172 (20.5%) were emergent. Overall mortality was 3.8% (n=32), and 371 patients (44.1%) experienced at least 1 postoperative complication, of which infectious complications (eg, septic shock, sepsis, wound infection, urinary tract infection) were most common (n=255, 30.3%). The RTOR rate was 10.3% (n = 87), the non–home discharge rate was 23.8% (n = 200), and the readmission rate was 30.9% (n = 260). The input variables most predictive of any adverse outcome were hematocrit (VIP, 24.9%), ASA class (VIP, 24.4%), creatinine (VIP, 17.4%), and prealbumin (VIP, 11.6%). The probability of any adverse outcome was 90.6% in the setting of hematocrit ≤27%, ASA class ≥3, creatinine ≥1.6 mg/dL, and prealbumin ≤3.1 mg/dL. All machine-learning models had an AUROC ≥0.99. Conclusions: Although nonpatient factors can contribute to unfavorable outcomes in patients with CDI following colorectal surgery, we identified 4 patient-specific variables that account for almost 80% of any adverse outcomes. Although further prospective study is needed, individuals with these preoperative risk factors could consider delaying their elective colorectal operations until they are medically optimized.

Funding: None

Disclosures: None

Background:C. difficile is the leading healthcare-associated pathogen. The C. difficile real-time polymerase chain reaction (PCR) stool test, used by >70% of hospitals, is highly sensitive but cannot differentiate colonization from infection. Inappropriate C. difficile testing may result in overdiagnosis and unnecessary treatment. Healthcare costs attributed to C. difficile are substantial, but the economic burden associated with C. difficile false positives in colonized patients is poorly understood. C. difficile PCR cycle threshold (CT) is as an inverse proxy for organism burden; high CT (≥30.9) has a high (>98%) negative predictive value compared to the reference gold standard, thus is a marker of colonization. Conversely, a low CT (≤28.0) suggests high organism burden and high specificity for true infection. Methods: A propensity score matching model for cost per hospitalization was developed to determine the costs of a hospital stay associated with C. difficile and to isolate the financial impacts of both true C. difficile infection and false positives. Relevant predictors of C. difficile positivity used in the model were age, Charlson comorbidity index, white blood cell count, and creatinine. We used CT data to identify and compare 3 inpatient groups: (1) true CDI, (2) C. difficile colonization, and (3) C. difficile negative. Results: A diagnosis of C. difficile adds significantly (>$3,000) to unadjusted hospital cost compared to a negative result. Propensity-adjusted analyses demonstrated that C. difficile colonization was associated with significantly increased (median, $5,000) hospital cost whereas any positive or true diagnoses of C. difficile were not associated with increased cost. Colonized patients also had significantly higher lengths of stay (1 day) and cost per length of stay ($218 per day). Conclusions:This is the first C. difficile cost analysis to utilize PCR CT data to differentiate colonization. Surprisingly, patients with a high CT had disproportionately higher hospital costs compared to matched C. difficile–negative patients, which was not seen among patients with a low CT or with any positive result. We hypothesize that this unexpected finding may be due to misdiagnosis and mistreatment of diarrhea not caused by C. difficile or unadjusted factors associated with high cost and non–C. difficile diarrhea. In addition, the discrepantly high cost attributed to C. difficile diagnosis cited in the literature ($3,000–11,000 per hospitalized case) could be explained by the common use of administrative data to identify C. difficile cases and controls as opposed to our study, which directly linked cost data to C. difficile–positive and –negative test results.

Funding: None

Disclosures: None

Background: Ceftolozane-tazobactam (C/T) and ceftazidime-avibactam (C/A) are new β-lactam/β-lactamase combination antibiotics that were approved by the FDA in 2014 and 2015, respectively, to treat complicated intra-abdominal and urinary tract infections. They are commonly used to treat multidrug-resistant Pseudomonas aeruginosa (MDRPA) and carbapenem-resistant Enterobacteriaceae (CRE) infections at any site. Both medications are also often used as salvage therapy when empiric therapy has failed or when the infectious organism tests resistant to all other available antibiotics. The purpose of this review is to present the clinical experience and reported clinical success rates of C/T and C/A. Methods: PubMed, EMBASE, and Google Scholar were searched from January 1, 2013, through October 1, 2019, for publications detailing clinical experience with C/T and C/A in patients with CRE and MDRPA infections. Included study designs were extended cases series and clinical observational studies. Information on infection type, bacterial agent, salvage therapy uses, clinical success, and resistance development during treatment were abstracted. Meta-regression analysis was used to determine the pooled effectiveness of C/T and C/A among included studies. Results: The literature search returned 1,645 publications. After exclusion criteria were applied, 16 publications representing 769 patients were retained. The study population was mostly male (pooled average, 62%). The major comorbidities represented in the pooled population were solid organ transplantation (20.0%), kidney disease (19.5%), cardiovascular disease (15.3%), and diabetes (15.3%). Pneumonia was the predominant infection type (41.4%) and MDRPA was the pathogen most frequently evaluated (57.7%). The pooled clinical success rate was 70.2% (95% CI, 64.5%–75.3%). Also, 10 studies explicitly evaluated C/A or C/T as salvage therapy. The pooled clinical success rate for salvage therapy studies was 75.2% (95% CI, 69.7%–80.0%). Development of resistance to C/T or C/A during or after treatment was reported for 2.0% of the population. Conclusion: Overall, these medications have a high clinical success rate in patients with severe and complicated infections and limited treatment options. Pooled clinical success rates were high (70.2%) and the medications were particularly effective as salvage therapy. Resistance rates were low, although this could have been biased by the small percentage of studies that reported on this outcome. More longitudinal studies comparing the effectiveness of C/T and C/A against other antibiotic regimens are needed.

Funding: None

Disclosures: None

Background: In The Netherlands, the national guidelines on Methicillin-Resistant Staphylococcus aureus (MRSA) prevention and control advocate screening of healthcare workers (HCWs) after unprotected exposure to MRSA carriers. Although this strategy is largely successful, contact tracing of staff is a time-consuming and costly component. We evaluated our contact tracing policy for HCWs over the years 2010–2018. Methods: A retrospective, observational study was performed in a Dutch teaching hospital. All HCWs who had unprotected contact with an MRSA carrier were included in contact tracing. When there had been a long period of unprotected admission prior to an MRSA finding, or when the index case was an HCW, the entire (nursing) team was tested. All samples of HCWs who were tested for MRSA carriage as part of contact tracing from 2010 until 2018 were included. A pooled nose, throat, and perineum swab was collected using the eSwab medium (Copan) and inoculated on chromID MRSA agar plates (bioMérieux) after enrichment in a broth. Molecular typing was performed using multiple-locus variable number of tandem repeat analysis (MLVA). Results: In total, we included 8,849 samples (range, 677–1,448 samples per year) from 287 contact tracings (range, 26–55 contact tracings per year). Overall, 32 HCWs were colonized with MRSA (0.36%; 95% CI, 0.26%–0.51%). None of them developed a clinical infection. Moreover, 8 HCWs (0.10%; 95% CI, 0.05%–0.19%) were colonized with the same MLVA type as the index case and were detected in 6 of 287 contact tracings (2%). In 4 of 8 of these cases, a positive HCW was the index for undertaking contact tracing. In 3 of 8 cases, it was clear that the HCW who was identified in the contact tracing was the source of the outbreak and was the cause of invasive MRSA infections in patients. Notably, a different MLVA type as the index case was found in 24 HCWs (0.27%; 95% CI, 0.18%–0.40%) of whom 7 of 24 HCWs (29.2%) were intermittent carriers. Conclusions: This study revealed a sustained low MRSA prevalence among samples in contact tracing of HCWs over 9 years. Furthermore, it shows that when MRSA contact tracing is performed according to the national guideline, only 1 of 1,000 samples results in a secondary case. This is similar to the population carriage rate of MRSA in The Netherlands. More frequently, an unrelated strain is found. These findings raise questions regarding the efficacy of the current strategy to perform contact tracing after unprotected exposure.

Funding: None

Disclosures: None

Background: The hallmark of Middle East respiratory syndrome coronavirus (MERS-CoV) disease is the ability to cause major healthcare-associated outbreaks with superspreading events leading to massive transmission and excessive morbidity and mortality. This abstract provides overview of MERS-CoV multi-healthcare facilities outbreak in Riyadh in June 2017, with focus on cluster reported and the control measures taken at King Saud Medical City. The outbreak began with a patient who presented with acute renal failure requiring hemodialysis and became a MERS-CoV superspreader, igniting the cluster of cases in several hospitals in King Saud Medical City. Methods: For epidemiologic investigation, a case was defined as any patient with laboratory-confirmed MERS-CoV infection with connection to the affected healthcare facilities. Contact tracing and testing were performed according to the Ministry of Health (MOH) guidelines. MERS-CoV testing was recommended for HCWs who had unprotected close contact with a confirmed case. Considering the superspreading phenomena, contact tracing was included all persons attended the same area with the positive case either as a patient, an HCW, or a patient’s visitor or companion. Laboratory confirmation was conducted using real-time RT-PCR. Genome sequencing and phylogenetic analysis were performed for available MERS-CoV rRT-PCR–positive samples by the CDC. The infection control measures applied included decreasing patient load through downsizing emergency department acceptance, maintaining low elective services, limiting inpatient admissions, and encouraging discharge. Early detection and quarantining of any suspected cases took place through extensive contact tracing, properly triaging all patients upon admission, consistent monitoring of inpatients and HCWs for any emerging acute respiratory illness, allocation of more single rooms inside the facility and staff dormitory, and extending the services of virology laboratory to get timely results. Further measures consisted of extensive education on infection control practices, monitoring healthcare worker adherence, reassuring the public by maintaining transparency of published reports, and launching a hotline to respond to HCW concerns. Moreover, travel restrictions applied to any person with a history of exposure to a confirmed MERS-CoV case during the 2-week monitoring period. Results: Overall, 44 cases of MERS-CoV infection were reported from 3 simultaneous clusters during the 2017 Riyadh outbreak, including 11 fatal cases. Among all of the cases, 29 cases were reported at King Saud Medical City. The outbreak at KSMC required 30 days to be controlled. Conclusions: High vigilance for early detection is a key control measure. To be more sensitive, point-of-care MERS-CoV testing is required because clinical suspicion is challenging in patients presenting with acute renal failure.

Funding: None

Disclosures: None

Background: Annual influenza immunization of healthcare workers (HCWs) is widely recommended to reduce the risk of healthcare-associated influenza (HAI). Although there is a clear association between higher HCW immunization and reduced incidence of HAI, data in acute care are lacking compared to the nursing home setting. Objective: Our goal was to assess the association between HCW influenza immunization and the incidence of HAI across 2 acute-care facilities. Methods: A multicenter prospective cohort study was undertaken at 2 acute-care hospitals including 1 university and 1 community-based academic hospital. Any patient prospectively identified with HAI was included between 2013–2014 and 2018–2019, whereas 2017–2018 was excluded due to vaccine mismatch. The HCW influenza immunization rate was defined as the proportion of HCWs (nurses and other allied and support staff but excluding physicians) immunized prior to December 15. A case of HAI was defined as laboratory-confirmed influenza A or B with symptom onset >72 hours after admission. The association between inpatient ward HCW influenza immunization rate and the incidence of HAI was compared using a Poisson regression analysis adjusting for hospital and influenza season. Results: Over 5 influenza seasons, the incidences of HAI at either facility were 0.24 and 0.22 per 1,000 patient days, whereas the median HCW influenza immunization rates were 57.3% (IQR, 42.5%–66.4%) and 66.6% (IQR, 50.6%–76.8%), respectively. When adjusting for hospital and influenza season in the multivariate analysis, HCW influenza immunization rates of 65% and 70% were not associated with HAI incidence. In contrast, HCW influenza immunization rates ≥75% was associated with a trend toward reduced HAI (IRR, 0.65; 95% CI, 0.39–1.08; P = .096) whereas inpatient wards above 80% immunization had significantly lower risk of HAI (IRR, 0.28; 95% CI, 0.089–0.89; P = .03). Conclusions: The risk of HAI across 2 acute-care hospitals was significantly lower among inpatient wards achieving HCW influenza immunization rates >80%. Acute-care facilities should aim for this minimum HCW immunization rate to protect patients from the complications of HAI.

Funding: None

Disclosures: None

Background: A critical step during outbreak investigations is actively screening for additional cases to assess ongoing transmission. In the healthcare setting, one widely used method is point-prevalence screening on the whole unit where a positive patient is housed. Although this point-prevalence approach captures the “place,” it can miss the “person” and “time” elements that define the population-at-risk. Methods: At University of North Carolina (UNC) Hospitals, we used business intelligence tools to build a query that harnesses the admission, discharge, and transfer (ADT) data from the electronic medical record (EMR). Using this data identifies every patient who overlapped in time and space with a positive patient. An additional query identifies currently admitted overlap patients and their current location. During an outbreak investigation, an analyst executes these queries in the mornings when surveillance screens are scheduled. The queries generate a list of patients to screen that are prioritized on the number of days they were in the same unit with the positive patient. This overlap methodology successfully captures the person, place, and time associated with possible disease transmission. We implemented the overlap method for the last 3 months following 1 year of point-prevalence approach screening during a novel disease outbreak at UNC Hospitals. Results: In total, 4,385 unique patients overlapped with previously identified infected or colonized patients, of which 781 (17.8%) from 40 departments were screened over 15 months. During a subsequent, currently ongoing, outbreak, we are utilizing the overlap method and in 6 weeks have already screened 161 of the 1,234 overlapping patients (13%). After 3 rounds of overlap screening, we have already been able to identify 1 additional positive patient. This patient was on the same unit as patient zero 4 months prior but was readmitted to a unit that would not have received a point-prevalence screen using the standard approach. Conclusions: Surveillance screening is a time-consuming, resource-intensive effort that requires collaboration between infection prevention, clinical staff, patients, and the laboratory. By harnessing EMR ADT data, we can better target the population at risk and more efficiently utilize resources during outbreak investigations. In addition, the overlap method fills a gap in the current CDC guidelines by focusing on patients who were on the same unit with any positive patient, including those who discharged and readmitted. Most importantly, we identified an additional positive patient that would not have been detected through a point-prevalence screen, helping us prevent further disease transmission.

Funding: None

Disclosures: None