Search

Background Understanding the burden of influenza is necessary to optimize recommendations for influenza vaccination. We describe the epidemiology of severe influenza in 50- to 64-year-old residents of metropolitan Toronto and Peel region, Canada, over 7 influenza seasons. Methods Prospective population-based surveillance for hospitalization associated with laboratory-confirmed influenza was conducted from September 2010 to August 2017. Conditions increasing risk of influenza complications were as defined by Canada's National Advisory Committee on Immunization. Age-specific prevalence of medical conditions was estimated using Ontario health administrative data. Population rates were estimated using Statistics Canada data. Results Over 7 seasons, 1228 hospitalizations occurred in patients aged 50–64 years: 40% due to A(H3N2), 30% A(H1N1), and 22% influenza B. The average annual hospitalization rate was 15.6, 20.9, and 33.2 per 100 000 in patients aged 50–54, 55–59, and 60–64 years, respectively; average annual mortality was 0.9/100 000. Overall, 33% of patients had received current season influenza vaccine; 963 (86%) had ≥1 underlying condition increasing influenza complication risk. The most common underlying medical conditions were chronic lung disease (38%) and diabetes mellitus (31%); 25% of patients were immunocompromised. The average annual hospitalization rate was 6.1/100 000 in those without and 41/100 000 in those with any underlying condition, and highest in those with renal disease or immunocompromise (138 and 281 per 100 000, respectively). The case fatality rate in hospitalized patients was 4.4%; median length of stay was 4 days (interquartile range, 2–8 days). Conclusions The burden of severe influenza in 50- to 64-year-olds remains significant despite our universal publicly funded vaccination program. These data may assist in improving estimates of the cost-effectiveness of new strategies to reduce this burden.

We enrolled 91 consecutive inpatients with COVID-19 at 6 hospitals in Toronto, Canada, and tested 1 nasopharyngeal swab/saliva sample pair from each patient using real-time RT-PCR for severe acute respiratory syndrome coronavirus 2. Sensitivity was 89% for nasopharyngeal swabs and 72% for saliva (P = .02). Difference in sensitivity was greatest for sample pairs collected later in illness.

Background Numerous factors that affect the presentation and severity of pneumococcal disease. Several studies in the pre-PCV era demonstrated that organism characteristics, including serotype, are associated with variability in disease presentation and severity. We undertook an analysis of population based surveillance for invasive pneumococcal disease (IPD) to assess whether herd immunity from PCVs will change the presentation and severity of IPD in adults Methods TIBDN has performed population-based surveillance for IPD in Toronto and Peel region (pop’n 4.5M) since 1995. All sterile site isolates of S. pneumoniae are reported to a central study laboratory, isolates are serotyped, and clinical and vaccination data are collected via patient and physician interview and chart review. Population data are obtained from Statistics Canada. Backwards stepwise logistic regression assessed patient characteristics, illness features, and isolate factors associated with clinical presentation and case fatality. Results Between 1995 and 2018, 8815 episodes of IPD were identified in adults. Patients infected with PCV10not7 serotypes were younger, more likely male and without underlying illness. Patients with infections due to non-vaccine types were more likely to be immunocompromised. Case fatality in IPD declined from 177/754 in 1995/6 to 113/554 in 2017/8; OR 0.67, 95%CI0.51-0.86, P< .0001) and in all serotype groups (Figure). In multivariable models adjusted for host factors, relative to infections caused PCV7 serotypes, those caused by PCV10not7 were less likely to be fatal (OR 0.65, 95%CI 0.46–0.91); those caused by PCV13not10 were more likely to be fatal (OR 1.6, 95%CI 1.3–1.9). Bacteremic pneumonia as a proportion of presentations is highest in IPD due to PCV10not7 and PCV13not10 serotypes (85% and 83%, respectively), and lowest in IPD due to non-vaccine serotypes and PCV20not15 (59% and 68%, P< .0001). Meningitis is least common in IPD due to PCV10not7 serotypes (2.6%), and highest in cases due to non-vaccine types and PCV20not15 (9.0% and 8.0%, respectively, P< .0001). FIgure Conclusion In our population, herd immunity from PCVs will result in a higher proportion of adult IPD occurring in immunocompromised cases, and a shift from bacteremic pneumonia to bacteremia without focus and meningitis. Disclosures Allison McGeer, MD, FRCPC, GlaxoSmithKline (Advisor or Review Panel member, Research Grant or Support)Merck (Advisor or Review Panel member, Research Grant or Support)Pfizer (Research Grant or Support)

Background Pneumococcal disease burden is difficult to quantify due to limited data regarding non-bacteremic disease. We assessed serotype-specific differences in pneumococcal disease presentations in adults in Toronto, Canada. Methods From 2003 to 2011, population-based surveillance for invasive pneumococcal disease was conducted and respiratory pneumococcal isolates collected in Metropolitan Toronto/Peel Region, Canada. Episodes of care were classified into disease categories. Results Of 3105 eligible cases of IPD, 2060 cases were bacteremic pneumonia, and 1045 bacteremia without pneumonia. Of 2751 eligible respiratory cases, 1542 (56.0%) were non-bacteremic pneumonia (NBPP), 467 (17.0%) were other acute respiratory infection (oARI), and 742 (27.0%) were isolates representing colonization. Serotypes 3 (11.3%), 19A (8.4%) and 22F (6.2%) were the most common; serotypes 1,5, and 8 were rare. Serotypes 4, 14, 7F, 9V, 12F, 14, 19A and 6C were over-represented in bacteremic disease, and serotypes 3, 6A, 11A, 19F, 23A, 23F, 35B, 35F were more common in NBPP. The proportion of cases due to PCV7 serotypes declined from 48.7% to 8.7% in bacteremic pneumonia, from 35.3% to 10.9% in NBPP, from 34.2% to 7.5% in oARI, and from 38.7% to 12.2% in colonizing isolates. In 2010–2011, PCV13 serotypes accounted for 62.6% of isolates associated with bacteremic pneumonia, 42.0% of bacteremia without pneumonia, 41.1% of NBPP, 25.7% of oARI, and 32.9% of colonizing isolates. Conclusions Serotype distributions differ significantly in different presentations of pneumococcal disease. Herd protection due to PCV7 has changed serotype distribution, but PCV13 serotypes remain important in all categories of disease.

Background The number of invasive group A Streptococcus (iGAS) infections due to hitherto extremely rare type emm74 strains has increased in several Canadian provinces since late 2015. We hypothesized that the cases recorded in the different provinces are linked and caused by strains of an emm74 clone that recently emerged and expanded explosively. Methods We analyzed both active and passive surveillance data for iGAS infections and used whole-genome sequencing to investigate the phylogenetic relationships of the emm74 strains responsible for these invasive infections country-wide. Results Genome analysis showed that highly clonal emm74 strains, genetically different from emm74 organisms previously circulating in Canada, were responsible for a country-wide epidemic of >160 invasive disease cases. The emerging clone belonged to multilocus sequence typing ST120. The analysis also revealed dissemination patterns of emm74 subclonal lineages across Canadian provinces. Clinical data analysis indicated that the emm74 epidemic disproportionally affected middle-aged or older male individuals. Homelessness, alcohol abuse, and intravenous drug usage were significantly associated with invasive emm74 infections. Conclusions In a period of 20 months, an emm74 GAS clone emerged and rapidly spread across several Canadian provinces located more than 4500 km apart, causing invasive infections primarily among disadvantaged persons.

Vancomycin-variable enterococci (VVE) are vanA-positive, vancomycin-susceptible enterococci with the ability to revert to a vancomycin-resistant phenotype on exposure to vancomycin. We sought to assess the prevalence of VVE and to determine clinical characteristics of patients infected with VVE. We prospectively collected Enterococcus faecium sterile site isolates from Toronto Invasive Bacterial Diseases Network hospitals from January 2015 to June 2016 and calculated VVE (defined as vanA-positive, vancomycin-susceptible isolates) prevalence among vanA-containing isolates. We performed chart reviews of VVE and vancomycin-resistant E. faecium (VRE) bacteremias identified from January 2012 to June 2016, and on a random sample of patients with bacteremia due to vanA/vanB-negative, vancomycin-susceptible enterococci (VSE) from January 2015 to June 2016. Clinical characteristics were compared and factors associated with mortality assessed. Because of the potential reversion from VVE to VRE, pulsed-field gel electrophoresis (PFGE) was performed for strains causing breakthrough bacteremia in order to identify relatedness among strains with different phenotypic resistance within the same patient. VVE comprised 47% (18/38) of vanA-positive isolates. The charts of 36 VRE, 25 VVE, and 79 VSE patients were reviewed. Central venous catheter associated bacteremia was more common in VVE (44%) and VRE patients (57%) than in VSE patients (28%) (P = 0.01). The Pitt bacteremia (OR 1.3, P = 0.002) and the Charlson score (OR 1.2, P = 0.008) were the only independent mortality predictors. PFGE of strains causing breakthrough bacteremia showed high within-patient clonality, irrespective of vanA-positivity or vancomycin-susceptibility. A substantial proportion of vanA-positive isolates are VVE and are therefore not detected with conventional selective culture methods. Bacteremia sources of patients with VVE are similar to those infected with VRE. We detected no association between VVE and 30-day mortality or breakthrough bacteremia.

Background In Ontario, Canada, PCV13 is covered for immunocompromised (IC) adults over 50y. PCV13 programs are thought not to be cost-effective in other adults because it is assumed that herd immunity from pediatric vaccination programs (PCV7 since 2005; PCV13 since 2010) will reduce PCV13 disease burden dramatically in adults. We analyzed data from the Toronto Invasive Bacterial Diseases Network (TIBDN) to ask whether PCV13-type invasive pneumococcal disease (IPD) in adults persists in our population. Methods TIBDN performs population-based surveillance for IPD in Toronto+Peel Region, Ontario (pop4.1M). All microbiology laboratories receiving specimens from residents report cases of IPD and submit isolates to a central study lab for serotyping; annual audits are conducted. Demographic, medical and vaccination information are obtained from patients, families and physicians. Population data are from Statistics Canada. Results Since 1995, 10,365 episodes of IPD have been identified; detailed medical information was available for 9,801 (95%) and serotyping for 9411 (91%). Among 8658 adult cases, 4,273 (49%) were in those aged 15–64 years, and 4,285 (51%) in those aged >645 years. The most common diagnoses were pneumonia (5,978/8,025, 74%) and bacteremia without focus (1,030, 13%); 470 (4.6%) cases had meningitis; the case fatality rate (CFR) was 21%. The incidence of disease due to STs in PCV13 in adults declined from 7.0/100,000/year 2001 to 2.9/100,000/year in 2015–2018 and was stable from 2015–2018 (Figure 1). The incidence was > 5/100,000/year in non-IC patients over 65 years, and younger patients with cancer and kidney disease (Figure 2). In IPD from 2015 to 2018, adult patients with PCV13 ST disease were younger (median age 64 years vs. 67 years, P = .03) than other patients; there was no significant difference in the proportion with at least one underlying chronic condition (253, 69% PCV13ST, vs. 541,74% other ST, P = 0.08), or in CFR (59, 16% PCV13 vs. 145, 20% other, P = 0.13). The ST distribution of cases due to PCV13 STs is shown in Figure 3. Conclusion A significant burden of IPD due to PCV13 serotypes persists in adults in our population despite 8 years of routine pediatric PCV13 vaccination. This burden needs to be considered in assessing the value and cost-effectiveness of PCV programs for adults. Disclosures All authors: No reported disclosures.

In multivariable analysis of associations between initial antibiotic therapy and clinical outcomes in 5005 patients with microbiologically confirmed Streptococcus pneumoniae infections, "discordant" empiric antibiotic therapy was not associated with 30-day mortality rate (hazard ratio, 0.94; 95% confidence interval, .67-1.32).

Background Monitoring the incidence and severity of disease due to varied pneumococcal (Pn) serotypes (STs) over time is important in assessing the benefit of Pn vaccines. We describe changes in adult IPD after the 2010 introduction of routine infant PCV13 in Ontario, Canada (PCV13 is funded only for immunocompromised adults ≥50 years as of 2015). Methods TIBDN has conducted population-based surveillance for IPD in Toronto/Peel, Canada (pop 4.3M) since 1995. Cases are reported to a central office; one isolate/case is serotyped. Demographic and clinical data are collected by chart review and patient/family physician interview. Results Of 6,275 episodes of adult IPD, 5,674 (90%) have STs and 6,007 (96%) detailed clinical data. Incidence of IPD decreased from 14.2/100,000/year in 1995 to 6.0/100,000/year in 2013–2017). One thousand two hundred and three (19%) adults with IPD were 15–44 years, 1,889 (30%) were 45–64 years, 3,182 (51%) ≥65 years. Figures 1 and 2 show rates over time by ST group and age. In multivariable analyses, there was no difference across vaccine ST groups (nonvaccine type (NVT) vs. PPV23 not PCV vs. PCV13) in patient age, proportion with ICU admission, requirement for mechanical ventilation (MV), death, length of stay (LOS) or diagnosis of meningitis, except that patients with NVT isolates were more likely to require ICU admission (OR 1.5, 95% CI 1.2,2.0), and to have meningitis (OR 1.9, 95% CI 1.1,3.3). Case fatality declined from 25% (480/1,949) 1995–2001 to 19% (148/763) in 2012–2017 (multivariable OR/year 0.98 95% CI 0.97,0.99); requirements for ICU admission (26–31%; OR/year 1.02, 95% CI 1.01,1.03) and MV (OR/year 18–22%; 1.02, 95% CI 1.01–1.03) increased, LOS did not change. From 2013 to 2017, the distribution of vaccine group STs has not changed: 37% PCV13 (383/1,031); 20% PCV20not13 (205); 9% PPV23not 20 (94), 34% NVT (349). NVT strains include over 23 ST, most commonly 23A (72, 21%), 15A (46, 13%), 35B (37,11%), 6C (36, 10%), 23B (20, 8%). Conclusion In our population, with infant but no routine adult PCV13, the incidence of adult IPD appears to have stabilized, with PCV13 ST strains contributing 37% of IPD. Case fatality has decreased; ICU admission increased. Adult vaccination may be required to further reduce PCV13 ST infections. Figure 1: IPD incidence, adults 15–65 years, by ST group, 2006–2017. Infant PCV7 started 2005 and PCV13 in 2010. Figure 1. IPD incidence, adults ≥65 years, by ST group, 2006–2017. Infant PCV7 started 2005 and PCV13 in 2010. Disclosures A. McGeer, Pfizer: Grant Investigator and Scientific Advisor, Research grant and Research support; Merck: Scientific Advisor, Research support; GlaxoSmithKline: Grant Investigator and Scientific Advisor, Research support.

Background Several studies have documented factors predictive of antimicrobial resistance (AMR) in invasive pneumococcal disease(IPD). However, the implementation of routine pediatric PCV programs, antimicrobial stewardship, and increasing immunocompromised in populating might be expected to change such factors. We report on predictive factors for AMR in IPD from 2012 to 2017. Methods TIBDN performs population-based surveillance for IPD in Toronto/Peel (pop 4.5M). IPD cases are reported to a central office and one isolate/case is serotyped and has antimicrobial susceptibility testing performed by broth microdilution to CLSI standards. Results 2459 cases of IPD were identified from January 2012 to December 2017. Overall rates of resistance to penicillin, macrolides, fluoroquinolones, and TMP-SMX were relatively stable over the course were stable over the study. Risk factors for infection with resistant to penicillin at meningitis breakpoints as opposed to penicillin- susceptible pneumococci were current residence at nursing home (odds ratio [OR], 2.30; P < 0.001), immune compromised status (OR, 1.41; P = 0.012), HIV infection (OR 2.13, P = 0.016), history of receiving PPV23 vaccine (OR 1.38; P = 0.007). Infection with TMP-SMX-resistant pneumococci was associated with HIV infection (OR, 3.2; P = 0.001) and current residence in a nursing home (OR 2.4, P = 0.002). Infection with macrolide-resistant isolates was associated with any use of macrolide 3 months prior to infection (OR, 3.24; P < 0.001), or macrolide treatment failure of the current episode (OR, 6.64; P = 0.003). Infection with levofloxacin-resistant pneumococci was associated with current residence in a nursing home (OR, 13.7; P < .001), and fluorquinolone treatment failure of the current episode (OR 49.4, P = 0.0034). Conclusion Previous same class antibiotic exposure remains a major predictive factor for macrolide resistance. History of treatment failure is a predictive factor for macrolide and fluoroquinolone failure. HIV infection and immune compromise are risk factors for IPD infection with penicillin resistant pneumococci. Hospital acquisition of infection is no longer a risk factor for fluoroquinolone resistance. Disclosures All authors: No reported disclosures.

Prospective population-based surveillance to assess the epidemiology of invasive pneumococcal disease (IPD) in hematopoietic stem cell transplant (HSCT) patients is limited and a comparison to the general population is lacking. By using a population-based Invasive Bacterial Diseases Network surveillance program, we studied the incidence, clinical significance, serotypes and antimicrobial resistance of IPD in a large cohort of adult HSCT patients and the general population. Streptococcus pneumoniae isolates and patient data were collected prospectively from 1995 to 2004. We identified 14 cases of IPD (based on sterile site isolates) in our HSCT population over a 10-year period. This translated to an incidence rate of 347 infections per 100 000 persons per year. This compared to an incidence of 11.5 per 100 000 persons per year in the general population (regression ratio=30.2; 95% confidence interval (CI) 17.8-50.8, P

BACKGROUND We conducted a prospective cohort study to assess the impact of antiviral therapy on outcomes of patients hospitalized with influenza in southern Ontario, Canada. METHODS Patients admitted to Toronto Invasive Bacterial Diseases Network hospitals with laboratory-confirmed influenza from 1 January 2005 through 31 May 2006 were enrolled in the study. Demographic and medical data were collected by patient and physician interview and chart review. The main outcome evaluated was death within 15 days after symptom onset. RESULTS Data were available for 512 of 541 eligible patients. There were 185 children (

Background In Ontario a publicly funded PCV7 infant program (3 + 1 schedule), was introduced in 1/2005, PCV10 in 10/2009 and PCV13 in 11/2010 (2 + 1 schedule with catch-up to 35m). TIBDN performs population-based surveillance for invasive pneumococcal disease (IPD) in Toronto/Peel to evaluate program impact. Methods IPD cases are reported to a central office and one isolate/case is serotyped. Demographic/ clinical data are collected by chart review and patient/physician interview. Results From 1/1995–5/2017, 9727 IPD cases have been identified. Among 910 IPD cases since 2015, 109 (12%) were aged

Background As indications for testing for influenza broaden, influenza is increasingly being diagnosed in association with hospitalization in adults. We report data on the burden of illness associated with laboratory confirmed influenza requiring hospitalization (LCI-H) in Toronto, Canada from 2010–2011 to 2015–2016. Methods TIBDN has performed population-based surveillance for LCI-H in adults (≥15years) in Toronto and Peel Region, Canada since 2005. All positive tests for influenza are reported, patients are approached for consent and data collected by chart review and patient/physician interview. Death within 30 days of hospitalization was considered associated with influenza. Population data were obtained from Statistics Canada, with data by underlying condition obtained from literature review and provincial health administrative data. Results Over 6 seasons, 5,591 LCI-H episodes were identified: 1,094 (20%) H1N1, 2,847 (51%) H3N2, 415 (7%) A(not typed), 1A(H3N2 and H1N1) and 1,235 (22%) B. The median age of patients was 71.4 years, with 69.3% of patients over 65 years of age; 3,015 (53.4%) were female, 2,618 (46.8%) had been vaccinated against influenza, and 683 (12.2%) were admitted from nursing homes. Incidence by age, influenza subtype and season is shown in Figures 1 and 2. The average annual incidence in healthy adults was 5 per 100,000, compared with 0.8 per 1,000 in adults with COPD or diabetes, 1.7 per 1,000 in adults with cardiac disease, and >2 per 1,000 in those with underlying kidney disease or immunosuppression. Overall, 12.8% of patients had a significant non-respiratory complication (eg. myocarditis, stroke, C. difficile infection). 720 (12.9%) required ICU admission, and 414 (7.4%) required mechanical ventilation. The median hospital length of stay was 9.58 days (IQR 3-11). None of 226 cases aged Conclusion Despite vaccination programs, influenza is a very common cause of hospitalization and death in our adult population. While surveillance for LCI-H underestimates the overall burden of influenza, it may nonetheless help to appropriately prioritize preventive programs. Disclosures J. Powis, Merck: Grant Investigator, Research grant. GSK: Grant Investigator, Research grant. Roche: Grant Investigator, Research grant. Synthetic Biologicals: Investigator, Research grant. A. Mcgeer, Hoffman La Roche: Investigator, Research grant. GSK: Investigator, Research grant. sanofi pasteur: Investigator, Research grant.

In 1995, a publicly funded pneumococcal vaccination program for 23-valent polysaccharide vaccine (PPV23) was introduced in Ontario. Conjugate vaccines were authorized in 2001 (PCV7), 2009 (PCV10) and 2010 (PCV13).From 1995-2011, active, population-based surveillance for invasive pneumococcal disease (IPD) was conducted in Metropolitan Toronto and Peel Region, Canada.6404 IPD cases were included. After PPV23 program implementation in 1995, IPD due to PPV23 strains decreased 49% in older adults prior to PCV7 introduction. Estimated PPV23 efficacy in vaccine eligible adults was 42.2% (95% CI; 28.6-53.2%). IPD incidence due to PCV7 serotypes in children5 years decreased significantly after PCV7 authorization and before introduction of a publicly funded PCV7 program. Seven years after PCV7 program implementation, the incidence of IPD due to PCV7 serotypes decreased to zero in children and by 88% in adults, however, overall IPD incidence remained unchanged in adults. In 2011, the incidence of IPD was 4.5 per 100,000 in adults aged 15-64 and 19.9 per 100,000 in adults aged over 65 years, with 45 serotypes causing disease. Between 1995 and 2011, the case fatality rate of IPD in adults decreased 2% per year (95% CI, -0.9% to -3.2%). In multivariable analysis, predictors of mortality included older age, chronic conditions, nursing home residence, current smoking, bacteraemia, and illness due to serotypes 3,11A, 19A, and 19F.While vaccination programs resulted in substantial public health benefits, herd immunity benefits of PCV7 were seen at low pediatric vaccination rates, and the case fatality rate of IPD has decreased, IPD will continue to be a cause of considerable morbidity and mortality in adults.

Prospective population-based surveillance to assess the incidence and impact of invasive pneumococcal disease (IPD) in organ transplant patients is lacking. By using a population-based Invasive Bacterial Diseases Network surveillance program, we studied the incidence, clinical significance, serotypes and antimicrobial resistance pattern of IPD in a large cohort of adult transplant patients and the general population. Streptococcus pneumoniae isolates and patient data were collected prospectively from 1995 to 2004. We identified 21 cases of IPD (based on sterile-site isolates) in our organ transplant population over a 10-year period. This translated to an incidence rate of 146 infections per 100 000 persons per year. This compared to an incidence of 11.5 per 100 000 persons per year in the general population (RR= 12.8; 95% CI 8.1–19.9, p < 0.00001). If nonsterile-site isolates (respiratory tract) were included, the incidence rate in transplant patients was 419 of 100 000 persons per year. Serotypes 23F and 22F were most common, and 85.0% had a serotype included in the 23-valent pneumococcal vaccine. The antimicrobial resistance rates were high, especially for penicillin and trimethoprim-sulfamethoxazole (TMP/SMX), but were not significantly different from the general population. Solid organ transplant recipients are at significantly greater risk for IPD than the general population. Preventative strategies are necessary.

Introduction: There is a paucity of data about the clinical characteristics that help identify patients at high risk of influenza infection upon ICU admission. We aimed to identify predictors of influenza infection in patients admitted to ICUs during the 2007/2008 and 2008/2009 influenza seasons and the second wave of the 2009 H1N1 influenza pandemic as well as to identify populations with increased likelihood of seasonal and pandemic 2009 influenza (pH1N1) infection. Methods: Six Toronto acute care hospitals participated in active surveillance for laboratory-confirmed influenza requiring ICU admission during periods of influenza activity from 2007 to 2009. Nasopharyngeal swabs were obtained from patients who presented to our hospitals with acute respiratory or cardiac illness or febrile illness without a clear nonrespiratory aetiology. Predictors of influenza were assessed by multivariable logistic regression analysis and the likelihood of influenza in different populations was calculated. Results: In 5,482 patients, 126 (2.3%) were found to have influenza. Admission temperature ≥38°C (odds ratio (OR) 4.7 for pH1N1, 2.3 for seasonal influenza) and admission diagnosis of pneumonia or respiratory infection (OR 7.3 for pH1N1, 4.2 for seasonal influenza) were independent predictors for influenza. During the peak weeks of influenza seasons, 17% of afebrile patients and 27% of febrile patients with pneumonia or respiratory infection had influenza. During the second wave of the 2009 pandemic, 26% of afebrile patients and 70% of febrile patients with pneumonia or respiratory infection had influenza. Conclusions: The findings of our study may assist clinicians in decision making regarding optimal management of adult patients admitted to ICUs during future influenza seasons. Influenza testing, empiric antiviral therapy and empiric infection control precautions should be considered in those patients who are admitted during influenza season with a diagnosis of pneumonia or respiratory infection and are either febrile or admitted during weeks of peak influenza activity.