Effectiveness of Influenza Vaccination During Pregnancy Against Laboratory-Confirmed Seasonal Influenza Among Infants Under 6 Months of Age in Ontario, Canada.

BACKGROUND: Randomized trials conducted in low- and middle-income settings demonstrated efficacy of influenza vaccination during pregnancy against influenza infection among infants <6 months of age. However, vaccine effectiveness (VE) estimates from settings with different population characteristics and influenza seasonality remain limited. METHODS: We conducted a test-negative study in Ontario, Canada. All influenza virus tests among infants <6 months from 2010 to 2019 were identified and linked with health databases to ascertain information on maternal-infant dyads. VE was estimated from the odds ratio for influenza vaccination during pregnancy among cases versus controls, computed using logistic regression with adjustment for potential confounders. RESULTS: Among 23 806 infants tested for influenza, 1783 (7.5%) were positive and 1708 (7.2%) were born to mothers vaccinated against influenza during pregnancy. VE against laboratory-confirmed infant influenza infection was 64% (95% confidence interval [CI], 50%-74%). VE was similar by trimester of vaccination (first/second, 66% [95% CI, 40%-80%]; third, 63% [95% CI, 46%-74%]), infant age at testing (0 to <2 months, 63% [95% CI, 46%-75%]; 2 to <6 months, 64% [95% CI, 36%-79%]), and gestational age at birth (≥37 weeks, 64% [95% CI, 50%-75%]; < 37 weeks, 61% [95% CI, 4%-86%]). VE against influenza hospitalization was 67% (95% CI, 50%-78%). CONCLUSIONS: Influenza vaccination during pregnancy offers effective protection to infants <6 months, for whom vaccines are not currently available.

Detection of OXA-181 Carbapenemase in Shigella flexneri.

We report the detection of OXA-181 carbapenemase in an azithromycin-resistant Shigella spp. bacteria in an immunocompromised patient. The emergence of OXA-181 in Shigella spp. bacteria raises concerns about the global dissemination of carbapenem resistance in Enterobacterales and its implications for the treatment of infections caused by Shigella bacteria.

Measuring waning protection from seasonal influenza vaccination during nine influenza seasons, Ontario, Canada, 2010/11 to 2018/19.

BackgroundWaning immunity from seasonal influenza vaccination can cause suboptimal protection during peak influenza activity. However, vaccine effectiveness studies assessing waning immunity using vaccinated and unvaccinated individuals are subject to biases.AimWe examined the association between time since vaccination and laboratory-confirmed influenza to assess the change in influenza vaccine protection over time.MethodsUsing linked laboratory and health administrative databases in Ontario, Canada, we identified community-dwelling individuals aged ≥ 6 months who received an influenza vaccine before being tested for influenza by RT-PCR during the 2010/11 to 2018/19 influenza seasons. We estimated the adjusted odds ratio (aOR) for laboratory-confirmed influenza by time since vaccination (categorised into intervals) and for every 28 days.ResultsThere were 53,065 individuals who were vaccinated before testing for influenza, with 10,264 (19%) influenza-positive cases. The odds of influenza increased from 1.05 (95% CI: 0.91-1.22) at 42-69 days after vaccination and peaked at 1.27 (95% CI: 1.04-1.55) at 126-153 days when compared with the reference interval (14-41 days). This corresponded to 1.09-times increased odds of influenza every 28 days (aOR = 1.09; 95% CI: 1.04-1.15). Individuals aged 18-64 years showed the greatest decline in protection against influenza A(H1N1) (aORper 28 days = 1.26; 95% CI: 0.97-1.64), whereas for individuals aged ≥ 65 years, it was against influenza A(H3N2) (aORper 28 days = 1.20; 95% CI: 1.08-1.33). We did not observe evidence of waning vaccine protection for individuals aged < 18 years.ConclusionsInfluenza vaccine protection wanes during an influenza season. Understanding the optimal timing of vaccination could ensure robust protection during seasonal influenza activity.

Effectiveness of maternal influenza vaccination during pregnancy against laboratory-confirmed seasonal influenza among infants under 6 months of age in Ontario, Canada.

BACKGROUND: Randomized trials conducted in low- and middle-income settings demonstrated efficacy of influenza vaccination during pregnancy against influenza infection among infants <6 months of age. However, vaccine effectiveness (VE) estimates from settings with different population characteristics and influenza seasonality remain limited. METHODS: We conducted a test-negative study in Ontario, Canada. All influenza virus tests among infants <6 months from 2010-2019 were identified and linked with health databases to ascertain information on maternal-infant dyads. VE was estimated from the odds ratio for influenza vaccination during pregnancy among cases versus controls, computed using logistic regression with adjustment for potential confounders. RESULTS: Among 23,806 infants tested for influenza, 1,783 (7.5%) were positive and 1,708 (7.2%) were born to mothers vaccinated against influenza during pregnancy. VE against laboratory-confirmed infant influenza infection was 64% (95% confidence interval [CI]: 50%-74%). VE was similar by trimester of vaccination (1st/2nd: 66%, 40%-80%; 3rd: 63%, 46%-74%), infant age at testing (0-<2 months: 63%, 46%-75%; 2-<6 months: 64%, 36%-79%), and gestational age at birth (≥37 weeks: 64%, 50%-75%; < 37 weeks: 61%, 4%-86%). VE against influenza hospitalization was 67% (95%CI: 50%-78%). CONCLUSIONS: Influenza vaccination during pregnancy offers effective protection to infants <6 months, for whom vaccines are not currently available.

Viral dynamics of the SARS-CoV-2 Omicron Variant in Paediatric Patients; A prospective cohort study.

BACKGROUND: There are limited data on the viral dynamics of SARS-CoV-2 in children. Understanding viral load changes over the course of illness and duration of viral shedding may provide insight into transmission dynamics to inform public health and infection control decisions. METHODS: We conducted a prospective cohort study of children 18 years and younger with PCR confirmed SARS-CoV-2 between February 1, 2022 and March 14, 2022. SARS-CoV-2 testing occurred on daily samples for 10 days; a subset of participants completed daily rapid antigen testing (RAT). Viral RNA trajectories were described in relation to symptom onset and resolution. The associations between both time since symptom onset/resolution and non-infectious viral load were evaluated using a Cox proportional hazards model. FINDINGS: Among 101 children aged 2 to 17 years, the median time to study-defined non-infectious viral load was 5 days post symptom onset, with 75% meeting this threshold by 7 days, and 90% by 10 days. On the day of and day after symptom resolution, 43 of 87 (49%) and 52 (60%) had met the non-infectious thresholds, respectively. Of the 50 participants completing RAT, positivity at symptom onset and on the day after symptom onset was 67% (16/24) and 75% (14/20). On the first day where the non-infectious threshold was met, 61% (n = 27/44) of participant RAT results were positive. INTERPRETATION: Children often met the study-defined non-infectiousness threshold on the day after symptom resolution. RAT tests were often negative early in the course of illness and should not be relied on to exclude infection. CLINICAL TRIALS REGISTRATION: NCT05240183.

Evaluating the Impact of Statin Use on Influenza Vaccine Effectiveness and Influenza Infection in Older Adults.

BACKGROUND: Older adults are recommended to receive influenza vaccination annually, and many use statins. Statins have immunomodulatory properties that might modify influenza vaccine effectiveness (VE) and alter influenza infection risk. METHODS: Using the test-negative design and linked laboratory and health administrative databases in Ontario, Canada, we estimated VE against laboratory-confirmed influenza among community-dwelling statin users and nonusers aged ≥66 years during the 2010-2011 to 2018-2019 influenza seasons. We also estimated the odds ratio for influenza infection comparing statin users and nonusers by vaccination status. RESULTS: Among persons tested for influenza across the 9 seasons, 54 243 had continuous statin exposure before testing and 48 469 were deemed unexposed. The VE against laboratory-confirmed influenza was similar between statin users and nonusers (17% [95% confidence interval, 13%-20%] and 17% [13%-21%] respectively; test for interaction, P = .87). In both vaccinated and unvaccinated persons, statin users had higher odds of laboratory-confirmed influenza than nonusers (odds ratios for vaccinated and unvaccinated persons 1.15 [95% confidence interval, 1.10-1.21] and 1.15 [1.10-1.20], respectively). These findings were consistent by mean daily dose and statin type. VE did not differ between users and nonusers of other cardiovascular drugs, except for ß-blockers. We did not observe that vaccinated and unvaccinated users of these drugs had increased odds of influenza, except for unvaccinated ß-blocker users. CONCLUSIONS: Influenza VE did not differ between statin users and nonusers. Statin use was associated with increased odds of laboratory-confirmed influenza in vaccinated and unvaccinated persons, but these associations might be affected by residual confounding.

A Practical Workflow for the Identification of Aspergillus, Fusarium, Mucorales by MALDI-TOF MS: Database, Medium, and Incubation Optimization.

There is an increasing body of literature on the utility of MALDI-TOF MS in the identification of filamentous fungi. However, the process still lacks standardization. In this study, we attempted to establish a practical workflow for the identification of three clinically important molds: Aspergillus, Fusarium, and Mucorales using MALDI-TOF MS. We evaluated the performance of Bruker Filamentous Fungi database v3.0 for the identification of these fungi, highlighting when there would be a benefit of using an additional database, the MSI-2 for further identification. We also examined two other variables, namely, medium effect and incubation time on the accuracy of fungal identification. The Bruker database achieved correct species level identification in 85.7% of Aspergillus and 90% of Mucorales, and correct species-complex level in 94.4% of Fusarium. Analysis of spectra using the MSI-2 database would also offer additional value for species identification of Aspergillus species, especially when suspecting species with known identification limits within the Bruker database. This issue would only be of importance in selected cases where species-level identification would impact therapeutic options. Id-Fungi plates (IDFP) had almost equivalent performance to Sabouraud dextrose agar (SDA) for species-level identification of isolates and enabled an easier harvest of the isolates with occasional faster identification. Our study showed accurate identification at 24 h for Fusarium and Mucorales species, but not for Aspergillus species, which generally required 48 h.

A Prolonged Outbreak of Human Adenovirus A31 (HAdV-A31) Infection on a Pediatric Hematopoietic Stem Cell Transplantation Ward with Whole Genome Sequencing Evidence of International Linkages.

A surge in hematopoietic stem cell transplantation (HSCT) human adenovirus A31 (HAdV-A31) infections was initially observed in late 2014/2015 at SickKids (SK) Hospital, Toronto, Canada. In response, enhanced laboratory monitoring for all adenovirus infections was conducted. Positive samples underwent genotyping, viral culture, and, in selected cases, whole-genome sequencing (WGS). HAdV-A31 specimens/DNA obtained from four international pediatric HSCT centers also underwent WGS. During the SK outbreak period (27 October 2014 to 31 October 2018), 17/20 HAdV-A31 isolates formed a distinct clade with 0 to 8 mutations between the closest neighbors. Surveillance before and after the outbreak detected six additional HAdV-A31 HSCT cases; three of the four sequenced cases clustered within the outbreak clade. Two SK outbreak isolates were identical to sequences from two patients in an outbreak in England. Three SK non-outbreak sequences also had high sequence similarity to strains from three international centers. Environmental PCR testing of the HSCT ward showed significant adenovirus contamination. Despite intense infection control efforts, we observed re-occurrence of infection with the outbreak strain. Severe but nonfatal infection was observed more commonly with HAdV-A31 compared to other genotypes, except HAdV-C1. Our findings strongly implicate nosocomial spread of HAdV-A31 over 10 years on a HSCT unit and demonstrate the value of WGS in defining and mapping the outbreak. Close linkages among strains in different countries suggest international dissemination, though the mechanism is undetermined. This large, extended outbreak emphasizes the pre-eminent role of HAdV-A31 in causing intractable pediatric HSCT outbreaks of severe illness worldwide.

Assessment of panfungal PCR performance with formalin-fixed paraffin-embedded tissue specimens†.

We reviewed the performance of a panfungal ITS-2 PCR and Sanger sequencing assay performed on 88 FFPE specimens at The Hospital for Sick Children (Toronto, Canada) in 2019. A potential fungal pathogen was identified by ITS PCR in 62.7 and 2.9% of positive and negative direct slide examination of tissue specimens, respectively. ITS amplicons were detected in 87/88 specimens, with 53/88 (60.2%) considered as 'positive-contaminants' and 34/88 (38.6%) as 'positive-potential pathogen' upon sequencing. Potential pathogens included Blastomyces dermatitidis (17.1%), Cryptococcus neoformans (17.1%), Histoplasma capsulatum (14.3%) and Mucormycetes (11.4%). Laboratories should only perform ITS PCR on FFPE tissues if fungal elements have been confirmed on histopathology slides. LAY SUMMARY: In this study, we examined how well a DNA-based test could detect DNA from fungi in archived human biopsy tissues. The best performance was achieved if fungi were seen in the tissue under a microscope before being tested. Our results indicate that we should only use this test if these conditions are met.

Influenza Vaccine Effectiveness Against All-Cause Mortality Following Laboratory-Confirmed Influenza in Older Adults, 2010-2011 to 2015-2016 Seasons in Ontario, Canada.

BACKGROUND: Older adults are at increased risk of mortality from influenza infections. We estimated influenza vaccine effectiveness (VE) against mortality following laboratory-confirmed influenza. METHODS: Using a test-negative design study and linked laboratory and health administrative databases in Ontario, Canada, we estimated VE against all-cause mortality following laboratory-confirmed influenza for community-dwelling adults aged >65 years during the 2010-2011 to 2015-2016 influenza seasons. RESULTS: Among 54 116 older adults tested for influenza across the 6 seasons, 6837 died within 30 days of specimen collection. Thirteen percent (925 individuals) tested positive for influenza, and 50.6% were considered vaccinated for that season. Only 23.2% of influenza test-positive cases had influenza recorded as their underlying cause of death. Before and after multivariable adjustment, we estimated VE against all-cause mortality following laboratory-confirmed influenza to be 20% (95% confidence interval [CI], 8%-30%) and 20% (95% CI, 7%-30%), respectively. This estimate increased to 34% after correcting for influenza vaccination exposure misclassification. We observed significant VE against deaths following influenza confirmation during 2014-2015 (VE = 26% [95% CI, 5%-42%]). We also observed significant VE against deaths following confirmation of influenza A/H1N1 and A/H3N2, and against deaths with COPD as the underlying cause. CONCLUSIONS: These results support the importance of influenza vaccination in older adults, who account for most influenza-associated deaths annually.

Solid organ transplant-specific antibiogram in a tertiary pediatric hospital in Canada.

SOT recipients are more vulnerable to infections with antimicrobial-resistant organisms, and therefore, it may be useful for transplant centers to create transplant-specific antibiograms to direct empirical antimicrobial regimens and monitor trends in antimicrobial resistance. SOT-specific antibiograms were created using antimicrobial susceptibility data on isolates from 2012 to 2018 at The Hospital for Sick Children, Toronto, Ontario, Canada. The CLSI guidelines were followed to generate the antibiograms except that results from 2 years of data were pooled on a rolling basis to achieve larger sample sizes. The 3 most frequent organisms in one analysis period of the SOT antibiogram were Escherichia coli (average sample size ±standard deviation; n = 28.7 ± 3.8), Staphylococcus aureus (n = 27.8 ± 5.0), and Pseudomonas aeruginosa (non-CF) (n = 19.8 ± 8.8). For E.coli, susceptibilities in the SOT antibiogram were significantly lower than those in the hospital-wide antibiogram in 2017-2018 for ampicillin (27% vs 47%; p = .014), piperacillin/tazobactam (55% vs 88%; p < .001), cefotaxime (59% vs 89%; p < .001), ciprofloxacin (71% vs 88%; p = .007), and trimethoprim-sulfamethoxazole (41% vs 69%; p = .001), but not significantly different for aminoglycosides and meropenem. In the SOT antibiogram of E. coli, decreased susceptibility trend was confirmed in some antibiotics, including piperacillin/tazobactam (83% in 2012-2013 vs 55% in 2017-2018). At our center, the solid organ transplant-specific antibiogram revealed important differences in E. coli susceptibilities and trends in antimicrobial resistance. Developing a SOT antibiogram will assist in revising and improving empiric treatment guidelines as well as monitoring antimicrobial resistance in this population.

Influenza Vaccine Effectiveness in Preventing Hospitalizations in Older Patients With Chronic Obstructive Pulmonary Disease.

BACKGROUND: Annual influenza immunization is recommended for people with chronic obstructive pulmonary disease (COPD) by all major COPD clinical practice guidelines. We sought to determine the seasonal influenza vaccine effectiveness (VE) against laboratory-confirmed influenza-associated hospitalizations among older adults with COPD. METHODS: We conducted a test-negative study of influenza VE in community-dwelling older adults with COPD in Ontario, Canada using health administrative data and respiratory specimens collected from patients tested for influenza during the 2010-11 to 2015-16 influenza seasons. Influenza vaccination was ascertained from physician and pharmacist billing claims. Multivariable logistic regression was used to estimate the adjusted odds ratio of influenza vaccination in people with, compared to those without, laboratory-confirmed influenza. RESULTS: Receipt of seasonal influenza vaccine was associated with an adjusted 22% (95% confidence interval [CI], 15%-27%) reduction in laboratory-confirmed influenza-associated hospitalization. Adjustment for potential misclassification of vaccination status increased this to 43% (95% CI, 35%-52%). Vaccine effectiveness was not found to vary by patient- or influenza-related variables. CONCLUSIONS: During the studied influenza seasons, influenza vaccination was at least modestly effective in reducing laboratory-confirmed influenza-associated hospitalizations in people with COPD. The imperfect effectiveness emphasizes the need for better influenza vaccines and other preventive strategies.

Comparison of Upper Respiratory Viral Load Distributions in Asymptomatic and Symptomatic Children Diagnosed with SARS-CoV-2 Infection in Pediatric Hospital Testing Programs.

The distribution of upper respiratory viral loads (VL) in asymptomatic children infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is unknown. We assessed PCR cycle threshold (Ct) values and estimated VL in infected asymptomatic children diagnosed in nine pediatric hospital testing programs. Records for asymptomatic and symptomatic patients with positive clinical SARS-CoV-2 tests were reviewed. Ct values were (i) adjusted by centering each value around the institutional median Ct value from symptomatic children tested with that assay and (ii) converted to estimated VL (numbers of copies per milliliter) using internal or manufacturer data. Adjusted Ct values and estimated VL for asymptomatic versus symptomatic children (118 asymptomatic versus 197 symptomatic children aged 0 to 4 years, 79 asymptomatic versus 97 symptomatic children aged 5 to 9 years, 69 asymptomatic versus 75 symptomatic children aged 10 to 13 years, 73 asymptomatic versus 109 symptomatic children aged 14 to 17 years) were compared. The median adjusted Ct value for asymptomatic children was 10.3 cycles higher than for symptomatic children (P < 0.0001), and VL were 3 to 4 logs lower than for symptomatic children (P < 0.0001); differences were consistent (P < 0.0001) across all four age brackets. These differences were consistent across all institutions and by sex, ethnicity, and race. Asymptomatic children with diabetes (odds ratio [OR], 6.5; P = 0.01), a recent contact (OR, 2.3; P = 0.02), and testing for surveillance (OR, 2.7; P = 0.005) had higher estimated risks of having a Ct value in the lowest quartile than children without, while an immunocompromised status had no effect. Children with asymptomatic SARS-CoV-2 infection had lower levels of virus in their nasopharynx/oropharynx than symptomatic children, but the timing of infection relative to diagnosis likely impacted levels in asymptomatic children. Caution is recommended when choosing diagnostic tests for screening of asymptomatic children.

Development and utility assessment of a machine learning bloodstream infection classifier in pediatric patients receiving cancer treatments.

BACKGROUND: Objectives were to build a machine learning algorithm to identify bloodstream infection (BSI) among pediatric patients with cancer and hematopoietic stem cell transplantation (HSCT) recipients, and to compare this approach with presence of neutropenia to identify BSI. METHODS: We included patients 0-18 years of age at cancer diagnosis or HSCT between January 2009 and November 2018. Eligible blood cultures were those with no previous blood culture (regardless of result) within 7 days. The primary outcome was BSI. Four machine learning algorithms were used: elastic net, support vector machine and two implementations of gradient boosting machine (GBM and XGBoost). Model training and evaluation were performed using temporally disjoint training (60%), validation (20%) and test (20%) sets. The best model was compared to neutropenia alone in the test set. RESULTS: Of 11,183 eligible blood cultures, 624 (5.6%) were positive. The best model in the validation set was GBM, which achieved an area-under-the-receiver-operator-curve (AUROC) of 0.74 in the test set. Among the 2236 in the test set, the number of false positives and specificity of GBM vs. neutropenia were 508 vs. 592 and 0.76 vs. 0.72 respectively. Among 139 test set BSIs, six (4.3%) non-neutropenic patients were identified by GBM. All received antibiotics prior to culture result availability. CONCLUSIONS: We developed a machine learning algorithm to classify BSI. GBM achieved an AUROC of 0.74 and identified 4.3% additional true cases in the test set. The machine learning algorithm did not perform substantially better than using presence of neutropenia alone to predict BSI.

The effect of human leukocyte antigen A1 and B35-Cw4 on sustained BK polyomavirus DNAemia after renal transplantation.

Human leukocyte antigen (HLA) class I presentation pathway plays a central role in natural killer (NK) cell and cytotoxic T-cell activities against BK polyomavirus (BKPyV) DNAemia. We determined the risk of sustained BKPyV DNAemia in 175 consecutive renal transplant recipients considering the simultaneous effect of donor/recipient HLA class I antigens and pre- or post-transplant variables. Median (IQR) age was 53 (44-64) years, and 37% of patients were female. 40 patients (22.9%) developed sustained BKPyV DNAemia [median (IQR) viral load: 9740 (4350-17 125) copies/ml]. In the Cox proportional hazard analysis, HLA-A1 (HR: 3.06, 95% CI: 1.51-6.17) and HLA-B35-Cw4 (HR: 4.63, 95% CI: 2.12-10.14) significantly increased the risk of sustained BKPyV DNAemia, while 2 HLA-C mismatches provided a marginally protective effect (HR: 0.32, 95% CI: 0.10-0.98). HLA-Cw4 is a ligand for NK cell inhibitory receptor, and HLA-B35 is in strong linkage disequilibrium with the HLA-Cw4 allele. The association between HLA-B35-Cw4 expression and sustained BKPyV DNAemia supports the important role of cytotoxic T cells and NK cells that would normally control BKPyV activation through engagement with immunoglobulin-like killer receptors (KIRs). Further studies are required to investigate the effect of HLA-C alleles along with NK cell activity against BKPyV DNAemia.

The impact of repeated vaccination using 10-year vaccination history on protection against influenza in older adults: a test-negative design study across the 2010/11 to 2015/16 influenza seasons in Ontario, Canada.

IntroductionAnnual influenza vaccination is recommended for older adults, but evidence regarding the impact of repeated vaccination has been inconclusive.AimWe investigated vaccine effectiveness (VE) against laboratory-confirmed influenza and the impact of repeated vaccination over 10 previous seasons on current season VE among older adults.MethodsWe conducted an observational test-negative study in community-dwelling adults aged > 65 years in Ontario, Canada for the 2010/11 to 2015/16 seasons by linking laboratory and health administrative data. We estimated VE using multivariable logistic regression. We assessed the impact of repeated vaccination by stratifying by previous vaccination history.ResultsWe included 58,304 testing episodes for respiratory viruses, with 11,496 (20%) testing positive for influenza and 31,004 (53%) vaccinated. Adjusted VE against laboratory-confirmed influenza for the six seasons combined was 21% (95% confidence interval (CI): 18 to 24%). Patients who were vaccinated in the current season, but had received no vaccinations in the previous 10 seasons, had higher current season VE (34%; 95%CI: 9 to 52%) than patients who had received 1-3 (26%; 95%CI: 13 to 37%), 4-6 (24%; 95%CI: 15 to 33%), 7-8 (13%; 95%CI: 2 to 22%), or 9-10 (7%; 95%CI: -4 to 16%) vaccinations (trend test p = 0.001). All estimates were higher after correcting for misclassification of current season vaccination status. For patients who were not vaccinated in the current season, residual protection rose significantly with increasing numbers of vaccinations received previously.ConclusionsAlthough VE appeared to decrease with increasing numbers of previous vaccinations, current season vaccination likely provides some protection against influenza regardless of the number of vaccinations received over the previous 10 influenza seasons.

Cumulative Antimicrobial Susceptibility Data from Intensive Care Units at One Institution: Should Data Be Combined?

Cumulative susceptibility test data (CSTD) are used to guide empirical antimicrobial therapy and to track trends in antibiotic resistance. The Clinical and Laboratory Standards Institute recommends reporting CSTD at least annually and sets the minimum number of isolates per reported organism at 30. To comply, many hospitals combine data from multiple intensive care units (ICUs); however, this may not be appropriate to guide empirical therapy because of variations in patient populations. In this study, susceptibility data for two different ICUs at a tertiary care hospital in Toronto, Canada, were used to create a traditional CSTD report, which combined data from different ICUs, and a rolling-average CSTD report, which pooled 2 years of data for each ICU separately. For simplicity, data for only the most common Gram-negative organisms (Escherichia coli,Pseudomonas aeruginosa) and the most relevant antibiotics (ciprofloxacin, piperacillin-tazobactam) were examined. With the rolling-average method, significant differences in susceptibility were seen between the ICUs in 50% of the organism-antimicrobial combinations. Furthermore, the 3% median year-over-year difference in susceptibilities seen for the 16 organism-antibiotic combinations by using the traditional method was lower than the 14% median difference seen for the 20 between-ICU within-year comparisons obtained using the rolling-average method. Changes in our selection of empirical antibiotics resulted from this revised approach, and our results suggest that pooling data from ICUs with different patient populations may not be appropriate. A rolling-average method may be an appropriate strategy for the creation of individual-unit CSTD reports.