Poster Viewing

Objectives: Stool storage is key to researching the association between gastrointestinal dysbiosis and disease states. Fecal microbiota transplantation using frozen filtrate is used for patients with recurrent Clostridioides difficile infection. This study determined the impact of storing stool and frozen filtrate on microbiome composition. Methods: Fresh stool was obtained from a high-diversity (HD) and low-diversity (LD) donor. Aliquots were stored at room temperature (RT), 5C, and at -20C for 24 and 48 hours, or processed immediately. Fresh stool was also homogenized with both 0.9N-sterile-saline and 0.9N-sterile-saline containing 10%-glycerol. Resulting filtrate aliquots were frozen at -20C and at -80C. At baseline and after 7, 9, 12, 18 and 24M storage, gDNA was isolated using the MO BIO PowerSoil® DNA Isolation Kit. 16s rRNA gene amplicon sequencing targeting the V4 hypervariable region was performed on the Illumina MiSeq. Low abundant OTUs were excluded. Results: Differences in microbiota profiles were observed in both HD and LD stool when stored immediately at –20C, more so with the LD stool. Differences were also noted for LD stool stored at room temperature and less so at 5°C. Storage at RT or 5°C had no impact for the HD stool. Long-term filtrate storage of the HD stool filtrate at –80C with 10%-glycerol best preserved the bacterial microbiota profile followed by –20C with 10%-glycerol then –80C and –20C without 10%-glycerol.  Differences were also observed in the LD stool filtrate when stored in any conditions, with the least impact being –80°C with glycerol. Conclusions: To preserve microbiota profiles, storing stool at 5°C until received in the laboratory is optimal. For researchers assessing microbiota correlations with disease states, immediate gDNA extraction from stool upon receipt is ideal. For those using frozen filtrate from healthy donors for fecal transplants, long-term storage is best at –80C with 10%-glycerol. 

Objectives: A recent review by Chan et al. reported that urogenital testing alone misses a significant percentage of Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) infections among men who have sex with men (MSM):14-85% of rectal and pharyngeal infections can be missed.  We provide estimates of CT/NG prevalence among gay, bisexual, and other MSM (gbMSM) by site in an urban center in Quebec. Method: From February 2017 to June 2018, sexually-active gbMSM ≥16 years were recruited via respondent-driven sampling (RDS). Pharyngeal samples were collected by trained nurses; urine and rectal samples by participants. All samples were analyzed using the cobas 4800 CT/NG assay (Roche Diagnostics). Prevalence proportions (95% CI) were RDS-adjusted. Results: Among the 1177 participants, mainly asymptomatic, CT infection (at least one positive sample) was found in 2.7% (1.3-4.2) and NG infection in 5.9% (3.1-8.7). Proportions for CT by site: 0.3% (0.1-0.4) urine samples, 2.3% (0.9-3.8) rectal samples, and 0.2% (0.0-0.5) pharyngeal samples. For NG, these were 0.3% (0.5-1.0), 2.9% (1.2-4.7) and 3.7% (1.3-6.1), respectively.  All three sites were sampled in 1145 participants (Figure). If only urine had been tested, 35/44 (79.6%) CT infections and 72/77 (93.5%) NG infections would have been missed. Conclusions: In our study, 88% (105/121) CT/NG infections would have been missed if only urine was tested. This represents a higher proportion than found in the literature, and higher than data from provincial notifiable diseases (which also includes cases among heterosexual men and cases detected though consultation for symptoms): in 2017, 1568/9656 (16%) CT cases and 2735/4676 (58%) NG cases in men were detected only from pharyngeal or rectal sites. Sampling all sites is crucial for screening and should be promoted, especially in gbMSM.

Number of positive samples for CT or NG among 1145 participants for whom all three sites were tested

Background: Men who have sex with men (MSM) are disproportionately affected by sexually transmitted infections (STIs) including chlamydia, gonorrhea, syphilis and HIV. In addition, many men remain undiagnosed and untreated with asymptomatic infections facilitating onward transmission. Barriers to traditional STI testing include healthcare-level stigma and homophobia, lack of confidentiality, long wait-times and limited access to testing sites. Patient-initiated strategies such as self-collection and self-testing have been proposed as alternatives to increase screening and treatment. We conducted a scoping study to summarize successful testing programs, as well as demographics of MSM who benefited, to help guide the creation of a local program. Methods: A literature search was conducted using controlled vocabulary (e.g., “Sexually Transmitted Diseases”, “Mass Screening”, “Internet”) and keywords (e.g., “STI”, “screening”, “self-initiated”). The authors reviewed extracted articles and included them in the study based on title and abstract content. Included articles addressed self-sampling or self-testing methods for sexually transmitted infections in men who have sex with men in developed countries.  Results: Forty-one articles were included. Testing was either conducted through downloaded laboratory forms at traditional collection sites, mail-in specimen collection kits or rapid self-testing devices. These strategies especially benefited older men, those with partners, and those unable to access traditional testing. They decreased concerns regarding stigma and homophobia. Barriers to accessing patient-initiated testing were person or care-specific, pertaining to fear of positive tests, concerns over confidentiality, and understanding test results. Some worried their testing would get lost. Care-specific concerns included the missed opportunities for assessment and counselling when performed remotely. Conclusions: Although traditional testing methods remain the standard of care, patient-initiated testing may bridge the gap in detection of STIs in MSM. A successful program may combine both self-collection and printed requisitions, with linkage to established clinics or telehealth. Continuity of care is paramount for program success.

Over recent years there has been a dramatic shift in the diagnostic testing methods available to identify Trichomonas vaginalis, from the lower sensitivity microscopic examination of a wet mount preparation of vaginal secretions, to highly sensitive molecular assays performed on vaginal and cervical swabs, as well as urine samples.  Since there are no current surveillance or reporting programs in place, and low sensitivity microscopy is still widely used for the diagnosis of T.vaginalis, the true prevalence of infection is unknown.  Based on extensive literature review, the prevalence of T.vaginalis however, is known to vary greatly between geographic areas and among different risk groups. Objective: To establish epidemiological prevalence data for T.vaginalis in a large community based population. Methods:  Analysis of positivity rates of T.vaginalis at two large laboratories over a nine month period to determine the prevalence of infection in the communities serviced by these laboratories. Both laboratories utilized molecular testing on the BD Viper XTR platform for the detection of T.vaginalis in urine specimens and vaginal and endocervical swabs. Results: Over the nine month period of analysis, a total of 42722 endocervical and vaginal swabs were submitted, and 483 of these tested positive for T.vaginalis by molecular testing resulting in a 1.13% positivity rate.  The positivity rate for the 30080 urine specimens tested for T.vaginalis was 1.87%. Conclusion: Compared to available prevalence estimates in the literature in various geographical regions and across risk groups, the positivity rates of Trichomonas vaginalis are lower in the community-based population analyzed in this study. 

Background: Gonorrhea is a sexually transmitted infection caused by Neisseria gonorrhoeae. Infection is usually limited to mucosal surfaces (e.g., urogenital tract, rectum, throat, eyes) but disseminated infections with systemic manifestations can occur. In Manitoba, Canada, the incidence of Gonorrhea increased from 96 cases/100,000 population in 2013 to ≈250/100,000 in 2018. Objective: To review the incidence and features of disseminated Gonorrhea infections. Methods: In Manitoba, screening and reference testing for sexually transmitted infections is centralized at Cadham Provincial Laboratory (CPL). Antimicrobial susceptibility testing and NG-MAST genotyping of cultured N. gonorrhoeae isolates is performed at the National Microbiology Laboratory. Disseminated infections were identified by querying the CPL laboratory information system (LabWare LIMS) for specimens that 1) tested positive for N. gonorrhoeae and 2) had a non-mucosal source (e.g., blood, joint fluid). Cases were reviewed to determine if screening for co-infections was performed at the time of diagnosis. Results: Between January 1, 2013 and December 31, 2018, 34 cases of disseminated Gonorrhea infection were identified. Incidence increased from 2-3 cases/year during 2013-2016, to 9 in 2017 and 15 in 2018.  Diagnostic specimens included blood (11 cases), joint fluid (19 cases), CSF (1 case), and multiple sources (3 cases). Nineteen (56%) cases also submitted urine samples for Gonorrhea and Chlamydia screening. Eleven were N. gonorrhoeae positive (including four co-infected with Chlamydia) and eight were N. gonorrhoeae negative (including two positive only for Chlamydia). HIV and Syphilis screening was performed for thirty (88%) and twenty-eight (82%) cases, respectively. None were positive.  Isolates from twenty-eight cases were available for susceptibility testing and genotyping. All isolates were susceptible to Azithromycin, Cefixime and Ceftriaxone. Nineteen (68%) were NG-MAST genotype ST-3671. Conclusions: The increasing incidence of Gonorrhea in Manitoba, Canada has been accompanied by an increase in cases of disseminated infection. The majority of cases involved ST-3671 strains. 

Objective: Few commercial assays for Mycoplasma genitalium (MG) are approved in Canada and data is limited with self-obtained samples. The objective was to compare 2 commercial assays on self-obtained vaginal swabs (SOVS) and first-void urine (FVU) from 300 sexually active young women. Methods:  Employing a research ethics-approved protocol, 171 consecutive women attending a clinic in Toronto, Canada collected a FVU and 3 SOVS. All patient samples were tested in a blinded fashion in a  SpeeDx Resistance Plus MG assay (SpeeDx Pty Ltd) in m2000sp (Abbott) and ABI 7500 (ThermoFisher) instruments, and the Seeplex STD6 ACE assay (Seegene Canada Inc) with EasyMag extraction and gel electrophoresis. Discordant samples were arbitrated with an Aptima MG assay (Hologic Inc) on a Panther instrument. Sensitivity (Sens), specificity (Spec), positive (PPV) and negative (NPV) predictive values were calculated. SpeeDx also provided identification of 5 mutations commonly associated with macrolide resistance. Results: In this interim analysis, 141 women were negative in SpeeDx and Seeplex in all samples and 30 were positive in at least one sample type by the 2 comparative tests. Aptima confirmed 25 true positives. The percent Sens, Spec, PPV and NPV estimates, respectively, were as follows: SpeeDx-SOVS 96, 100, 100, 97.9; SpeeDx-FVU 64, 100, 100, 94.2; Seeplex-SOVS 44, 97.3, 73.3, 91; Seeplex-FVU 20, 97.9, 62.5, 97.9. SpeeDx determined that 56% (14/25) of infections possessed mutations associated with macrolide resistance. Conclusions: The SpeeDx Resistance Plus MG assay demonstrated high accuracy in identifying MG-infected women by testing SOVS, which proved to be a more reliable sample than FVU. Providing simultaneous macrolide resistance testing in SpeeDx provided real-time value for treatment. The results from the Seeplex STD6 ACE assay may be a reflection of level of detection, cross reactions or the subjective nature of identifying bands in a gel.

Objective: Sexually active women may be infected with Mycoplasma genitalium (MG) with macrolide resistant mutations (MRM). Commercial assays measuring both the presence of MG and MRM have been developed and require clinical evaluation. The objective was to compare Allplex^TM  MG and AziR assays (Seegene Canada Inc.) to the SpeeDx Resistance Plus MG assay (SpeeDx Pty Ltd) on vaginal swabs. Methods: Two vaginal swabs (VS) in transport media specific for each assay were collected from 190 women. Multiplex assays were performed following the manufacturers’ instructions. Both Allplex STD4 assay which detects MG plus 3 other pathogens and the Allplex MG AziR assay which detects MG and 6 MRM extracted VS on a Microlab STARlet IVD (Hamilton) with subsequent PCR amplification on a CFX96 (Bio-Rad). The SpeeDx assay which detects MG and 5 MRM performed extraction on an m2000sp (Abbott) and PCR in an ABi 7500 instrument (ThermoFisher). Agreement and Kappa statistic compared positive and negative results for the detection of MG. Wild-type (WT) and MRM were also compared for the 2 assays. Results: Allplex assays recorded 27 MG-positives and 163 negatives compared to 25 and 165 in SpeeDx. Agreement of positives was 85.7% and negatives 97.6%. Overall agreement was very good (97.89%), Kappa 0.911 (95% CI 0.825-.997). Each MRM assay recorded 58.3% (14/24) of the samples to be resistant (R) with 8 discordants: MG AziR R and SpeeDx WT (n=4); SpeeDx R (n=4) and MG AziR (3 WT and 1 negative). Allplex assay MRM were A2058G (n=13), A2058T (n=1) and A2059G (n=4). Conclusions: All assays demonstrated very good agreement for the detection of MG in vaginal samples, and were easy to perform. MG detection and MRM results can be provided at the same time. R and WT discordancy was not due to the extra A2059T mutation included in the Allplex assay.

Objective: A pipeline of epidemiological, laboratory, and genomic investigation was used for real-time detection and investigation of a cluster of carbapenemase-producing organisms (CPO) in a community setting.

Methods:Three patients at an acute care facility tested positive for carbapenemase-producing E. coli over a short period of time.  Infection prevention and control (IPC) followed up on the cases to identify risk factors for CPO acquisition. Their whole genome sequence data was cross-referenced against a database of CPOs identified in BC since 2008, using core genome SNP-based dendograms to visualize isolate relatedness. MinION long-read plasmid sequencing data were used to supplement MiSeq data to generate a genomics-based case definition, which was used to confirm subsequent cases suspected by IPC to be part of the cluster. Results: IPC follow-up initially identified none of the usual CPO risk factors. The isolates were E. coli whole-genome MLST type 405, and carried a novel NDM-1-producing plasmid, previously unseen in BC. On a dendogram, these isolates, along with two additional clinical isolates from a community laboratory, clustered within 4 SNPs.  In-depth epidemiological investigation revealed shared exposure to two private retirement facilities in the community. The novel plasmid was an IncI1-alpha plasmid. This genomic case definition was used to confirm case inclusion in the ensuing outbreak investigation. It also allowed the identification of two additional cases, not previously known to be related, without known ties to the retirement facilities or the acute care facility. All identified cases  clustered very closely by core genome (mean SNVs = 5.61). Conclusion: Using WGS to support epidemiological investigation allowed early detection of a community-based cluster, and provided hypotheses for transmission mechanisms. While little is known about the epidemiology of CPO in the community, the close clustering of cases in this outbreak suggested clonal transmission from a point source.

Objective: Increasing antimicrobial resistance (AMR) is a growing concern as it limits the ability to treat infections.  Canadian Antibiotic Resistance Alliance (CARA) data from 14 tertiary care hospitals reported an increase % non-susceptibility for ciprofloxacin and ceftriaxone from 2016 to 2017 (about 650 isolates per year).  We compare two years of standardized E. coli antibiogram data from CNISP sentinel hospitals across Canada. Methods: Hospitals participating in the Canadian Nosocomial Infection Surveillance Program (CNISP) were asked to submit annual antibiogram data for all E. coli isolates in a standardized format for the years 2016 and 2017.   Percent (%) non-susceptibility was calculated for reported antibiotics. Results: Fifty (50) hospitals from 9 provinces submitted data for 2016, and 63 hospitals from 8 provinces submitted for 2017.  The highest number of isolates were tested for ampicillin (51,965 in 2016 and 58,362 in 2017).  From 2016 to 2017 a significantdecrease in % non-susceptibility was evident for: ampicillin 44 & 41.9%; ceftriaxone 9 & 8.3%; and cotrimoxazole 23 & 21%; whereas, ciprofloxacin 18.8 & 19%; amoxicillin/clavulanic acid 16.7 & 16.4%; piperacillin-tazobactam 4.7 & 4.5%; and meropenem 0.2 and 0.1% remained stable.  Comparing regional to national rates for 2017, Central Canada (Ontario and Quebec) had the highest non-susceptibility rates for amoxicillin/clavulanic acid (23%), ceftriaxone (9.4%), ciprofloxacin (20.3%), and cotrimoxazole (24.1%).  Eastern Canada had a significantly higher piperacillin-tazobactam non-susceptibility (6%), but at the same time, the lower non-susceptibility for amoxicillin/clavulanic acid (11.7%), ceftriaxone (6.2%), ciprofloxacin (12.7%) and cotrimoxazole (16.3%). Conclusions: Based on CNISP data, no concerning changes in AMR for E. coli in Canada were observed between 2016 and 2017.  One limitation is allowing either “urine isolates” (typically 90% of all) or “all isolates” submissions interchangeably.  This was a robust AMR surveillance project and is likely to increase in organisms tested and subpopulations including blood cultures.

Background: Antibiotic resistance is a growing concern. Preliminary data suggests fecal microbiota transplantation (FMT) may help decolonize gastrointestinal antimicrobial resistant organisms (AROs) through competition with susceptible organisms. The objective of this study was to evaluate the effectiveness of FMT by enema to eradicate ARO rectal colonization. Methods: 12 patients treated with FMT by enema for recurrent Clostridiodes difficile infection (rCDI) and 7 patients treated with FMT from a prior rCDI clinical trial who had pre- and post-FMT stool or rectal swabs were included in this study.  Pre- and post-FMT stool aliquots stored at -80C were thawed and tested for ARO [methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant enterococci (VRE), ESBL, and carbapenemase-producing organisms (CPO)] by planting onto Oxoid Denim Blue agar, Oxoid Brilliance VRE agar, and Oxoid MacConkey/cefpodixime agar, respectively following clinical laboratory operating procedures. Microbiota diversity was determined using 16s rRNA sequencing. Correlation with clinical outcomes was completed. Results: 5/19 patients had AROs detected pre-FMT (VRE n=2; ESBL n=2; VRE and ESBL n=1). 2/5 (40%) [95%CI: 0.12-0.77] of patients had complete ARO clearance post-FMT. 2/3 (67%) [95%CI: 0.20-0.94] of ESBL colonized patients and 1/3 (33%) [95%CI: 0.06-0.80] of VRE colonized patients had ARO clearance post-FMT. 2/3 (67%) [95%CI: 0.20-0.94] of patients whose rCDI symptoms were cleared, had ARO clearance. 2/2 (100%) [95%CI: 0.29-1.00] of patients with sustained uptake of diverse donor microbiota, eradicated their AROs post-FMT. Conclusions: FMT is a promising ARO eradication treatment with an estimated efficacy of 100% [95%CI: 0.29-1.00] in patients with sustained uptake of diverse donor microbiota. A prospective evaluation of FMT on ARO eradication is ongoing.

Background: The development of Canadian healthcare associated infections (HAI) and infection prevention and control (IPC) guidelines facilitates a national approach for responding to public health emergencies and maintaining safe healthcare delivery. The objective of this poster is to describe the role of the National Advisory Committee on Infection Prevention and Control (NAC-IPC) which informs national infection control guidelines. Methods: The NAC-IPC is composed of members with various expertise, including infectious disease, medical microbiology, and infection prevention and control. They lead individual task groups in the development of IPC guidelines. Topic selection for evidence based guideline development is based on a number of criteria including public health response to emerging issues, epidemiology and research findings, changes in practice and impact upon the health of Canadians. Guidelines are developed using a standardized process involving systematic reviews or environmental scans, and evidence grading where applicable. Results: Some examples of documents developed by the NAC-IPC are shown in Table 1. Recommendations are informed by evidence and collective expert opinion where evidence is sparse. 

Table 1: Types of documents produced by the NAC-IPC

Conclusions: The NAC-IPC is well-positioned to facilitate development of national guidelines for HAIs and emerging pathogens; inform federal-provincial-territorial public health networks; and provide opportunities for international collaboration, knowledge exchange and mobilization.  

Background: CPE are a rapidly evolving problem. This study aims to describe the epidemiology of CPE in south-central Ontario to inform CPE control programs. Methods:  The Toronto Invasive Bacterial Diseases Network performs population-based surveillance for CPE colonization/infection in Toronto/Peel. Microbiology laboratories report all CPE isolates to the study; annual audits are conducted. Incidence calculations use first isolates as numerator; population estimates are from Statistics Canada. Results: CPE  incidence has increased from 0 in 2006 to 1.5/100,000 in 2016-2018 (Figure). Bacteremia incidence has increased to 0.36 per 100,000 in 2018.  Among 783 incident cases, median age is 70y (IQR 57-79yrs); 450 (57%) are male. Most common species are E coli (n=345; 43%), K. pneumoniae (319; 39%) and Enterobacter spp. (79; 10%); most common genes are bla_NDM (±OXA, 432; 53%); bla_KPC (143; 18%), bla_OXA-48-like (186; 23%) and bla_VIM (34; 4%). Among 345 patients with only rectal colonization when first identified, 40 (12%) have had a subsequent clinical isolate (6 blood, 7 other sterile, 26 non-sterile sites). Among 646 persons with documented history of hospitalization and travel in the year prior to identification, 199 (31%) had been hospitalized in the Indian subcontinent, 77 (12%) hospitalized elsewhere outside Canada, 83 (13%) had travelled to the Indian subcontinent without hospitalization, 217 (34%) had been hospitalized in Canada without hospitalization elsewhere or travel to the Indian subcontinent, 34 (5%) had not been hospitalized, but had travelled to countries outside North America and Northern Europe, and 30 (5%) had no travel or hospitalization history. Among 34 non-hospitalized, non-Indian subcontinent travelers, 7 (all OXA) had travelled to Egypt (5) or other Eastern Mediterranean countries. Conclusions: CPE are increasing in incidence in Ontario. More than one-third of cases appear to be acquired in Canadian hospitals. Travel to some non-Indian subcontinent countries without hospitalization may pose an exposure risk.

Background: Norovirus is one of the most common viral pathogens implicated in gastroenteritis outbreaks in community and health care settings. Short incubation period and high attack rate allow rapid spread through inpatient wards to patients, staff and visitors. Early identification and implementation of infection prevention and control measures is essential to interrupt transmission. Methods: Our centre is a 250-bed tertiary care Pediatric and Women’s hospital serving the Maritimes. We describe a norovirus outbreak in our 24-bed, single room Pediatric Medical Unit. 

Hospital-acquired norovirus definition:

Patients admitted ≥48hrs with lab-confirmed norovirus AND ≥1 of:

Acute onset diarrhea.

OR

≥2 of: nausea, vomiting, abdominal pain, fever, or headache.

In 2017 FilmArray Gastrointestinal (GI) Panel was introduced in our Microbiology Laboratory. Since then, stool samples sent for viral, bacterial, or parasitic testing are evaluated by PCR. The panel tests for 22 GI analytes, with a 2-hr turnaround time. Previously, in-house stool viral testing was limited to adeno- and rotavirus antigen. Patient characteristics were collected and analyzed. Results: Patient 1 had new-onset diarrhea and vomiting on day 0; Patients 2 and 3 became symptomatic on day 1. Patient 3’s parents were previously symptomatic and had used the ward kitchen. Two care-givers of Patient 2, and 1 staff were symptomatic over days 0 to 2. Outbreak was over on day 6. Patients 1, 2 and 3 all tested norovirus positive in stool on day 1. On days 2-3, 6 additional patients with diarrhea tested norovirus negative. Symptomatic patients were immediately placed on contact precautions, ward cleaning frequency increased, and hand hygiene was reinforced. Common areas were closed until the outbreak was over. All patients with diarrhea were tested during the outbreak. Conclusion: FilmArrayGI panel enabled same-day identification of norovirus in this single-ward outbreak and permitted real-time identification of the termination of the outbreak.

Background: Certain surgical procedures may pose a risk of exposing patients to the blood of a healthcare worker (HCW). Although rare, HCW-to-patient transmissions of a bloodborne virus (BBV) have been documented. This Guideline was developed to provide a national framework for policies on the management of HCWs infected with BBV(s). Methods: Systematic reviews of the literature (1995-2016) were conducted to inform the transmission risk of hepatitis B (HBV) and C (HCV) and human immunodeficiency (HIV) viruses from infected HCWs to patients. Grey literature informed sections on disclosure of HCW’s serologic status, Expert Review Panels, and lookback investigations. National stakeholder partners were consulted and a Task Group provided technical expertise. Results: Provided HCWs adhere to Routine Practices, the risk of HCW-to-patient BBV transmission is negligible, except during exposure-prone procedures where HCW injury may expose a patient’s open tissues to the HCW's blood. Reported transmission rates were 0-3% (HIV, Table 1), 0.04-3.7% (HCV) and 0.06–11.11% (HBV). Rates vary with source viral load, nature of exposure, and IPC breaches. Current antiviral therapy informed guideline recommendations, with viral load thresholds defining fitness for practice.

Table 1: HIV transmission risks during exposure-prone procedures

Conclusions: The guideline provides a pan-Canadian approach for managing HCWs infected with a BBV, with recommendations directly impacting clinical practice related to preventing and controlling healthcare-associated infections.

Objectives: Infection prevention and control (IPC) guidelines for extended-spectrum beta-lactamase (ESBL)-producing organisms vary across organizations and jurisdictions. The primary objective of this study was to review the published literature about the use of contact precautions for ESBL-positive inpatients in acute care facilities in non-outbreak settings to aid in the development of provincial evidence-based recommendations. Methods: Literature searches were performed in PubMed and EMBASE from beginning to August 2018. A grey literature search was also done to determine current practices, recommendations, and guidelines. The key search words were ‘Extended Spectrum Beta Lactamase’, ‘ESBL’, ‘patient’, ‘contact’, ‘precaution’ and ‘isolation’. Results: After removing duplication, there were 119 relevant articles. Based on the scoping review, there was strong support for continued attention to IPC routine practices, including hand hygiene. There was no evidence that additional contact precautions or patient isolation practices reduce transmission of ESBL organisms among inpatient populations in non-outbreak settings. In addition, there were no differences in recommendations for high-risk patient populations such as burns, oncology and critical care. Institution and enhancement of antimicrobial stewardship programs and practices was strongly supported. Conclusion: We recommend against routine institution of contact precautions for inpatients colonized or infected with ESBL-producing organisms in non-outbreak settings. Rather, there should be a strong focus on IPC routine practices and enhanced institutional support for antimicrobial stewardship programs. As more research becomes available on this topic, meta-analysis will need to be conducted on comparable data.

Objective(s): MRSA has been a pathogen under public health surveillance in Alberta since 2005 and the epidemic clone CMRSA10/USA300 has remained high in community settings. This study focuses on the characterization of all blood MRSA isolates received by the Provincial Laboratory for Public Health (ProvLab) in the last five years in Alberta. Methods:  MRSA blood isolates from first clinical cases were submitted from all laboratories in Alberta to ProvLab from January 1, 2013 to June 30 2017. Clinical information provided on the requisition was recorded. Molecular characterization was performed using PFGE and spa typing and epidemic types were assigned. Results: Of 758 MRSA isolates, 54.9% were designated as CMRSA10/USA300, followed by CMRSA2/USA100/800 (22.2%), CMRSA7/USA400 (12.1%), CMRSA8 (3.8%), and 5.9% were not assigned a Canadian epidemic type. CMRSA1, 3/6, 4, and 5 were at insignificant numbers (1.1% of all MRSA). For demographic distribution, 47.6% of CMRSA2/USA100/800 isolates originated in the Calgary Zone while similar proportions of CMRSA10/USA300 (46.2%) and CMRSA7/USA400 (45.7%) were from the Edmonton Zone. Most isolates submitted (70.2%) originated from inpatient locations, including 69.0% of CMRSA10/USA300 and 70.8% of CMRSA2/USA100/800. For gender distribution of all prototypes, 61.5% were from males. The highest incidence of CMRSA10/USA300 was in the 30-40 year age group (20.2%) while the majority of CMRSA2/USA100/800 and CMRSA8 patients were in the 70-80 year age group. A total of 101 unique spa patterns were observed among the isolates tested, including t008 (CMRSA10/USA300; n=379), as well as 12 novel spa types. Ten CMRSA8 isolates matched novel spa type t6443 (8/10 male; average age of 74.3 years). Conclusion: The majority of the MRSA blood isolates in Alberta are related to the CMRSA10/USA300 epidemic type with a predominant spa type of t008. 

Objective: The long-term health burden of Lyme disease (LD) remains poorly characterized. Our objective was to systematically review the long-term sequelae and health-related quality of life (HRQoL) associated with LD. Methods: We performed systematic literature searches in Medline, Embase, Scopus, CINAHL, PsycInfo and Environment Complete up to July 2017 following PRISMA guidelines. The protocol describing study eligibility criteria was published on PROSPERO (CRD42017068765). We included North American and European observational studies measuring attributable health burden: long-term sequelae, HRQoL, and prognostic factors. We excluded studies with unclear LD diagnosis criteria, co-infected patients, or did not have a non-LD control group. Two reviewers independently completed screening, data extraction and quality appraisal. Results: We screened 8,698 articles and included 35 primary studies conducted between 1994 and 2016. Most studies were conducted in the United States (74%), used retrospective cohorts (54%), and used LD diagnosis based on, or adapted from, the CDC case definition (66%). Studies investigated patients with varying LD stages (1 early localized, 9 early disseminated, 2 late disseminated, 11 post-treatment LD syndrome, and 9 mixed). Studies reported sequelae (79%), HRQoL (42%) and prognostic factors (11%). Mortality was not reported. Arthralgia (8.3%), memory impairment (3.3%), facial nerve palsy (2.9%), and sleep difficulty (4.3%) were the most commonly reported physical, cognitive, neurologic, and functional sequelae, respectively. Most HRQoL studies used Short Form-36 (67%) and reported physical and mental component scores. Mean follow-up duration (range, 2.1–15.4 years) and HRQoL varied with LD stage. Approximately half (52%) of the included studies met 80% of their respective quality appraisal criteria. Conclusions: Our review highlights the presence of long-term sequelae and reduced quality of life associated with certain stages of Lyme disease. Outcomes reported can support clinical management and will be useful for future clinical and economic evaluations of interventions targeting LD treatment and prevention.

Objective: Carbapenemase-producing Enterobacteriaceae (CPE) are a subgroup of carbapenem-resistant bacteria that pose a significant global health threat. We conducted a systematic literature review on the health-related quality of life (HRQoL), health outcomes, and long-term sequelae attributable to CPE infection. Methods: We followed PRISMA reporting guidelines and published our review protocol on PROSPERO (CRD42018097357). We searched four electronic databases: Medline, Embase, CINAHL and the Cochrane Library between January 2008 and May 2018 for concepts related to: CPE, quality-of-life, complications, and mortality. We included primary studies with a carbapenem-susceptible control group, conducted in Organization for Economic Co-operation and Development countries. We excluded studies that were not published in English, used an inappropriate control, or reported inappropriate outcomes. Quality appraisal was completed using Joanna Briggs Institute checklists. We qualitatively summarized the most frequently reported sequelae and conducted a meta-analysis. Results: We identified 8,671 studies of which 17 met the eligibility criteria for inclusion into this review. All studies reported health outcomes, none reported HRQoL. Most studies were conducted in teaching or university-affiliated hospitals (76%), from Europe (65%), and used case-control designs (53%). Mortality was the most commonly reported consequence of CPE-infections, with in-hospital mortality as the most often reported outcome (62%). Our meta-analysis (n=5 studies) estimated a risk difference of the in-hospital mortality rate of 0.25 (95% CI, 0.17 – 0.32). Duration of antibiotic therapy (range, 4-29.7 days vs. 1-23.6 days) and length of hospital stay (range, 21-87 days vs. 15-43 days) were relatively higher for CPE-infected patients compared to carbapenem susceptible patients. Overall, most studies (82%) met over 80% of their respective quality appraisal criteria. Conclusions: Health outcome studies associated with CPE infection are focused on short-term (e.g. in-hospital) outcomes; long-term sequelae and quality-of-life are not well studied. Future opportunities exist for longer follow-up to assess the clinical outlook for CPE infections.

Objectives: Identify causes and confounding factors associated with Clostridioides difficile deaths over a two-year period in a large, tertiary care, teaching hospital. Methods: Retrospective chart review of deaths directly related to C. difficile or in which C. difficile was a contributing factor at the Royal Alexandra Hospital (RAH) in Edmonton, Alberta, from 01 January 2017 to 31 December 2018. Results: There were 19 deaths due to laboratory-confirmed C. difficile identified during the study period (male, 10/19; mean age, 78.9 years). Initial diagnosis was ‘severe’ in 7 and 16 cases by 2010 and 2018 Infectious Diseases Society of America classification, respectively. All were primary episodes, 15/19 were hospital-acquired to RAH. Recent antibiotic exposure was identified in 16 cases (indication sepsis, 8/16; pneumonia, 7/16) with 12 patients receiving at least two antibiotic classes. Initial management was guideline concordant in 14 cases (by either 2010 or 2018 classification). Pre-printed care order sets were utilized in 8 cases; initially 5/8 were guideline concordant with 2 corrected upon recognition of severity (7/8). Sixteen patients had treatment escalation within 72 hours; 8 were guideline discordant due to inappropriate escalation of oral vancomycin dose. Delay in symptom recognition (>48hr before testing) occurred in 8 cases, and treatment initiation delay (>24hrs) in 1 case. There was a lack of symptomatic response to treatment in 15 cases (average 2.5 days until death). Renal failure was the most common worsening comorbidity identified as contributing to death (9/19). Conclusions: Mortality from C. difficile is a quality indicator for C. difficile management, and the health care system. Review of local data indicate that deteriorating health states and increasing comorbidities influenced poor outcomes. Lack of adherence to updated guideline-recommended treatment, lower uptake of pre-printed care order sets and improved symptom recognition and timely testing are targets for intervention.

Background: In 2015, 80,000 BC residents are living with HCV with the number of hepatitis related deaths expected to increase by one third by the end of 2022. In 2016, HIV incidence in BC was 5.1/100,000 with highest rates in Vancouver Coastal in the nation. HIV and HCV continue to be recognized as a major public health threat. Method: A data base collected from testing fairs at high risk locations Downtown Vancouver was analyzed. Participants self reported risk factors, history of HIV/HCV infection and current infectious status anonymously before screening. Known positive status were not screened. Positive participants were offered private onsite specialist counselling. Results: Data on 421 participants was collected: 247 male, 53 female and 121 chose not to identify themselves. 58% (244) reported screening >6months ago of which 85% (207) tested positive for either HIV/HCV. 88% (372) indicated a history of recreational drug use, of which 36% (135) had a history of IVDU. 41% (173) engaged in unprotected sex, of those 31% (54) were in a relationship. 8% (34) were newly screened and likely to be male. Participants were asked if they were already known to have any of the following illnesses: HIV, HCV, and HBV. There were 8 self-reported cases of HIV, 67 HCV, and 7 HBV. There were 3 new cases of HIV and 23 new cases of HCV. 4 HIV/HCV co-infected cases, 4 cases of HIV, and 81 cases of HCV reported being engaged in healthcare. History of IVDU was associated with 6 times more likelihood of being HIV positive. PWID were 15 times more likely to have HCV. Conclusion: An active approach to HIV/HCV screening in high-risk population can be used as an effective tool in decreasing disease incidence and eradication of HCV and HIV in Canada.

Introduction: Restriction of patient movement is an important component in management of hospital GI and ILI outbreaks; however, no standard definitions or practices exist. Objectives: Our objectives were to (1) collect information about outbreak-related unit closure decisions in acute care settings across Alberta from the perspective of Alberta Health Services (AHS) Medical Officers of Health (MOHs) and Infection Prevention and Control (IPC) specialists, and (2) investigate provincial variability in unit closure definitions and practices, with the goal of informing policy. Methods: We designed and performed a novel, cross-sectional, confidential, voluntary survey of AHS MOHs and IPC physicians and operational leads. Unit closure approaches were defined separately for Admissions-Transfer to-Transfer out as “No” – prohibiting movement, “Yes” – allowing movement, and “Selective” – allowing movement according to the delineated criteria. The “more-restrictive-approach” was determined as any combination of “No” and “Selective” approaches. Descriptive statistics, Fisher’s exact test, and chi-square were used. Results: The response rate was 38.2% (n=21; IPC, n=15; MOHs, n=6). The majority (n=8, 38.1%) defined unit closure as No-No-Selective. Seventeen responders reported utilizing a more-restrictive-approach (80.9%; p<0.001). Preference for more-restrictive-approach did not differ among health zones (c² (2)=0.103; p=0.949). IPC inclined towards prohibiting admission and transfer to the unit in 80.0% and 73.3% of the cases (MOHs in 33.3% and 33.3%), but were more flexible in transferring out (prohibition, 13.3% versus 33.3%). Considering ILI and GI outbreaks separately, there was no difference between proportions of specialists with more-restrictive versus less-restrictive approach (n=17, 80.9% for each; p=1.000). Conclusions: Our findings support similar approaches to unit closure definitions and practices across Alberta; the vast majority of specialists apply restrictions to patient movement during hospital ILI and GI outbreaks. Exploring the effect of various unit closure practices on outbreak outcomes will assist in improving outbreak control policies and processes. 

Objectives: This study aimed to (1) review evidence from studies, systematic reviews, and guidelines that designate ward closure as a control measure for both influenza-like illness (ILI) and gastrointestinal illness (GI) outbreaks, (2) determine the effectiveness of ward closure in controlling outbreaks in acute care settings based on the most plausible evidence, and (3) provide practical ward closure implementation recommendations based on the current evidence. Methods: Electronic searches were conducted in PubMed, EMBASE, Google Scholar, and the Cochrane Library. A grey literature search was completed to determine current practices, guidelines, and recommendations. The key search words were ward closure, outbreaks, and effectiveness. Results: The term “ward closure” was used variably with no universal definition across the literature. However, in most instances, it referred to restrictions on patient movements into and out of a ward.  The effectiveness of ward closure in controlling ILI and GI outbreaks is unclear, with variable results on outbreak duration and severity. Most studies were observational, and ward closure was usually one of multiple outbreak control measures implemented concurrently. Conclusions: Evaluating effectiveness of ward closure as a sole intervention is difficult as it is usually implemented with other infection prevention and control (IPC) measures. The observational nature of the involved studies did not provide concrete evidence for effectiveness of ward closure. Therefore, we recommend that the current state of practice be maintained; that is, that ward closure be considered as a possible outbreak control measure which may be used based on the IPC risk assessment for a particular ILI or GI outbreak.  Accordingly, more controlled research studies are needed to determine the effectiveness of ward closure in managing acute care ILI and GI outbreaks.

Background: Prior to the recent outbreak in 2015, Zika virus (ZIKV) infections in adults were considered to be largely asymptomatic, and characterized by a short, febrile illness. Increasing reports of symptomatic neurologic disease amongst infected adults warrant further investigation into the manifestations and long-term outcomes associated with ZIKV in adults. Objective: To synthesize the literature on clinical manifestations and sequelae of ZIKV infection in adults. Methods: We conducted a systematic search of the MEDLINE, Embase, PubMed, CINAHL, LILACS and WHO's ICTRP clinical trials databases using the search terms “Zika virus” and “Zika infection”. Abstracts and conference proceedings were excluded, along with editorials, letters and news articles. Case series/reports with less than 10 participants and any animal studies were also excluded. Two reviewers followed PRISMA and MOOSE reporting guidelines. We used Joanna Briggs Institute Critical Appraisal tools for quality appraisal. Conflicts were resolved by consensus or consultation with third reviewer. Results: We identified 6248 references in our initial search up to April 30, 2018, of which 38 studies were included [Cross-sectional/descriptive (n=28), case-control (n=5), case reports/series (n=3), cohort (n=2)]. The majority of studies originated from North and South America, with publications from Brazil (n=7), Colombia (n=6), and the USA (n=5) constituting more than half of the data set. The most common outcomes reported were Guillain-Barre syndrome, encephalitis and myelitis. Approximately 37% of the studies reviewed met 80% or more of the quality assessment criteria for their respective study design. We will finalize complete analysis of the data prior to the conference. Conclusion: Based on preliminary results, there is evidence of adverse neurological clinical manifestations and long-term sequalae associated with ZIKV infection in adults. More targeted studies need to be conducted in non-pregnant adults to better understand clinical manifestations and longer term outcomes in this population. 

Objective: 13-valent pneumococcal conjugate vaccine (PCV13) was licensed in Canada for the prevention of vaccine-type (VT) pneumonia in adults in July 2015. Dramatic reductions in VT-disease have been observed in older adults post implementation of pediatric PCV7 programs. Similar herd effect pattern has been assumed post PCV13-implementation, resulting in lack of age-based adult PCV13 recommendation. Recent surveillance suggests reductions of VT-IPD are significantly lower and have plateaued, leaving a persistent and substantial burden of potentially preventable PCV13-type disease in adults. The purpose of this work is to estimate the impact of a PCV13 immunization program for Canadian adults aged ≥65 years on community-acquired pneumonia (CAP) hospitalization, using current data. Methods: A model was constructed to estimate the number of CAP hospitalizations averted in the Canadian population aged ≥65 years over 5 years, based upon Canada-specific disease incidence and published estimates of PCV13 effectiveness and durability. Model parameters included: i) the size of Canadian population aged ≥65 years, ii) all-cause CAP incidence, iii) proportion of VT-CAP, iv) median length of stay, v) PCV13 effectiveness, and vi) duration of PCV13 protection over a five-year time horizon.  Impact on multi-drug resistant CAP was also evaluated. Assumptions included: constant rates of all-cause CAP, proportion of VT-CAP, PCV13 effectiveness and 5% annual all-cause mortality rate for the 5-year period. Results: Based on model assumptions, PCV13 use in Canadian adults aged ≥65 years would lead to 62 (11–77) less hospitalizations per 100,000 persons, per year. This reduction, applied to the entire Canadian population of older adults, would avert an estimated 17,274 (3,037–21,711) hospitalizations and 138,192 (24,298–173,690) hospital days over a 5-year period. Conclusion: Despite herd effects from the routine pediatric program, direct PCV13 immunization of older adults in Canada could result in considerable additional reduction in hospitalizations for pneumonia.

Background: Hepatitis C virus (HCV) elimination requires a marked, contemporaneous scale-up of both HCV diagnostics and therapeutics. In publicly funded health care systems, finding maximum efficiencies is paramount. In particular, revision of medically complex HCV care models is needed. One strategy is task shifting non-nursing roles to other skilled individuals, such as clerical staff. Our objective is to determine the effect of skilled clerical staff on overall HCV wait times and treatment initiations. Methods: A provincial HCV elimination strategy was developed where key elements included: centralized triage and referral, rapid treatment start, inclusion of incarcerated individuals, harm reduction strategies, and patient self-referral. Intrinsic to the low cost business plan is each individual working to full scope of practice. In particular, a medical clerk with advanced skills in database management and telephone based patient engagement was added to the HCV clinical team to complement an experienced nurse and physician. The required amount of time to engage each patient, as well as the wait times before and after engagement were measured. Results: We collected data from 3 time periods: pre-clerk (Jan - Dec 2017), clerk (Dec 2017- Aug 2018), and post-clerk (Oct 2018). Treatment initiations were an average of 2.25, 15.3, and 3 patients per month respectively during each of the periods. Median wait time, in days, for the pre-clerk and clerk period was 140 (IQR 84-235) and 35 (IQR 21-63.75), respectively. First appointment attendance was 54.5%, 82.2%, and 14.3% (p<0.05, clerk vs post clerk period) respectively for the pre-clerk, clerk, and post-clerk time periods. Conclusions: Task shifting initial HCV patient engagement from a nurse to skilled clerical personnel is a successful tool to gain rapid scale-up of patient treatment initiation. Models that encourage task shifting need to be supported longitudinally to maintain benefit in HCV elimination.

Introduction: Sharing of P. aeruginosa (Pa) strains between cystic fibrosis (CF) patients with chronic infection is relatively common. It is unclear how common Pa strain sharing and multiple lineage infection are in new-onset Pa infections occurring earlier in CF, when infections are aggressively treated with antibiotic eradication therapy (AET). We performed whole genome sequencing of Pa isolated from sputum of children prior to initiation of inhaled AET, to determine the frequency of mixed infection and strain sharing, and impact on AET failure. Methods: We sequenced genomes of 342 Pa isolates collected from 65 children with 75 episodes of new-onset Pa between 2012 and 2016. Up to 10 isolates were sequenced per episode. An initial phylogenetic tree was constructed using a mash genome distance based method. Closely related strains were further investigated by mapping genomes to a closely related reference followed by generation of maximum likelihood phylogenetic trees. Strain sharing was inferred based on detection of appropriate topological signal in the trees. Univariate logistic regressions were used to assess associations between mixed infection, strain sharing and AET failure. Results: A large number of patients shared Pa strains (N=25/65, 40%). Mixed infection (two or more strains present in sputum sample concurrently) occurred in 12/75 episodes (16%). Having a mixed infection was significantly associated with sharing of Pa strains (unadjusted OR 10.7, 95% CI 2.2; 53.7, p <0.01) but was not associated with AET failure. Furthermore, strain sharing was not associated with AET failure. Conclusions: A substantial proportion of patients with new-onset Pa infection were infected with a strain shared with other patients. Mixed lineage Pa infections were relatively frequently observed in new-onset episodes and were associated with strain sharing between patients; however, mixed infections and strain sharing were not associated with AET failure in this cohort.

Objective: Sepsis from tuberculosis in people living with HIV is uncommon in Canada. We present a case of tuberculosis sepsis in an HIV-positive woman in B.C. and examine factors contributing to delay in diagnosis until after her death. Methods: Retrospective chart review of clinical notes and investigations. Results: A 38-year-old HIV-positive woman presented to hospital with two weeks of fevers, back pain, and abdominal pain. She was non-adherent with HIV treatment (CD4+ T-cells of <33 cells/mm³ for the previous 3 years), actively injecting drugs, and was partially treated for tricuspid valve endocarditis one year ago. Imaging demonstrated right lung tree-in-bud nodularity and ground glass opacities, L5-S1 osteomyelitis, and a mediastinal mass adjacent to the distal esophagus. She underwent urgent endoscopy and thoracoscopy for possible esophageal perforation, which showed extensive esophageal candidiasis. Vancomycin, piperacillin-tazobactam, and fluconazole were started. Blood, urine, and sputum cultures were negative. She left against medical advice on day three of admission, re-presenting two days later with ongoing fever and new productive cough with dyspnea. Antimicrobials were restarted. Repeat imaging demonstrated pleural and pericardial effusions, mild ascites, and mediastinal lymphadenopathy. Due to rapid decline in respiratory status, she was transferred to ICU and underwent urgent bronchoscopy with bronchoalveolar lavage specimens sent for bacterial culture. She continued to deteriorate despite maximal interventions. After discussion with family, she was transitioned to comfort care and passed away soon after. One week after her death, mycobacterial blood culture grew Mycobacterium tuberculosis. Bronchoscopy specimens were then sent for mycobacterial culture and found to be 3+ AFB smear positive, eventually also growing M. tuberculosis. Conclusion: Sepsis from TB can be missed due to competing comorbidities and anchoring biases, especially in marginalized populations. A high index of suspicion and low threshold for empiric TB treatment should be maintained, particularly for immune-compromised individuals. 

Background: Mendelian susceptibility to mycobacterial disease (MSMD) is a rare primary immunodeficiency (PID) disorder caused by impairment of interferon-gamma mediated immunity. It is characterized by predisposition to disease caused by weakly virulent Mycobacteria, such as Bacillus Calmette-Guérin (BCG) vaccine, environmental non-tuberculous mycobacteria (NTM), and Salmonella species. Interleukin 12 receptor β1 (IL-12Rβ1) deficiency is the most common disease of Mendelian susceptibility to mycobacterial disease. Objectives: To highlight the importance of clinical suspicion for early diagnosis and treatment of suspected primary immunodeficiency especially in areas with high rate of consanguineous marriages to optimize the patient's outcomes. Methods: A case report on Inherited IL-12Rβ1 Deficiency with a literature review on its epidemiology and treatment. Results: We report a 6-year-old boy of consanguineous parents, who presented at 7 months of age with chest infection, generalized lymphadenopathy and inflammation at the Bacille Calmette- Guérin (BCG) inoculation site (BCGitis). Gastric aspirates for Acid Fast Bacilli (AFB) smear, and Polymerase chain reaction (PCR) amplification of M.tuberculosis DNA were positive. Mycobacterial culture and TB genotyping confirmed Mycobacterium bovis (BCG) vaccine strain. The diagnosis of Disseminated Bacillus Calmette-Guérin (BCG) infections were established. Patient were started on anti-TB medications, few months later he presented with recurrent lymphadenopathy; Lymph node biopsy culture revealed a salmonella non typhi. Initial immunological workup were normal, however in presence of persistent mycobacterial and salmonella infections genetic analysis was sent which identified a homozygous mutation in IL12RB1 gene, p.R173W (c.517C>T). This mutation leading to complete IL-12Rβ1deficiency. Both parents and sibling are heterozygous for this mutation. The patient still on antimicrobial therapy with good response although he developed generalized skin rash, biopsy revealed vasculitis responded to adding more anti-salmonella therapy. Conclusion: Although the incidence of Inherited IL-12Rβ1 Deficiency is rare, early testing and diagnosis is crucial as well as prolonged effective management of disseminated Mycobacteria and salmonella infections. 

Objective(s): The mainstay diagnosis of rickettsial disease remains acute and convalescent serologies, with extensive cross-reactivity occurring within the Spotted Fever group (SFG).  We report a biologic false positive SFG rickettsial serology case due to an unrelated infection. Methods: A 56-year male with controlled chronic HIV infection presented with three days of left-sided hearing loss, 2-weeks of blurry vision, and generalised macular rash involving his forehead, chest, and back.  Ophthalmic examination of the left eye revealed optic disk oedema with retinal exudates, flame haemorrhages, a hemi-macular star, and cotton wool spots. Results: MRI of the brain/orbits revealed hyperintensity of the left optic nerve with no infarcts.  A titre of 1:4096 to SFG and negative for Bartonella henselae were reported.  Lumbar puncture revealed a pleocytosis of 16 x 10**6/L (83% lymphocytes) and protein 0.63 g/L.  A diagnosis of early syphilis with neurologic manifestations (neuroretinitis and otosyphilis) was made.  Other markers showed a a high RPR (1:256), positive TPPA.  CSF VDRL was negative but FTA-ABS reactive (3+).  After 14 days of intravenous penicillin, RPR declined to 1:2 at 4 months.  He never received treatment specific for rickettsial disease. Conclusion(s): The elevated SFR-titre was considered a biologic false-positive based on the temporal occurrence of events, lack of tick bite, eschar, and no treatment for rickettsial disease: reports in the literature are rare.  Acute syphilis is recognised to cause potential biologic false positives amongst other immunological assays due to the production of a myriad of treponemal and non-treponemal antibodies. We hypothesise that the non-treponemal antibodies (towards the cardiolipin-lecithin-cholesterol antigen), could potentially cross-react with the rickettsial antigen in the IFA assay.  The initial manifestations were reasonably suggestive of Rocky Mountain Spotted Fever, however the lack of locally acquired cases in Alberta or tick bite eventually ruled out a serologic diagnosis of RMSF.

Background: AFM (Acute Flaccid Myelitis) is commonly due to viral causes:  polio virus, nonpolio enterovirus, West Nile virus, St. Louis encephalitis virus, Japanese encephalitis virus, and adenovirus. [2, 1] Currently there is an outbreak of Enterovirus D68 causing AFM in 24 states in the US  (396 cases since 8/2014 to 10/2018 ) primarily in children. Methods: We present a case report of AFM in an immunocompromised host that highlights the necessary diagnostic workup for AFM, and potential pitfalls in diagnostic testing in immunocompromised hosts. Results: A 70 female presented with fever, generalized weakness and acute confusion. She had a history of marginal cell lymphoma treated with 5 cycles of Bendamustine and Rituximab since 2016 and received her first maintenance dose of Rituximab 2 weeks prior to her recent admission. LP for viral studies and a paraneoplastic work up: LP 21 cells 87 % lymphocytes negative HSV 1 and 2 PCR, negative VZV PCR, negative Enterovirus PCR, negative West Nile (WMV) IgM, and negative for bacterial and fungal culture. BAL negative for virology, and MRI head showed postoperative changes with no evidence of lymphoma recurrence. In the ICU she progressed with declining level of consciousness and developed acute flaccid paralysis. EMG studies supported an acute motor axonal neuropathy with active denervation of muscles (Figure 1). This patient was presenting with seronegative West Nile infection due to effect of Rituximab on B cells, which was subsequently confirmed positive with a stored LP sample sent for WNV PCR on the CSF. Conclusions: Patients with compromised immune systems are more susceptible to viral infections and more likely to develop severe complications.  A review of the literature in lymphoma patients highlights the impact of Rituximab on B cell function with susceptibility to viral infections and false negative serological testing.[6] 

Background: Acute otitis media (AOM) is uncommon in the adult population, and while complications are rare, they can be severe. These complications include infectious seeding to adjacent structures resulting in brain abscess, significant neurological deficits and sinus thrombosis. The objective of this case is to highlight the importance of prompt diagnosis and treatment of AOM in adults to prevent serious complications, as well as to consider oral-pharyngeal organisms when treating brain abscess. Clinical Case: A 50-year-old male presented with a one week history of right-sided vision loss and headache. He reported chronic heavy alcohol use, was a long-standing smoker, had poor dentition, and had a two month history of untreated AOM with purulent otorrhea. Shortly after presenting to hospital, he had a first-episode tonic-clonic seizure. The patient was intubated and transferred to a tertiary care center. A head CT showed a left occipital lesion measuring 5.7cm x 3.3cm. An empiric antibiotic regimen was initiated, including vancomycin, metronidazole and ceftriaxone, and an intraventricular shunt was placed to drain the abscess. A follow-up head MRI showed left external auditory canal thickening and extension of the abscess into the lateral and fourth ventricle. Purulent drainage demonstrated gram negative rods on gram stain and Aggregatibacter aphrophilus was identified using MALDI mass spectrometry. Following drainage and ongoing treatment with ceftriaxone, the patient made a full recovery with no residual neurological deficits. Discussion/Conclusion: This case demonstrates the rare but serious potential sequelae of untreated AOM in the adult population, as well as the importance of prompt diagnosis and treatment. Aggregatibacter aphrophilus is a normal oral-pharyngeal organism that is a rare cause of brain abscess. It is important to consider oral-pharyngeal flora in the differential and empiric treatment for brain abscess.

Objective(s): Management of periprosthetic joint infections (PJIs) may involve a two-stage revision. The first stage involves removing all components and placing an antibiotic-impregnated cement spacer that provides high intra-articular concentrations of antibiotic(s) above the minimum inhibitory concentration of the infecting pathogen. Methods: A 59-year old female underwent the first stage of a two-stage procedure to treat a late-onset Escherichia coli PJI of the left knee. A custom-made cement spacer containing piperacillin/tazobactam, vancomycin, and tobramycin mixed with commercial methyl methacrylate, and gentamicin co-polymer was inserted. A bone abscess cavity was also filled with tobramycin-containing commercial bone filler. One week later, the patient developed progressive renal dysfunction despite hydration and exclusion of other causes, with a serum peak of 151 umol/L (baseline 75) at two weeks. At this point, the spacer was removed, and a new arthroplasty was implanted. Within ten days, her creatinine returned to baseline. Results: Iatrogenic acute kidney injury secondary to aminoglycoside toxicity due to persistent leeching from the antibiotic-impregnated spacer, and possibly the bone filler. Tobramycin serum levels ranged between 1.5-2.2 mg/L for 3 to 14 days post-operatively. Renal function normalised and aminoglycoside levels became undetectable only after the antibiotic spacer was removed and the second stage of the revision completed. Conclusion(s):Intra-articular spacers containing potentially nephrotoxic agents should be considered carefully when managing PJIs. Adding multiple antibiotics to cement mixtures may prevent proper setting, leading to increased and/or prolonged leaching of antibiotics. In our patient, spacer removal was the only course of action. Data to support better patient outcomes when multiple antibiotics are used with or without an aminoglycoside for synergy in cement spacers is scarce. Future directions include evaluation of cement mixture stability when multiple antibiotics are used, in order to determine their utility in treating PJIs.

Objective: Clostridium difficile infection continues to be a challenge for healthcare-associated communities.  While testing is done on symptomatic individuals, asymptomatic carriers of Clostridium difficile are a potential risk to transmission and symptomatic infection. Identifying at risk populations and methods for detection can help determine the epidemiological importance of asymptomatic individuals and provide potential strategies for interventions. Methods: Rectal swabs originally for Methicillin-resistant Staphylococcus aureus (MRSA) surveillance were collected from four units across four acute-care hospital sites. The populations of the units consist of medical, hemodialysis, and pediatric patients. Residual ESwab transport medium from these rectal swabs, was tested with a validated laboratory-developed, loop-mediated amplification (LAMP) assay based on the tcdC gene. The specimens that tested positive with the LAMP assay, along with a random sample of negatives were further tested by polymerase chain reaction (PCR) to confirm LAMP test results. Results:  We used our laboratory-developed LAMP assay to test 288 rectal swab specimens for C. difficile colonization. Utilizing an established clinical cut off of <45 minutes, 31 LAMP positives were identified. Prevalence for each unit is as follows; 0% (0/20) to 6.7% (2/30) for the two medical units, 7.9% (9/114) for hemodialysis patients, and 6.5% (8/124) for pediatrics.  Across the four hospitals, there is an overall prevalence of 6.6% (19/288).Conclusion: For the four hospital sites, there was a moderate prevalence among the various patients undergoing MRSA screening. Outcome studies will be needed to establish prognosis of asymptomatic carriers and the effect of carriage on hospital transmission. LAMP testing was an efficient method for colonization screening, however studies are needed to further optimize detection. 

Objective: Banked human milk (BHM) is commonly used in countries around the world as preferred feeding for preterm infants when mother’s own milk is unavailable. However, BHM could incur some inherent infectious risks, even when pasteurized. Because of the ubiquity of Bacillus cereus in the environment and its ability to resist the Holder pasteurization process, there is a concern that BHM might lead to severe B. cereus infections. We reviewed observed and published cases to determine the potential causal role of BHM as the source for these infections. Methods: Two infants in the province of Quebec (Canada) developed a B. cereus neonatal infection, both had received BHM. We conducted bacteriological studies to compare clinical isolates and those found in the involved BHM. Results: After extended culture of BHM retention lots, Bacillus cereus were found in involved batches in the first case. However, molecular typing showed that the strain found in the BHM was different from the clinical isolate, therefore excluding BHM as the source of contamination. In the second case, a Brevibacillus spp. was isolated, a species distinct from the clinical isolate. Conclusion: Based on these cases and others reported in the literature there has never been a documented causal link between B. cereus contaminated BHM and preterm neonatal infection. Therefore, the risk that BHM can cause this infection remains theoretical. Given the widespread presence of Bacillus cereus in the hospital environment and its capacity to resist standard cleaning procedures, it seems likely that airborne or direct/indirect contact are the main sources of most, if not all cases of severe B. cereus neonatal infections, even in babies exposed to BHM.

Background: Neonatal mortality accounts for nearly half of all deaths among children under five. With 46 neonatal deaths per 1,000 live births, Pakistan has the highest neonatal mortality rate in the world, bolstering the need for interventions that improve newborn survival. A recent cluster randomized controlled trial (cRCT) estimated the effect of an integrated neonatal care kit (iNCK) on neonatal mortality compared to standard of care in Rahimyar Khan, Pakistan. Objective: To strengthen the evidence-base towards wide-scale implementation of the iNCK, we evaluated the cost-effectiveness of iNCK distribution compared to standard care from the healthcare payer perspective. Methods: We performed a cost-utility analysis using a Markov model based on cRCT trial data and a comprehensive literature review. The base case was either standard care or distribution of the iNCK to pregnant mothers whose infant was followed over a lifetime time horizon. Outcome measures were life years, disability-adjusted life years (DALYs), costs and incremental net monetary benefit (INMB, at a cost-effectiveness threshold of USD 15.50), discounted at 3%. We conducted deterministic sensitivity analysis to assess parameter uncertainty. Results: At a cost-effectiveness threshold of USD 15.50, distribution of the iNCK resulted in lower expected DALYs (28.70 versus 29.54 years) at lower expected cost (USD 72.41 versus 86.11), translating to an INMB of USD 27 per iNCK distributed. These results were sensitive to the baseline risk of infection, cost of the kit and the relative risk of infection associated with iNCK use. Below a relative risk of infection of 0.83 and cost of the iNCK less than USD 32, the iNCK remained cost-effective compared to standard of care. Conclusions: The distribution of the iNCK dominated the current standard of care, i.e., is less costly and more effective, with most of the effectiveness attributable to a reduction in neonatal infection.

Objectives: There are very few studies on acute encephalitis with onset during the neonatal period. The objectives of this study were to investigate the etiology and salient clinical features of neonatal encephalitis. Methods: Neonates with possible infectious encephalitis (IE) were prospectively enrolled. Inclusion criteria included encephalopathy (altered/fluctuating level of consciousness ≥24 hours) plus ≥2 of: fever/temperature instability; seizure(s); focal neurologic findings; CSF pleocytosis; EEG abnormalities consistent with encephalitis; neuroimaging abnormalities consistent with encephalitis. Neonates with a clear diagnosis of post-perinatal asphyxial encephalopathy or culture proven bacterial meningitis were excluded. Results shown as absolute numbers, proportions or medians [interquartile range] as appropriate. Results: Fifty-nine neonates fulfilled the inclusion/exclusion criteria (June 2013 – November 2018). Empiric acyclovir was initiated in 49 (83.1%) cases. An infectious etiology was identified in 25 (42.4%): enteroviruses (n=15), HSV (n=5), HHV6 (n=2), parainfluenza 3 (n=1), influenza A (n=1), CMV (n=1). A non-infectious cause was confirmed in 20 (33.9%): missed hypoxic-ischemic encephalopathy (n=10), genetic/metabolic disorders (n=7), ischemic/hemorrhagic stroke (n=3). No specific etiology was identified in 14 (23.7%). Thirteen (52%) neonates with IE either died (n=7) or suffered neurologic sequelae (n=6). Deaths were attributable to HSV (n=4), enteroviruses (n=2) and HHV6 (n=1). Neurocognitive sequelae were documented in one case each of enterovirus, HSV2, HHV6, CMV, parainfluenza 3 and influenza A. Differences between neonates with and without IE, respectively, included age in days of symptom onset (7 [6, 10] vs. 1 [0, 3]; p<0.001), gestational age (37.0 [36.0, 39.0] vs. 38.6 [37.6, 40.0]; p=0.045), peripheral leukocyte count (10.5 [IQR 5.9, 14.6] vs. 14.3 [IQR 10.7, 21.7]; p=0.008) and CSF glucose (2.80 [IQR 2.3, 3.2] vs. 3.10 [2.8, 3.8]; p=0.003). Conclusion: Enteroviruses and HSV are the predominant causes of neonatal IE. Outcome of neonatal IE is poor with approximately half dying or suffering neurologic sequelae.

Objective: Competence assessment, training and teaching (CATT) are integral in developing knowledgeable competent staff and essential in meeting accreditation requirements for a Microbiology Laboratory. The introduction of WASP™/WASPLab™ automation in our laboratory supports effective methodologies to provide the required consistency in specimen processing, incubation and digital imaging. The objective of this study is to use culture images and gram slides generated by the WASP™/WASPLab™ automation for competence assessment, training and teaching programs. Methods: WASP™/WASPLab™ automation and automated staining was used to provide competence assessment to 49 Medical Laboratory (ML) Technologists and 13 ML Assisstants. ML Technologists analyzed digital images and read WASP™ prepared slides from positive blood cultures stained with an automated stainer. ML Assistants performed maintenance tasks. Results: Upon direct observation, all13 MLAs performed the appropriate maintenance tasks on the WASP™ slide prep module as well as demonstrating problem solving skills. Using WASP™/WASPLab™, 5 urine specimens were analyzed by 49 MLTs. There were varied images and a simulated choice of button selection in the screening module. All MLTs were successful in their selections, image analysis and reporting. The same MLTs were given  5 gram slide preparations of positive blood cultures, made by the WASP™ and stained with the automated stainer. All of the MLTs were correct in their blood culture gram reporting. Conclusions: Laboratory automation provides standardized sample preparation, processing and digital imaging which aligns itself perfectly for competence assessment as well as training and teaching opportunities. Digital imaging is stored and available for viewing over time as assessing staff over a long period of time across different shifts is challenging. Laboratory automation can help to develop highly skilled competent personnel who provide consistent, predictable and high quality results. Stored WASPLab™ images can be used for training new personnel or teaching students and residents.

Background: Ceftolozane-tazobactam (C-T) has been approved for treating adults with complicated urinary tract infections, acute pyelonephritis, and complicated intra-abdominal infections in combination with metronidazole.  In the current study, isolates were collected from US patients hospitalized with pneumonia (PIHP) from 2015-2017. Methods: A total of 4,337 GN isolates, including 2,102 Enterobacteriaceae (ENT) and 1,528 Pseudomonas aeruginosa (PSA) isolates, were collected in 2015-2017 from 30 US hospitals and tested for C-T susceptibility (S) by CLSI broth microdilution at JMI Laboratories. Only 1 isolate per patient per infection episode was included. Other antibiotics tested were amikacin (AMK), cefepime (FEP), ceftazidime (CAZ), colistin (COL), levofloxacin (LVX), meropenem (MEM), and piperacillin-tazobactam (TZP). Resistant (R) phenotypes analyzed were ENT R to doripenem, imipenem, or meropenem (CRE) and extended-spectrum β-lactamase (ESBL, non-CRE). Multidrug-resistant (MDR) isolates were identified as nonsusceptible (NS) to 3 or more antimicrobial classes. PSA phenotypes analyzed were CAZ-NS, MEM-NS, and TZP-NS. Results: Of the 4,337 GN isolates, 3,820 (88.1%) had a C-T MIC ≤ 8 mg/L. The 3 most prevalent GN species isolated from PIHP were PSA (n=1,528; 35.2%), Klebsiella pneumoniae (KPN, n=562; 13.0%), and Escherichia coli (EC, n=434; 10.0%). The %S of C-T and comparators for the top 3 pathogens are shown in the table. C-T showed activity against these isolates with %S of 96.5%, 88.6%, and 97.5% against EC, KPN, and PSA, respectively. Conclusions: C-T demonstrated activity against the most prevalent contemporary GN isolates from PIHP in the US. C-T was the only β-lactam that had >88%S against all 3 species: EC, KPN, and PSA. C-T and COL were the only agents tested that had >95%S for EC and PSA pathogens in PIHP. For PSA, C-T maintained activity against isolates resistant to CAZ, TZP, and MEM. These data suggest that C-T may be a useful treatment for GN infections causing PIHP.

Objective: A discrepancy in susceptibility rates between our laboratory and another prompted a review of historical Streptococcus pneumoniae disc diffusion results for tetracycline and doxycycline. Methods: Disc diffusion zone sizes for 2680 clinical isolates from 2013 to 2017 were extracted from our lab database. Susceptibility rates were calculated based on CLSI breakpoints and by anatomic site.  Zone sizes were plotted to visualize the wild-type distribution of isolates. EUCAST breakpoints were also applied.Results: When CLSI breakpoints became available for doxycycline in 2014, our lab switched from tetracycline to doxycycline testing by disc diffusion, resulting in a contradictory increase in resistance to doxycycline. Visualization of zone size distribution suggests that CLSI doxycycline breakpoints distinguish poorly between wild-type and non-wild-type isolates of Streptococcus pneumoniae; EUCAST tetracycline breakpoints more accurately distinguish. Sputum isolates of Streptococcus pneumoniae demonstrate higher doxycycline resistance than isolates from eyes or other anatomic sites. Conclusions: CLSI doxycycline disc diffusion breakpoints appear to fall within the wild-type distribution in our patient population and may account for antibiogram differences between our laboratory and others. An additional study is planned to determine whether MIC testing is preferable. Sputum isolates of Streptococcus pneumoniae were more resistant to doxycycline than isolates from other anatomic sites. 

Introduction: Dalbavancin, a second generation lipoglycopeptide with a half-life of 14 days, was recently approved for use in Canada for acute bacterial skin and skin structure infections (SSTIs) caused by susceptible gram-positive organisms. There are limited publications describing its use for other indications such as infective endocarditis, bacteremia, and osteomyelitis/joint infections. Objective: We describe two cases of dalbavancin use through the Special Access Programme (SAP) preceding its formal Notice of Compliance, including treating Bacillus cereus endocarditis and diabetic foot infection with osteomyelitis. Case Presentation: The first patient was a 42 year-old male with a history of opioid substance use disorder and multiple infective endocarditis episodes, including several hospitalizations in one year with Bacillus cereus group bacteremia and mitral valve endocarditis. During admissions he was not compliant with vancomycin therapy and had linezolid treatment failure after discharge with persistently positive blood cultures. After obtaining SAP approval, an initial loading dose of 1,000 mg and then three 500 mg weekly infusions of dalbavancin were administered via an outpatient parenteral therapy clinic. The patient attended all appointments and post-treatment blood cultures were negative. The second patient was a 56 year-old male with schizoaffective disorder and diabetes admitted with acute mania, psychosis, and polymicrobial diabetic foot osteomyelitis of the right great toe with pathogens including methicillin-susceptible Staphylococcus aureus and Group B Streptococcus. He required involuntary psychiatric care and restraints for administration of standard intravenous antibiotic therapy. Oral antibiotics were unsuitable due to drug interactions. SAP approval was granted for four weekly inpatient infusions of dalbavancin, dosed as for the first patient, allowing for conservative osteomyelitis management and limb preservation. Conclusion: Dalbavancin, a long-acting parenteral agent with weekly dosing, is a new antimicrobial in Canada for treating SSTIs. These cases of off-label treatment of endocarditis and osteomyelitis illustrate its successful use in challenging compliance scenarios.

Objective: The emergence of MDR/XDR P. aeruginosa has led to a reliance on suboptimal agents (Poly/AG) for the management of infections due to this pathogen. C/T is a novel agent with excellent in vitro activity against resistant P.aeruginosa that is indicated for cUTI and cIAI and being reviewed for VABP. The purpose of this study was to assess comparative rates of clinical cure, mortality, and acute kidney injury (AKI) among patients treated with C/T versus a Poly/AG based regimen for P. aeruginosa infections.  Methods: This was a retrospective, multi-site cohort of adult inpatients with infections due to MDR/XDR Pseudomonas. Patients treated for >/= 48 hours with C/T or Poly/AG-based regimen were eligible for inclusion. Patients with a creatinine clearance < 20 mL/min, or those requiring renal replacement therapy at baseline were excluded. Bivariate comparisons for baseline clinical characteristics and outcomes were assessed.  Results: A total of 117 (57 C/T, 60 Poly/AG) patients were included. Baseline characteristics, infection source, severity of illness, and time to appropriate therapy were similar between the treatment groups. The most common infections were ventilator associated (54%) or hospital acquired (17%) pneumonia. Combination therapy was more frequently used in the Poly/AG group (72% vs. 12%; p < 0.001) Treatment with C/T was associated with a higher rate of clinical cure (79% vs. 62%; p = 0.046) and a lower incidence of AKI (7% vs. 33%; p < 0.001) compared to Poly/AG based therapy. In hospital mortality rates were similar (28% vs. 37%; p =0.33). No patients receiving C/T had hypersensitivity reactions, neurological adverse events, or C. difficile infections. Conclusion: This multi-center retrospective analysis provides real world data supporting improved outcomes with C/T compared to Poly/AG based regimens for invasive infections due to MDR/XDR P. aeruginosa.

Background: Ceftolozane-tazobactam (C-T) is approved by the US Food and Drug Administration and by the European Medicines Agency to treat complicated urinary tract infections, acute pyelonephritis, and complicated intra-abdominal infections. The Program to Assess Ceftolozane-Tazobactam Susceptibility (PACTS) monitors gram-negative (GN) isolates resistant to C-T worldwide. In the current study, isolates were collected from patients hospitalized with bloodstream infections (BSIs) from 2015-2017 within the United States (US). Methods: A total of 3,377 prevalence-based BSI GN isolates, including Escherichia coli (EC; 1,422), Klebsiella pneumoniae (KPN, 630), and Pseudomonas aeruginosa (PSA; 344), were collected during 2015-2017 from 32 PACTS hospitals in the US. Isolates were tested for C-T susceptibility by CLSI broth microdilution method in a central monitoring laboratory (JMI Laboratories). Other antibiotics tested were amikacin (AMK), cefepime (FEP), ceftazidime (CAZ), colistin (COL), levofloxacin (LVX), meropenem (MEM), and piperacillin-tazobactam (TZP). Antibiotic-resistant phenotypes analyzed (CLSI, 2018) for EC and KPN included carbapenem-R (CR) and non-CR extended-spectrum beta-lactamase (ESBL); as well as CAZ-nonsusceptible (CAZ-NS), MEM-NS, and COL-NS PSA. Results: Of the 3,377 BSI GN isolates, 3,219 (95.3%) had a C-T MIC ≤ 4 mg/L. The 3 most prevalent GN species isolated from BSIs were EC (42.1%), KPN (18.7%), and PSA (10.2%). The %S of C-T and comparators for the top 3 pathogens are shown in the table. C-T showed activity against these isolates with %S of ≥96.0% against all 3 species. Of the comparators tested, AMK and COL also had high %S against these isolates. Conclusions: C-T demonstrated activity against the most prevalent contemporary GN isolates from BSIs in the US. C-T was the only beta-lactam that had ≥96%S against all 3 species: EC, KPN, and PSA. For PSA, C-T maintained activity (>90%S) against isolates resistant to CAZ, TZP, and MEM. These data suggest that C-T may be a useful treatment for GN BSI.

Introduction:  Plazomicin, a next-generation aminoglycoside, retains in vitro activity against Enterobacteriaceae possessing common aminoglycoside modifying enzymes.  Enterobacteriaceae that harbor a 16S rRNA methyltransferase demonstrate phenotypic resistance to plazomicin, but these remain rare in Canada at present.  A patient with a 16S rRNA methyltransferase-producing Klebsiella pneumoniae isolate recovered on urine culture is described.  Methods:  A 64 year-old previously healthy male was hospitalized in India with obstructive uropathy.  He subsequently returned to Canada and was admitted to hospital here for further evaluation.  Abdominal imaging demonstrated a large bladder neoplasm.  A biopsy was consistent with a high grade carcinoma.  Bilateral nephrostomy tubes were inserted, and the patient received 14 days of antimicrobial therapy for a possible urine infection.  At the completion of treatment, a follow-up urine specimen for culture was obtained and a pan-aminoglycoside (gentamicin, tobramycin, amikacin) resistant K. pneumoniae isolate was recovered.  Plazomicin susceptibility testing was performed by broth microdilution, as described by the Clinical and Laboratory Standards Institute.  The isolate underwent whole genome sequencing (National Microbiology Laboratory, Winnipeg, Canada) to assess for the presence of a 16S rRNA methyltransferase enzyme.  Results:  The K. pneumoniae isolate had a plazomicin MIC of >64 µg/ml (resistant by the US FDA breakpoint).  The rmtF gene, which encodes a 16S rRNA methyltransferase enzyme, was detected by whole genome sequencing.  Molecular testing also demonstrated the presence of a NDM-1 carbapenemase and a CTX-M-15 extended-spectrum beta-lactamase.  As the patient described here did not have any symptoms consistent with a urinary tract infection at the time the isolate was identified, antimicrobial therapy was not prescribed. Conclusions:  Hospitalization outside of Canada is a risk factor for colonization with antimicrobial-resistant bacteria.  While much of the literature has focused on acquisition of beta-lactamase producers, isolates with genes conferring resistance to other antimicrobials (including novel agents such as plazomicin) may also be acquired.   

Background: Escherichia coli causes 80-85% of acute episodes of uncomplicated cystitis. Rates of resistance have undergone considerable variations, and consequently the empirical treatment requires regular updating of the antibiograms for different geographical areas and institutions. Objective: To compare susceptibility patterns of E.coli urinary isolates in out-patients and long term care (LTC) patients in local health integration networks (LHIN) in Ontario. Method: Retrospective data analysis was done on E.coli isolates from urine samples in 2016-2017 for out-patients and LTC patients, excluding hospitals. Susceptibility to Ciprofloxacin, Trimethoprim-sulfamethoxazole and Nitrofurantoin and prevalence of Extended Spectrum β lactamase (ESBL) were determined. Result: A total of 203,336 from 2017 and 191,748 isolates from 2016 were analyzed. Median overall susceptibility to Ciprofloxacin in Ontario was 69%, Trimethoprim-Sulfamethoxazole 74%, and Nitrofurantoin 97%, ESBL prevalence was 12.5%. For patients in LTC facilities, median susceptibilities were 51%, 67.5%, 93%, respectively and ESBL prevalence was 20.6%. For out-patients, median susceptibility to Ciprofloxacin was 84%, Trimethoprim-Sulfamethoxazole 79% and Nitrofurantoin 97%, ESBL prevalence was 5.5%. Lowest ciprofloxacin susceptibility for out-patients was from Central West LHIN at 79%. For LTC, lowest Ciprofloxacin susceptibility was from Champlain LHIN at 40%. For Trimethoprim–sulfamethoxazole, lowest susceptibility in out-patients was in Central West LHIN at 74%, and for LTC was in Champlain LHIN at 59%. Highest prevalence for ESBL in out-patients was from Central West LHIN at 8.9%, and for LTC was from North West LHIN at 31.3%. Susceptibility results were not significantly different in 2017 compared to 2016. Conclusion: E.coli isolates from urine have more resistance to antibiotics in patients from LTC compared to patients in the community. Separation of the patient population into community and LTC subgroups is helpful when selecting empiric antibiotic therapy. Resistance to Ciprofloxacin and Trimethoprim–sulfamethoxazole was high, highlighting the importance of relevant, local antibiograms.

Objective(s): Cefazolin has excellent activity against methicillin-susceptible Staphylococcus aureus (MSSA), but some strains show cefazolin inoculum effect (CzIE) which may reduce the clinical effectiveness of this drug. Currently, MSSA strains with CzIE are identified by the micro broth-dilution method (MBDM) using a high inoculum, but this method is cumbersome and cannot be used routinely. This study sought to evaluate a more simple method to identify MSSA with CzIE using standard disk diffusion method (DDM). Method: 201 non-duplicated MSSA isolated from blood cultures were evaluated. The cefazolin MIC was determined by the MBDM using the standard inoculum [10⁵ colony forming units (CFU)/mL] and high inoculum (10⁷ CFU/mL). MBDM testing was performed according to CLSI guidelines. S. aureus references strains which exhibit high-inoculum effect or not were used as controls. CzIE was defined as an increase in minimum inhibitory concentration (MIC) to cefazolin >16 mg/L when tested with the high inoculum.  Zones of inhibition for cefazolin were determined by DDM using a 30μg cefazolin disk as per CLSI guidelines. Results: Of 201 MSSA, 62 isolates had a cefazolin MIC of ≥16 mg/L, thus were positive for CzIE. One hundred thirty-nine strains did not show CzIE using MBDM. Using a zone of ≤29 mm as a cut-off, the sensitivity of cefazolin DDM was 91.93% (57/62), but the specificity was low at 77.69% (108/139). Conclusion: This study shows that the DDM of cefazolin is reasonably sensitive for identifying MSSA with CzIE. The method is easy to perform and may identify patients for whom cefazolin may not be an optimal drug. Further analyses are underway to validate the robustness of these conclusions.

Objective: To evaluate the impact of weekly antimicrobial stewardship (ASP) rounds for general internal medicine (GIM) patients, first led by an ASP and later general internal medicine (GIM) pharmacists. Methods: Weekly interdisciplinary ASP rounds attended by physicians, trainees and pharmacists from both the ASP and GIM teams were conducted at a tertiary care hospital in Hamilton, ON. All GIM patients on anti-infectives and not followed by the infectious diseases service were reviewed. Rounds were first led by an ASP pharmacist (1-6/2018), and then by the GIM pharmacists (7-11/2019). The ASP pharmacist was available for consultation as needed and occasionally attended rounds. The transition was supported by an 11-module educational ASP credentialing program. Process measures were collected and statistical process control (SPC) was used to evaluate changes in antimicrobial utilization (monthly DDD per 1000 patient days). Results: Process measures captured in detail between 1-8/2018 showed that the non-teaching teams attended 19/26 (73%) and teaching teams attended all scheduled rounds, and 567 patients were discussed. ASP intervened on an average 1 in 4 patients and 143/159 (89.9%) interventions were accepted; the majority focused on duration (n=74; 46.5%) and discontinuation (n=35; 22%). Compared to 2016/2017, there was an 18% reduction in utilization (from 559 in 2016/2017 to 460 DDD/1000 patient days). Five months were below 2 control limits, and the monthly utilization was below the 2016/2017 average in each month since the implementation of ASP rounds meeting SPC criteria as a special cause change (‘significance’). Conclusion: Dedicated interdisciplinary ASP rounds to discuss patients admitted under a GIM service facilitated optimization of anti-infective therapy, and resulted in a significant reduction in antimicrobial utilization. These gains were sustained when GIM pharmacists took over leading the rounds, suggesting this is a feasible sustainability model for antimicrobial reviews.

Objectives: Antimicrobial resistance (AMR) is a global health concern. Our health region lacks an antimicrobial stewardship program (ASP), an intervention strongly recommended to curb AMR. We explored perceptions and attitudes towards antimicrobial prescription, resistance, and stewardship and whether these varied by practitioner type; and identified prescriber preferences in developing a local ASP. Methods: A 29 item-questionnaire (19 Likert-scale, 7 multiple-choice and 3 open-text questions), adapted from a published survey was disseminated using online software.  401 invitations were emailed to attending physicians and residents within medical specialties, surgical specialties, family medicine and critical care. The survey was open for 6 weeks. Comparisons between groups were performed by Fisher’s Exact tests.
Results: We received 122 completed responses, a response rate of 30.4%. Most (76.9%) respondents were from medical subspecialties or family medicine. 56.5% of participants were residents. Over 95% of participants agreed that broad spectrum agents contributed to AMR, and that prescription optimization limits its development. While only 7.6% of participants felt they had overprescribed antimicrobials, 48.8% felt these were over-prescribed by clinicians other than themselves.  82% of participants felt that an ASP had potential to improve patient care; this attitude did not vary by resident vs. attending status (p=0.311), between prescribers in academic or community settings (p=0.561), by clinical specialty (p=0.508) or by frequent or infrequent antibiotic prescribers (p=0.650). Most respondents favored audit and feedback from an infectious disease physician. Conclusion: Surveyed attending and resident physicians acknowledged the importance of antimicrobial optimization in limiting AMR. There may exist a lack of awareness among prescribers regarding over-prescription habits. All prescriber groups exhibited receptiveness towards implementation of a formal ASP and agreed it would benefit patients. In developing a formal program in our region, prospective audit and feedback with communication from infectious disease physicians would likely be viewed favorably.

Objective:  Managing urinary tract infections (UTIs) caused by antibiotic- resistant E.coli is challenging given limited oral therapeutic options. Current guidelines recommend basing empirical treatment of UTIs on local or regional susceptibility data. The aim of this study was to evaluate in vitro susceptibility of urinary isolates of Escherichia coli to oral antimicrobial agents in patients at two acute care hospitals in the year 2016. Methods: Patients aged ≥18 years with positive urine cultures for E. coli from January to December 2016 were included. In vitro susceptibility testing of E. coli to oral antimicrobial agents under review were tested using the commercial VITEK 2 automated identification/susceptibility testing instrument, and when indicated, by disk diffusion in accordance with Clinical and Laboratory Standards Institute (CLSI) guidelines.  Descriptive statistics were used to summarize the result of the study. Results: A total of 1999 urinary isolates of E. coli were included for analysis. The rates of E. coli susceptibility to nitrofurantion, trimethoprim‐sulfamethoxazole, and ciprofloxacin were 96.3%(1925/1999), 73.9%(1478/1999), and 72.3%(1446/1999) , respectively. The rates of extended spectrum β-lactamase (ESBL)- and AmpC- β -lactamase - E. coli were 11.6% (231/1999) and 2.3%(45/1999), respectively. The rates of susceptibility of ESBL-E.coli to nitrofurantoin, trimethoprim‐sulfamethoxazole, and ciprofloxacin were 91.3% (211/231), 30.3%( 70/231), and16% (37/231), respectively. The rates of susceptibility of AmpC- β -lactamase-E coli to nitrofurantoin, trimethoprim‐sulfamethoxazole, and ciprofloxacin were 91.1%(41/45), 71.1% (32/45), and 68.9%(31/45), respectively. Conclusions:  Nitrofurantoin is the most active oral agent against urinary isolates of E. coli including ESBL- and AmpC- β -lactamase-producing strains. Trimethoprim‐sulfamethoxazole and ciprofloxacin should not be used as empiric treatment of UTIs, particularly in patients at risk of infections with ESBL- and AmpC- β -lactamase- producing E coli given resistance rates of Enterobacteriaceae spp exceed 20%.

Objectives: Prolonged antibiotic duration of therapy is common in long-term care (LTC) settings and is associated with increased risk of harm for residents, including the development of antimicrobial resistance. Although there is a growing body of evidence evaluating the factors influencing the initiation of antibiotic therapy, little is known about the influencers of prolonged duration. To identify potential antibiotic stewardship opportunities aimed at prolonged duration of therapy, this study examined barriers and enablers to using shorter courses of antibiotic therapy in the LTC setting. Methods: Semi-structured interviews were conducted with prescribers in LTC home settings. A total of eight LTC clinicians participated in the study (4 physicians and 4 nurse practitioners). Questions and clinical scenarios (addressing urinary tract infection, pneumonia and cellulitis) explored the factors influencing prescribers’ decisions about duration of therapy. Interview data were analysed deductively, using the Theoretical Domains Framework (TDF). The coding structure reflected the 14 domains of the TDF. Two reviewers independently analysed the data and disagreements were resolved with consensus. Results: The most commonly cited domains (and frequency of occurrence) influencing duration of antibiotic therapy in LTC were environmental context and resources (n=84); knowledge (n=32); beliefs about consequences (n=29); social influences (n=21); and behavioural regulation (n=19).  Specific concerns described by participants included the perceived lack of evidence to support shorter courses in LTC residents, the misconception that shorter courses could lead to greater rates of resistance, and the strong role of habit and prior experience in selecting antibiotic duration. Conclusions: There are several factors affecting antimicrobial duration prescribing behaviour aside from the clinical scenario itself. Tackling misconceptions and providing educational support may be helpful approaches to address prolonged treatment. These findings provide theory-informed evidence to support the development of antimicrobial stewardship interventions aimed at improving duration of antibiotic therapy. 

Objective(s): In April 2018, Lions Gate Hospital (LGH) transitioned from paper charts to a new electronic medical record system. With the new system, automatic stop dates for antimicrobials were replaced by soft stop dates and soft stop reminders. Antimicrobial utilization data from the legacy and electronic system were analyzed and compared to identify the new system’s impact on the duration of antimicrobial therapy. Methods: An antimicrobial point prevalence survey was conducted at LGH between September to November 2018. The survey included all inpatients admitted at the hospital at 8:00am and receiving antimicrobial therapy within the past 24 hours of the survey dates. The duration of the current active antimicrobial prescription order(s) (orders with the same drug, dose, route, and frequency) was recorded. Previous doses were considered a separate order and excluded from the survey. Results: 220 patients were surveyed of whom 60 patients (26.9%) met the inclusion criteria for a total of 79 antimicrobial prescriptions (average 1.32 antimicrobials per patient receiving antimicrobials). Comparing 2017 to 2018, the number of patients receiving antimicrobials (61 vs. 60) and antimicrobial prescriptions (78 vs. 79) remained consistent; however, the survey identified an increase in average duration of antibiotic prescriptions on the day of survey from 3.0 to 4.9 days. Duration of oral antimicrobials increased from 2.5 to 5.5 days while IV antimicrobials increased from 3.4 to 4.5 days. Conclusion(s): These results suggest that the loss of automatic stop dates for antimicrobials due to implementation of an electronic health record system may have resulted in an increase in antimicrobial prescription duration. This presents future opportunity for improvement with implementation of the new electronic system and antimicrobial stewardship initiatives. Further point prevalence surveys need to be done to assess the impact of these initiatives.

Objectives: Optimal prescribing of pre-operative antimicrobial prophylaxis is complex and multidisciplinary.  Previous audits have identified opportunities for improvement in general surgery and orthopedics, but no data is available for the majority of surgical sub-specialties.  We wish to investigate pre-operative antibiotic prescribing concordance with administered antibiotics in the OR across all surgical subspecialties and assess guideline concordance of those antibiotics received in the OR. Methods: As part of a larger quality improvement initiative, audits were conducted on patients from each of the 13 surgical subspecialties at a large, urban teaching hospital system. Patients were randomly selected for prospective audit conducted during the OR followed by retrospective review of pre-admission consults and antibiotic prescribing.   All administered antibiotics were adjudicated for concordance by comparing administered antibiotics to local best practice guidelines.  All analysis was conducted using descriptive statistics.  Criteria for concordance assessment were initial choice, dose or timing as well as redose choice, dose or timing. Results: 200 patients were audited during the study period (August to November 2018).  Discordance was common between antibiotics prescribed on initial assessment by surgeons prior to the OR and those given in the OR (49%).  The most common reasons for discordance were 1) no record of antibiotics prescribed in initial surgical assessment (30%); 2) antimicrobial dose change (27%) and 3) antimicrobial choice change (17%).   Of all antibiotics administered in the OR, 39% were guideline discordant.  The majority of these (56%) were due to discordant initial antibiotic dose followed by initial antibiotic choice (17%) and re-dose dose (11%). Conclusions:  A high degree of discordance exists between both the prescribed and OR administered antimicrobials administered as well as between OR administered antimicrobials and local practice guidelines.  Opportunities for both process and outcome improvements exist and require further investigation.

Daptomycin and linezolid are indicated by Alberta Health Services (AHS) guidelines for resistant Gram-positive infections. The first-line agent is vancomycin; however, daptomycin and linezolid are frequently used for dosing or administration route convenience. Objectives: (1) Evaluate the number of guideline concordant prescriptions of daptomycin and linezolid in a 6-month period and (2) investigate reasons for guideline discordant prescriptions. Methods: We conducted a retrospective chart review of all inpatient, emergency, and outpatient antimicrobial therapy (OPAT) clinic prescriptions for daptomycin and linezolid at an inner-city tertiary care hospital in Edmonton, AB, over a 6-month period (July to December 2017). Results: There were 45 courses of daptomycin and 24 courses of linezolid prescribed. 26 of 45 (57.8%) daptomycin prescriptions and 11 of 24 (45.8%) linezolid prescriptions were discordant with AHS guidelines. The Infectious Diseases consult service was involved in the majority of these prescriptions (daptomycin, 40/45, 89%; linezolid, 18/24, 75%). 8/11 discordant linezolid prescriptions were initiated due to lack of or discontinuing intravenous (IV) access due to substance use. 14/26 discordant daptomycin prescriptions were initiated to increase compliance (once daily dosing) or facilitate discharge without the need for home parenteral therapy. Conclusion: There is a high rate of discordance for daptomycin and linezolid when using the AHS formulary guidelines. However, the majority of guideline discordant prescriptions were due to valid extenuating circumstances in this unique patient population that are not included in the AHS guidelines (e.g., need for once daily or oral options) that necessitated the use of daptomycin or linezolid rather than first-line vancomycin (either as an inpatient or via home parenteral therapy). Additionally, use of daptomycin in the OPAT clinic instead of vancomycin in the inpatient setting is a cost-saving and patient-centered measure. 

Objective(s): Global Point Prevalence Survey (PPS) of Antimicrobial Consumption and Resistance is a project collecting international data to monitor rates of antimicrobial prescribing in hospitalized patients. Antimicrobial utilization data from Fall 2017 and Fall 2018 were compared with the goal of identifying stewardship opportunities. Methods: A point prevalence study was performed according to the Global Point Prevalence Survey. In September-November 2017, data were collected from paper charts while in September-November 2018 they were gathered from the newly implemented electronic medical record system. Via a snapshot of all patients receiving antimicrobial therapy at 0800h on the survey day, this project characterizes the overall antimicrobial use pattern (administration, route, and guideline adherence). Results: In 2017 and 2018, a similar number of patients received antimicrobial therapy on the survey day, 61 (28.8%) and 60 (26.9%), respectively. While only 64.1% (50 prescriptions) of the orders were guideline-compliant in 2017, an improvement was seen in 2018 with 84.8% (67 prescriptions) guideline-compliance. While the percentage of orders with documented stop date increased from 34.7% (25 prescriptions) to 41.8% (33 prescriptions) from 2017 to 2018, the proportion of oral antimicrobials decreased from 46.2% to 35.4%. Conclusion(s): The PPS results reveal that the total usage of antimicrobial agents was similar between two consecutive years. While there seems to be an improvement in guideline adherence, the percentage of orders with documented stop date remains low. As a result, documenting treatment duration and IV to PO stepdown remain two key opportunities for the antimicrobial stewardship program.

Background: US data suggests that a two EIA algorithm (2EIA) (whole cell (WC) EIA followed by C6 EIA) has improved sensitivity but equivalent specificity to the recommended two tier algorithm (TTA) (EIA followed by Western blot (WB)) for the serologic diagnosis of LD. From 2011 to 2014, the QEII laboratory used a WC EIA for LD serology.  Positive or indeterminate results were then tested with the C6 EIA by the National Microbiology Laboratory.  Only the sera with positive or equivocal C6 results were tested with IgG and IgM WB, providing the opportunity to evaluate the real world performance of the 2EIA compared to TTA in a Nova Scotian population. Methods: A retrospective chart review was performed on patients testing positive with both WC and C6 (2EIA approach). Patients were classified as having LD if they had 1. a positive TTA result; 2. A negative TTA result but had symptoms consistent with LD; or 3. Evidence of seroconversion between consecutive specimens. Specificity was calculated based on standard 4x4 table. Results: From 2011-2014, 10253 specimens were tested for LD, 9790 were negative.  Of the 256/463 positive charts reviewed to date, 213 and 43 specimens were classified as coming from patients with and without LD respectively.  The number of 2EIA positive patients without LD (false positive results) was 43.  Of the 213 results from patients with LD there were: 119 positive IgG WB; 57 positive IgM WB with negative IgG WB; 17 negative specimens with IgG/IgM WBs. The remaining 20 specimens were IgG WB borderline or negative, with negative/borderline/not tested IgM WB. Calculated specificity is 99.56% (99.41-99.68%). Conclusion: Preliminary analysis suggests the 2EIA has excellent specificity and is more sensitive than the TTA.  A further chart review is required to accurately define the sensitivity of 2EIA in our population.

Objectives: In 2018 the Canadian Public Health Laboratory Network recommended detecting Shiga toxin (Stx)-producing E. coli (STEC) by one of three methods: nucleic-acid testing, chromogenic agar, and toxin antigen detection after pre-incubation in broth. In this study, the sensitivities of the latter two methods were compared prospectively. Methods: Feces submitted for culture were tested for STEC using the CHROMagar™ STEC. The production of Stx by mauve-coloured colonies was confirmed by SHIGA TOXIN QUIK CHEK™. If the requisition indicated hematochezia, hemolytic uremic syndrome, or a request to rule out STEC, feces were additionally incubated in MacConkey broth, then tested using the SHIGA TOXIN QUIK CHEK™ (brothEIA). Enrichment culture to obtain a STEC isolate from brothEIA-positive-only specimens was performed at the Alberta Provincial Laboratory for Public Health. Isolates were serotyped at the Public Health Agency of Canada-National Microbiology Laboratory. Culturing a STEC isolate from the CHROMagar or on subculture from brothEIA-positive-only specimens was considered the reference standard for a positive result. Results: From June 11 to December 31, 2018, there were 12,882 fecal culture specimens. Of 1599 specimens meeting criteria to be tested by both CHROMagar and the brothEIA methods, 41 positives were detected. For two specimens, the brothEIA was toxin positive but failed to grow on enrichment culture, therefore 39 specimens were included in the sensitivity calculations. Each method had a sensitivity of 87% (95% CI 72.8-94.8): 5 false negatives for each method. The most common serotype was O157 (26%), followed by O26:H11 (10%). No serotype was disproportionately represented in the false negatives for either the plate or the brothEIA method. Conclusions: The CHROMagar and brothEIA each exhibits acceptable sensitivity in our study. While using both methods simultaneously would increase the number of STEC detected, the additional costs may not be justifiable in a cost-benefit analysis.

Background: Rapid identification of methicillin-resistant Staphylococcus aureus (MRSA) is important for patient care and infection control. Several studies have evaluated the performance of the Bruker^TM MALDI-TOF MRSA Subtyping Module, based on the detection of phenol-soluble modulin (PSM), from routine cultures; however, few have tested selective or chromogenic media. The purpose of this study is to evaluate the subtyper using common MRSA surveillance media. Methods: 108 MRSA and 101 methicillin-sensitive S. aureus (MSSA) isolates were randomly selected from surveillance cultures. Isolates were subcultured to two chromogenic media (Bio-Rad MRSASelect^TM II and Alere Colorex MRSA Agar), mannitol-salt agar (MSA), and blood agar (BA) and were confirmed as MRSA by PBP2a’ latex agglutination. Isolates were spotted on a target plate using the direct transfer method and sequentially run on two MALDI-TOF machines. PSM peak detection was done automatically by the Bruker^TM software. Results: Performance varied depending on the media. From BA, the sensitivity was 35-38% and specificity was 100%. Sensitivity and specificity from MSA was 81-85% and 36-58%, respectively. Sensitivity from the Alere chromogenic agar was 84% but was 0% from the Bio-Rad chromogenic agar. Specificity was not determined as MSSA are unable to grow on chromogenic media. Conclusion: Using nonselective media (BA), the subtyper exhibited limited sensitivity but excellent specificity; however, from selective media (MSA), sensitivity increased but the specificity was poor. This suggests that the rate of false positives can be high depending on the media, an observation that has been underappreciated in the literature. Variable performance was also observed between the two chromogenic plates but suggests that the subtyper can be used to confirm presumptive MRSA colonies from the Alere agar. Taken together, our results show that the Bruker^TM MRSA subtyper exhibits media-dependent performance and highlights that extensive validations are needed before adapting rapid identification methods for routine resistance detection. 

Objectives: The CANWARD study assesses the pathogens causing infections in patients affiliated with Canadian hospitals and evaluates the prevalence of antimicrobial resistance in these isolates. Methods: In 2018, twelve tertiary-care centres across Canada submitted clinical isolates from patients attending clinics (C), emergency rooms (ER), medical and surgical wards (W), and intensive care units (ICU). Antimicrobial susceptibility testing was performed using CLSI broth microdilution methods. Results:  A total of 2,963 isolates were collected: 40.1%, 39.8%, 10.1%, and 10.0% from blood, respiratory, urine, and wound/IV site specimens, respectively. Patient demographics were as follows: patient gender, 54.1% male and 45.9% female; and patient age, 6.6% ≤17 years, 44.7% 18-64 years, and 48.7% ≥65 years. Isolates were from patients on W 39.3%, ER 24.4%, ICU 19.6%, and C 16.7%. The most common pathogens were S. aureus 21.1% (16.5% MSSA / 4.6% MRSA), E. coli 18.7%, P. aeruginosa 9.7%, K. pneumoniae 6.5%, and S. pneumoniae 4.2%. Resistance rates for E. coli were: 0% for meropenem and tigecycline, 0.5% ertapenem, 3.7% piperacillin/tazobactam, 8.3% gentamicin, 10.2% ceftriaxone, 20.4% ciprofloxacin, and 23.6% trimethoprim/sulfamethoxazole (SXT). For P. aeruginosa, resistance rates were 2.7% colistin, 3.4% ceftolozane/tazobactam, 12.9% gentamicin, 13.1% ciprofloxacin, and 17.0% piperacillin/tazobactam and meropenem. Resistance rates for MRSA were: 0% vancomycin, ceftobiprole, daptomycin, linezolid, and SXT, 7.1% tigecycline, 35.7% clindamycin, 64.3% ciprofloxacin, and 73.8% clarithromycin. Overall, the prevalence of MRSA, vancomycin-resistant enterococci (VRE), and extended-spectrum beta-lactamase (ESBL)-producing E. coli was 21.6%, 13.0%, and 11.3%, respectively. Conclusions:  In Canada, resistance rates for E. coli remain lowest for meropenem, tigecycline, ertapenem and piperacillin-tazobactam, while for P. aeruginosa, rates are lowest with colistin, ceftolozane/tazobactam, and gentamicin. No resistance was observed in MRSA with vancomycin, ceftobiprole, daptomycin, linezolid, or SXT.

Background. Herpes simplex virus 1 (HSV-1), HSV-2 and Varicella-Zoster virus (VZV) are responsible for a variety of human diseases, of which cutaneous and mucocutaneous infections are the most common. Differentiation between these viruses based on clinical presentation lacks both sensitivity and specificity. Culturing viruses is a lengthy, insensitive and time-consuming process. This study compares the viral culture and a commercial PCR assay to the Solana HSV 1+2/VZV assay for detection of HSV-1, HSV-2 and VZV in cutaneous or mucocutaneous specimens. Methods. From June to October 2018, 302 cutaneous or mucocutaneous specimens for which HSV-1, HSV-2 or VZV viral culture or PCR detection have been requested were randomly selected and kept at -80°C. The Solana HSV 1+2/VZV assay was performed according to the manufacturer’s instructions. Discrepancies between the results of the culture and the Solana assay were resolved with a commercial PCR assay (RealStar alpha Herpesvirus PCR Kit). Results. Compared to culture (n=247), the Solana assay had a sensitivity of 100% (123/123). The Solana assay detected 27 false negative specimens using culture (4 HSV-1, 11 HSV-2 and 12 VZV) that were confirmed positive by the arbiter PCR assay, conferring a specificity of 100%. Compared to the commercial PCR assay (n=55), the Solana assay had a sensitivity of 93.8% (30/32) and a specificity of 100% (23/23). Conclusions. The Solana HSV 1+2/VZV assay was more sensitive than the viral culture for the detection of HSV-1, HSV-2 and VZV in cutaneous or mucocutaneous specimens. It also performed comparably to a commercial PCR assay. This molecular assay is Health Canada approved for the simultaneous detection of HSV-1, HSV-2 and VZV in the same tube. The Solana assay could be completed in 60 minutes for up to 12 specimens, with approximately 1 minute hands-on time per specimen.

Objective: In early 2017, a Pertussis outbreak of 50 cases led to a 20-fold increase in testing requests from a baseline of 5 tests a month.  Case tracking was hampered by offsite testing delay made worse by poor weather conditions.  Our objective was to develop a rapid cost-effective process for both outbreak and routine testing. Methods: The Health Canada Approved AmpliVue (Quidel) using Helicase Dependent Amplification in a hand-held cartridge was subjected to an expanded validation of 100 tests (24 from the manufacturer, 15 from a proficiency panel, 46 from a provincial laboratory using a mock universal transport media (UTM), and 15 locally acquired random samples).  At the same time, the BDMAX (Becton Dickinson) with BioGx reagents on a multiplex PCR platform was verified against a panel of 287 samples (The AmpliVue expanded panel and an additional 182 local samples and 5 additional provincial lab samples.  The national microbiology laboratory assisted with discordant results. Results: Both platforms were easy to use. The AmpliVue assay had 100% sensitivity, 92% specificity for B. pertussis only. The BDMAX test demonstrated 100% sensitivity, 100% specificity for B. pertussis. B. parapertussis showed a sensitivity of 71%, specificity of 100%.  Sensitivity and specificity of B. holmesii was 100%.  There was a 9.0% unresolved rate for B. pertussis on the BD Max, which may be due to DNA degradation and sample dilution during the mock UTM preparation.  Routine testing turnaround time improved from 5.3 days to 1.5 days after implementation (January to March 2016 vs. 2018). Conclusions:  Both platforms performed well and are highly complementary by allowing rapid turnaround time, decreased wastage, and confirmatory PCR molecular testing with multiplexing and surge capacity. A Bordetella spp. proficiency panel should be provided to Canadian laboratories on a more frequent basis to enhance quality assurance of various testing methods.

Objective: Streptococcus pyogenes (Group A) causes “Strep throat”. The objective of this study was to validate the use of WASP™/ WASPlab™ to seed and analyze Colorex™ Strep A agar (CHROMagar™) to screen for Streptococcus pyogenes in throat specimens. Methods: In this study 159 clinical throat specimens were collected with ESwab™ kits and processed on a WASP™ using Colorex™ Strep A (CHROMagar™) agar plates and incubated in the WASPLab™ for 20 hours in CO2 at which point imaging analysis was performed. Vitek MS (Maldi-ToF) and PathoDx were performed on target and non-target colour colonies isolated. Results were compared to the same samples set up on Blood agar incubated at 35 degrees C anaerobically for 20 hours. The samples had all been tested for S. pyogenes by LAMP PCR. Results: Of the 159 clinical specimens, 120 were positive for S. pyogenes Group A by LAMP PCR. Of those 120 specimens, 116 grew on  Colorex™ Strep A agar and 109 showed beta haemolysis on blood agar. 56 target positive colonies were tested with Vitek MS (Maldi-ToF) and all 56 identifies as S. pyogenes. The other 60 target positive colonies were tested with PathoDx using A and C. All 60 tested A positive C negative. White colonies identifies as other Streptococcus species. Colorex™ Strep A agar (CHROMagar™) showed a sensitivity of 96.7% (95%CI 0.92-0.99) and a specificity of 100% (95%CI 0.95-1) as compared to LAMP PCR. Conclusions: Results showed Colorex™ Strep A (CHROMagar™) had a significantly greater sensitivity than Blood agar in isolating S. pyogenes from throat specimens. The use of the WASP for set up provides efficent and consistent processing and WASPlab™imaging allows for high resolution digital imaging analysis. Several images appeared negative until zooming into the image showed few target positive colonies.

Objective(s): The Serious Outcomes Surveillance Network of the Canadian Immunization Research Network (CIRN SOS) has been performing active influenza surveillance since 2009 (ClinicalTrials.gov identifier: NCT01517191). Influenza A and B viruses are identified and characterized using real-time RT-PCR, and on a subset of patients, multiplex testing was performed to identify other respiratory virus etiologies. Since both methods could identify influenza A and B, a direct comparison was performed. Methods: Validated real-time RT-PCRs from the World Health Organization (WHO) to identify influenza A and B viruses, characterize influenza A viruses into the H1N1 or H3N2 subtypes, and to describe influenza B viruses belonging to the Yamagata or Victoria lineages. In a subset of patients, the Seeplex RV15 One-Step ACE Detection assay (RV15) kit was also used for detection of other respiratory viruses. Results: In total, 1111 nasopharyngeal swabs were tested by RV15 and real-time RT-PCRs for influenza A and B identification and characterization. For influenza A, RV15 showed 98.0% sensitivity, 100% specificity, and 99.7% accuracy. Performance characteristics of RV15 were similar for Influenza A subtypes H1N1 and H3N2. For influenza B, RV15 had a 99.2% sensitivity, 100% specificity and 99.8% accuracy, with similar assay performance for both the Yamagata and Victoria lineages. Conclusions: Overall, detection of circulating subtypes of influenza A and lineages of influenza B by RV15 was similar to real-time RT-PCR. Multiplex testing with RV15 allows for a more comprehensive assessment of respiratory virus surveillance in hospitalized adults, without significantly compromising the reliability of influenza A or B virus detection.

Objective: Rapid molecular multiplex testing has revolutionized enteric diagnostics and timely treatment. Nonetheless, maintaining organism viability in the pre-analytic phase remains necessary for isolation, antimicrobial susceptibility testing and serological typing/identification. The new FecalSwab system (Copan Diagnostics, Brescia, Italy) is a convenient alternative to bulk stool for collecting, transporting, and processing specimens for the diagnosis of enteric pathogens. However, FecalSwab suitability for molecular platforms and maintaining bacterial viability at different temperatures has not been well evaluated. Therefore, we assessed the effect of temperature and time on the performance of the FecalSwab system for culture and molecular detection of enteric pathogens. Methods: A clinical stool specimen previously characterized as positive for one of the bacterial (Salmonella spp., Shigella spp., Yersinia enterocolitica, Campylobacter coli/jejuni and Shiga toxin-producing E. coli) and viral (norovirus, rotavirus, and adenovirus) targets was swabbed according to manufacturer guidelines, and stored at 4⁰C, 22⁰C, or 37⁰C for up to 7 days. Conditions were performed in triplicates for each pathogen. Samples of the FecalSwab transport medium were collected at baseline, 24h, 48h, and 7 days following storage to determine colony counts and assessed for molecular diagnostic suitability with a lab developed multiplex enteric assay run on the BD Max system (BD Diagnostics, Baltimore, MD). Results: Following 7 days of storage at the specified temperatures, each bacterial isolate was recovered from the FecalSwab with the exception of Campylobacter spp. which rapidly lost viability. Evaluation of the FecalSwab system to support molecular diagnostics revealed that the target microorganisms could be detected without a significant loss in sensitivity (≤3.2 cycle threshold) from baseline at all time points. Conclusion: Copan FecalSwab is a suitable device to collect and store stool specimens for molecular diagnostics using the BD Max system, as well as for culturing the majority of the tested bacterial enteric pathogens. 

Background: Rapid molecular assays for influenza A/B detection are increasingly being adopted in healthcare facilities.  We investigated the limit of detection for the cobas® Influenza A/B&RSV (Roche Molecular Diagnostics) assay using the Liat® system in comparison to our laboratory-developed real-time PCR (LDT). Methods:  A comparison of the Liat® to the LDT was performed utilizing two previously positive nasopharyngeal samples (influenza A-H1N1 and influenza B) and purified influenza A/PR/8/34(H1N1) (Advanced Biotechnologies Inc). The patient samples were serially diluted either with blank universal transport media (UTM) or UTM from samples previously testing negative to the expected Ct (35, 38.3, 41.6), while the standard was diluted to 25copies/mL. For the LDT, samples were extracted on the MagNA Pure Compact, and amplified on the LightCycler® 480 Instrument II (Roche Molecular Diagnostics). Primer/probes were based on the CDC influenza A/B protocol. Results: For each dilution of patient sample in blank UTM or negative patient UTM, four replicates were tested (n=96). The LDT recovered all influenza A/B at Ct35 and 38.3, regardless of UTM (16/16 and 16/16 respectively). However, a discrepancy was identified on the Liat based on the diluent at Ct38.3: blank UTM (8/8) and negative patient UTM (3/8). Recovery at Ct41.6 was inconsistent for both methods: LDT blank UTM (3/8), LDT negative patient UTM (2/8), Liat blank UTM (1/8) and Liat negative patient UTM (0/8). Serial dilutions of the purified standard with blank UTM revealed identical results between the Liat® and LDT: 250copies/mL (5/5;5/5), 125copies/mL (4/5;5/5), 62.5copies/mL (6/6;6/6) and 25copies/mL (4/6;4/6). Conclusions: Laboratories validating rapid molecular PCR systems should carefully consider dilution media when assessing limit of detection.  We identified a difference of approximately 1log in the detection of virus when directly comparing testing performed using blank UTM as diluent versus UTM from negative patient samples.  

Objective: Early initiation of antiviral therapy for individuals with influenza infection is important for improving outcomes. Rapid molecular influenza assays may reduce diagnostic uncertainty and improve patient outcomes by providing faster results compared to traditional batched real-time PCR. The objective of this study is to determine the utility of implementing a rapid influenza PCR compared to the standard of care in a hemodialysis unit. Methods: From November 1, 2017 to March 31, 2018, we assigned samples collected from a single center, hemodialysis unit to be processed using a rapid influenza PCR (cobas® Influenza A/B & RSV assay)  or the standard of care (in-house developed multiplex real-time PCR). Samples were assigned to the rapid PCR if the patient received dialysis treatment in the morning dialysis shift, while the remainder were processed as per standard of care. Study outcomes included the time to result of nasopharyngeal swab, prescription of influenza antiviral therapy, time to receiving prescription, and the need for emergency room visit or hospitalization within two weeks of presentation. Results: During the study period, 44 patients were assessed (14 with the rapid PCR and 30 with the standard of care assay). Rapid PCR significantly reduced time to result (2.3h vs 22.6h, p<0.0001). Individuals who were tested using the rapid PCR had a trend to shorter time to receiving antiviral prescriptions (0.7 days vs 2.1 days, p=0.11), and fewer emergency room visits (7.1% vs 30%, p=0.13) and hospitalizations (14.3% vs 30%, p=0.46) within 2 weeks of testing. Conclusions:Rapid influenza molecular testing in the hemodialysis unit was associated with a shorter time to a reportable result and may be associated with reduced time to prescription of antiviral therapy and fewer hospitalizations/emergency room visits. Further study with a larger cohort is needed to confirm these findings.

Background: The new Xpert Xpress Flu/RSV (Cepheid Inc.) assay is a rapid, on-demand real-time (RT) PCR assay that detects influenza A/B and respiratory syncytial virus (RSV) using a single-test cartridge system with minimal hand-on-time. We evaluated the performance of the Xpert Xpress Flu/RSV assay in comparison with our in-house influenza A/B and RSV RT PCR. Materials/methods: We selected a panel of convenience, archived nasopharyngeal (NP) swab specimens tested by in-house PCR. We included positive specimens for each target covering a range of PCR Ct values (17 – 35) and known negative specimens. All specimens were tested by the Xpert Flu assay and re-tested in parallel by our in-house RT PCR. For discordant samples BioFire FilmArray Respiratory Panel (FA RP) was the resolving test. Results: We included 89 NP specimens of which 69 were positive (23 Influenza A, 23 Influenza B, 23 RSV) and 20 were negative. For 88 of the 89 specimens a concordant result was obtained (98.8 % agreement). One discordant RSV result (in-house PCR positive [Ct=40]; Xpert Xpress Flu/RSV negative) was observed and was re-tested by the FA RP. The discordant result was resolved in favor of the Xpert assay. One previously negative specimen had a low positive influenza A result by both tests (Xpert assay [Ct=35], re-tested in-house PCR [Ct=36]) and was considered a concordant influenza A positive result. After resolution of one discordant result, the Xpert Flu assay correctly detected all 70 positive specimens and all 19 negative specimens for a positive percent agreement and negative percent agreement of 100 % for all three targets. The Xpert Xpress Flu/RSV assay was able to detected all of the 9 specimens with relatively high Ct values (Ct>32). Conclusions: The Xpert Xpress Flu/RSV assay offers accurate, rapid, on-demand influenza A/B and RSV testing.

Objective: To evaluate performance of the Roche Cobas e602 electrochemiluminescence immunoassay against a previously-validated Siemens ADVIA Centaur chemiluminescence immunoassay for syphilis screening. Methods: 98 frozen clinical isolates previously-characterized by the Siemens Centaur assay (with a range of positive, negative, and equivocal values) were tested on the Roche Cobas system (resulting in positive or negative values). Precision and interference studies were also performed. Results: Discrepant results were found in 17 of 98 samples (17.3%), 14 of which (82.4%) tested as equivocal by the Siemens assay but non-reactive by the Roche assay. Using additional testing information (repeat EIA and Treponema pallidum particle agglutination assay (TPPA), a provincial test interpretation algorithm, and expert opinion), 15 of 17 discrepant results were further classified as positive or negative. 2 results were excluded from analysis, one due to a lack of confirmatory testing and another because the patient was recently treated for syphilis. All samples that tested as equivocal by the Siemens assay but non-reactive by the Roche assay were classified as negative. The Roche assay was accurate for 94 of 96 results (97.9%). Within-run, between-run and near-cut-off precision testing was satisfactory and there was no evidence of interference with elevated bilirubin, hemoglobin or lipid levels. Conclusions: The Roche Cobas e602 CLIA assay showed good performance as a syphilis screening test when compared to the Siemens Centaur assay. Limitations of this study include comparison to an existing test method and algorithm rather than the gold standard for all samples tested. 

 Objective: Lyme disease (LD) is the most prevalent tick-borne infection in parts of Europe and Asia and at least five genospecies within the Borrelia burgdorferi sensu lato complex can cause disease in people.  A number of laboratories in Canada are now using the Zeus pepC10/VlsE IgG/IgM ELISA (pepC10) as a screening assay for LD serology.  To-date, there has not been a comprehensive evaluation of this kit’s ability to detect antibodies to the European genospecies of B. burgdorferi nor are these performance characteristics available through the manufacturer.  Therefore, we evaluated the performance characteristics of pepC10 as a screening test for European LD. Method: A total of 262 diagnostic samples (LD suspect patients with travel to Europe/Asia) and 64 European external proficiency (EP) samples were tested by Immunetics C6 IgG/IgM (C6) and pepC10 ELISAs. Diagnostic (n=75) or EP (n=36) samples that were equivocal or positive by either ELISA were subsequently tested by EUROIMMUN Anti-Borrelia EUROLINE (B. afzelii) IgG and EUROIMUN B. garinii IgG Western blot assays. The blot results were considered as the gold standard. Results: The analytical sensitivity of the pepC10 was much higher than that of the C6 on diagnostic samples (93.1% vs 65.5%), while the analytical specificity of both kits was quite poor at around 50%. In contrast, both assays performed extremely well on the external proficiency samples of European origin with 100% sensitivity to pepC10 vs 97.1% for C6. The specificity was 100% and 96.6% for pepC10 and C6, respectively. Conclusion: Based on this evaluation, pepC10 ELISA assay is able to detect antibodies to European genospecies within the B. burgdorferi complex. It is comparable, if not superior, in sensitivity and specificity relative to the C6 ELISA.  Laboratories in Canada can use this assay for screening patients suspected of LD acquired outside of North America.

Objective(s): Central nervous system (CNS) infections such as meningitis and encephalitis cause significant morbidity and mortality worldwide, but their etiology remains unknown in a large proportion (15-70%) of cases. The diagnosis of CNS infections is challenging because the diversity and number of pathogens that may cause CNS infections greatly outnumber available test methods. In this study, we aimed to develop and validate a metagenomic NGS (mNGS) approach for broad-range detection of pathogens associated with CNS infections. Methods: Cerebrospinal fluid (CSF) specimens (n=3) spiked with control bacterial and viral pathogens at varying concentrations were simultaneously assessed by qPCR and mNGS on an Illumina MiSeq platform to determine the analytical sensitivity of mNGS for pathogen detection. For clinical validation, residual CSF specimens (n=36) from patients with suspected CNS infections previously tested by culture and PCR, were retrospectively analyzed by mNGS. Data were analyzed by using MetaPhlAn2 followed by alignment with the viral genome database using Bowtie2. Results: In simulated specimens, mNGS assay detected all spiked pathogens that were detectable by PCR. In clinical specimens, the observed agreement with conventional methods was 92% (Kappa = 0.816; 95% CI = 0.618-1). Confirmatory PCR results on 3 specimens that gave discrepant results were in favor of mNGS assay. MetaPhlAn2 results were more specific than those obtained by other available bioinformatics tools but MetaPhlAn2 was less sensitive for viral detection compared to other tools. Expert assessment of mNGS results is important to exclude potential contaminants and background noise. Analysis of additional CSF samples is currently underway for further validation of the mNGS approach. Conclusion(s): Application of mNGS may detect pathogens that cannot be identified by the existing methods. However, careful interpretation of mNGS results is necessary to prevent reporting of false positive results.

Objective: Molecular methods enable more rapid and sensitive detection of herpes simplex virus (HSV) than viral culture. Three commercial molecular methods, all of which detect both HSV-1 and HSV-2, were compared to viral culture for the detection of HSV from swab specimens. Methods: Pediatric and adult patient viral swab specimens were cultured for HSV. Residual swab fluid was frozen at –80°C until tested with the 3 molecular methods: the Quidel Solana HSV 1+2/VZV Assay, the Focus Diagnostics Simplexa HSV 1 & 2 Direct Assay and the Luminex Aries HSV 1&2 Assay. A true positive was defined as positive by culture or positive by ≥ 2/3 molecular methods. Results: 177 specimens were studied. The sensitivity of culture was 81.3% (61/75, 95% CI 70.7-89.4%) and specificity was 100% (102/102, 95% CI 96.4-100%). The sensitivities of both the Solana and Simplexa were 100% (75/75, 95% CI 95.2-100%) and specificities were also both 100% (102/102, 95% CI 96.4-100%). The Aries had a sensitivity of 98.7% (74/75, 95% CI 92.8-99.97%) and specificity 99.0% (101/102, 95% CI 94.7-99.98%). All three molecular methods were significantly more sensitive than culture (p ≤ 0.0005 for Solana and Simplexa and p ≤ 0.0012 for Aries). Conclusion: All the molecular methods studied provided a significantly higher sensitivity than culture. In addition, the molecular methods took 1-2 hours to perform compared to a mean of 2 days for culture results. Use of any of the three molecular methods could lead to improved patient care. 

Background:  New technologies in the microbiology laboratory have the potential to reduce turn-around-time and improve accuracy, antimicrobial stewardship and patient outcomes. However, there is a paucity of data on their adoption by Canadian laboratories. The objective of our study was to investigate the patterns of practice of adopting new technologies amongst the Institute for Quality Management in Healthcare (IQMH) participant laboratories and summarize post-implementation quality indicators. Methods: In June 2017, a web-based patterns of practice qualitative survey on the adoption of novel technologies was conducted by the IQMH across all 73 microbiology laboratories that participate in the IQMH bacteriology proficiency testing program. Results: 69 of the 73 (94.5%) laboratories responded [8 university hospital (univ-hosp), 50 community hospital (comm-hosp) and 11 non-hospital (non-hosp)]. Only 50% of univ-hosp and 8% of comm-hosp reported having dedicated personnel for methods evaluation compared to 82% of non-hosp. Of all laboratories, 30% reported implementing MALDI-TOF (100% univ-hosp, 18% com-hosp, 36% non-hosp) with improved turn-around-time, time to appropriate treatment, and costs reported as benefits. Automated specimen processors were implemented by 14% (62.5% univ-hosp, 10% comm-hosp, 27% non-hosp), while only 4% had implemented total laboratory automation (12.5% univ-hosp, 2% comm-hosp, 9% non-hosp) with post-implementation reduction in turn-around-time, error rate, and costs cited. Syndromic (either respiratory, gastrointestinal, or meningoencephalitis) multiplex testing had been implemented by 7%  (25% univ-hosp, 6% comm-hosp, 0% non-hosp) with improved turn-around-time and diagnostic yield reported post-implementation. Conclusions: There is a wide range of adoption of new technologies among the different laboratory categories.  Post-implementation quality indicator data provided may be useful for peer laboratory business case development. The lack of dedicated methods evaluation personnel in comm-hosp and less so in univ-hosp compared to non-hosp laboratories is striking and may be partly responsible for the varying degree of uptake of novel technologies.

Objectives: Local confirmatory testing for carbapenemase producing organisms (CPO) is optimal when isolated from a patient population where such organisms are infrequently isolated. Presumptive reports of CPO place additional strain on health care facilities operating in overcapacity bed situations and create additional stress for Infection Control and Prevention programs and care providers. Methods: Twenty previously characterised isolates including Enterobacter cloacae (7), Escherichia coli (5), Klebsiella  pneumoniae (4), Klebsiella oxytoca (1), Klebsiella variicola (1), Enterobacter(Klebsiella) aerogenes (1), and Citrobacter freundii (1) were included . All isolates were previously molecularly confirmed by BCCDC for carbapenemases. To be eligible, each organism had to be resistant to at least 2 carbapenems by two methods (BD Phoenix and etest). All isolates were also tested using NG-Test CARBA 5. (NGBiotech, France). This kit confirms the presence of K-KPC, O-OXA48, V-VIM, I-IMP, N-NDM using a monoclonal based lateral flow assay in 15 minutes. Results:  Eighteen isolates correlated directly with the PCR result from BCCDC – eight positive for NDM, one positive for KPC, and nine that were negative for all carbapenemases. Regarding the two discrepant results, the Citrobacter freundii isolate tested weak positive for NDM by the NG-Test CARBA 5, but the result from BCCDC was negative. One strain of Enterobacter cloacae was difficult to emulsify and did not flood the column rendering the test invalid. Conclusions: Our health authority has a robust admission CPO screening program. Screening or clinical isolates that initially test as presumptive CPO trigger patient isolation and contact screening.  It is desirable to perform confirmatory testing promptly to focus infection prevention efforts appropriately. NG-Test CARBA demonstrated excellent sensitivity and specificity for the confirmation of 5 carbapenemase enzymes in organisms testing resistant to carbapenems.  

Objectives: ESBL-producing Enterobacteriaceae are an increasing threat. We have described the rise in ESBL-producing E. coli in the Toronto region.  Herein, we describe the associated multi-locus sequence types (MLST) and geo-temporal epidemiology using whole genomic sequencing (WGS) of bloodstream ESBL-producing E. coli. Methods: All adult inpatients from four tertiary-care hospitals in Toronto with bloodstream ESBL-producing E. coli from 2006 through 2012 were included (n=174). Corresponding isolates, one/patient/year, were recovered from -80^oC and genomic DNA was extracted using DNeasy Blood and Tissue Kit (Qiagen). WGS was completed on the NextSeq500 with 2x150 base paired-end reads using the Nextera XT DNA Library Prep Kit (Illumina).  Genomes were de novo assembled with SPAdes. MLST were determined using the Bacterial Analysis Pipeline with MLST 2.0.1 (Center for Genetic Epidemiology).   Linear trend analysis and geo-temporal mapping using the patient’s residential postal code was completed using Tableau. Results: There was a significant rise in the total number of bloodstream ESBL-producing E. coli between 2006 and 2012 (P=0.0005) with a corresponding rise in the proportion of ESBL-producing E. coli from 6.4% to 12.7%. This rise was driven by a significant rise in ST131 (P=0.0002) (Figure), with significant rises noted in three hospitals (P=0.004, 0.02, 0.01) and a non-significant trend in the fourth (P=0.08). The change in non-ST131 E. coli was not significant (P=0.09) (Figure). ST131 E. coli was first noted in patients residing in Brampton, a city in which 30.6% identify as having East Indian ethnicity (Figure), notable as the proportion of ESBL-producing E. coli has been reported as high as 70% in South East Asia. Conclusions: There was a significant rise in bloodstream ESBL-producing E. coli in the Toronto region between 2006 and 2012 led by the introduction and increase in ST131 E. coli. Further characterization of the factors associated with the endemicity of ST131 is ongoing.

Objective: Varicella-zoster virus (VZV) is the cause of chicken pox and shingles. Molecular detection is the diagnostic standard for detection of this important pathogen. Many laboratory-developed tests involve significant manual processing steps. For high volume testing laboratories, introduction of automated, random-access testing platforms offers potential time and cost savings. The purpose of this study was to compare an automated real-time PCR assay using analyte-specific reagents (ASR) for detection of VZV in swabs from lesions with clinical performance of a previously-validated lab-developed test (LDT) routinely used in the laboratory. Methods: Swabs collected in virus transport medium were tested using a PCR LDT and a novel PCR utilizing ASRs for VZV. ASR VZV primer probe recon solution (primers: 0.75 µM, probes: 0.5 µM, potassium chloride: 81.25 mM, magnesium chloride: 5 mM, Tris: 10 mM, water, topped with oil layer), enzyme cartridges (lyophilized enzymes, nucleotides, buffer), and extraction reagent packs were loaded onto a Panther Fusion instrument and amplified using the following thermal cycling conditions: 2 minutes at 95°C; 45 cycles of denaturation at 95°C for 8 seconds, and annealing/extension at 60°C for 25 seconds. MyAccess software was used for data analysis. Results: A novel VZV assay was successfully developed using ASR reagents, with full automation of extraction, amplification, and detection. Of 89 samples tested with both assays, 23 (25%) were positive for VZV by both methods. The overall agreement between the methods was 100%. Conclusions: Based upon preliminary results, the VZV PCR assay on the Panther Fusion instrument has similar performance characteristics to the LDT assay currently in use. An automated instrument with continuous random access offers potential for significant reductions in hands on- and turnaround times. However, further study will be required to confirm these findings and to evaluate potential workflow benefits.

Objective(s): The Aptima Herpes Simplex Virus (HSV) 1&2 Assay recently received Health Canada approved for detection and differentiation of HSV-1 and HSV-2 from anogenital sites. This assay uses target capture, transcription mediated amplification (TMA), and real-time detection of messenger RNA (mRNA) produced in host cells during HSV infection. To evaluate its performance, the Aptima assay was compared to another Health Canada approved assay, the BD ProbeTec HSV 1&2 Qx Amplified DNA Assay which used strand displacement amplification (SDA) technology. Methods:  As recommended by the manufacturers, the Aptima assay was performed on a Panther system, and the BD Probetec assay was performed on a Viper instrument. Analytical sensitivity and specificity were assessed using 10-fold serial dilution of viruses in viral transport media (VTM), and nucleic acids extracted concentrated from other viruses including all members of the Herpesviridae family. The clinical sensitivity and specificity were assessed prospectively using 158 swabs from oral and anogenital sites collected in VTM. Discrepant results were resolved with real-time PCR using the RealStar HSV 1-2 assay (Altona Diagnostics). Results: Both the Aptima and Viper assays showed excellent clinical and analytical specificity, without any false positive reactions. However, the Aptima HSV assay failed to detect HSV in specimens with low viral loads, resulting in reduced sensitivity for HSV-1 of 85.0% (34/40) and 95.8% (23/24) for HSV-2. The analytical analyses coincided with reduced sensitivity of approximately 10-fold for both HSV-1 and HSV-2 for Aptima compared to the Viper Probetec assay. Conclusions: This study demonstrated that detection of HSV mRNA using the Aptima HSV assay was less sensitive than HSV DNA detection through SDA technology on the Viper system. It is unclear whether this difference is attributed to the methodology itself, or limitations of mRNA-based detection for the diagnosis of infection with DNA viruses like HSV.  

Background: Cervical cancer screening relies on Pap cytology to detect precancerous lesions. A majority of women with abnormal cytology have either atypical squamous cells of undetermined significance (ASCUS) or low-grade squamous intraepithelial lesion (LSIL), and in most, these are not predictive of cancer risk. However, all such cases are followed with repeat cytology or colposcopy because some may have an underlying high-grade disease. ASCUS-HPV triage is recommended to better identify those at increased risk. In this regard, LSIL-HPV triage may also be helpful.  We assessed the usefulness of LSIL-HPV triage as part of an ongoing study investigating the application of CINtec PLUS (Roche), a dual-stain biomarker test, in LSIL triage. Methods: LSIL cases seen at the colposcopy clinic, Juravinski Hospital, Hamilton were prospectively enrolled with informed consent. Cervical specimens were collected at enrolment in ThinPrep for routine Pap. The remnant from the ThinPrep vials was used for HPV testing utilizing cobas 4800 assay (Roche). Biopsy confirmed cervical intraepithelial neoplasia grade 2 or worse (≥CIN2) served as the clinical endpoint. Results: Preliminary analysis was based on 347 patients (target, n=600). Ages ranged from 19-76 (median 33), with 204 (58.8%) >30 years of age. Of the 347, 188 (54.2%) tested HPV+, and 159 (45.8%) HPV-. There were 34 cases of ≥CIN2, and 33 had tested HPV+ (sensitivity, 97.1%). Of 313 without >CIN2, 158 tested HPV- (specificity, 50.5%; negative predictive value, 158/159 = 99.4%). Among the 188 testing HPV+, most were positive for high-risk oncongenic types other than 16 and 18 regardless of biopsy result. Conclusions: LSIL-HPV triage may have the potential to safely relegate half of women to routine screening with a very high negative predictive value, while maintaining superb sensitivity to detect ≥CIN2. Longitudinal studies could provide additional clinical data to assess the long term negative predictive value of LSIL-HPV triage.

Background: A Gram-negative coccobacillus (GNCB) unidentifiable by the sender using by MALDI-TOF/other methods was received at the Special Bacteriology Unit (SBU) on a Thursday in September 2018 for ‘urgent testing’.  The isolate had been recovered from pleural fluid of a 62y male residing in eastern Canada and hospitalized with empyema. We describe here significant biosafety issues which arose as a result of this case. Methods: SBU’s ‘urgent testing’ protocol dictated that the sender be immediately contacted for additional information. 16S rRNA gene sequencing was initiated directly from growth provided by the sender. Coincidently, subculturing of the sample in the BSC was observed by two new students. Assembly / BLASTing was done by usual NML methods. Results: The referring lab had been asked if the patient possibly had tularensis or brucella, based on their description of “tiny” GNCB.  These were thought to be highly unlikely due to zero prevalence provincially. On Friday pm, the bacterium was identified as Francisella tularensis by 16S sequencing. Materials were sealed and either autoclaved or turned over to the BADD unit for further analysis in CL3, which rapidly identified it as the RL3 agent, F. tularensis subsp tularensis. The sender was contacted with results.  SBU staff were interviewed by biosafety officers regarding handling practices (for a RL3 agent in a CL2). It was ultimately deemed that the isolate had been processed with minimal hazard to staff. Conclusions. Negligible prevalence of tularensis in that province and empyema / pleural fluid as source, contributed to having a minimal suspicion for a RL3 agent.  This referral created potential occupational hazardous incidents for staff at the NML and sender sites. F. tularensis must be characterized using advanced methods within a CL3 by staff with expertise. Labs must also be aware that ‘standard’ MALDI-TOF libraries cannot ID these taxa.

Objective: Candida auris is an emerging yeast that is associated with high rates of antifungal resistance and healthcare-associated outbreaks that can be difficult to control. Our objective was to carry out genomic characterization of all known C. auris cases in Canada to monitor the emergence of this species. Methods: Sixteen isolates of C. auris identified from 2012 - 2018 were submitted to the Canadian Mycology Reference Centre which was established this year by the National Microbiology Laboratory (NML). Isolates from all 16 cases were subjected to whole genome sequencing (WGS) on the Illumina Nextseq platform. Phylogenetic analysis based on single nucleotide variations (SNV) was carried out with the SNVPhyl v1.1 pipeline and assemblies were performed with SPAdes v1.6. Results: The isolates were obtained from axilla/groin (n=5), blood (n=4), ear (n=3), and other sites (n=4). All isolates of C. auris fell within the four known genomic lineages (clades) named after their apparent geographical origin. Isolates from the South Asian clade were identified in British Columbia (n=9) and Manitoba (n=1). C. auris was also identified in Ontario (South American clade; n=4), Quebec (South African clade; n=1), and Alberta (East Asian clade; n=1).  The four geographical clades were highly divergent (average 16,000 - 65,000 SNV differences between clades) but Canadian isolates within each clade appeared clonal. Canadian isolates from the South American clade differed by 14-26 SNVs while those from the South Asian clade differed by 1 - 128 SNVs. We identified mutations associated with fluconazole resistance (ERG11 Y143R) and voriconazole resistance (ERG11 Y132F) but not echinocandin resistance (FKS1 hotspot mutations). Conclusions: The 16 cases of C. auris in Canada represent all four known genomic lineages. Isolates tended to be clonal within each clade but high resolution WGS may be helpful in discriminating between patient transmission and separate introductions into a healthcare facility.

Objectives: Rapid detection of carbapenemase-producing organism (CPO) is essential for containment and patient management. We evaluated the ability of the NG-Test CARBA 5 immunochromatographic assay (NG Biotech, France) to detect five common carbapenemases using well-characterized (phenotypic and PCR/sequencing) Gram-negative bacilli (GNB). Methods: 297 GNB including 257 CPO (250 targeted-CPO: 127 KPC, 69 NDM, 32 OXA, 11 NDM+OXA, 8 VIM, 3 IMP; 8 non-targeted CPO (3 GES, 3 SME, 1 NMC), and 40 non-CPO were tested comprising 289 Enterobacteriacaeae (115 Klebsiella pneumoniae, 79 Escherichia coli, 51 Enterobacter cloacae, 44 other) and 8 non-Enterobactericeae GNB (2 Acinetobacter baumannii, 3 Pseudomonas species, 2 Aeromonas hydrophila, and 1 Shewanella putida).  Isolates were recovered from -80^oC under selective pressure (MacConkey with ertapenem disc) and plated to Oxoid MacConkey-cefpodoxime/MacConkey-meropenem CPO screening bi-plates. As directed, a fresh 18hx37^oC colony was collected with a loop and suspended in extraction buffer; 100μL was added to the sample well with results read at 15minx21^oC. The reader was blinded and discrepancies were repeated. Results: Of 297 tests, results were easy to interpret and all but three (2 OXA, 1 mucoid KPC) were available within 5 minutes. CARBA 5 initially identified 257/261 (98.5%; 96.0-99.5) targeted CPO-proteins [125/127 (98.4%; 94.1-99.9) KPC; 79/80 (98.7%; 92.6->99.9) KPC; 42/43 (97.7%; 86.8->99.9) OXA, 8/8 (100%; 62.8-100) VIM; 3/3 IMP (100%; 38.0-100%)].  Four (1.5%) targeted-CPO that were initially missed were positive on repeat testing suggesting too low an inoculum was possibly initially used; one had notable poor growth on initial testing. All 48 (100%; 91.1-100) non-targeted-CPO/non-CPO were negative. Final CPO-detection sensitivities/specificities were 100% for all targets; respective 95%CI were: KPC 96.5-100/96.6-100; NDM 94.5-100/97.5-100; OXA48-like 90.2-100/ 96.6-100; VIM 62.8-100/97.9-100; IMP 38.0-100/98.2-100. Conclusions: The NG-Test CARBA 5 was easy to use and provided highly-accurate (100% sensitive/specific) rapid detection for KPC, NDM, OXA48-like, VIM and IMP CPO. 

Objectives:  We aimed to determine the duration of CPE colonization to better inform providers, patients and infection control programs about prognosis. Methods:  Participants were recruited from population-based surveillance.  Eligible persons were colonized/infected with CPE, had stable housing, and were expected to survive and reside in the area for ≥1year. Participants were screened (groin & rectal swabs, and urine samples) at 1 and 3 months after identification, then quarterly until ≥3 negative sets of screening specimens were obtained. Specimens were incubated in BHI broth overnight then subjected to direct PCR to identify carbapenemase genes, with culture of PCR positive specimens. Decolonization was defined as occurring when 3 complete sets of swabs were PCR negative, and no later swabs were positive or clinical isolates identified. Time of decolonization was defined as date of first qualifying negative swabs. Results: 284 (76%) of 385 eligible persons participated: 87 completed follow-up, 113 are being followed, and 84 have incomplete data (30 died; 54 withdrew/were lost to follow-up). Median age is 70 years, 164 (58%) are male, 182 (64%) have at least one underlying comorbidity, and 99 (35%) initially had a clinical specimen (vs. screen only).  Most common organisms were E. coli (152,54%), Klebsiella spp. (92,32%) and Enterobacter spp. (25,9%); most common genes were bla_NDM  (±OXA) (164,58%), bla_KPC (24,8%), and bla_OXA-48-like (82,29%). The figure shows time to decolonization. Men (OR 0.53 95%CI 0.34,0.81), persons colonized/infected with Klebsiella spp. versus other bacteria (OR 0.50, 95%CI 0.24,1.02), those with clinical isolates (OR 0.49, 95%CI 0.29,0.81), and those with more sites positive at enrolment (OR for 2v1 site  0.30, 95%CI 0.14,0.67) were less likely to become decolonized. Conclusions: Most CPE colonized/infected persons appear to clear their organism over time, although about 1 in 5 remain colonized at 2 years. Individual characteristics significantly affect duration of colonization.

Background: Cardiac surgery is procedure on the valves or septum of heart. It does not include coronary artery bypass graft, surgery on vessels, heart transplantation, or pacemaker implantation. The objective of our study is to answer three questions: a) What is the risk of surgical site infection (SSI) after cardiac surgery? b) What are risk factors for SSI after cardiac surgery? c) What are the effects of antibiotic prophylaxis on SSI risk? Methods: Surveillance data based on NHSN/CDC protocols were collected during five years (2013-2017), from 7 hospitals at Belo Horizonte, Brazil. Outcome: SSI and total length of hospital stay. 23 independent variables were analyzed by univariate and multivariated methods. Results: A sample of 3,827 patients submitted to cardiac surgery was analyzed: SSI risk = 2.6% (I.C.95%=2.1%;3.2%). Hospital length of stay in non-infected patients (days): mean=19, median=13, std.dev.=21. Hospital stay in infected patients: mean=32, median=25, std.dev.=36 (p<0.001). Main risk factors for SSI: less than five surgical healthcare professionals at surgery (SSI=3.3%, RR=1.7, p=0.008); preoperative hospital length of stay more than four days (SSI=3.9%, RR=1.8, p=0.003); failure on prophylactic antibiotics (SSI=7.7%, RR=3.5, p<0.001). Patients with preoperative hospital length of stay more than four days that did not receive prophylactic antibiotics have 13.7% SSI (RR=4.4, p<0.001). Conclusion: If the length of preoperative hospital stay is less than or equal to four days, the risk of surgical infection in cardiac surgery will be reduced approximately 50%, from 3.9% to 2.1%. To prevent infections is crucial to maintain the length of preoperative hospital stay as short as possible and, most important, do not failure on antibiotic prophylaxis.