Student Poster (SP) Judging

Introduction: Although a vaccine-preventable disease, influenza causes approximately 3 to 5 million cases of severe illness and about 290 000 to 650 000 deaths worldwide, which occur primarily among people 65 years and older. Nonetheless, prevention of influenza and its complications rely mainly on vaccination. Objective:We aimed to systematically evaluate influenza vaccine effectiveness at reducing healthcare utilization in the elderly, defined as the reduction of outpatient visits, ILI and influenza hospitalizations, utilization of antibiotics and cardiovascular events by vaccination status during the influenza season. Methods:We searched MEDLINE, EMBASE, CINAHL, Cochrane Library and considered any seasonal influenza vaccine, excluding the pandemic (2009-10 season) vaccine. Reviewers independently assessed data extraction and quality assessment. Results: Of the 8,308 citations retrieved, 22 studies were included in the systematic review. Overall, two studies (9%) were deemed at moderate risk of bias, thirteen (59%) at serious risk of bias and seven (32%) at critical risk of bias. For outpatient visits, we found modest evidence of protection by the influenza vaccine. For all-cause hospitalization outcomes, we found a wide range of results, mostly deemed at serious risk of bias. The included studies suggested that the vaccine may protect the elderly against influenza hospitalizations and cardiovascular events. No article meeting our inclusion criteria explored the use of antibiotics and ILI hospitalizations. The high heterogeneity between studies hindered the aggregation of data into a meta-analysis. Conclusion: The variability between studies prevented us from drawing a clear conclusion on the effectiveness of the influenza vaccine on healthcare utilization in the elderly. Overall, the data suggests that the vaccine may result in a reduction of healthcare utilization in the elderly population. Further studies of higher quality are necessary.

Background: Corynebacterium diphtheriae can cause significant disease, including wound infections and bacteremia. In 2015, CLSI M-45 susceptibility breakpoints for Corynebacterium species and penicillin were lowered (≤1mg/mL to ≤0.12mg/mL). We reviewed the impact of updated breakpoints on C. diphtheriae susceptibility at our institution and investigated resistance mechanisms through whole-genome sequencing (WGS). Methods: A retrospective review was performed on C. diphtheriae clinical isolates recovered from March 2015 – June 2018. Susceptibility testing was conducted using Etest (bioMérieux, Marcy-l'Étoile, France) and interpreted using the CLSI M-45 2^nd (2010) and 3^rd (2015) editions. WGS was performed by next-generation sequencing (MiSeq, Illumina, San Diego, CA). MLST and antimicrobial resistance markers were analyzed from WGS utilizing 5 public databases (ARG-ANNOT, CARD, MEGARes, ResFinder and SRST2-ARG-ANNOT) using ARIBA. Results: 56 non-toxigenic C. diphtheriae isolates were identified: blood (1), throat (1), and wound (54). MICs were available for 39 patients, and 48 isolates were available for WGS. Using 2010 breakpoints, all isolates (39/39) were considered penicillin susceptible, but all were reported as non-susceptible (intermediate) using 2015 breakpoints. Distribution of penicillin MICs did not change over time. One isolate was resistant to erythromycin and clindamycin, with interpretations unchanged between 2010 and 2015. WGS identified a predominant strain: ST-76 (45/48). Other ST types included ST-05 (1/48), ST-32 (1/48) and 1 novel ST (most similar to ST-444/ST-442/ST-441). No mutations associated with beta-lactam resistance were identified for samples with sufficient sequencing depth (n=37). Conclusions: Application of 2015 CLSI interpretive criteria resulted in all C. diphtheriae isolates recovered at our institution to be classified as penicillin non-susceptible; conversely, 2010 CLSI criteria would have classified all isolates as penicillin susceptible. WGS did not reveal any molecular basis for penicillin non-susceptibility. Further investigation is required to understand the generalizability of these findings to other strains of C. diphtheriae, and more broadly, the potential clinical significance of penicillin susceptibility misclassification.

Objective: Calgary Laboratory Services (CLS) performs MALDI and VITEK directly from positive blood cultures for organism identification (ID) and antimicrobial susceptibility testing (AST). Our objective was to compare the performance of direct MALDI/VITEK to a commercial blood culture ID/AST platform, Accelerate PhenoTest™ BC Kits (AXDX), in the ID/AST of clinical blood cultures positive for Gram-negative bacilli (GNB) and in blood cultures inoculated with multi-drug resistant GNB (MDR GNB). Methods: Blood cultures positive for GNB were collected at CLS and tested using AXDX, direct MALDI/VITEK, and compared to conventional methods (plate incubation followed by MALDI and VITEK). A subset of sterile blood cultures were inoculated with MDR GNB. Discrepancies in very major errors (VME) and major errors (ME) were confirmed with microbroth dilution. Results: Twenty-eight clinical samples and 30 inoculated samples were analyzed. In the clinical samples, direct MALDI had higher ID failures (31.0%) compared to AXDX (3.4%). Time to ID was 1.5 hours, 6.1 hours, and 22.2 hours for AXDX, direct MALDI and conventional methods, respectively (p<0.001). Time to AST was 6.6 hours, 16.8 hours, and 33.4 hours for AXDX, direct MALDI and conventional methods, respectively (p<0.001). In the clinical samples, AXDX had EA/CA >95%, 0 VME/ME, and 3.7% minE. Direct VITEK had EA/CA >99%, 0 VME/ME, and 0.65% minE. In the inoculated samples, AXDX had EA, CA, VME, ME and minE of 85.6%, 87.9%, 1.6%, 1.4%, and 10.4%, respectively. Direct VITEK had EA, CA, VME, ME and minE of 97.6%, 94.8%, 0.8%, 0%, and 4.6%, respectively. Conclusions: Direct MALDI/VITEK and AXDX performed well on clinical samples but direct MALDI/VITEK outperformed AXDX when challenged with MDR GNB. AXDX had fewer ID failures and faster results, though its results were not based on real-world settings as was the direct MALDI/VITEK.

Objective: Persons who inject drugs (PWID) being treated for infective endocarditis (IE) remain at risk of bloodstream infections (BSIs) due to ongoing intravenous drug use, often through central catheters inserted for prolonged antimicrobial treatment. We sought to characterize BSIs in this patient population and determine the clinical factors associated with their development. Methods: We conducted a nested case-control study of episodes of definite IE based on the modified Duke criteria in PWID ≥18 years of age, admitted to tertiary care centres in London, Ontario from March 1 2007 to March 31 2018. We identified and characterized cases of new BSIs among this population, and compared them against episodes of IE without new BSIs. We also compared the incidence of inpatient versus outpatient BSIs. Results: There were 424 episodes of IE among PWID, and 81 (19.1%) were complicated by BSIs. There were 138 BSIs with 280 unique isolates, of which 156 (55.7%) were gram negative bacilli, 75 (26.8%) were fungi and 49 (17.5%) were gram positive cocci. The most common bacteria included ESKAPE organisms associated with nosocomial antibiotic resistance. Factors associated with BSIs included previous IE, right-sided IE, opiate and polysubstance use, ongoing inpatient drug abuse and peripherally inserted central catheter placement. BSIs were more commonly identified in PWID receiving inpatient treatment (9.60 BSIs per 1000 days of intravenous access, 95% CI 7.95–11.5) than outpatient treatment (5.23 BSIs per 1000 days of intravenous access, 95% CI 3.50–7.46). Conclusion: BSIs are a common complication in PWID being treated for IE with parenteral antimicrobials, and empiric therapy should cover resistant gram negative bacteria and fungi. Although detection bias could have decreased outpatient BSI rates, carefully selected PWID with IE may be safely treated with outpatient parenteral antimicrobial therapy, reducing the harms and costs of extended hospitalization.

Objectives: Detection of group A streptococcus (GAS) in pharyngeal swab specimens is important to help prevent acute rheumatic fever (ARF) as well as local suppurative complications. Molecular methods are becoming increasingly popular for GAS detection, as they are considerably faster than culture and have greater sensitivity than rapid antigen detection tests. We therefore compared three commercially available GAS nucleic acid amplification tests with bacterial culture. Methods: The three molecular methods assessed were the Quidel Solana GAS assay, the Luminex Aries Group A Strep assay and the Focus Diagnostics Simplexa Group A Strep Direct assay.  We defined a true positive result as one positive by culture or positive by ≥ 2/3 molecular methods. Samples collected were set up for routine bacterial culture and the remainder of the swab fluid (BD E-swab) was frozen at–80°C until molecular testing was performed. Results: 286 throat swabs (206 children, 80 adults) were collected from patients with suspected pharyngitis. The sensitivity of culture was 84.8% (95% CI 77.7-90.3%) with a specificity of 100% (95% CI 97.5-100%). Culture was significantly less sensitive than the true positive definition based on the molecular assays (p = 0.0001). The sensitivity of the Solana assay was 94.2% (95% CI 88.9-97.5%) and the specificity was 98.7% (95% CI 95.2-99.8%). Simplexa assay sensitivity was 99.3% (95% CI 96.0-99.9%) and the specificity was 95.3% (95% CI 90.6-98.1%).  For the Aries assay, the sensitivity was 96.4% (95% CI 91.8-98.8%) and the specificity was 98.0% (95% CI 94.2-99.6%). For a single specimen, the Solana assay took approximately 40 minutes to complete, the Simplexa approximately 1.25 hours, and the Aries approximately 2 hours. Conclusions: All three commercial methods were more sensitive and also much more rapid than culture for GAS detection from throat swabs.

Objectives: In the case of antibiotic-associated hemorrhagic diarrhea that is not accounted for by common enteric pathogens, Klebsiella oxytoca is often an overlooked cause. This is a Canadian case presentation of antibiotic-associated hemorrhagic diarrhea caused by K. oxytoca, with a review of the literature providing clearer indications for K. oxytoca testing. Case presentation: A 57-year old woman was hospitalized for with new-onset bloody diarrhea (up to 25 episodes daily) and mild lower abdominal pain following a three-day course of amoxicillin for gingivitis. Previously, the patient had no significant past medical history and was taking no medications. Throughout admission, the patient was afebrile and her vitals were stable. Laboratory test results demonstrated mildly elevated leucocytes (12.0*10⁹/L) and her abdominal X-ray was negative for signs of thumb-printing or obstruction. The initial stool culture and C. difficile toxin assays were negative; however, repeat stool cultures for K. oxytoca were positive. The patient was treated with metronidazole and improved gradually. She was discharged after four days with follow-up colonoscopy. Conclusions: Our case report describes a 57-year old woman with antibiotic-associated hemorrhagic diarrhea secondary to K. oxytoca infection. Though a rare cause of hemorrhagic diarrhea, K. oxytoca needs to be considered when traditional stool cultures and C. difficile toxin assays are negative, especially given a history of recent antibiotic use with amoxicillin or other penicillin derivatives. 

Objectives:Bacteremia is associated with significant morbidity and mortality, particularly given increasing antimicrobial resistance. This Canadian ward surveillance study, CANWARD, assesses (1) the epidemiologic profile of blood culture pathogens and (2) their antimicrobial susceptibility profiles. Methods: Between 2007 and 2016, each participating Canadian tertiary care centre submitted 100 consecutive, clinically-significant isolates of aerobic and facultative bacteria from positive blood cultures. Susceptibility testing was performed by CLSI broth microdilution method. Results: 18686 blood isolates were submitted, of which 57.3% were from males. When categorized by location, 35.9% were from emergency rooms, 33.6% from medical units, 15.9% from intensive care units, 7.8% from surgical units and 6.8% from clinics. The most common pathogens were Escherichia coli (23.2%), Staphylococcus aureus (17.7%) [methicillin-susceptible (MSSA), 13.9% and methicillin-resistant (MRSA), 3.8%], Klebsiella pneumoniae (7.4%), Streptococcus pneumoniae (4.8%), Enterococcus faecalis (4.2%), Pseudomonas aeruginosa (3.8%) andmethicillin-resistant Staphylococcus aureus (MRSA) (3.8%). Of the 3297 S. aureus isolates, 21.3% were MRSA with 32.5% possessing community-acquired genotypes and 63.3% possessing healthcare-acquired genotypes. MRSA rates have decreased over 10 years (P<0.0001), while rates of vancomycin-resistant enterococci (VRE), extended-spectrum beta lactamase-producing (ESBL) E. coli and ESBL K. pneumoniae have increased (P<0.007, <0.0001 and <0.0001 respectively, Cochran-Armitage test for trend). The most active agents against MRSA were ceftobiprole (100% susceptible) linezolid (100%), vancomycin (99.6%) and daptomycin (99.7%). The most active agents against P. aeruginosa were ceftolozane-tazobactam (99.3%), tobramycin (96.6%) and colistin (94.3%). Piperacillin-tazobactam, meropenem and ciprofloxacin susceptibility rates to Pseudomonas aeruginosawere 90.0%, 84.7% and 84.8% respectively. Conclusions: E. coli, S. aureus and K. pneumoniaeare consistently the top pathogens isolated from blood culture specimens. Rates of MRSA decreased from 2007 to 2016, while rates of ESBL E. coliand ESBLK. pneumoniae increased. Vancomycin, ceftobiprole, daptomycin and linezolid are the most active agents against Gram-positive cocci. Effective agents against Gram-negative bacilli varied depending on speciation, however carbapenems and piperacillin-tazobactam demonstrated consistent activity.

Objective: Fluoroquinolones (FQNs) are broad-spectrum antibiotics with diverse indications. In January 2017, Health Canada warned that FQNs should be used with caution given risks for significant adverse effects including tendinopathy, peripheral neuropathy, and neuropsychiatric disorders. Prospective audit and feedback (PAF) is an effective tool to reduce inappropriate use of antimicrobials. The objective was to study the impact of PAF on inpatient FQN prescriptions. Methods: A multi-center pre-post quasi-experimental design was used to compare pre-intervention (June to August 2017) and intervention (September 2017 to February 2018) inpatient FQN use. Chart reviews were conducted on all patients prescribed FQNs to gather patient demographics, indication for and appropriateness of prescriptions. The intervention consisted of PAF on all FQN prescriptions, with written and verbal feedback from the Antimicrobial Stewardship team to optimize prescribing. The primary outcome was quantity of FQN use and appropriateness. Results: 1107 patients (pre-intervention = 425, intervention = 682) were evaluated. PAF resulted in an overall reduction from 4.14 to 2.9 days of therapy/100 patient days and 5.36 to 4.4 defined daily doses/100 patient days (p<0.001). Ciprofloxacin was used primarily for genitourinary or intra-abdominal infections and accounted for two-thirds of FQN prescriptions. Following the intervention, ciprofloxacin use for non-specific symptoms attributed to UTI decreased from 60% to 46% (p=0.049). The primary indication for Levofloxacin was respiratory tract infections with a reduction in median days of therapy from 5 to 4 (p=0.009) following the intervention. The appropriateness of all FQN prescriptions increased from 68% to 88% post-intervention (p<0.001). Fewer patients were prescribed unnecessary therapy (p<0.001) or identified to be at high risk for a FQN-associated adverse event (p<0.001). Conclusion: PAF can significantly decrease overall inpatient FQN use and improve appropriateness of prescribing, making this a useful tool in light of increasing concerns for FQN associated adverse events.

Background: IL-7 is a key cytokine in CD8⁺ T-cell survival and proliferation. The IL-7 receptor, CD127, is expressed both membrane-bound (mCD127) and soluble (sCD127). In tuberculosis and sepsis, the level of sCD127 in the plasma has been correlated with disease progression. Two factors are known to affect sCD127 levels. IL-7 induces sCD127 secretion from CD8⁺ cells, while C allele rs6897932 is correlated with higher sCD127 levels. We hypothesized that rs687932 C- allele is associated with higher sCD127 secretion in response to IL-7. Correspondingly, the response to IL-7 of CD8⁺ T-cells from the genotyped donors was measured. Methods: CD8⁺ cells from the peripheral blood of healthy donors were stimulated with IL-7.   mCD127 and sCD127 were measured by flow cytometry, and ELISA respectively.  The genotype of donors was determined with a PCR. Results: Of 11 donors, 9 were CC, 3 were CT and 0 were TT genotype. As shown previously, the percentage of CD8⁺ cells expressing mCD127 decreased over time of IL-7 stimulation. In the individuals with the CT genotype, mCD127 levels returned to baseline levels at 72hr of stimulation with 10 ng/mL IL-7. No difference in mCD127 between genotype was detected at 8hr, 24hr or 48hr with either 10ng/mL or 1ng/mL IL-7. As expected, sCD127 secretion increased with time of IL-7 stimulation. No difference between genotypes was seen at any time point with 10ng/mL IL-7. Interestingly, with 1ng/mL IL-7, sCD127 secretion was lower in CT donors compared to CC donors [CC: 2030 pg/mL 73 vs CT: 1188 pg/mL 75; mean SEM ].     Conclusion:  The results suggest the presence of a host factor that may predict the response to IL-7 therapy, in the setting of, TB, sepsis or other diseases where sCD127 plays a role.

Background: Inappropriate antimicrobial use (AU) is recognized as a leading cause of antimicrobial resistance. However, quantifying AU in hospitals is challenging due to variability in information systems. Point prevalence surveys (PPS) provide a means to quantify AU in a cross-sectional manner within and between institutions. The aim of this study was to describe and compare the patterns of AU across Canadian pediatric hospitals. Methods: A PPS of AU for all inpatients (excluding mental health and mother-newborn units) was conducted between November 6-13, 2018 in 13 Canadian pediatric hospitals. Data was entered on a REDCap database including co-morbid conditions, admitting service, antimicrobial(s), reasons for AU and pathogen(s) identified. Results: In total, 1493 patients-days were surveyed. The mean proportion of children receiving at least one antimicrobial was 497/1493 (33.3%) [range 21.4% to 43.6%]. Of 757 antimicrobials prescribed, the three most common were aminopenicillins (15.5%; 117), third generation cephalosporins (12.8%; 97) and antipseudomonal penicillins (10.9%; 83). The proportion used as targeted, empiric or prophylactic therapy was 25% (189), 53.1% (402) and 21.1% (160), respectively. The frequency of carbapenems and vancomycin use was 3.8% (29) and 7.4% (56), respectively. Of the antimicrobials used for targeted or empiric therapy (n = 591), 131 (22.2%) were for pneumonia, 106 (17.9%) for abdominal infections and 111 (18.8%) for fever ‘without source’. Ampicillin was used as the empiric treatment of community-acquired pneumonia (CAP) in 34.1% (14/41) of cases. Conclusions: Approximately 1/3 of children hospitalized in Canadian pediatric hospitals were prescribed at least one antimicrobial in our study. Carbapenems were prescribed infrequently. However, ampicillin was prescribed only for 1/3 of empirically treated CAP. More detailed analysis of the rationale for AU (e.g. lack of adherence to CAP guidelines) is required to fully understand antimicrobial prescribing in pediatric hospitals in order to prioritize measures aimed at optimizing AU.

Objective:  Heparin is often used as a lock solution for central venous catheters (CVC), but the incidence of catheter-related complications remains high. The purpose of this systematic review was to summarize findings about locks, in particular in the pediatric population, and to compare their characteristics to identify which one should be used in our hospital. Method: We searched the PubMed database from January 1^st, 2007 to December 31^st, 2018 for clinical trials, observational studies and review articles discussing prophylactic lock solutions for CVCs. We collected data relative to study type, study time, study population, catheter type, lock solution types, outcomes and biases of the clinical trials. The outcomes recorded were catheter-related infections (CRI), adverse events and catheter-related thrombosis, occlusions and dysfunctions. Results: We identified 56 clinical trials, 6 observational studies and 23 reviews. Ethanol seemed more effective than heparin in preventing CRIs but was associated with a high risk of catheter malfunction. Taurolidine-citrate-heparin and antibiotics (mostly gentamicin) coupled with heparin seemed to be more effective than heparin alone in preventing CRIs and seemed to have similar antithrombotic effects. No antimicrobial resistance was reported. The 0.9% saline locks were similar to heparin in preventing infections and caused a higher rate of catheter thrombosis or occlusions. Other locks such as 4% tetra-sodium EDTA, minocycline-EDTA and tinzaparin were researched, but the limited number of studies and patients included did not allow for any conclusions to be drawn. Conclusion: Given the risk of antimicrobial resistance with long-term use of antibiotic locks, taurolidine-citrate based locks seemed to be the most effective lock solution. However, most studies included a small population size, decreasing the ability to draw conclusions. Larger scale, multicenter randomized controlled trials are needed.

Objective: To provide updated evidence for carbapenemase-producing organisms (CPO) contact and unit screening practices to inform the revision of Alberta Health Services (AHS) provincial guidelines. Methods: Literature searches were performed in the PubMed, EMBASE, Google Scholar, and Cochrane Library electronic databases. A grey literature search was also done to determine current practices, guidelines, and recommendations. The key search words were carbapenemase-producing organisms, screening, criteria, close, contacts, unit, and ward. Results: Review of 15 studies revealed that the most common definitions for contacts were shared room or unit (10), shared healthcare provider (6), and household/environmental exposure to a previous or current CPO-positive case (4). Other risk factors were duration of exposure, geographic proximity, concomitant infection, antimicrobial therapy, invasive procedure(s), and mechanical ventilation. A scoring system has been proposed based on these definitions and risk factors to support the risk assessment and decision-making for screening of CPO contacts. Twenty studies regarding unit screening for CPO were reviewed. The two major criteria for unit screening were the presence of a CPO outbreak on the unit, and the presence of a known CPO-positive patient on a high-risk unit (e.g., ICU). Thirteen of the examined studies recommended weekly unit screening; however, studies showed considerable variation and heterogeneity in the criteria to define a unit as free of CPO transmission. Most commonly, either two or three unit-wide negative results were recommended.  Conclusions: Screening of close and unit contacts of CPO-positive patients relies on an infection prevention and control risk assessment; however, criteria vary across jurisdictions.  Further characterization of risk factors, and development of risk scoring systems, will aid in the decision-making process. Further study is required to determine the number of screens required to conclude a CPO outbreak on a unit.

Objective: Following a poultry industry ban on preventative use of ceftiofur in 2014, increased use of gentamicin and lincomycin-spectinomycin was reported. Simultaneous increases in gentamicin resistance (gen-R) rates in Salmonella Heidelberg from human and chicken sources were also observed. Our objective was to carry out a One Health genomic investigation of gen-R in human and chicken isolates of S. Heidelberg to determine potential transmission. Methods: The Canadian Integrated Program for Antimicrobial Resistance Surveillance (CIPARS) collected Salmonella isolates from provincial public health laboratories, as well as broiler chicken farms, abattoirs, retail stores and other sources across Canada.  Antimicrobial susceptibility testing was carried out by broth microdilution using the Sensititre™ Complete Automated System (ThermoFisher Scientific) and whole genome sequencing was carried out by Nextseq™ (Illumina). Phylogenetic analyses were carried out with SNVPhyl pipeline v1.1, genomes were assembled with SPAdes v0.5 and resistance genes were identified with the staramr v0.3 tool. Results: In human isolates of S. Heidelberg (n=5974), the proportion of gen-R increased 3-fold from 2.3% in 2003-2015 to 6.8% in 2016-2017, while in chicken isolates (n=3510) from core CIPARS, CIPARS targeted studies, FoodNet Canada, and other monitoring programs, the proportion of gen-R increased 6-fold from 1.8% in 2003-2015 to 11.9% in 2016-2017. Amongst gen-R S. Heidelberg isolates from humans (n=66) and chickens (n=27) collected between 2014-2017, three gen-R resistance genes were found in both sources: aac(3)-VIa [81.8% of human isolates and 86.2% of chicken isolates], aac(3)-IId [16.7% human and 3.4% chicken], and ant(2'')-Ia [1.5% human and 3.4% chicken], while aac(6')-Ib3 was identified in chickens only (6.9%). Phylogenomic analyses showed several related clusters of human and chicken isolates whose genomes differed by only 1-30 nucleotides. Conclusion: Gen-R was most commonly attributable to aac(3)-VIa and isolates from both humans and chickens were closely related suggesting potential transmission. 

Objectives: Daptomycin is approved for skin and soft tissue infections due to Staphylococcus and selected Streptococcus spp., and S. aureus bacteremia with or without right-sided, native valve infective endocarditis. Since 2015, daptomycin usage has increased at our institution. The study objectives were to describe prescribing patterns of daptomycin, and to identify the rate and predictors of clinical failure overall and in the vancomycin-resistant Enterococcus (VRE) subgroup. Methods: This was a retrospective cohort study of adults prescribed daptomycin for >48 hours (without exposure in the preceding 3 months) between April 2015 and March 2017. Primary outcomes were the patient, infection, and treatment factors associated with daptomycin prescribing. Secondary outcomes were rate and predictors of clinical failure (a composite of in-hospital mortality, daptomycin discontinuation due to toxicity or lack of efficacy, and readmission for or retreatment of the index infection). Results: Among the 81 patients enrolled, the median age was 60 years and 42.0% of patients had a Charlson Comorbidity Index of ≥5. Of those with bacteremia, 18.4% had a Pitt Bacteremia Score of ≥4 and 52.6% had bacteremia for >4 days. Daptomycin was prescribed off-label in 88.9% of cases. VRE was isolated in 50.6% of patients. The median daptomycin dose was 6.0 mg/kg overall and in the VRE subgroup. The rate of clinical failure was 30/81 (37.0%) overall with 13 deaths, and 19/41 (46.3%) in the VRE subgroup with 9 deaths. Bacteremia for >4 days was associated with an increased risk of clinical failure (OR 6.36, 95%CI 1.34-30.17, p=0.02). Conclusions: Daptomycin was largely prescribed off-label for VRE. The median dose of 6.0 mg/kg was consistent with manufacturer recommendations for S. aureus infections. The rate of clinical failure in the VRE subgroup exceeded that of the overall population, suggesting antimicrobial stewardship opportunities to evaluate formulary restrictions, dosing, and indications for daptomycin.

Objectives: Invasive pneumococcal disease (IPD) is a significant source of morbidity and mortality, particularly in children and adults ≥65 years. This study identified changes in serotype distribution and antimicrobial susceptibilities of IPD isolates collected from Canadian adults ≥65 years. Methods: The SAVE study (a collaboration between CARA and NML) collected 9166 IPD isolates from 2011-2017, including 3529 obtained from adults ≥65 years. Serotyping was performed by the Quellung reaction and antimicrobial susceptibility testing was performed using CLSI methods. Multidrug/extensive-drug resistance (MDR/XDR) was defined as resistance to ≥3/≥5 antimicrobial classes, respectively. The Cochran-Armitage trend test was utilized to calculate the significance of changes in susceptibility rates and serotype distribution over time. Results: Overall, the most common serotypes identified within the ≥65-year age group were 22F (11.0%), 3 (8.8%), 19A (7.1%), 7F (5.6%) and 15A (5.4%). The proportion of serotypes 7C, 8, 9N, 19F, 23F, 24F, 31 and 38 demonstrated increasing trends over time, while 6C, 7F and 19A demonstrated decreasing trends (P<0.05). Of note, the trends for vaccine serotypes 7F, 19A and 19F were also identified in patients aged <18 and 18-64 years while the increasing proportion of vaccine serotype 23F was only seen in the ≥65-year age group. Overall susceptibility rates <90% were noted for clarithromycin (75.0%), doxycycline (88.7%) and trimethoprim-sulfamethoxazole (88.5%). No individual antimicrobial demonstrated a significant susceptibility trend from 2011-2017, although clarithromycin susceptibility increased and trimethoprim-sulfamethoxazole susceptibility decreased over time (P=NS). The prevalence of both MDR and XDR IPD decreased over time, although only the XDR trend was statistically significant (P=0.039). Conclusion: Within the ≥65-year age group, key vaccine serotypes decreased in prevalence from 2011-2017, accompanied by a decrease in MDR/XDR isolates. However, the prevalence of other vaccine/non-vaccine serotypes have increased, warranting continued surveillance of IPD in this crucial age group. 

Objectives: FMT is an effective treatment in patients with recurrent rCDI. Less is known regarding how the gut microbiota is influenced by bowel lavage prior to FMT and what impact multiple FMTs may have on the gut microbiota. Methods: Nine rCDI patients undergoing FMT by enema were included in this study. Rectal ESwabs were obtained immediately before and after PICO-SALAX®. Patients received 2-3 FMTs over a week following PICO-SALAX® administration. Rectal ESwabs were obtained prior to the second and third as well as at 1-month following FMT. Rectal ESwabs were stored neat in -80°C. The MO BIO PowerSoil® DNA Isolation Kit was used to isolate gDNA; 16s rRNA gene amplicon sequencing (V4 hypervariable region) was performed on the Illumina MiSeq. Low abundant OTUs (<0.001%) were excluded from the dataset. Microbiota measures were compared using Wilcoxon signed-rank test with Benjamini-Hochberg false discovery rate correction. Results: There were no significant differences in proportion of phyla or genera in the pre- versus post-PICO-SALAX® microbiota but there were significant decreases in richness (ACE, Chao1) and number of observed taxa and increases in the diversity (Simpson and Shannon). Significant differences in the patient’s baseline microbiota (increases in Bacteroidetes and Proteobacteria but decreases in Actinobacteria and Fusobacteria) were observed after three FMT at the 1-month follow-up but not after the first nor second FMT. While changes in alpha diversity were observed post-FMT, none were considered significant. Conclusions: The microbiota was only significantly changed consistently at the 1-month follow-up suggesting the need for multiple FMT when delivering FMT by enema. Bowel lavage with PICO-SALAX® significantly reduced the number of observed taxa and modified diversity indices but did not otherwise have a substantive impact on the microbiota. Whether bowel lavage is needed to improve uptake of donor microbiota requires further evaluation.

Background: The study aim was to assess the coverage of invasive Streptococcus pneumoniae (SPN) by PCV-10 and -13 across Canadian geographic regions. Methods: As part of a collaboration between CARA and the NML, SPN isolates were collected from jurisdictions across Canada. Serotyping was performed by the Quellung reaction and susceptibility testing was performed in accordance with CLSI methods. Multidrug resistance (MDR) was defined as resistance to ≥3 distinct antimicrobial classes. Results: A total of 9166 isolates were collected as part of the SAVE 2011-17 study. Of the total, 66.6% (n=6110), 21.7% (n=1985) and 11.7% (n=1071) of isolates were obtained from Central, Western, and Eastern regions, respectively.  Nationally, PCV-13 provided significantly greater coverage (31.7%, n=2905) than PCV-10 (14.3%, n=1309, P=<0.0001). By region, PCV-13 delivered 33.7% (n=2060), 23% (n=456) and 36.3% (n=389) coverage for Central, Western, and Eastern regions, respectively. For PCV-10, coverage was significantly reduced (P=<0.0001) at 15.4% (n=939), 10.1% (n=201) and 15.8% (n=169) for Central, Western, and Eastern regions, respectively. The most common PCV-10,-13 serotype was 7F, accounting for 8.2% (750/9166, 4^th most common) of isolates; however, a significant decreasing trend (P=<0.0001) was observed. The most common PCV-13 serotypes were 3 and 19A, representing 8.5% (781/9166, 2^nd most common) and 8.3% (757/9166, 3^rd most common) of isolates, respectively. Serotype 19A demonstrated a significant decreasing trend (P=<0.001), whereas serotype 3 demonstrated a significant increasing trend (P=0.008). PCV-13 provided significantly greater national coverage (53.9%, 280/519) of MDR isolates than PCV-10 (14.8%, 77/519, P=<0.0001). Regional coverages of MDR isolates by PCV-10 and (PCV-13) were: 14.6% (46.4%), 17.2% (73.7%) and 13% (62.3%) for Central, Western, and Eastern regions, respectively. Conclusion: Overall, PCV-13 provided significantly greater national and regional coverage of invasive SPN compared to PCV-10. Notably, PCV-13 provided significantly greater coverage of MDR SPN.

Objective: We encountered an 85-year-old lady with infective endocarditis possibly due to inadequate duration of therapy for bacteremia. In 2017, she was found to have Staphylococcus lugdunensis bacteremia thought to be secondary to a tunneled dialysis catheter. The catheter was exchanged over a guidewire and she received cefazolin for 16 days. Ten months later, blood cultures again grew S. lugdunensis. A transthoracic echocardiogram showed presence of a vegetation. Through molecular and matrix assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS) typing methods, we aimed to establish whether she had the same strain of S. lugdunensis during both admissions. Methods: Three of patient’s S. lugdunensis isolates as well as ten random S. lugdunensis clinical isolates from Eastern Ontario from 2017-2018 were compared. MALDI-TOF MS was conducted on Bruker Microflex LT benchtop instrument operated by FlexControl software. Mass spectrum analysis was performed using the BioTyper 2.0.  A gel view representation of reference strain spectra was visually examined to identify peaks with variable occurrence among strains. Hierarchical clustering of these profiles was performed using the online software DendroUPGMA. The results were internally validated by comparison with polymerase chain reaction (PCR) primed with OPA-18, OPA-2, RAPD1, ERIC1, and ERIC2. The PCR products were electrophoresed on agarose gel.  Hierarchical clustering of DNA banding patterns of the PCR was analyzed using the software PyElph. Results: Results from both typing methods showed that all three patients’ S. lugdunensis isolates were closely related to one another. However, these samples showed similarities with two of the ten random S. lugdunensis isolates. Conclusion: This is a preliminary indication that these closely related isolates are likely identical strains. We need to investigate the possibility of prevalent clone in our region by repeating the study on all S. lugdunensis isolates collected in 2014-2018.

Background: Patients with cystic fibrosis (CF) can become chronically infected with species within the Burkholderia cepacia complex (BCC), leading to worsening lung function and quality of life and increased mortality. One member of BCC, B. cenocepacia causes a severe decline in lung function, which can develop into a life-threatening systemic infection known as ‘cepacia syndrome’. Reduction in the prevalence of B. cenocepacia has been achieved through social distancing of CF patients and implementation of stringent infection control measures. The identification of B. cenocepacia is currently performed by molecular methods at a reference laboratory, with a 7-day turn-around-time. In this study, we evaluated the utility of matrix-assisted laser desorption/ionisation-time of flight mass spectrometry (MALDI-TOF MS) for the rapid identification of B. cenocepacia in comparison with molecular identification. Methods: 65 previously characterised B. cenocepacia isolates and 12 isolates belonging to other BCC species were retrieved from -80^oC stock and passaged twice on blood agar plates. MALDI-TOF MS was undertaken directly from isolated single colonies using a Bruker Microflex LT instrument (Billerica, MA) and spectra analysed using Bruker Biotyper with Compass version 4.1.80 (Billerica, MA). Each isolate was spotted in triplicate. Species-level identification was accepted if the score was ≥2.00. Results: 61/65 isolates were concordant and produced a score ≥2.00. 3/65 isolates were discordant (in only one replicate spot). None of the 12 non-cenocepacia isolates was mis-identified as B. cenocepacia, resulting in an overall accuracy of 96%. Conclusions: MALDI-TOF MS showed excellent concordance with reference methods in the identification of B. cenocepacia. While preliminary, this work suggests a potential use of MALDI-TOF MS for rapid identification of B. cenocepacia. Earlier identification of B. cenocepacia will ensure rapid implementation of infection control measures and may prevent additional transmission events.  

Background: In light of the increasing need to detect carbapenemase producing Enterobacteriaceae (CPE) in clinical laboratories, EUCAST recently revised its carbapenem zone diameter breakpoints. Since October 2018, our laboratory discontinued screening by ertapenem disk diffusion and adjusted its meropenem zone diameter screening breakpoint to reflect this change. A retrospective analysis was conducted to estimate the effect these updated breakpoints would have had on CPE detection. Methods: From September 2015 to September 2018, our laboratory screened isolates for CPE based on the 2013 EUCAST zone diameter screening breakpoint of <25mm for meropenem or ertapenem (10µg disks). Isolates which screened positive were subjected to confirmatory phenotypic and molecular testing. For this study, we re-evaluated these previously screened positive isolates with the updated 2017 EUCAST meropenem screening breakpoint of <28mm, when accompanied by a temocillin 30µg disk zone diameter <11mm, and assessed the impact of discontinuing ertapenem screening on CPE detection. Results: 877 potential CPE isolates were detected from 417 (48%) screening and 460 (52%) clinical specimens using 2013 EUCAST breakpoints. 111/877 isolates were later confirmed CPEs, producing 766 false positive screens (positive predictive value, PPV 13%). If we defined CPE detection using only an ertapenem zone diameter <25mm, all 111 CPEs would be detected, but at a cost of 757 negative confirmatory tests (PPV 13%). Alternatively, if we defined detection using only a meropenem zone diameter <25mm, we would miss one OXA-48 producer while generating 159 negative confirmatory tests (PPV 41%). When using the 2017 EUCAST meropenem breakpoint coupled with temocillin testing, the missed OXA-48 was detected, and there was no difference in PPV (41%). Conclusion: Applying the 2017 EUCAST zone diameter screening breakpoints for meropenem and discontinuing ertapenem screening can reduce unnecessary CPE confirmatory testing by up to 80% without otherwise affecting the overall detection of CPEs identified in our laboratory.

Objectives: Delivery of outpatient antimicrobials is complex, particularly for vancomycin which usually requires therapeutic drug monitoring, dose adjustments, slow infusion time, and monitoring of kidney function. In areas without outpatient parenteral antimicrobial therapy programs, safe and appropriate use of vancomycin can be challenging. The purpose of this study is to characterize and compare vancomycin use before and after an audit and feedback based intervention. Methods: This study was an observational study of the barriers to optimal outpatient vancomycin use in a small health authority zone. After identifying these barriers and summarizing outpatient vancomycin use, we performed a quality improvement intervention to address these barriers. The use of vancomycin was compared in the post-intervention period. All patients in the health region prescribed outpatient vancomycin were included in the study. Results: The main barriers for optimal outpatient use were physician knowledge of treatment guidelines and microbiology results and limited outpatient nursing resources. In the post-intervention period, there were 60% less patients (25 vs 63) on outpatient vancomycin. Overall there was a reduction in days of vancomycin therapy per 100 days by 29%. The average vancomycin level was approximately 8 and 10 mg/L before and after the intervention, respectively. Conclusion(s): By identifying the local drivers of excessive outpatient vancomycin prescriptions, we were able to reduce outpatient vancomycin use. Utilizing local data to provide feedback to stakeholders was effective in decreasing vancomycin use in a small health authority zone.

Objective: Pneumocystis jirovecii is an opportunistic fungus associated with respiratory infections. It is most strongly associated with an impaired immune response due to HIV infection; however, non-HIV associated immunosuppression such as malignancy or transplant can also predispose to infection. Microscopic examination of GMS-stained slides from respiratory specimens is currently the test method performed in our laboratory. With the future goal of introducing PCR for P. jirovecii diagnosis, we sought to review test requests to determine if ordering practices are appropriate. Methods: All requests for Pneumocystis jirovecii investigation for January and June 2018 were extracted from the laboratory information system. We reviewed the corresponding requisitions and recorded the following data: the presence of clinical information, the extent of microbiology test requests and the method of specimen collection. Additionally, we reviewed the provincial electronic health record for the presence of qualifying risk factors. Results: During the review period, a total of 233 samples from 184 patients were submitted for examination. Pneumocystis jirovecii cysts were identified in 4 specimens. HIV infection was documented in 11 patients (13 specimens). Most specimens were collected during bronchoscopy (95%, n=221) with 56% (n=130) and 39% (n=91) submitted as bronchial wash and bronchial-alveolar lavage samples, respectively. 61% (n=142) of requisitions were devoid of information relating to immunosuppression. In 42% (n=97) of specimens submitted, no risk factors for immunosuppression could be identified from the electronic health record. 57% (n=133) of submitted specimens also had all other listed microbiology tests requested. More than 1 specimen was submitted for 42 patients resulting in an additional 49 (21%) specimens. Conclusions: We conclude that laboratory resources for the identification of Pneumocystis jirovecii are not being utilized effectively. Possible strategies to improve test utilization prior to PCR implementation may include audit and feedback, and requisition design to discourage testing without consideration of risk factors. 

Objectives: To describe the prescribing practices in an outborn, surgical neonatal intensive care unit (NICU) setting and inform targeted antimicrobial stewardship initiatives toward decreasing the antimicrobial burden. Methods: We performed a retrospective review of all antibiotics prescribed at an outborn level III NICU between November 1, 2017 and August 31, 2018. For all antibiotic courses, a complete electronic chart review was performed to identify the antibiotic indication, microbiological data, days of therapy (DOT), duration and appropriateness of therapy. Results: Antibiotics were prescribed in 173 neonates (57% of all admissions) at a rate of 519 DOT per 1,000 patient-days. The most common indication was empiric therapy for suspected early-onset sepsis (EOS), followed by antibiotic prophylaxis and empiric therapy for suspected late-onset sepsis (LOS) (Figure 1).  Antibiotics for EOS were initiated in 94% (83/88) of neonates prior to transfer. Of the 25 patients with CNS symptoms (seizures or hypoxic ischemic encephalopathy) started on empiric antimicrobial therapy, 13 (52%) remained on antimicrobials even after an alternative diagnosis was assigned. Among the 311 DOT (38 courses) prescribed for prophylaxis, 192 (62%) were for surgical prophylaxis.  Eighteen courses (72%; 110 DOT) were deemed to be inappropriate because of prolonged use, with a median therapy of 3 antibiotic days (IQR 2-8). General surgery (90 DOT), followed by ENT (73 DOT) and cardiovascular surgery (17 DOT) prescribed the majority of these courses. Conclusion: Prolonged therapy for suspected EOS and surgical prophylaxis are key drivers for antibiotic utilization in the outborn, surgical NICU setting. The decision to initiate these courses is often made by care providers prior to transfer to the receiving neonatal team. Thus, antimicrobial stewardship initiatives should target efforts to discontinue antibiotics by 36 hours of culture incubation, or earlier if a unifying non-infectious diagnosis is confirmed, and to decrease prolonged post-operative antibiotics prophylaxis.

Background: Sepsis and pneumonia are main causes of unexpected death in ICU. Although the main processes related to these infections are already known, risk prediction models can be used to identify unexpected cases. Methods: Multiple logistic regression models were built to predict central line-associated bloodstream infections (BSI) and ventilator-associated pneumonias (VAP). Independent variables: age, ICU length of stay (STAY), severity scores at admission (APACHE, TISS-28, SOFA), and Braden score for pressure ulcers. BSI and VAP were identified by using NHSN/CDC protocols, collected between Jan-Jun/2018 from one Medical/surgical ICU. Unexpected infections: any infection that is 5% or less likely to occur according to its model that eventually occurs, will be classified as UNEXPECTED. Results: A total of 532 patients were analyzed (16 BSI [3.0%], 48 VAP [9.0%], 34 deaths [6.4%]). Hospital death in patients without BSI was 5% which increased to 63% in infected patients (p<0.001). While mortality rate in patients with VAP was 44%, in non-VAP mortality was only 3% (p<0.001). Only 1% of patients without BSI and VAP died. Logistic model coefficients for BSI: -7.3 (constant); 0.07 TISS-28 (p=0.03); 0.13 SOFA (p=0.01); 0.04 STAY (p=0.01); Area under the ROC Curve = 0.74. From the 16 BSI patients, five were classified as unexpected BSI. VAP logistic model coefficients: -3.7 (constant); 0.03 TISS-28 (p=0.03); 0.05 APACHE (p=0.02); 0.03 STAY (p=0.02); -0.11 BRADEN (p=0.01); Area under the ROC Curve = 0.69. From the 48 VAP patients, just one was classified as an unexpected VAP. Conclusions: The models built allow    us to identify unexpected infections. Those  cases were deeply analysed by the London protocol, which showed gaps in the ICU nosocomial infections prevention.

Objective:  A systematic review of the current literature to determine the fetal and neonatal outcomes of pregnancies infected with Zika virus (ZIKV). Methods: We conducted a systematic review following PRISMA guidelines and Cochrane systematic review methodology. MEDLINE, Embase, PubMed, CINAHL, LILACS, and WHO's ICTRP clinical trials registries database were searched using terms “Zika virus” and “Zika infection”. Editorials, letters, news articles, and experimental animal studies were excluded. Case reports or series with less than 10 cases were excluded. Two independent reviewers conducted title/abstract screening, full text screening, and data extraction using a pre-specified form. Conflicts were resolved by consensus or consultation with a third reviewer. JBI tools were used to assess quality. Results: The initial search (up to 30/04/2018) identified 7394 references, of which 69 studies met inclusion criteria (92,882 cases of possible ZIKV in pregnancy). There were 47 case series, 7 case-control, 9 cohort and 6 cross-sectional studies. Studies were predominately from Brazil (45), the USA (11) and Colombia (5). There was wide variation in diagnostic criteria for ZIKV exposure due to local availability of serological testing and pre/postnatal neuroimaging. 6.5% (258/3772) of exposed fetuses developed birth defects consistent with ZIKV. Anomalies were most common after first trimester exposure (8.5% vs. 6.3% vs 5.2% per trimester). Microcephaly and intracranial abnormalities predominated, but extracranial CNS, cardiac and growth abnormalities were also frequently present in affected fetuses/infants. The effect of timing of exposure, maternal symptoms, testing approaches and infant follow up duration were analyzed individually. Final analysis will include all publications to 31/12/2018. Conclusion: ZIKV infection during pregnancy leads to intracranial and extracranial abnormalities in affected fetuses and infants. Understanding the nature, timing and frequency of these abnormalities will allow development of screening, diagnostic and clinical guidelines for ZIKV. 

Objective: To describe the serotype distribution and clinical spectrum of Haemophilus influenzae (Hi) bacteremia in children admitted to 7 PICNIC centres. Methods: All cases of Hi bacteremia were identified in children admitted between 2013 and 2017 to participating centres. Disease was defined as complicated if the following occurred: a) ≥ 2 sites were affected, b) surgical intervention was required, c) organ failure, d) ICU admission, e) seizures, f) sensory or motor deficits, g) treatment-related complications, or h) death. Results: There were 87 eligible cases of Hi bacteremia. Preliminary analysis was limited to 53 cases from 4 centres with complete clinical and microbiological data. Male to female ratio was 31:20 and median age was 1.0 year (range: 0-15) years. Twenty-three (45%) were infants with median age 6 months (range 0 – 11 months). Hi serotypes included: a (N=18; 34%), b (N=8; 15%), f (N=7; 13%), c (N=1; 2%), e (N=1; 2%), non-typeable (N=16 cases; 30%) and unknown (N=2; 4%). Clinical foci included: bacteremia without a focus (N=22; 41%), meningitis (N=17; 32%), cellulitis (N=4; 8%), septic arthritis (N=3; 6%), pneumonia (n=3; 6%), epiglottitis or sinusitis (N=2; 4%) and endovascular infection (n=2; 4%). Complicated disease occurred in 19 (36%) cases; there was one (2%) death. Of the 17 cases with serotyping available, complication rates were: 62%, 42%, 13% and 13% for Hia, Hif, Hib and nontypeable Hi, respectively. Factors associated with complicated disease included typeable Hi (p=0.029) and a CNS focus (p<0.001). Conclusion: In the era of efficacious conjugate Hib vaccines, serotype a has emerged as the leading cause of typeable Hi disease and is associated with more complicated disease as compared to other serotypes. Strategies for preventing Hi disease should be directed at improving vaccine uptake rates to control Hib disease and developing an effective vaccine for preventing serotype a disease.

Background: Visiting therapy dogs are growing in popularity within long term care facilities, psychiatric wards, prisons and hospitals. However, concerns regarding zoonotic disease (particularly MRSA) and infection control often limit the application of the visits. The objective of this study was therefore to determine the frequency of colonization of therapy dogs with zoonoses (Staphylococcus pseudintermedius, S. aureus and Escherichia coli), and the antimicrobial susceptibility of these organisms. Methodology: In October and November 2018, pharyngeal and rectal swabs were taken from 38 dogs registered with the St. John’s Ambulance Therapy Dog program. Samples were selectively cultured for each organism using CHROMagar Staph aureus and CHROMagar Orientation, and the antimicrobial MICs were determined by broth microdilution. A survey was administered to handlers to gather metadata regarding therapy dog activities, duties and health status. Results: 28 dogs (74%) were colonized with S. pseudintermedius of which 3 (10.7%) were MRSP. 13% of dogs were colonized with S. aureus none of which were MRSA; all visited hospitals or special care facilities. All dogs were colonized with E. coli including 4 carrying ESBL-producers. Penicillin + ampicillin was the most common resistance phenotype identified in 66.7% of S. pseudintermedius and 80% of S. aureus isolates. 89.7% of E. coli isolates were pan-susceptible. The survey revealed that visits to hospitals were common, 13 dogs (34%) reported visiting a hospital in the previous 3 months. Hand hygiene was variably practiced; 31% of handers reporting using hand sanitizer at every therapy dog visit, 46% sometimes and 23% never. When used, hand sanitizer was most commonly applied both before and after dog contact (70%). Conclusion: While therapy dogs carry potential pathogens, the magnitude of the zoonotic risk associated with these animals is ill defined. Although this study had a small sample size, it was encouraging that no MRSA were identified. 

Objectives: Aerococcus urinae is a Gram-positive organism, initially identified in 1992.  It is known to cause urinary tract infections (UTIs), bacteremia, and endocarditis in humans.  There is limited data regarding the susceptibility of A. urinae to first-line antimicrobials indicated for the treatment of UTIs.  In 2016, A. urinae was isolated in 125 urine samples processed by four hospitals in the study region.  The unfamiliarity with this organism and the lack of local antimicrobial susceptibility rates presents a challenge for clinicians and often results in the unnecessary use of broad-spectrum antibiotics.  The primary objective of this study was to establish the susceptibility rate of A. urinae urinary isolates to cefazolin, ampicillin, nitrofurantoin, fosfomycin, and ciprofloxacin.  The secondary objective was to identify demographic characteristics associated with A. urinae bacteriuria in our patient population. Methods: Urinary samples received by the laboratory from October 2017 to June 2018 underwent routine identification as per physician orders.  Samples that grew A. urinae were included in this study and subjected to susceptibility testing.  Susceptibility testing was conducted via disk diffusion and results were interpreted based on published breakpoints for zone diameters. Results: A total of 72 isolates were included.  Susceptibility rates for cefazolin, ampicillin, nitrofurantoin, fosfomycin, and ciprofloxacin were 100%, 99%, 99%, 96%, and 65%, respectively. The average age of patients was 79 years, 63.9% were female, 31.9% were recently hospitalized, and 44.4% were residents of a long-term care facility. Conclusion: Cefazolin, ampicillin, nitrofurantoin, and fosfomycin demonstrated good in vitro activity against A. urinae.  In contrast, ciprofloxacin demonstrated decreased activity against this organism. Currently recommended first-line agents for the management of uncomplicated UTIs could be utilized to treat this organism.  Characteristics of patients with A. urinae bacteriuria are consistent with risk factors predisposing to UTIs.

Background: Treponemal and nontreponemal serologies remain the cornerstone of syphilis diagnosis and management. In Alberta, the reverse screening algorithm is utilized with a treponemal enzyme immunoassay (EIA) used as the screening test and the Treponema pallidum Particle Agglutination assay (TPPA) performed as the confirmatory test. If the EIA is positive and the TPPA is indeterminate on repeat testing the clinician can request an INNO-LIA (Innogenetics multiparameter line immunoassay), a type of treponemal pseudo-Western blot which is performed at the National Microbiology Lab (NML). Prior to October 2016, PLNA performed the INNO-LIA in-house as the confirmatory test but, due to prolonged turnaround time (TAT), a high number of indeterminate results and clinician dissatisfaction, it was replaced by the TPPA. In this study, we compare these two methods over similar time periods. Methods: INNO-LIA results from April 2015 to March 2016 and TPPA results from October 2016 to December 2017 were extracted from our laboratory information system and analyzed. Results: Out of 2578 INNO-LIA, 480 (19%) were invalid. From the initiation of TPPA, there were 3904 TPPA run, of which 198 (5%) were indeterminate which was significantly less (p<0.0001) than the INNO-LIA. Fourteen (7%) of those tests were arbitrated by INNO-LIA at NML. 13 (93%) INNO-LIA tests were reactive and 1 test was indeterminate. Supply cost for the INNO-LIA was $35.54/test and TPPA was $2.35/test. TAT for the INNO-LIA was 10 days whereas TPPA was 72 hours. Conclusions: We demonstrate that in Alberta confirmatory syphilis testing by TPPA produces significantly less indeterminate results than INNO-LIA, is cost saving and ultimately, improves care and management of patients and their close contacts. The one disadvantage of the TPPA is the poor sensitivity in early and primary syphilis.

Background: Microbiology point-of-care tests (POCT) are simple-to-use, automated assays performed outside of a laboratory infrastructure that can improve diagnostic accessibility and turnaround times, with potential benefits for antimicrobial stewardship and patient flow. Commercial availability and performance of POCT have rapidly evolved. Accreditation standards for implementing POCT exist for laboratories but if POCT are implemented without laboratory knowledge, these standards may not be followed. The goal of this survey was to describe laboratory awareness and involvement in POCT use. Methods: In January 2018, a web-based patterns-of-practice qualitative survey was conducted by the Institute for Quality Management in Healthcare across all 73 laboratories participating in their bacteriology proficiency testing program. Questions addressed laboratory awareness and involvement in assuring accreditation standards were followed regarding POCT implementation. Results: All 73 participants completed the survey. 12% of respondents reported POCT use within their hospital, while 5% reported adoption across affiliated outpatient settings. Notably, 11% and 18% were unsure, respectively. Of those aware of POCT use, 45% were not involved in the decision to introduce POCT on site, and 40% did not participate in the device selection process. Similarly, 40% of participants were unaware of any verification completed prior to the routine use of the device, and 20% noted the absence of standard operating procedures. 20% of participants noted absence of initial training and 30% were unaware of longitudinal competency assessments. Ongoing device maintenance was lacking in 89% of participating institutions. Additionally, 70% of participants stated that there was no overall monitoring of outcome measures after POCT implementation. Conclusion: Our survey results indicate a low amount of laboratory awareness and involvement with microbiology POCT and a concerning proportions of institutions lacking standard quality management of POCT. In anticipation of the expanding adoption of POCT, establishing national guidelines requiring laboratory oversight of POCT should be a priority.

Objectives: To describe antimicrobial prescribing decision-making by ICU clinicians before and after implementation of Stewardship at Bedside Rounds (SABeR). Methods: Assessment of antimicrobial decision-making was conducted in a 24-bed tertiary care ICU using a standardized audit tool before and after implementation of the intervention. The audit tool captured discussion of antimicrobial decision-making using a framework developed by the antimicrobial stewardship (AMS) program consisting of eight antimicrobial prescribing decision nodes. The intervention consisted of transitioning the ICU inter-professional team from thrice-weekly AMS team-led rounds to standardized infection reporting and review during daily bedside rounds.  This included introducing “infection” (temperature, white blood cell count, and antimicrobial therapy) as a body system by nurses during head to toe patient report.  ICU physicians and pharmacists were educated to expect inclusion of infection as a separate system as an AMS intervention, and to move stewardship discussions to the bedside. All outcomes were assessed using descriptive statistics. Results: Ninety-five (95) pre-intervention observations occurred between May and October 2017 and 217 post-implementation observations occurred between October 2017 and August 2018.  Table 1 illustrates the difference in antimicrobial decision-making discussion of patients prescribed antimicrobials. Discussion of all decision–making nodes increased post-implementation except tailoring of therapy.

Table 1: Antimicrobial Therapy Discussion Pre- and Post-Implementation

Conclusion: Antimicrobial prescribing decision-making improved after the implementation of SABeR.

Background: Patients with cystic fibrosis (CF) suffer from chronic lung infections that frequently involve Pseudomonas aeruginosa. Antibiotics cannot eradicate the P. aeruginosa lung infection due to the high degree of tolerance displayed by the bacteria. Tobramycin combined with fumarate (TOB-FUM) has been previously shown to promote significant killing of antibiotic tolerant P. aeruginosa whereas tobramycin alone is ineffective. TOB-FUM is currently being investigated in clinical trials to treat chronic P. aeruginosa infections in CF patients. However, P. aeruginosa genes that are required for TOB-FUM susceptibility have not been previously documented. Objective: Demonstrate that the rpoN gene is required for TOB-FUM killing of P. aeruginosa. Methods: Antibiotic tolerant cultures of P. aeruginosa wild-type and ΔrpoN strains were incubated for 4 hours with or without fumarate and increasing concentrations of tobramycin. Bacterial survival was assessed by plate counting. Results: Minimal killing of wild-type and ΔrpoN cells was observed when cultures were treated with either tobramycin or fumarate alone. Treatment of wild-type cells with TOB-FUM resulted in a 3-log10 decrease in cell viability. In contrast, TOB-FUM treatment was unable to kill ΔrpoN cells. This phenotype could be complemented in trans with a plasmid-borne copy of rpoN. TOB-FUM treatment of mixed wild-type and ΔrpoN cultures (as might be observed in a CF lung) led to preferential killing of the wild-type cells while the ΔrpoN cells were spared.  Conclusion: The rpoN gene is required for susceptibility to TOB-FUM. Importantly, loss of RpoN function is commonly observed in P. aeruginosa CF clinical isolates. Future work will confirm that clinical isolates lacking RpoN function are not susceptible to TOB-FUM. Clinical trials that test TOB-FUM efficacy might need to stratify patients based on the frequency of rpoN mutant isolates in their lungs to determine if patients with a lower frequency of rpoN mutants benefit more from TOB-FUM than patients with a higher burden of rpoN mutants.

Background/Objective: The primary focus of pediatric Zika virus (ZIKV) research has been the sequelae of congenital infection.  ZIKV acquired post-natally has generally been described as asymptomatic or a self-limited febrile illness; however, there has been minimal research into the clinical course and potential complications. Animal studies suggest that ZIKV acquired in infancy may have neurodevelopmental sequelae. We aim to summarize the literature on health outcomes in post-natally acquired pediatric ZIKV infection using systematic review methodology. Methods: We are conducting a systematic review following PRISMA guidelines and Cochrane systematic review methodology. MEDLINE, Embase, PubMed, CINAHL, LILACS, and WHO's ICTRP clinical trials registries database were searched using terms “Zika virus” and “Zika infection”. Editorials, letters, news articles, and experimental animal studies were excluded. Case series required 10 or more cases for inclusion. Two independent reviewers conducted title/abstract screening, full text screening, and data extraction. Conflicts are resolved by consensus or consultation with a third reviewer. Results: Initial search up to April 30, 2018 recovered 7394 references. 9 case series met inclusion criteria and present data from Brazil (N=1), Colombia (N=1), Dominica (N=1), Singapore (N=2), and US (N=4). N ranged from 11-18,576, and age from 1 month-18 years. Laboratory confirmation of ZIKV infection was inconsistent across regions. Overall, there was poor reporting of clinical symptoms. The most common symptoms were fever (71-93%) and rash (94-100%), in part reflective of case definitions. Severe complications were rare (Guillain-Barre Syndrome 0-0.2%; meningitis/encephalitis 0-0.09%; mortality 0-0.05%). No neurodevelopmental outcomes were reported.  Literature search will be updated to December 31, 2018 and data analysis completed by the time of the conference. Conclusion: This up-to-date systematic review of health outcomes in post-natally acquired pediatric Zika virus infection will summarize current knowledge and identify research gaps for optimization of clinical care and public health interventions. 

Objectives: The Canadian Pediatric Society no longer recommends the use of universal ocular prophylaxis with erythromycin ointment 0.5% to prevent neonatal gonococcal ophthalmia. Instead, screening for Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) in all pregnant women is considered the most effective way of preventing vertical transmission and neonatal conjunctivitis. The aim of this study was to assess compliance with Quebec pregnancy screening guidelines. Methods: The list of all women who delivered at a tertiary care hospital in the province of Quebec, between April 2015 and March 2016, was cross-referenced with the list of samples tested for CT/NG. Maternal medical records were reviewed for demographic, prenatal and diagnostic information. Results: Amongst 2688 women, 432 (16%) weren’t screened during pregnancy and 38 (1.4%) had an invalid result reported. Among the 2218 (82.5%) women with at least one valid result, infection was detected in 45 (2%): CT (43; 1.9%) and NG (4; 0.2%); two women were co-infected. Prevalence of CT infection was significantly higher among women aged <25 years old (9.3%; 28/301) than among those aged ≥25 yo (0.9%; 17/1926; p<0.001). Of the 2 177 women with an initial negative test result for CT and NG, 8% (170/2177) were retested: 36/272 (13.2%) among women aged <25 yo and 134/1905 (7%; p<0.0001) among those aged ≥ 25 yo. Subsequent infection was detected in four (2.4%) women, three of whom being <25 yo. Conclusions:
Compliance with CT/NG screening guidelines is insufficient to stop current universal
ocular prophylaxis. Repeating universal screening later in pregnancy should be
considered: in addition to identifying women who become infected later in pregnancy, this
strategy could decrease the number of women who are not screened at all during
pregnancy.

Objective: Total laboratory automation (TLA) smart incubators have the potential to reduce incubation times of primary cultures which would also improve time to reporting for susceptibility testing. Our goal was to verify the performance of susceptibility testing by KB from clinical isolates recovered from primary cultures with reduced incubation times. Methods: Inoculum for KB testing of clinical isolates was prepared from cultures incubated for 12 and 15 hours in the Kiestra smart incubators and compared to susceptibility testing performed after 18 hours of incubation. Antibiotic disks were dispensed on inoculated Mueller-Hinton agar and incubated according to CLSI recommendations. Zones of inhibition were measured using the Kiestra zone measurement function. Results: Susceptibility testing was performed on 117 Enterobacteriaceae, 32 non-fermentors, 46 S. aureus, 54 coagulase negative Staphylococcus and 75 Enterococcus species. There was no statistically significant difference in the zone of inhibition for any drug-organism combination when KB was performed from isolates recovered from 12 or 15 hour cultures compared to 18 hours. For Enterobacteriaceae there was 1 very major error (VME) (tobramycin at 12 hours), no major errors (ME) and 43 minor errors (MNE). There were 21, 12 and 11 MNE with isolates recovered from cultures incubated for 12, 15 or 12 and 15 hours respectively. Categorical agreement for all organism-drug combination when performed from 15 hour cultures was higher than if performed from 12 hour cultures (p<0.001). There were 2 MNE among non-fermentors (12 hours) and no discrepancies for the Gram positive organisms. Conclusion: Reducing incubation times of primary cultures did not adversely affect the outcome of KB susceptibility testing. Correlation with 18 hours was better if primary cultures were incubated for 15 hours compared to 12 hours.

Objective(s): The objectives of this study are to describe the epidemiological, microbiological, clinical, echocardiographic and outcome variables of patients with intravenous drug use-associated infective endocarditis (IVDU-IE) admitted to X hospital, estimate the prevalence of IVDU-IE in X city, and estimate the rates of complications and mortality among patients with IVDU-IE admitted to X hospital. Methods: A retrospective chart review was conducted on patients who met the following three criteria: (1) International Classification of Diseases, Tenth Revision (ICD-10) codes with a discharge diagnosis of IE, (2) IVDU within 3 months of IE, (3) admitted to X Hospital between January 1, 2012 and December 31, 2017. Data was collected on cases meeting these criteria on the following categories: epidemiology, microbiology, clinical signs and symptoms, echocardiography, complications throughout admission, and outcomes. Results: 42 cases were identified that met our inclusion criteria. The majority of patients (72.4%) were male, with opioids being the most common injection drug used (79.3%). The most common clinical sign exhibited by cases was fever (90.5%) and Staphylococcus aureus (61.9%) was the most common microorganism isolated. The tricuspid valve was most commonly affected (58.5%) and 50% of cases had heart failure as a complication throughout admission. 45.2% of cases required a valve replacement, and all patients received antibiotics. 31.0% of patients died during the study period. Conclusion(s): Despite the relatively young age of this patient population, IVDU-IE is a disease associated with significant morbidity and mortality. Implementing effective harm reduction strategies for these patients are always a worthwhile effort. We hope that the information gathered in this study will be used by all healthcare professionals who work with this population to better understand the clinical characteristics and the outcomes of these patients.

Background: Beta-lactam (BL) antibiotics are considered first-line treatment for a range of infections. Patients who develop acute interstitial nephritis (AIN) secondary to BL exposure are often prescribed less optimal antibiotics or those reserved for resistant organisms to avoid the potential risk of recurrent AIN with other BLs. In the medical literature, therapies used following development of BL-induced AIN have not been well-defined, making it difficult to standardize antibiotic alternatives. Objectives: To describe the management and evaluate the clinical course of patients following BL-induced AIN. Methods: Retrospective cohort study at a tertiary hospital from 2012 to 2017. Patients with AIN were identified using ICD-10 code “N10” for “Acute Tubulo-Interstitial Nephritis,” and search parameters “acute kidney injury,” “acute renal failure”, and “AIN.”  Demographics, antibiotic treatment regimens, symptoms, and clinical outcomes were collected. Results: Fourteen patients (male 64%/female 36%) were diagnosed with BL-induced AIN (biopsy-proven 14%, clinically-proven 86%) from 2012 to 2017. Mean increase in serum creatinine was 192 µmol/L secondary to BL-induced AIN, with a mean decrease to 130 µmol/L within 30 days of discontinuation of the BL. Classical symptoms of AIN included rash (14%), fever (57%) and eosinophilia (50%, n=10), however, presenting symptoms varied amongst the population. Carbapenems (36%), fluoroquinolones (21%) and vancomycin (14%) were the three most common alternative antibiotics following development of BL-induced AIN, and no further modifications were made to the initial change in antibiotic. Three patients did not require additional antibiotic therapy. All patients experienced microbiological cure. Conclusion: The incidence of BL-induced AIN appears low. Discontinuation of the offending BL antibiotic improved signs and symptoms of AIN within 30 days. Use of alternative antibiotics, including BLs, did not appear to worsen clinical outcomes, which is suggestive that these antibiotics can be used. Further studies are required to confirm use of alternative BLs in BL-induced AIN.

Background: Clostridium difficile infection (CDI) is an important cause of nosocomial diarrhea. Previous studies have suggested that metronidazole and vancomycin are equally effective for the treatment of non-severe CDI, however, recent guidelines have recommended the initial use of vancomycin. Our study objective was to identify clinical predictors of adverse outcomes and the impact of first-line vancomycin for treatment of non-severe, inpatient CDI. Methods: We conducted a retrospective chart review of all adult inpatients with first episode CDI at our institution from January 2013 to May 2018. CDI was defined as a positive C. difficile Loop-mediated isothermal amplification assay, in conjunction with ≥3 Type 5–7 stools on the Bristol stool scale. We abstracted comorbidities, medications, and relevant outcomes (recurrence, relapse, and death). Results: A total of 737 cases were included. Patients had a median age of 72.3 years (Q1: 61.2, Q3: 83.3) and 628 (85.2%) were classified as non-severe CDI. Predictors of relapse and all-cause 30-day mortality for the overall cohort were: hospital-acquired infection (OR_adj: 2.08; 95%CI: 1.50–2.89; P<0.001), age ≥65 (OR_adj: 1.98; 95%CI: 1.37–2.88; P<0.001), and white blood cell count >15×10⁹/L (OR_adj: 1.76; 95%CI: 1.26–2.46; P=0.001). Amongst patients with non-severe CDI, relapse, recurrence, and mortality rates were 17.4%, 7.0%, and 11.4% respectively when treated with initial metronidazole, compared to 18.6%, 3.1%, and 7.8% respectively when treated with initial vancomycin. The use of first-line vancomycin for treatment of non-severe CDI was not associated with relapse, recurrence, or 30-day mortality alone. However, in an adjusted analysis, the use of first-line vancomycin for treatment of non-severe CDI was associated with a reduction in the composite outcome of recurrence or 30-day mortality (OR_adj: 0.51; 95%CI: 0.28–0.94; P=0.03). Conclusions: First-line vancomycin was associated with reduced recurrence or all-cause 30-day mortality in the treatment of inpatients with non-severe, first episode CDI. 

Background: Viral respiratory infections are acute self-limited illnesses; however those with chronic underlying illnesses are at greater risk of morbidity. Objective: To understand the epidemiology and usage of preventative measures to help reduce transmission events.  Method: A retrospective chart review was performed on patients admitted to high-acuity units (MICU, SICU, PICU, NICU) with positive molecular tests from Oct.1^st, 2016-May 30^th, 2017 at 7 hospitals in Winnipeg, MB. Results: 307 (188 adult [18+y]; 119 pediatric) patients had specimens submitted for multiplex viral testing; at least one virus was detected in 112 (39.4%). Among adults, viruses identified were: influenza (12, 6.4%), parainfluenza (11, 5.9%), rhinovirus (10, 5.3%), coronavirus (7, 3.7%), HMPV (5, 2.7%), RSV (3, 1.6%) other (3, 1.6%). Among children, viruses identified were: RSV (27, 23%), rhinovirus (15, 13%), parainfluenza (6, 5.0%), influenza (4, 3.4%), coronavirus (3, 2.5%), other (6, 5.0%) (distribution P<.001 compared to adults). Overall, 4 (7.8%) of adult infections and 3 (5.1%) of pediatric infections were hospital-acquired. No significant difference in the distribution of viruses was noted based on geographic region or urban vs. rural environments.  Influenza vaccination history for the current season was documented in 7.8% of adults and 5, 4.2% of children (p=0.001). 42.3% of patients had documented infection control orders. Among children, 34 (58%) had orders, 31 (92%) meeting infection control guidelines. Among adults, 11 (22%) had orders, 4 (36%) of which met guidelines. Conclusions: Viral respiratory tract infections are common in high-acuity units, and causative viruses appear to differ between adult and pediatric patients. Hospital-acquired cases are uncommonly diagnosed. Infection control orders were sub-optimal in both adult and pediatric settings. This study highlights the need for further research into respiratory viral illness in ICUs, as well as barriers with regards to the implementation of infection control measures. 

Objectives. The Champlain local health integration network offers health services to 1.3 million residents. Within an 18000 sq. km area, 16 community and tertiary care hospitals collect and send blood cultures to the Regional Microbiology Reference Laboratory situated in a tertiary care University hospital. A retrospective audit of blood cultures over a one-year period assessed time from collection to loading of blood cultures bottles in, and time to detection by the BD BACTEC blood culture system for each site. Methods. A total of 59122 negative blood cultures were reviewed to assess time from collection to loading, and 10535 positive blood cultures were assessed for average time to positivity. For selected sites, hourly distribution of number of samples was compared to daily courier schedules. Results. Median time to loading for samples originating from within the Ottawa city-limits was 3 hours compared to 7.6 to 22.7 hours for blood cultures collected from client sites outside of the Ottawa City limits. For most sites, 75% of samples reached the laboratory within 20 hours. Average time to positivity for all cultures did not vary significantly between inner-city sites (22.5 hours) and distant sites (25.6 to 31.8 hours). However, there was a trend of decreasing time from loading to positivity for distant sites (13.5 to 16.1 hours) compared to inner-city sites (18.5 hours). Conclusions. Despite prolonged delayed entry of blood cultures from remote client sites, bacterial growth may have reduced the time to detection of positive cultures by the BD BACTEC blood culture system and mitigated some of the negative impact of delays in transportation. Successful laboratory regionalization may largely depend upon optimization of courier routes. Adequate coordination of courier service with peak collection times may reduce delays.

Background: Sink drains are a reservoir of gammaproteobacteria which may be transmitted to patients. We sought to assess factors associated with aerosolization from ICU sinks. Methods: Sink tailpiece and faucet interior surfaces in 7 ICUs were cultured semi-quantitatively by swabbing defined surface areas, inserting swabs into 1ml of neutralizing broth, plating different volumes onto MacConkey3 with crystal violet (Mac3CV), incubating aerobically at 37ºC for 18-24hours, then counting gammaproteobacterial colonies. 850L air samples were collected by impaction onto Mac3CV with the sampler held at a defined location 20cm from faucets, and plates incubated aerobically at 37ºC for 18-24hours before counting colonies. For sinks without electronic controls, air samples were collected first with cold, then hot, water running. Logistic regression with generalized estimating equations was used to assess factors associated with detectable gammaproteobacteria in air. Results:  Of 382 tailpiece swabs, 70 had no growth (NG), 144 growth 1-750cfu/cm², and 168 growth >750cfu/cm². Among air samples, 247 had NG, 126 growth 1-235cfu/1000L, and 9 growth >235cfu/1000L. In multivariable analysis, gammaproteobacteria were more likely be detected in air if tailpiece surface growth was 1-750cfu/cm² (OR=3.5, 95%CI 1.3-9.5) or >750cfu/cm² (OR 5.5, 95% CI 2.0-15) compared to NG, and in summer (April-September) versus winter (OR 6.0, 95%CI 2.0-18). The probability of detection of gammaproteobacteria in air also differed in different hospitals (growth in air detected in 4.7%-90% of samples, P=0.0002), but was not affected by growth from faucet samples. Of 301 matched air samples with hot and cold running water, 94 (31%) and 67 (22%) had detectable growth, respectively (OR=1.6, 95% CI 1.1-2.3). Conclusions: Gammaproteobacteria are more often detected in air adjacent to sinks when higher concentrations are present in sink tailpieces, when  hot versus cold water is running, and in summer. Further data are needed to understand whether such aerosolization creates patient risk.

Background: Antimicrobial stewardship (AS) aims to optimize appropriate use of antimicrobials. Education of medical trainees is an important strategy to achieve this goal. Our study assesses resident physicians on their previous educational experiences with AS and subsequent preparedness towards prescribing antimicrobials appropriately. Methods: Resident physicians from all levels in training from all postgraduate residency programs at a Canadian university-affiliated teaching hospital were invited to complete a 16-item online survey (April – October 2018). Results: 127 (20%) of the 630 residents completed the survey. 85% of residents (108/127) are familiar with the term “Antimicrobial Stewardship”. Only 57% (72/127) are familiar with in-hospital strategies employed by AS, including stewardship rounds, local antibiograms and restricted antimicrobial formulary. Even fewer (31%) residents have participated in AS initiatives, most of which were passive activities such as AS lectures and learning modules during undergraduate medical studies or attending in-patient AS rounds. Over 90% of residents believe that AS should be integrated as part of their education during both medical school and residency. However, only 53% and 60% of residents felt their medical school and residency programs have prepared them well with antimicrobial use in practice, respectively. Conclusion: Most residents have an awareness of AS but only a minority report participation in AS endeavours. Residents recognize the importance of formative education on AS principles and appropriate use of antimicrobials. Our survey highlights a perceived need by residents for enhanced education at both the undergraduate and postgraduate levels. Formal integration of AS into the undergraduate curriculum is planned by our AS team to address this.

Background: Malaria is a common illness in people migrating to the West. Hospital care, including intensive care unit stays, can result in significant financial costs. Pre-departure anti-malarial treatment of migrants to Canada from malaria-endemic countries could prevent morbidity and reduce costs to Canada’s universal health care system. Methods: Health records for children diagnosed with malaria at the Children’s Hospital of Eastern Ontario were retrospectively reviewed from 2010-2017. Patient demographics, details of care, and costs of inpatient and outpatient care related to malaria were determined and compared between migrants (immigrants/refugees) versus children who acquired malaria while traveling to visit friends and relatives. Results: 24 migrants and 9 VFRs with malaria were identified. 20/24 (83%) migrants and 9/9 (100%) VFRs had Plasmodium falciparum malaria; 18/24 (75%) migrants and 3/9 (33%) VFRs were from East Africa, and 5/24 (21%) and 6/9 (67%) were from West/Central Africa. Migrants were median 12.0 years (IQR 8.8, 14.0) versus VFRs 6.4 years (4.0, 11.0) (p=0.09). Time from arrival to Canada to onset of symptoms was similar between groups (migrants – median 5 days; VFRs – 3 days). Similar proportions of migrants and VFRs were admitted to hospital (80% and 71%) and admitted to intensive care unit (27% and 29%). Total cost of care for migrant malaria over the study period was $174,485 CAD, and for VFRs was $61,656 CAD. Conclusions: A significant burden of malaria among migrant children was found in this study, shortly after arriving to Canada. Assuming a cost of empiric pre-departure treatment for malaria of $3/person, the cost of treating all newcomers from Africa to Canada in the same 8-year period (194,267) would be ~$582, 800 CAD. This is only three times the cost of treating children at a single tertiary care center, indicating an apparent cost benefit to pre-departure treatment.

Background: We compared the Allplex Respiratory Panel (RP) Assays 1, 2, and 3 (Seegene, Republic of Korea), one-step real-time PCR detecting 15 respiratory viruses with influenza A subtyping, against a 12-target multiplex PCR laboratory-developed assay (LDA) used at our institution. Methods: In Jan.-Feb. 2018, 325 nasopharyngeal specimens underwent the routine LDA. Specimens were stored at 4°C and within 24 hours underwent automated nucleic acid extraction using the STARLET system (Hamilton, USA), and Allplex RPs on a CFX96 (Bio-Rad, USA) thermocyler. Another 68 samples tested by the LDA in July-Sept. 2018 containing underrepresented viral targets were stored at -80°C, then tested by Allplex RPs in Sept. 2018. Percent agreements and Cohen’s kappa values were calculated. Discrepant results were examined. Results: Ten samples were excluded for failed extraction or invalid results on the LDA. 207/383 (54.0%) samples yielded positive results by Allplex RPs and 177/383 (46.2%) by the LDA. Positive percentage agreement were between 83.3% to 100%, except for human enterovirus (HEV) and parainfluenza virus (PIV) 2 (66.7% and 58.3%). Negative percentage agreements ranged from 93.1% to 100%. Kappa values ranged from 0.56 to 1, and was lowest for human rhinovirus (HRV). Discordant results were identified in 69 samples. Thirty specimens contained targets only identified by Allplex RPs. Twelve were positive only by the routine assay, of which five were PIV-2 from frozen samples. Twenty-seven samples positive by both assays contained discrepancies, of which the majority were polyviral. Seventeen involved one assay detecting HEV/HRV while the other detected one of the two. Conclusions: In this partially prospective evaluation, the Allplex RPs and our LDA show high percentage agreement and kappa values. Discrepant results mainly involve HEV, HRV and PIV-2, which may result from testing of frozen samples and/or cross-reactivity between picornaviruses, requiring further testing to resolve.

Objective: Inappropriate use of antimicrobials results in unnecessary costs, Clostridium difficile infections, longer length of hospital stay (LOS), adverse drug events, and antimicrobial resistance. As such, an antimicrobial stewardship program should measure these and other outcomes for quality improvement of patient care. The objective of this study was to assess the feasibility of specific patient-centred outcome measures that would be possible targets for ongoing prospective analysis. Methods: We reviewed the electronic medical records of patients admitted to the internal medicine and family practice services at our tertiary referral centre from June to August 2018. Patients who received audit and feedback interventions were matched to controls by age, gender, antibiotic, and indication. The following outcomes were collected: acceptance of stewardship intervention, ICU admission, represcription, bloodstream infection with any antibiotic-resistant organisms, fungemia, renal toxicity, neutropenia, readmission, and mortality. Linear regression was used for analysis of numerical outcomes and logistic regression was used for analysis of categorical outcomes. Results: The 31 patients in the intervened group and the 31 patients in the control group were similar with respect to their baseline characteristics of age (median 83.5 years, IQR 73.75-90), gender (71% males), Charlson Comorbidity Index (median score 8, IQR 6-10), and pre-intervention LOS (median 2 days, IQR 0-5). Twenty-three interventions (74%) were accepted. Compared to the control group, the intervened group had significantly lower 30-day mortality (Odds Ratio 0.31, p=0.0485), adjusted for comorbidity score and pre-intervention LOS. Other outcomes were not significantly different between the two groups. Conclusions: Audit and feedback interventions were associated with significantly lower 30-day mortality. Ultimately, mortality represents the most objective clinical outcome measure and is readily extractible from electronic medical records for analysis and inclusion in dashboard metrics and quality improvement cycles.

Background: Ontario has one of the largest groundwater-reliant populations in Canada, with approximately 1.6 million households utilising private wells. Unlike municipal water supplies, well owners are the primary agents responsible for managing their drinking water, including source maintenance, resource protection, and water testing. A previous Ontarian study found that just 11-12% of well owners complied with provincial water testing guidance during any year between 2008 and 2012. This finding suggests significant gaps in knowledge and/or tools for stewardship, representing a major concern due to the ubiquity of contaminant sources in rural areas (e.g. agricultural run-off, septic tanks, etc.). The current study sought to identify and assess the gaps associated with private well water stewardship; namely, knowledge, attitudes and practices (KAP), to contribute to improved health behaviours among rural Ontarian residents. Methods: A province-wide online survey was undertaken over the 4-month period May to August 2018. The survey was designed to quantify information among Ontario’s well owners based on their awareness, perceptions and behaviours in relation to their personal source and local sources of contamination. Results: The survey was completed by 1030 respondents (99% CI 4.02%). Preliminary findings indicate that previous experiences (i.e. residential presence during well construction, previous case(s) of acute gastrointestinal illness within household) significantly influence both owner awareness (p<0.001, p=0.038, respectively) and perception of local groundwater contamination risk (p=0.017, p<0.001, respectively). Additionally, increased awareness (p=0.018) and positive attitudes (p=0.006) towards personal well water supplies were associated with increased likelihoods of testing. Conclusion: Findings illustrate that experiences influence both respondent awareness and risk perception, with increased levels of awareness and positive attitudes enabling health behaviours. Results will provide public health agencies with a framework for designing strategies and policies for increasing awareness and addressing the drivers of, and barriers to, protective actions among well users in Ontario, and further afield. 

Background: Mycobacterium avium complex (MAC) are a group of ubiquitous non-tuberculous mycobacterium (NTM) known to cause disseminated infection in individuals with impaired cell mediated immunity, such as advanced HIV. Recently, disseminated MAC infection has been reported in otherwise healthy individuals with impaired functioning of the interferon-gamma (IFN-γ) pathway, particularly Asian populations. In patients with anti-IFN-γ autoantibodies, rituximab has been used successfully to treat refractory cases. Case Presentations: Case 1: A 44-year-old Filipina woman presented with five months of constitutional symptoms, lymphadenopathy, elevated liver enzymes and pancytopenia. She was found to have disseminated MAC involving her lungs, liver, bone marrow and lymph nodes. Initial immunodeficiency work-up including HIV-1/2 antibodies was negative. Autoantibodies to IFN-γand IL-17a were positive. She was treated with daily azithromycin, ethambutol and rifabutin but had clinical and microbiological progression at four months.  Rituximab was initiated with immediate symptomatic improvement. Case 2: A 53-year old previously healthy Cambodian woman presented with nausea, vomiting, and abdominal pain, and was found to have disseminated intra-abdominal MAC. Autoantibodies were positive to IFN-γ. She was treated with parenteral azithromycin, rifampin, amikacin and moxifloxacin, but failed to respond. After three months without improvement, rituximab was added with good clinical response. Case 3: A 52-year Filipino man with a history of non-typhoidal salmonella bacteremia presented with cough, back pain and constitutional symptoms, and was found to have disseminated MAC involving the lungs, lymph node and bone. Autoantibodies to IFN-γ were positive.  He is currently on azithromycin, ethambutol and rifabutin with clinical response to therapy at five months. Conclusion: We present the first reported cases of disseminated MAC associated with anti-IFN-γ autoantibodies in Canada. Refractory disease was common and improved with rituximab therapy. These cases illustrate the importance of investigating the IFN-γ pathway in otherwise healthy patients with disseminated NTM disease.  

Background: Approximately 1.5 million individuals in Ontario are supplied by private wells. Unlike municipalities, private well water quality remains unregulated; therefore, owners are responsible for testing, treating, and maintaining their own systems. It is estimated that contamination of private wells is responsible for approximately 80,000 cases of acute gastrointestinal illness per year in Canada, highlighting their significant impact of this water source on human health. Methods: Well water sample submission data from 2010-2017 were analyzed for Escherichia coli(E. coli)to study the relationship between sampling frequency and E. colidetection rates in Ontarian private wells. Detection rates were further analyzed relative to geological (consolidated and unconsolidated aquifers) setting to determine how hydrogeology impacts E. colidetection (via transport). Power curves were used to estimate the number of samples required to achieve a number of sentinel detection rates (e.g. 25%, 50%, 75%). Results: Province wide, 897,378 samples were analyzed, with detection rates found to increase in concurrence with sample number per well. Power curves indicate that a 50% detection rate (i.e. E. colipresent at least once in half of sampled wells) would occur if each well were sampled 12 times. Statistically significant differences were found between detection rates in consolidated and unconsolidated aquifers (p = 0.00023), highlighting geological structure highlighting geological differences and their potential impact on well water contamination. In consolidated aquifers, E. coliwere detected in 2.1% of wells sampled once and 6.2% of those sampled twice. In unconsolidated aquifers, significantly more samples are required to achieve analogous detection rates. Conclusions: A site specific approach is required for private well testing recommendations in Ontario. Current practises (e.g. ≤1 test per annum) put well owners at risk of exposure to waterborne pathogens, likely due to a poor understanding of pathogen occurrence and movement in and to groundwater sources.

Objective: There has been evolving literature on appropriate duration and step down to oral therapy for treatment of Gram-negative bacteremias.  The primary objective of this study was to describe the practice variability in management of uncomplicated Gram-negative bacteremias in patients who did and did not have an infectious diseases (ID) consult. Methods:  This was a retrospective cohort study of adult patients admitted to hospital between December 2014–2017 with a bacteremia secondary to non-multiple drug resistant Gram-negative bacteria.  Exclusion criteria included death, discharge, or change to palliative status within 48 hours of positive culture.  Febrile neutropenia patients were also excluded.  In addition to descriptive statistics, chi-square test, Wilcoxon two-sample test, and Student t test were done as appropriate. Results: Sixty-one patients were enrolled, with 24.6% (n=15) having an ID consult.  Average age was 68.7±15.3 years, with 45.9% (n=28) being male.  Community-acquired bacteremia was most common (61.7%, n=37), followed by hospital (20.0%, n=12) and long-term care (18.3%, n=11) associated infections.  The most common microorganism was Escherichia coli (n=31) followed by Klebsiella pneumoniae (n=7), with genitourinary source the most frequent etiology (n=31).  Piperacillin/tazobactam was the initial antibiotic in 49.2% (n=30) of patients.  Total days of therapy averaged 16.1±5.4 versus 13.3±3.7 in patients with and without an ID consult, respectively (p=0.087).  The percent of total therapy provided intravenously was 52.7% in the ID consult group versus 44.5% in those without a consult (p=0.1334).  Both mortality in-hospital (p=0.5637) and readmissions to hospital at 30 days (p=0.1763) were not statistically significant between groups. Conclusions:  Practice patterns in regards to total duration of therapy and proportion of therapy provided intravenously were similar in patients with uncomplicated Gram-negative bacteremias with and without an ID consult.  The majority of patients were treated initially with broad spectrum agents with total durations longer than 7 days.  

Objective: Susceptibility testing of HI is problematic and does not correlate well with PBP3 mutations in BLNAR strains. We tested BLN HI strains for the presence of recognized PBP3 mutations and evaluated the EUCAST Penicillin screen test for predicting Amp resistance. Method:  Amplification products of part of the ftsI gene of PBP3 for 120 BL negative HI was sequenced as previously described and compared to the ftsI gene of the parent Rd strain. Screening for PBP3 mutations using the 1U Pen disk was performed as per EUCAST. Broth microdilution by Sensititre (Thermofisher, CA) was performed as per CLSI. Results: There were 57 BLN HI with no amino acid substitution (AAS) in fstI and for 63 there were 20 AAS patterns organized into 9 groups. The common AAS R517H and N526K were present in 3/40 and 39/40 strains respectively with reduced Amp susceptibility (median MIC 1μg/mL). The sensitivity, specificity and agreement of the Pen screen to predict PBP3 AAS was 95%, 93% 92% respectively. For 32/40 Pen screen positive isolates the Amp MIC was 0.5 – 2.0μg/mL, for 1 >8μg/mL and for 7 <0.25μg/mL. Of these 4 (10%) isolates were non-susceptible (intermediate or resistant) to Amp by CLSI or EUCAST interpretations respectively. Conclusion: The Pen screen test was accurate at predicting the presence of one or more AAS. Although isolates with AAS had reduced susceptibility to Amp compared to wild type strains the clinical relevance of these mutations is not clear since 90% of the BLN strains with a PBP3 mutation were susceptible to Amp by broth microdilution. However, the Pen screen test would be a rapid and cost effective approach for screening HI from sterile sites for Amp resistance for further susceptibility testing.

Background: Surgical site infection associated with shunt placement, treatment for hydrocephalus, is the most common complication and cause of morbidity and mortality. The objective is to answer three questions: a) What is the risk of meningitis after ventricular shunt placement?  b)  What are the risk factors for meningitis? c) What main microorganisms cause meningitis? Methods: Data based on NHSN/CDC protocols were collected between Jul/2015-Jun/2018 from 12 hospitals at Belo Horizonte, Brazil. Outcomes: meningitis, hospital death and total length of hospital stay. We evaluated 26 independent variables by univariate and multivariate analysis. Sample size= 926. Results: 71 cases of meningitis were diagnosed (risk = 7.7% [I.C.95% = 6.1%;9.6%]). Mortality rate in patients, without infection was 10% while hospital death of infected patients was 13% (p=0.544). Hospital length of stay in non-infected patients (days): mean = 21, median = 9, std.dev. = 28; hospital stay in infected patients: mean = 34, median = 27, std. dev. = 37 (p=0.025). Three main risk factors were identified by logistic regression model: age beneath two years (Odds Ratio – OR = 3.20;p<0.001), preoperative hospital length of stay greater than four days (OR = 2.02;p=0.007) and four days post hospital admission, the risk of meningitis is increased from 9% to 18% (p=0.026). From 71 meningitis, in 45 (63%) the etiologic agent was identified: Staphylococcus aureus (33%), Staphylococcus epidermidis (22%), Acinetobacter sp (7%), Enterococcus sp (7%), Escherichia coli (7%), Pseudomonas sp (7%), and other (18%). Conclusion: Two intrinsic risk factors for meningitis post ventricular shunt (under two years old and multiple surgeries), and one extrinsic risk factor, preoperative length of hospital stay, were identified.  Incidence of meningitis decreases when patients receive urgent surgical treatment.

Background: Alveolar echinococcosis (AE) caused by the tapeworm Echinococcus multilocularis is endemic to Canada. While human infection is rare, most cases of AE in humans present in the liver. We describe a rare case of disseminated AE in a patient who presented with focal seizures. Methods: A case encountered in clinical practice is presented. Results: A 74-year-old woman from Saskatchewan presents with focal seizure involving the right hand and face. There is no history of travel outside of the province. The patient has regular contact with hunting dogs. Brain imaging reveals a multicystic mass in the left frontal lobe measuring 28 x 12 x 28 mm with surrounding edema (Figure A). The patient undergoes work-up for potential malignancy revealing multiple cystic lesions in the liver, spleen, kidneys, and pancreas. Open biopsy of the brain lesion is reported as meningioma. Three weeks post-biopsy the patient develops worsening right hemiplegia. MRI brain reveals the left frontal lobe mass has tripled in size (Figure B). Near-complete surgical excision is performed. Histopathology initially queries neurocysticercosis; tissue polymerase chain reaction returns positive for Echinococcus multilocularis. The patient is placed on long-term albendazole therapy with periodic imaging to monitor residual lesions. Conclusion: Echinococcus multilocularis is a parasitic infection endemic to Canada. Diagnosis is based on appropriate history and examination in addition to imaging. Definitive diagnosis can be made by histopathology and molecular confirmation. Complete surgical excision (where possible) and medical therapy are essential to halting further progression of disease. Consideration of the diagnosis is essential to guide proper management and avoid improper technique sampling of lesions which can cause worsening of the disease. 

Background: Salmonella enterica serovar Heidelberg is the third most frequently isolated serovar in Canada and is of particular interest due to its resistance to the cephalosporin class of antimicrobials. Through national surveillance of Canadian S. Heidelberg isolated from poultry and human infections we previously demonstrated that human and animal derived isolates were genetically similar (ST15), and that structurally similar IncI1 plasmids harbouring the bla_CMY-2 gene were predominantly responsible for cephalosporin-resistance. Here, we focus on the impact of a well-characterized and widely disseminated IncI1plasmid (p12-2460) on the core-physiology of a S. Heidelberg isolated from the Canadian national surveillance. Methods: The plasmid was transferred via conjugation to a susceptible and previously whole-genome sequenced (WGS) S. Heidelberg isolate (N13-01291). Illumina WGS was performed on the parent and transconjugant to monitor genetic alteration introduced during conjugation. Growth curves were conducted in technical triplicate over 24-30 hr periods under various conditions including growth at 37°C, 30°C, or 42°C in Luria-Bertani broth, and growth at 37°C in M9 minimal media. For proteomic analysis, proteins were extracted from mid-log phase cultures (OD₆₀₀ ≈ 0.6) and trypsin digested overnight. The peptides were then labeled with tandem mass tags, fractionated, and underwent LC-MS/MS analysis. Results: WGS analysis revealed that the strain pairs were isogenic except for a threonine to serine change in TrmE in the transconjugant. Plasmid p12-2460 reduced the growth rate and maximum cell density under all growth conditions tested, including minimal media, rich media, and different temperatures. 85 chromosomally-encoded proteins had a 2-fold or greater change in the presence of the plasmid. Upregulated proteins included 3 from the Tol/Pal system which is involved in outer membrane integrity and downregulated proteins included 2 porin proteins. Conclusion: The p12-2460 plasmid conferred a growth disadvantage in various conditions and caused changes in chromosomally-encoded protein expression of the host.

Objectives: Determinants and outcomes from serious infections may differ between urban residents as compared to those living in rural and remote areas. The purpose of this study was to compare the population incidence and outcomes of Staphylococcus aureus bloodstream infections (SaBSI) among those residing within a primary tertiary care center versus referral/secondary catchment areas. Methods:  A population-based surveillance cohort design was used. All residents of the western interior of British Columbia who developed an incident community-onset SaBSI between April 1, 2010 and March 31, 2017 were included. The Kamloops local health area (population 2017, 115,482) was classified as the tertiary region and all other areas of the western interior as the non-tertiary region (population 2017: 67, 359).  Results: A total of 288 cases of SaBSI were identified for an overall incidence of 23.1 per 100,000/year; the tertiary and non-tertiary rates were 24.1, and 21.5 per 100,000/year, respectively (incidence rate ratio (IRR): 1.1, 95% confidence interval (CI):0.9-1.4, p-value: 0.2). There was an overall increasing incidence observed over the study period and this was similar for both cohorts. The proportion methicillin-resistant Staphylococcus aureus (MRSA) was higher among tertiary versus non-tertiary cases (22.9% versus 13%; p=0.04). Non-tertiary residents were similar to tertiary residents in age (median 64.6 versus 61.9 years; p=0.7), and median Charlson comorbidity scores (1.5 versus 1; p=0.4). The 30-day case-fatality rate was higher among non-tertiary (19%) as compared to tertiary area (12%) residents but this was not statistically significant (p=0.1). Conclusion: Residence within the tertiary catchment area was associated with similar overall incidence and fatality but a higher proportion of MRSA infection.

Objective: Clinical syndromes associated with Bartonella quintana infection can be insidious and difficult to diagnose for multiple reasons. Clinically, B. quintana can manifest as asymptomatic bacteremia or with subtle subacute constitutional symptoms. The mainstay of literature surrounding B. quintana endocarditis is from Europe and developing nations. Herein we describe a case of native valve endocarditis secondary to B. quintana in a homeless male with pre-existing valvular disease and undertake a comprehensive literature review of documented B. quintana endocarditis in North America. Methods: A retrospective analysis via a comprehensive literature search was completed using MEDLINE publications from 1946-present and the PubMed database using the keywords “Bartonella quintana” and “endocarditis.” All published cases of B. quintana endocarditis from North America underwent content analysis and data reduction in an attempt to identify common characteristics.  Such cases were analyzed while redundant cases were excluded. Results: Twelve patients (median age 52 (IQR 44-56, 83.3% male) each had an episode of B. quintana endocarditis. Of the 7 cases where outcome was known, one patient died and six were cured. All cases of B. quintana endocarditis had aortic valve involvement whereas only three cases had pre-existing valvulopathies. Of those with B. quintana endocarditis, 16.7% were HIV positive, 50.0% were homeless and 63.5% met criteria for alcohol abuse. In 8 patients who received antibiotics, median duration of therapy was 122 days (IQR 42-172) with 58.3% of cases going to surgery. Conclusions: This review represents the most up-to-date and comprehensive summary of B. quintana endocarditis cases within North America. It should be considered in homeless individuals even without pre-existing cardiac valvulopathy. Outbreaks of asymptomatic chronic Bartonella bacteremia within the homeless population coupled with challenges in finances and medication compliance makes infection prevention and control and public health measures important considerations. Surgery is almost always needed for cure. 

Objective: Mycobacterium kansasii is considered as more pathogenic amongst non-tuberculous Mycobacteria species (NTM). We aimed to characterize clinical and microbiological features of M. kansasii cases in B.C. Methods: Retrospective chart review was conducted on patients with positive M. kansasii cultures from 2006 to 2018. Student’s T-test and Fisher’s exact test were used for statistical analyses. Results: There were 48 cases of M. kansasii infection, of which 43 had clinical information available. Twelve patients (28%) underwent treatment; median duration was 12 months. Of the 8 patients with known follow-up, 6 reported clinical improvement, 1 did not improve, and 1 one was always asymptomatic. Between the treated and untreated cases, there were no statistically significant differences in respiratory and constitutional symptoms, radiographic findings or underlying lung disease (Table 1). The most common reasons for not treating M. kansasii infection were absent/improved symptoms (10 of 21), or severe comorbidities precluding therapy (7 of 23). The latter trended towards significance (p=0.07) as compared to treated group. All three isolates with susceptibility testing were sensitive to rifampin. Empiric anti-mycobacterial regimens were chosen for the remaining patients. Conclusion: M. kansasii infection in B.C. is uncommon and few diagnosed patients receive therapy. Isolates tested to date are susceptible to rifampin, but most clinicians choose treatment regimens empirically. Table 1: Characteristics of treated and untreated cases of M. kansasii. Subset of cases where information was available indicated in brackets.

Background: Dengue is an arboviral infection of worldwide importance, mainly transmitted by the mosquito Aedes aegypti, found in tropical and subtropical regions of the world. Objective: To report a case of neonatal dengue as a differential diagnosis with neonatal sepsis, which must be considered in endemic areas. Clinical case: This case report is a male newborn, 9 days old, full term, weighing 3400 grams at birth, admitted to the Emergency Room with fever, jaundice, and rash. Mother with suspected chikungunya. The neonate presented severe thrombocytopenia (64000 platelets / mm³), increased C-reactive protein and positive anti-dengue immunoglobulin M (IgM). This last result was possible only after six days at the hospital. During this period, the neonate was treated with ampicillin and gentamicin according the Institution protocol of neonatal sepsis. After 8 days of hospital stay, the patient was discharged under medical guidelines. Conclusions: The reported cases in the literature show the importance of suspecting the disease in pregnant women, since even when the risk of vertical transmission is low, the prevalence of congenital infection could generate an important demand of health services in endemic areas of the disease. Given the epidemiological situation of dengue in the world and the possibility of complications of the disease, this report emphasizes the importance of the pediatrician to be aware of the possibility of vertical transmission of the virus. With a high index of suspicion, early diagnosis, close monitoring, timely intervention and critical consideration, a successful outcome is possible.

Objectives:  Compare the incidence of O157 STEC and non- O157 STEC detected using a multiplex enteric PCR versus culture methods and describe the clinical characteristics of patients diagnosed with non-0157 STEC. Method: Stool specimens submitted from June to November 2018 from the Hamilton-Niagara Region and North Toronto underwent multiplex bacterial PCR. The molecular targets used were STX1, STX2 and E. coli O157.  A retrospective chart review of patients with 0157 and non-O157 STEC was performed using a standardized data extraction form. Results: Out of a total of 2795 stool specimen, 26 (0.93%) were positive for STEC. Of the 26, 20 (76.9%) were non O157 and 6 were O157. Serotyping was available for some isolates, and included O 113: H4, O26:H11 and O118:H2. In 2017, out of 2795 stool specimens tested by routine culture, four were positive for E. coli O157: H7. There were no non -O157 STECs detected since only sorbitol MacConkey agar plates were used for culture and many non-O157 STEC ferment sorbitol.Clinical information was available for 11 patients of which 9 were positive for non-0157 STEC.  Patients ranged in age from one year three months to 86 years. Several of these patients were < 18 years. Clinical information included: bloody stools in 6 (54.5%), acute kidney injury in 2 (18%), fever in 2 (18%), Hemolytic Uremic Syndrome in 1 (9%). There was travel history in only one patient with non-O157 STEC. Four patients (36.3%) received antibiotics for treatment of diarrheal symptoms before receiving PCR results. Conclusions: In patients presenting with bloody diarrhea a diagnosis of STEC infection should be considered, especially among pediatric patients and antibiotics should be avoided until laboratory confirmation is obtained. Non O157 STEC infections are more common than O157 and laboratories that do not routinely detect non-0157 STEC will miss the majority of STEC infections.                  

Introduction: The natural course of intestinal colonization with CPO is unknown, but carriage may persist for years. Standardized re-screening criteria are currently not available. Objectives: We aimed to (1) perform a review of the evidence-based literature; (2) compare proportions of CPO-colonized individuals at various timepoints after the initial diagnosis of intestinal colonization; and (3) estimate a predictive value of remaining colonized at those timepoints. Methods: PubMed, EMBASE, Web of Science, and Cochrane Library (January 2000 to August 2018) were searched. Randomized-control trials, retrospective and prospective cohort, case series and cross-sectional studies were included. The primary outcomes were proportions of patients who remained CPO-colonized at 1, 3, 6, 9 and 12 months. Mean proportions (µ±SD) were compared using t-test. Studies with the most complete and homogeneous data were chosen to calculate the predictive value of remaining CPO-colonized at subsequent timepoints versus baseline. Re-colonized patients were removed from the pool. The missing data were replaced based on the proportions of CPO-colonized to compensate for drop-out. Results: Twenty-five studies were analyzed. As shown in Figure 1, mean proportions of CPO-colonized versus CPO-decolonized individuals differed significantly at 6 (25.0±16.3% vs 75.0±16.3%; p≤0.0001) and 12 months (21.3±13.9% vs 78.7±13.9%; p≤0.0001). There was no difference between the mean proportions of CPO-colonized individuals at 6 and 12 months (25.0±16.3% vs 21.3±13.9%; p=0.562). Five studies’ results were included in the calculation of the predictive value of remaining CPO-colonized at 6 months versus baseline (16.4%; 95%CI=14.5-18.3%); and at 12 months (13.3%; 95%CI=11.6-15.1%). Conclusions: Study findings suggest similar intestinal colonization rates and predictive values of remaining CPO-colonized at 6 and 12 months. These results will enable the development of an evidence-based and justified rationale to consider CPO re-screening at 6 months, which promises to be a more patient-centered and cost-effective strategy.

Background: Direct, overt observations are the current gold standard for evaluating hand hygiene (HH) compliance rates. However, the Hawthorne Effect affects the accuracy of these values. The aim of this study was to quantify the contribution of the Hawthorne Effect on HH compliance rates at a tertiary care teaching hospital by comparing HH rates from covert and overt direct observations by indication for HH, and across healthcare profession groups, hospital areas, and clinical programs. Methods: Covert HH observations were made by two students over a two-month period at the Royal Alexandra Hospital in Edmonton, Alberta. Students received the same training as HH reviewers performing overt observations but were disguised as hospital staff to ensure they could move freely throughout the units without being questioned. Overt observations were simultaneously occurring on the same units. Compliance was defined as appropriate hand washing with either alcohol-based hand rub or soap and water during one of the four moments of HH as per the Alberta Health Services provincial guidelines (based on the Canadian Patient Safety Institute guidelines). Results were assessed using the two-tailed Fisher’s exact test, with a significance level of P<0.05. Results: There were a total of 3,078 covert observations on 27 inpatient units and the Emergency Department. Overall compliance was 57.9% (1,782/3,078). There were 26,253 overt observations over the previous 12-month period with a compliance rate of 86.4% (22,683/26,253; P<0.05). The covert HH compliance rates were significantly lower for each moment of HH, across healthcare profession groups and clinical programs, and in different hospital units. Conclusion: The Hawthorne Effect contributed to a 28.5% overestimation of HH compliance rates using overt observations. This must be considered both when interpreting HH data using the overt method, and when considering subsequent interventions to improve HH rates.   

Alveolar echinococcosis (AE) is a life-threatening disease caused by the zoonotic cestode Echinococcus multilocularis. In North America, Echinococcus multilocularis infection is rare, with only three reports of autochthonous cases of AE outside of Alaska and Northern Canada in the literature. We describe two cases of locally acquired AE in southwestern Canada, which, in combination with cases in dogs and high prevalence in Saskatchewan wildlife, suggest that a more pathogenic strain of the parasite may be expanding geographically.  

Two patients presented within a month to the Infectious Disease service in our centre. The first case was a male with a history of chronic lymphocytic lymphoma identified with an enlarging liver mass. After biopsy revealed pathology consistent with AE, the mass was surgically resected, and the diagnosis confirmed by duplex PCR based on nad1 and rrnS loci. Peritoneal metastases were identified at surgery and the patient is currently on indefinite albendazole.

The second case was a female on long-term immunosuppression for a previous diagnosis of transverse myelitis. Further, she had a prior diagnosis of primary biliary cirrhosis based on liver biopsy. The liver mass was identified incidentally after she presented with pyelonephritis. A biopsy was done, and histopathology was identical to the first case. She was placed on albendazole and the lesion was treated with microwave ablation.

Neither patient had a history of travel to a region endemic to AE, and both patients reside in a similar area in southwestern Canada. The source of their infections has not yet been identified. 

Although lack of AE reporting makes it challenging to assess incidence of this disease, its geographic distribution appears to be expanding and diagnosis can be elusive. AE should therefore be considered in immunosuppressed patients with liver masses in Canada, even when there is no travel history to endemic areas.

Background: Hepatitis C (HCV) is a viral illness that infects about 1% of the Canadian population. Recently, there have been positive strides in improving treatment and at mitigating barriers to treatment such as opening treatment options to all individuals despite their level of liver fibrosis.  Methods: This is qualitative study using interview style surveys to understand the patient perspective of their HCV treatment delay before the change in policy. Convenience sampling was used to interview 15 patients about their risk factors, when they were diagnosed and whether or not they were treated for HCV. We compiled answers into various themes. Interviews were conducted from October 2017 until February 2018. Results: Of the 15 study participants, 13 were male and 2 were female. All patients were above the age of 40. Out of the sample, 9 were not undergoing treatment despite having HCV. Patients were allowed to give multiple reasons for their HCV treatment delay. The most common self-reported reasons were addiction to recreational substances, their disease not being progress enough, and too much paperwork. Conclusion: This study brings to light a different perspective when discussing HCV treatment barriers and delay. Our findings conclude that there is a need for greater education in those that engage in risky behaviours that increase risk to HCV infection. Similarly, education should be targeted at changing risky behaviours after diagnosis to avoid reinfection.

Objective(s): The minimum inhibitory concentration (MIC) of vancomycin has been used as a marker for the response to anti-staphylococcal penicillins in methicillin-susceptible Staphylococcus aureus (MSSA) bacteremia.  This study evaluated whether an increase in vancomycin MIC in MSSA isolates would be associated with a concurrent rise in the MIC of the other antibiotics used to treat MSSA bacteremia, including telavancin, daptomycin, oxacillin and cefazolin. Methods: 305 MSSA strains recovered from hospitalized patients with bacteremia were tested for their susceptibility to vancomycin, telavancin, daptomycin, oxacillin and cefazolin by the Etest according to the manufacturer's instructions. MIC range, MIC mean, MIC50, MIC 90 and MIC 100 for all antibiotics tested were recorded, and data was categorized according to vancomycin MIC (0; 0.75; 1.0; 1.5; and 2.0) or oxacillin MICs (0.038, 0.094, 0.125, 0.19, 0.25, 0.38, 0.5, 0.75, 1.0, and 1.5) to evaluate the effect on MICs to other antibiotics tested. Results:  When the MICs of the comparator antibiotic stratified by vancomycin MIC, a positive association in both means and ranges of telavancin and daptomycin were noted, but no associated was seen between vancomycin and oxacillin or cefazolin. However, when MSSA isolates were stratified by oxacillin MICs, the MICs for cefazolin followed the same trend, but not telavancin or daptomycin. Conclusions: This experiment proved a rise in vancomycin MIC of MSSA is associated with a parallel rise in the MIC of telavancin and daptomycin, and that an increase in MIC to oxacillin in associated with a parallel rise in MIC to cefazolin. Physicians should be aware of these associations when treating MSSA with increasing MICs to vancomycin or oxacillin. Clinical outcomes analyses are underway in cases fitting MSSA bacteremia with elevated MICs to vancomycin or oxacillin. 

Objectives: Streptococcus anginosus is a common pathogen known to cause abscesses in various body systems. However, osteomyelitis is a rare manifestation and S. anginosus causing an extra-osseous mass has not been described. Methods: We present a case of a 62-year old immune-competent male with left thigh pain, fevers and constitutional symptoms. He had an elevated leukocyte count at 22x10⁹/L and CRP of 200mg/L. A MRI found an 18cm mass along his left femur initially thought to be a malignant sarcoma or lymphoma. The diagnosis was questioned when initial biopsy was negative for malignancy cells and showed chronic inflammation. Results: Repeat biopsy of his extra-osseous femoral mass was culture positive for Streptococcus anginosus. Fungal and mycobacterial cultures, along with serology for HIV, Hepatitis, Syphilis were negative. He was initially treated with intravenous Ceftriaxone and received a percutaneous drain insertion. Ultimately, the patient underwent an en-block resection of his femur with sequestrectomy of the bone and antibiotic cement placement.

Figure 1: 18cm extra-osseous, non-cystic soft tissue mass along left femur on MRI

Conclusions: We present to our knowledge the first case of a solidified extra-osseous mass caused by S. anginosus. This was initially mistakenly diagnosed as a presumptive malignant tumour. Through this case, we review the common causative agents for osteomyelitis as well as the rare complications from streptococcal osteomyelitis.

Background: Recently, the WHO has highlighted the need for improved epidemiological surveillance and a better understanding of the health burden imposed by non-influenza RNA respiratory viruses. Human coronaviruses (CoVs) are a major cause of respiratory and gastrointestinal infections with associated morbidity and mortality. The objective of our study was to characterize the epidemiology of CoVs in our tertiary care health centre, and identify clinical correlates of disease severity. Methods: Nasopharyngeal and mid-turbinate swabs and bronchoalveolar lavages were tested for CoVs (OC43, 229E, NL63 and HKU1) by multiplex PCR (xTAG RVP, xTAG RVP FAST v2 or RPP, Luminex). Demographic and clinical data was obtained from the charts of patients admitted between 2010 and 2016, and a univariate analysis was performed. A number of variables consistent with a severe disease burden were evaluated and included (but not limited to): patient outcome, ICU admission, number of symptoms and length of stay. Results: During our study period CoVs represented 11.2% (542/4660) of all positive respiratory virus samples. OC43 was the most commonly identified CoV, followed by 229E, NL63 and HKU1. In contrast to what has been reported in US-based studies, no co-infections with multiple CoVs or other respiratory viruses were detected. The average length of stay for our cohort was 13.5 days, and it was noted that 17.5% required admission to the ICU (mean ICU admission time = 13 days). Interestingly, increased number of symptoms was found to correlate with ICU admission (OR 1.293, 95% CI 1.019-1.640). Overall mortality in our cohort was 7%, although no statistically significant difference in mortality or ICU admission was associated with any specific CoV strain. Conclusions: This study highlights the underappreciated burden of CoVs in a hospital setting and suggests that more comprehensive study of CoV infections at the provincial and national level is necessary.

Objectives: To develop and evaluate the performance of a LAMP assay for detecting Trichomonas vaginalis (TV) in vaginal swabs and compare to microscopy and PCR. Method:  We used 150 selected e-swab vaginal specimens sent for routine TV microscopy at the Hamilton Health Sciences and 7 External Quality Assurance specimens. For LAMP assay, 100μl e-swab fluid was mixed with 100μl lysis solution, then boiled for 10 mins and 5μl of the extract was used as template. A 182bp fragment of TV-specific repeated DNA sequence was amplified. LAMP was carried out at 65 ⁰C for 30 mins using a standard reaction mixture. Amplification was detected using SYBR^® Green in a Genie^® II instrument. The Cut-off detection time was ≤25 min. Limit of detection (LOD) was determined, using known concentrations of serial dilutions of TV culture. For PCR, 200μl e-swab liquid was extracted and eluted in 55μl buffer using esayMag and 10μl was used for PCR. Two PCR methods were used for comparison. An LDT PCR amplified a 96bp 18S rDNA using QuantiTect^® SYBR^® Green-PCR Kit. The Altona RealStar^® Trichomonas vaginalis PCR Kit 1.0 was used according to the Results: 45 clinical and 6 EQA specimens were positive for TV by LAMP and both PCRs. There was no discordance between LAMP and PCR. Only 15 of 45 LAMP/PCR positives (33.3%) were positive by microscopy. The LAMP LOD was 10³ organisms /ml. The turn-around-time for LAMP is under 1 hour as compared to 3 hours for PCR. Conclusion: Detection of TV by LAMP and PCR is more sensitive than microscopy. The performance of LAMP is comparable to PCR. However, LAMP is faster and more cost-effective than PCR. This assay provides promising results for accurate and faster detection of TV compared to PCR at a lower cost.