CC BY
0Vaccine xxx (2017) xxx-xxx
Contents lists available at ScienceDirect
Vaccine
journal homepage: www.elsevier.com/locate/vaccine
Burden of vaccine-preventable pneumococcal disease in hospitalized adults: A Canadian Immunization Research Network (CIRN) Serious Outcomes Surveillance (SOS) network study
Jason J. LeBlanca'*, May ElSherifa, Lingyun Yea, Donna MacKinnon-Cameron a, Li Lia, Ardith Ambrose a, Todd F. Hatchettea, Amanda L. Langa, Hayley Gillis a, Irene Martin b, Melissa K. Andrewa, Guy Boivinc, William Bowie d, Karen Green e, Jennie Johnstonef, Mark Loebf, Anne McCarthy g, Allison McGeerh, Sanela Moracaa, Makeda Semreth, Grant Stiverc, Sylvie Trottierc, Louis Valiquette1, Duncan Websterj, Shelly A. McNeil a,\ on behalf of the Serious Outcomes Surveillance (SOS) Network of the Canadian Immunization Research Network (CIRN)
a Canadian Center for Vaccinology (CCfV), IWK Health Centre, Nova Scotia Health Authority (NSHA), and Dalhousie University, Halifax, Nova Scotia (NS), Canada b National Microbiology Laboratory (NML), Winnipeg, MB, Canada c Centre Hospitalier Universitaire de Québec, Québec, Québec (QC), Canada d Vancouver General Hospital, and University of British Columbia, Vancouver, BC, Canada e Mount Sinai Hospital, Toronto, ON, Canada fMcMaster University, Hamilton, ON, Canada g Ottawa Hospital General Campus, Ottawa, ON, Canada h McGill University Health Centre, Montreal, QC, Canada
i Centre Intégré Universitaire de Santé et de Services Sociaux de l'Estrie - Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, QC, Canada j Saint John Regional Hospital, St. John, NB, Canada
ARTICLE INFO
ABSTRACT
Article history:
Received 14 October 2016
Received in revised form 11 May 2017
Accepted 16 May 2017
Available online xxxx
Keywords:
Pneumococcal
Streptococcus pneumoniae
Serotype
Burden
Background: Pneumococcal community acquired pneumonia (CAPSpn) and invasive pneumococcal disease (IPD) cause significant morbidity and mortality worldwide. Although childhood immunization programs have reduced the overall burden of pneumococcal disease, there is insufficient data in Canada to inform immunization policy in immunocompetent adults. This study aimed to describe clinical outcomes of pneumococcal disease in hospitalized Canadian adults, and determine the proportion of cases caused
by vaccine-preventable serotypes.
Methods: Active surveillance for CAPSpn and IPD in hospitalized adults was performed in hospitals across five Canadian provinces from December 2010 to 2013. CAPSpn were identified using sputum culture, blood culture, a commercial pan-pneumococcal urine antigen detection (UAD), or a serotype-specific UAD. The serotype distribution was characterized using Quellung reaction, and PCR-based serotyping on cultured isolates, or using a 13-valent pneumococcal conjugate vaccine (PCV13) serotype-specific UAD assay.
Results and conclusions: In total, 4769 all-cause CAP cases and 81 cases of IPD (non-CAP) were identified. Of the 4769 all-cause CAP cases, a laboratory test for S. pneumoniae was performed in 3851, identifying 14.3% as CAPSpn. Of CAP cases among whom all four diagnostic test were performed, S. pneumoniae was identified in 23.2% (144/621). CAPSpn cases increased with age and the disease burden of illness was evident in terms of requirement for mechanical ventilation, intensive care unit admission, and 30-day mortality. Of serotypeable CAPSpn or IPD results, predominance for serotypes 3, 7F, 19A, and 22F was observed. The proportion of hospitalized CAP cases caused by a PCV13-type S. pneumoniae ranged between 7.0% and 14.8% among cases with at least one test for S. pneumoniae performed or in whom all four diagnostic tests were performed, respectively. Overall, vaccine-preventable pneumococcal CAP
* Corresponding author at: Division of Microbiology, Nova Scotia Health Authority (NSHA), Room 404B, Mackenzie Bldg, Queen Elizabeth II Health Sciences Centre 5788 University Ave, Halifax, NS B3H 1V8, Canada.
E-mail addresses: jason.leblanc@nshealth.ca (J.J. LeBlanc), shelly.mcneil@nshealth.ca (S.A. McNeil). 1 Address: Canadian Center for Vaccinology (CCfV), IWK Health Centre, 4th Floor Goldbloom Pavillion, 5850/5980 University Ave., Halifax, NS B3K 6R8, Canada.
http://dx.doi.org/10.1016/j.vaccine.2017.05.049 0264-410X/© 2017 The Author(s). Published by Elsevier Ltd.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
and IPD were shown to be significant causes of morbidity and mortality in hospitalized Canadian adults in the three years following infant PCV13 immunization programs in Canada.
© 2017 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4XI/).
1. Introduction
Community acquired pneumonia (CAP) is the leading infectious cause of adult mortality in developed countries, and represents an illness of considerable public health concern [1-4]. While the etiology often remains elusive, Streptococcus pneumoniae is the most frequently identified bacterial pathogen, and can represent up to 15% of all-cause CAP [2,5-7]. In addition to CAP, the spectrum of pneumococcal diseases includes non-invasive conditions (e.g. otitis media), and invasive pneumococcal diseases (IPD) such as bac-teremia, meningitis, and empyema [4,5]. Both CAP and IPD are major causes of morbidity and mortality, with approximately 14.6 million cases and 1.6 million deaths occurring worldwide each year [3-5,8]. Incidence rates and mortality risks are highest in children, elderly adults, or individuals with co-morbidities [2,8,9]. Fortunately, pneumococcal vaccines are available to mitigate the burden of pneumococcal infection.
In Canada, both conjugated and the non-conjugated 23-valent pneumococcal polysaccharide vaccines (PPV23) are recommended for children at increased risk for pneumococcal disease for the prevention of IPD [10]. After the introduction of the 7-valent pneumococcal conjugate vaccine (PCV7) in childhood immunization programs in all Canadian provinces by 2005, IPD cases by PCV7-serotypes declined over time [11-13]. However, non-PCV7 sero-types emerged to become predominant, leading to the introduction of the 13-valent pneumococcal conjugate vaccine (PCV13) by 2010 [10,14]. In Canadian adults, PPV23 is broadly recommended for adults aged >65 years and adults of any age at increased risk of invasive pneumococcal infection due to underlying comorbidities, behavioral, and environmental factors [10]. PCV13 was authorized for use in Canadian adults aged >50 years for the prevention of IPD in 2012. In 2013, the National Advisory Committee on Immunization (NACI) recommended the use of PCV13 in adults of any age who are immunocompromised by medications or disease for the prevention of IPD [10].
Recently, the Community-Acquired Pneumonia Immunization Trial in Adults (CAPiTA) demonstrated that PCV13 was effective in preventing both vaccine-type CAP and IPD in immunocompetent adults aged >65 years, leading the US Advisory Committee on Immunization Practices (ACIP) to recommend that all adults >65 years receive a dose of PCV13 followed by PPV23 [15,16]. In 2015, Health Canada expanded the approved indications for PCV13 in adults to include the prevention of vaccine-type CAP and IPD in adults aged >18 years. In 2016, NACI published interim guidance recommending the use of PCV13 in immunocompetent adults aged >65 years, on an individual basis, for the prevention of vaccine-type CAP and IPD [17]. NACI deferred recommendations on the universal use of PCV13 for immunocompetent older adults in publicly funded immunization programs citing the need to better understand the impact of infant PCV13 immunization programs on the serotype distribution of S. pneumoniae in CAP and IPD cases in adults [17].
The Serious Outcomes Surveillance (SOS) Network of the Canadian Immunization Research Network (CIRN) has been conducting active surveillance for CAP and IPD in hospitalized adults since December 2010; here we describe the clinical outcomes associated with all-cause CAP, pneumococcal CAP (CAPSpn), bacteremic CAPSpn, and IPD (non-CAP) [18]. In addition, the proportion of pneumococcal disease caused by vaccine-preventable serotypes and non-vaccine types (NVT) are described.
2. Materials and methods
2.1. Study eligibility
The CIRN SOS Network, formerly the Public Health Agency of Canada/Canadian Institutes of Health Research Influenza Research Network Surveillance (PCIRN), was established in 2009 for influenza virus surveillance [18]. Active CAP and IPD surveillance was initiated in December 2010, and conducted in nine hospitals across five Canadian provinces (BC, ON, QC, NB, and NS). On a daily basis, dedicated surveillance monitors reviewed all adult admissions (aged >16 years) to medical wards or intensive care units (ICU) to identify patients with an acute respiratory illness, who were admitted from the community or another acute care hospital with an admitting diagnosis of influenza, CAP, asthma or acute exacerbation of chronic obstructive pulmonary disease, or who presented with any other respiratory tract infection or symptoms. Patients were eligible for enrollment if they met the study case definition for CAP or had laboratory evidence of IPD. This study reports on cases admitted between December 1, 2010 and December 31, 2013.
2.2. Case definitions
A case was considered CAP if a hospitalized patient presented within 72 h of admission with a new or evolving pulmonary infiltrate on chest radiograph suggesting pneumonia (as interpreted by the treating physician or radiologist) associated with two or more of the following signs or symptoms of pneumonia: temperature >38 °C; cough; sputum production; shortness of breath; pleuritic chest pain; crackles, or consolidation upon chest examination. Patients were excluded if they did not meet the CAP case definition or were discharged prior to identification, refused consent, or were institutionalized such as a long term care (LTC) facility or if incarcerated. CAPSpn was defined as a CAP case with laboratory confirmation of S. pneumoniae using urine antigen detection (UAD) or positive culture from sputum or bronchial alveolar lavage, or from sterile sites [e.g. blood, cerebrospinal fluid, synovial fluid, pleural fluid or peritoneal fluid). Bacteremic CAPSpn was defined as isolation of S. pneumoniae from blood culture in a confirmed CAP case. A case of IPD (non-CAP) was defined as isolation of S. pneumoniae from a normally sterile site in a patient without CAP. Institutionalization was not an exclusion criterion for IPD. The decision to exclude residents of LTC facilities for CAP and not IPD (non-CAP) was made to ensure consistency with accepted case definitions of CAP versus healthcare associated pneumonia as described in the IDSA/ATS Consensus Guidelines on the Management of Community Acquired Pneumonia in Adults (https://www.idsociety. org/uploadedFiles/lDSA/Guidelines-Patient_Care/PDF_Library/CAP %20in%20Adults.pdf).
2.3. Ethics
This study was approved by the research ethics boards (REB) at each participating hospital. Eligible patients or their legally authorized representative provided written informed consent for participation in the study. ln the event that a patient died or consent could not be sought, clinical and outcome data were retrospectively collected from the patient medical record in accordance with local REB approval (n = 606).
J.J. LeBlanc et al./Vaccine xxx (2017) xxx-xxx
2.4. Data collection
Data was collected by trained SOS Network surveillance monitors from the medical record and by patient interview; for those patients enrolled without consent, only data available in the medical record was collected. Details of the presenting respiratory illness signs and symptoms, chest radiography findings, admitting diagnosis, patient demographics, co-morbidities, pneumococcal and influenza immunization status, social history, and functional status defined by the Frailty Index (based on Comprehensive Geriatric Assessment) [19] were collected. In addition, markers of severity and disease outcome were collected including requirement for mechanical ventilation, admission to an Intensive Care Unit (ICU), length of hospital stay (LOS), mortality (assessed at 30 days post-discharge), and difference between the baseline and 30-day frailty index.
2.5. Specimen collection and laboratory testing
When available, respiratory, blood, and sterile site specimens were cultured according to routine practices at each CIRN site laboratory. S. pneumoniae isolates were subjected to Quellung at the National Microbiology Laboratory (Winnipeg, MB), and PCR-based serotyping at the CIRN SOS Network central laboratory (Halifax, NS) [20]. Urine specimens were not part of standard of care, but were collected from CAP cases that provided consent. Urine was stabilized by adding 25 mM PIPES buffer, pH 6.8 (Boston BioProducts, Ashland, MA) prior to freezing. All available urine was tested at the CIRN SOS Network central laboratory (Halifax, NS) using the pan-pneumococcal (UADSpn) Binax NOW Streptococcus pneumoniae Urinary Antigen Test according to manufacturer recommendations. Urine was also tested on a Luminex 2.0 instrument by Pfizer Vaccines Research (Pearl River, NY) using a validated PCV13-specific assay UAD (UADPCV13) [15,21,22]. All specimens and S. pneumoniae isolates were stored at -80 °C at each SOS Network site, and shipped in batches on dry ice to the SOS Network central laboratory in Halifax, NS.
2.6. Calculation of CAP attributed to S. pneumoniae
The proportion of all-cause CAP attributed to S. pneumoniae was calculated based on cases that underwent sputum culture, blood culture, UADSpn or UADPCV13. The proportion of serotypeable S. pneumoniae was defined based on Quellung serotyping and PCR on blood or sputum culture isolates, as well as UADPCV13. Non-serotypeable results were either S. pneumoniae that were isolated, but not available (NA) or not viable (NV) following re-culture, non-typeable (NT) by Quellung or PCR, or were identified solely by UADSpn.
2.7. Statistical analysis
Patient demographics and clinical outcomes were summarized for all-cause CAP, CAPSpn, bacteremic CAPSpn, and IPD. Multivariate analyses were used to evaluate confounding variables using a backward selection model. T-test and ANOVA were used for comparing continuous variables, and chi square was used to compare categorical variables. A p value <0.05 was considered statistically significant. All analyses were performed using Statistical Analysis Software (SAS) version 9 4 (SAS Institute, Cary NC).
3. Results
3.1. Proportions of vaccine-preventable CAP and S. pneumoniae serotype distribution
From December 2010 to December 2013, 11,575 patients were identified and screened for study eligibility. Of these, 800 were
excluded based on lack of consent or were discharged prior to data collection, 877 refused consent, 3802 were ineligible based on case definitions, and 1246 met exclusion criteria. A total of 4769 cases of all-cause CAP and 81 IPD (non-CAP) cases were enrolled (Table 1). Only one IPD case was admitted from LTC.
A laboratory test for S. pneumoniae was performed in 80.8% (3851/4769) of CAP cases; 3110 (65.2%) had blood cultures, 1941 (40.7%) were tested using UADSpn, 1917 (40.2%) were tested using UADPCV13, and 1640 (34.3%) had a sputum culture (Table 2). The proportion of S. pneumoniae-positive results for each individual test ranged from 7.3% to 10.3%; however, laboratory test combinations increased the diagnostic yield. Of the patients having had any tests, 549/3851 (14.3%) were identified as CAPSpn. For the smaller subset CAP cases (n = 621) who had all four tests, S. pneumoniae was identified in 23.2% (Table 2). CAPSpn cases were categorized based on vaccine serotype, age, and disease classifications (Table 2, and Supplementary Tables S1 and S2). The proportion of CAPSpn cases increased with age, and most occurred in the >65 and >50 age cohorts (Table 2).
The predominant serotypes in CAP and IPD were vaccine-preventable serotypes: serotypes 3, 7F, 19A, and 22F (Fig. 1). The serotype distribution in CAPSpn or IPD cases consisted of 7.1% (31/438) PCV7 serotypes, 66.2% (290/438) PCV13 serotypes, and 84.5% (370/438) PPV23 serotypes. Only 12.8% (56/438) were NVTs. Overall, the proportion of hospitalized CAP cases caused by a PCV13-type S. pneumoniae was between 7.0% and 14.8% among cases with at least one or all four diagnostic tests for S. pneumoniae, respectively (Table 2).
3.2. Patient demographics
While the mean age of all-cause CAP cases was >65 years, patients hospitalized with CAPSpn, bacteremic CAPSpn, or IPD were younger (P = 0.01) (Table 1). Approximately 47% and 78% of the CAPSpn cases were in the >65 and >50 age cohorts, respectively. The majority of patients in all groups had underlying medical co-morbidities. Overall, a higher proportion of patients in the all-cause CAP group had pulmonary, cardiac, vascular, and endocrine co-morbidities, than those in the CAPSpn group. Compared to CAP groups, IPD cases were more frequent in immunocompromised patients (P = 0.007). Smoking was less prevalent in the IPD (p =0.007). Patients admitted with all-cause CAP, CAPspn, and bac-teremic CAPspn were moderately frail at baseline. Concomitant influenza infection was diagnosed in approximately 15.0% of CAP cases. In all groups, only approximately half of patients had received a prior pneumococcal vaccine (with PPV23), and approximately 70% had received the current season influenza vaccine. No significant difference was noted between CAP and IPD cases for pneumococcal vaccine (P = 0.55), influenza vaccine (P = 0.55), or receipt of antibiotics within four hours of admission (P = 0.24).
3.3. Clinical outcomes
Overall, 30-day mortality was 12.0% in patients with all-cause CAP, 9.7% in patients with CAPSpn, 20.0% in patients with bacteremic CAPSpn, and 21.0% in patients with IPD (see Table 3). Mortality was highest in the older age groups; the >65 and >50 age cohorts accounted for 66% and 85% of deaths in the CAPSpn group, respectively. Mean LOS ranged from 11.5 days in patients with all-cause CAP to 15.0 days in patients with IPD and, like mortality, increased with age. Compared to patients with all-cause CAP, pneumococcal CAP and IPD were associated with more severe outcomes including admission to ICU, requirement for mechanical ventilation, development of in-hospital complications, and 30-day mortality.
Table 1
Demographics and clinical characteristics of hospitalized adults with CAP or IPD.
Variable Disease category
All-cause CAPa CAP, not Spnb CAPSpnc (n = 549) Bacteremic CAPSpnd IPD (non-CAP)e
(n = 4769) (n = 3302) (n = 238) (n = 81)
Age; mean ± SD (range) 68.7 ±17.0 (17-104) 68.5 ±16.7 (17-104) 62.4 ±17.1 (20- 60.4 ±17.9 (20-99) 62.3 ±16.9 (19-97)
Age 16-49; % 13.9 (662/4769) 13.5 (447/3302) 100) 22.2 (122/549) 28.6 (68/238) 21.0(17/81)
Age 50-64; % 22.0 (1050/4769) 22.6 (745/3302) 30.8 (169/549) 30.3 (72/238) 33.3 (27/81)
Age 65-74; % 21.1 (1005/4769) 21.6 (714/3302) 20.4 (112/549) 17.2 (41/238) 22.2 (18/81)
Age 75-84;% 24.9 (1186/4769) 24.8 (819/3302) 17.5 (96/549) 15.1 (36/238) 14.8 (12/81)
Age 85 + ; % 18.2 (866/4769) 17.5 (577/3302) 9.1 (50/549) 8.8 (21/238) 8.6 (7/81)
Age > 65; % 64.1 (3057/4769) 63.9 (2110/3302) 47.0 (258/549) 41.2 (98/238) 45.7 (37/81)
Age > 50; % 86.1 (4107/4769) 86.5 (2885/3302) 77.8 (427/549) 71.4 (170/238) 79.0 (64/81)
Gender (male); % 53.5 (2553/4769) 53.7 (1772/3302) 53.9 (296/549) 55.0 (131/238) 49.4 (40/81)
> 1 co-morbidity; % 93.4 (4456/4769) 93.8 (3098/3302) 88.2 (484/549) 86.1 (205/238) 87.7 (71/51)
Pulmonary; % 53.8 (2568/4769) 53.7 (1774/3302) 48.3 (265/549) 40.3 (96/238) 27.2 (22/81)
Asthma; % 12.4 (591/4769) 12.7 (421/3302) 14.4 (79/549) 15.6 (37/238) 12.3 (10/81)
Cardiac; % 46.0 (2195/4769) 46.0 (1522/3302) 30.4 (167/549) 29.4 (70/238) 28.4 (23/81)
Vascular; % 63.5 (3030/4769) 63.5 (2097/3302) 51.2 (281/549) 50.4 (120/238) 49.4 (40/81)
Endocrine; % 27.0 (1286/4769) 28.1 (928/3302) 20.9 (115/549) 21.8 (52/238) 25.9 (21/81)
Diabetes mellitus (uncomplicated); % 21.7 (1034/4769) 22.4 (739/3302) 17.5 (96/549) 17.6 (42/238) 23.5 (19/81)
Diabetes mellitus (complicated); % 5.3 (252/4769) 5.7 (189/3302) 3.5 (19/549) 4.2 (10/238) 2.5 (2/81)
Immunocompromisedf; % 29.7 (1416/4769) 31.2 (1030/3302) 27.9 (153/549) 29.8 (71/238) 44.4 (36/81)
Cancer'; % 25.6 (1219/4769) 26.6 (877/3302) 23.3 (128/549) 25.2 (60/238) 42.0 (34/81)
Solid tumor; % 55.0 (670/1219) 55.4 (486/877) 44.5 (57/128) 35.0 (21/60) 35.3 (12/34)
Solid tumor, treated within 5y; % 23.2 (283/1219) 23.4 (205/877) 23.4 (30/128) 20.0 (12/60) 17.6(6/34)
HIV; % 1.4 (46/3240) 1.4 (31/2260) 3.4 (12/349) 3.5 (5/143) 3.6 (2/56)
Splenectomy/functional asplenia; % 0.6 (21/3240) 0.7 (16/2260) 1.1 (4/349) 2.1 (3/143) 7.1 (4/56)
Bone marrow transplant; % 0.7 (23/3240) 0.8 (17/2260) 1.1 (4/349) 2.1 (3/143) 5.4 (3/56)
Solid organ transplant; % 1.5 (49/3240) 1.9 (44/2260) 0.3 (1/349) 0.7 (1/143) 0.0 (0/81)
Sickle cell disease; % 0.2 (5/3240) 0.2 (5/2260) 0.0 (0/349) 0.0 (0/143) 0.0 (0/56)
Current/past smoker; % 69.0 (3023/4384) 68.0(2096/3081) 70.6 (370/524) 64.6 (144/223) 52.3 (34/65)
Obesity (body mass index >30); % 25.1 (1014/4032) 25.8 (734/2841) 21.3 (102/479) 22.9 (46/201) 23.8 (15/63)
Frailty Index, day 0g; Mean ± SD 0.2 ± 0.1 (2307/4769) 0.2 ±0.1 (1688/3302) 0.2 ± 0.1 (214/549) 0.2 ±0.1 (73/238) 0.2 ±0.1 (25/81)
Concomitant influenza infection; % 15.4 (336/2179) 13.9 (220/1588) 15.6 (43/276) 12.5 (12/96) 10.0 (2/20)
Pneumococcal vaccineh; % 53.3 (1606/3014) 53.7 (1200/2235) 40.4 (163/403) 37.7 (60/159) 47.8 (22/46)
Time to first antibiotic <4 h; % 67.3 (2957/4393) 68.3 (2073/3033) 72.6(371/511) 76.0 (171/221) 60.3 (41/68)
Time to first antibiotic <8 h; % 85.8 (3771/4393) 86.3 (2618/3033) 88.3 (451/511) 92.0 (207/221) 77.9 (53/68)
a All-cause CAP meeting clinical case definition. b CAP testing negative by sputum culture, blood culture or UAD. c CAP with positive UAD or S. pneumoniae culture (blood/sputum/sterile site). d CAP with blood culture positive for S. pneumoniae.
e Disease without CAP, but culture positive for S. pneumoniae from a sterile site. f Excludes cancer not treated within 5 years.
g The Frailty Index is defines by scoring 40 health variables, and a score of 0 signifies the absence of deficits and 1 represents presence of all deficits [19]. h The date of PPV23 vaccine administration was not provided in most cases.
Risk factors for serious outcomes including death and requirement for ICU admission or mechanical ventilation in hospitalized patients with CAP are shown in Tables 4 and 5. In the adjusted logistic regression model, increasing age, male gender, presence of underlying comorbidities and immunocompromise were significant predictors of 30-day mortality (Table 4). Only male gender was noted as a significant predictor of ICU admission or mechanical ventilation. Receipt of antibiotics within 4 h of admission was significant but likely a proxy of severity upon admission rather than a predictor of ICU admission or requirement for mechanical ventilation. The fact that increasing age is associated with a decrease in risk of ICU admission likely means that older patients were just not admitted to ICU, and this in itself could be explained by the higher mortality in older individuals (Table 4). In fact, increasing age as well as a history of receipt of pneumococcal vaccine was associated with lower risks for ICU or requirement of mechanical ventilation (Table 5).
4. Discussion
Surveillance of pneumococcal CAP and IPD, and monitoring the epidemiology of S. pneumoniae, is important to assess the impact of pneumococcal vaccines on the burden of pneumococcal disease.
While national surveillance in Canada provides routine monitoring of serotype distribution in IPD cases, little data describing the burden of pneumococcal CAP is available. In this study, performed at the onset of Canadian PCV13 infant routine immunization programs, CAPSpn were shown to contribute markedly to burden of disease in hospitalized Canadian adults, particularly amongst those aged >50 and >65 years. The risk factors for death or severe outcomes among CAP cases were consistent with previous studies, and included increasing age, underlying co-morbidities, and immune compromise [9]. Interestingly, almost one third of CAPSpn cases and half the IPD cases in this study had an underlying condition leading to immunocompromise, which would have already given them an indication for PCV13 and PPV23.
The burden of disease of all-cause CAP in hospitalized Canadian adults was significant and increased with age [1]. This trend was also observed for CAPSpn, and is consistent with a recent incidence reports from the US and UK [2,5,23,24]. The >50 years and >65 year age groups were of particular interest, and showed the highest proportions of CAPSpn in this study. The US ACIP has recommended the routine use of both PCV13 and PPV23 in all adults aged >65 years based upon the PCV13 effectiveness data from the CAPiTA trial while in Canada, NACI has recommended, on an individual basis, that immunocompetent adults aged >65 years (not
J.J. LeBlanc et al./Vaccine xxx (2017) xxx-xxx 5
Table 2
Proportions of pneumococcal CAP among hospitalized adults by laboratory tests performed.
Variable Laboratory test
UADpcvi3 (%) UADspn (%) BC(%) SC (%) Any pneumococcal testf All pneumococcal testg
(%) (%)
Tested/CAP cases 40.2 (1917/ 40.7 (1941/ 65.2 (3110/ 34.3 (1640/ 80.8 (3851/4769) 13.0 (621/4769)
4769) 4769) 4769) 4769)
S. pneumoniae positive/ 10.3 (197/1917) 9.1 (177/1941) 7.4 (238/3110) 7.3 (119/1640) 14.3 (549/3851) 23.2 (144/621)
tested
Serotypeable results
Totala 100.0(197/197) 65.5 (116/177)e 78.6 (187/238) 44.0 (59/119) 69.9 (384/549) 80.6 (116/144)
PCV7-typeb 10.7 (21/197) 8.6 (10/116)e 2.1 (4/187) 6.8 (4/59) 7.6 (29/384) 2.6 (3/116)
PCV13-typeb 98.5 (194/197) 77.6 (90/116)e 57.8 (108/187) 59.3 (35/59) 70.1 (269/384) 73.0 (92/116)
PPV23-typeb 93.9 (185/197)e 90.5 (105/116)e 85.6 (160/187) 76.3 (45/59) 86.5 (332/384) 84.5 (98/116)
NVTb 0.5 (1/197)e 6.9 (8/116)e 12.8 (24/187) 20.3 (12/59) 10.4 (40/384) 10.3 (12/116)
Non-serotypeable results
Totala 0.0 (0/197) 34.5 (61/177) 21.4 (51/238) 50.4 (60/119) 30.1 (165/549) 19.4 (28/144)
NAc 0.0 (0/197) 0.0 (0/61) 98.0(50/51) 95.0 (57/60) 60.6 (100/165) 10.7 (3/28)
NTc 0.0 (0/197) 100.0(61/61) 2.0 (1/51) 5.0 (3/60) 1.8 (3/165) 85.7 (24/28)
NVc N/A 0.0 (0/61) 0.0(0/51) 0.0 (0/60) 37.6 (62/165) 3.6 (1/28)
Observed proportions
PCV7-typed 1.1 (21/1917) 0.5 (10/1941) 0.0 (4/3110) 0.2 (4/1640) 0.5 (29/3851) 0.5 (3/621)
PCV13-typed 10.1 (194/1917) 4.6 (90/1941)e 3.5 (108/3110) 2.1 (35/1640) 7.0 (269/3851) 14.8 (92/621)
PPV23-typed 9.7 (185/1917)e 5.4 (105/1941)e 5.1 (160/3110) 2.7 (45/1640) 8.6 (332/3851) 15.8 (98/621)
NVTd 0.0 (1/1917)e 0.4 (8/1941)e 0.8 (24/3110) 0.7 (12/1640) 1.0 (40/3851) 1.9 (12/621)
Abbreviations: Blood culture (BC); invasive isolates excluding blood (INV); not available (NA); not applicable (N/A); not viable (NV); not typeable (NT); non-vaccine type (NVT); pan-pneumococcal urine antigen detection (UADSpn), PCV13-specific urine antigen detection (UADPCV13); 7-valent pneumococcal conjugate (PCV7), 13-valent
revealed that 26% were immunocompromised, and thus had an indication for pneumococcal vaccination based on current guidelines. (Table S5). Further analyses are underway to assess the potential need for consideration of age-based vaccination recommendations in this age group. In Western Europe, children and adults over the age of 50 years are recognized as being at highest risk of pneumococcal infection, and recommendations for pneumococcal vaccine use in adults aged >50 years are gradually being adopted [25].
It should be noted that the burden of CAPSpn in this study only captured hospitalized adults, and it is possible that some CAPSpn cases were missed if they did not meet the case definition, refused consent, or were discharged prior to enrollment. The timing of specimen collection, prior antibiotics use, prior vaccination (pneumococcal or influenza), and the inherent differences in test characteristics, could all lead to underestimates of CAPSpn disease [26,27]. However, the more challenging aspect of this study is that not all tests for the detection of S. pneumoniae were performed on every case, even if consent was obtained and patients met the case definitions. In older individuals, the proportion of CAPSpn may have been underestimated as these patients were less likely to have had a urine collected for antigen testing given the logistical difficulty of collecting urine from frail, hospitalized adults (Tables S1 and S5). While sputum and blood cultures are routinely recommended in some hospitalized patients such as those in an ICU, urine for pneu-mococcal antigen is not standard of care in most institutions, and CAP is often treated empirically without culture. As such, despite the encouragement of study staff, only on 81% (3851/4769) of CAP cases had at least one test for S. pneumoniae. Of these, S. pneu-moniae was confirmed in only 14.3%. Our best estimate of the contribution of S. pneumoniae to hospitalized CAP of 23.2% was obtained from the 621 patients in whom all tests were performed.
pneumococcal conjugate (PCV13); pneumococcal polysaccharide vaccine (PPV23). a Number of results/total number of S. pneumoniae positive results. b Number of results/total serotypeable results. c Number results/total non-serotypeable results.
d The proportion of results based on the serotypeable CAPSpn cases only. e Deduced from UADPCV13 or culture and Quellung reaction. f Cases where UADPCV13, UADSpn, SC or and BC were performed. g Cases where all four tests were performed.
Fig. 1. S. pneumoniae serotype distribution in hospitalized adults with CAP or IPD. Serotypes obtained from S. pneumoniae culture isolates alone (n = 233; in red); UADPCVi3 alone (n = 132; in blue), and concordant results between isolates and urine (n = 72; in purple). The serotype coverage of pneumococcal vaccines (PCV7, PCV13, and PPV23) and non-vaccine serotypes (NVT) are is indicated.
previously immunized against pneumococcal disease) be given PCV13 followed by PPV23, for the prevention of CAP and IPD [17]. In our study, approximately half (47%) of CAPSpn cases were aged >65. However a greater proportion (78%) of CAPSpn cases was aged >50. Given that 26% of CAPSpn cases occurred in the 50 to 64 year age group, careful consideration of the burden of disease and underlying risk factors for pneumococcal disease in this age group is warranted. Preliminary assessment of this age group
Table 3
Clinical course and outcomes of adults hospitalized with CAP or IPD.
Clinical outcome Disease classification
All CAPa(n = 4769) CAP, not Spnb CAPspnc (n = 549) Bacteremic CAPSpnd IPD (non-CAP)e
(n = 3302) (n = 238) (n = 81)
30-day mortality; N (%) 12.0 (574/4769) 11.4 (375/3302) 9.7 (53/549) 20.0 (28/238) 21.0(17/81)
Age 16-49; % of total deaths 5.4 (31/574) 4.8 (18/375) 9.4 (5/53) 7.1 (2/28) 29.4(5/17)
Age 50-64; % of total deaths 14.6 (84/574) 16.0 (60/375) 18.39 (10/53) 25.0 (7/28) 17.6(3/81)
Age 65-74; % of total deaths 17.6 (101/574) 17.3 (65/375) 22.6 (12/53) 21.4 (6/28) 23.5 (4/81)
Age 75-84; % of total deaths 28.4 (163/574) 30.7 (115/375) 13.2 (7/53) 17.9 (5/28) 5.9 (1/81)
Age 85+; % of total deaths 34.0 (195/574) 31.2 (117/375) 35.8 (19/53) 28.6 (8/28) 23.5 (4/81)
Age > 65; % of total deaths 80.0 (459/574) 79.2 (297/375) 66.0 (35/53) 67.9 (19/28) 52.9 (9/81)
Age > 50; % of total deaths 94.6 (543/574) 95.2 (357/375) 84.9 (45/53) 92.9 (26/28) 70.6 (12/81)
LOS in days; Mean (range; Q75) 11.5 (l-384; 13) 11.8 (1-384; 14) 12.5 (1-105; 14) 13.6 (1-72; 17) 15.0(1-112; 17)
LOS age 16-49; Mean (range; 9.3 (1-85; 6) 10.0(1-85; 6) 8.8 (2-66; 5) 9.1 (2-66; 5) 9.2 (1-30; 6)
median)
LOS age 50-64; Mean (range; 10.9 (1-251; 7) 10.8 (1-251; 7) 13.7 (1-101; 8) 15.2 (1-72; 11) 16.0(1-58; 13)
median)
LOS age 65-74; Mean (range; 11.6 (1-372; 7) 12.2 (1-372; 8) 12.4 (1-81; 8) 15.2 (1-65; 10) 15.0 (3-68; 8)
median)
LOS age 75-84; Mean (range; 11.9 (1-114; 8) 12.3 (1-114; 8) 13.7 (1-76; 8) 17.6 (1-69; 12) 24.5 (3-112; 17)
median)
LOS age 85+; Mean (range; median) 13.3 (1-384; 9) 13.6 (1-384; 9) 15.9 (1-105; 11) 12.6 (1-27; 11) 9.1 (6-11; 10)
ICU admission; N(%) 17.3 (826/4769) 17.7 (584/3302) 29.3 (161/549) 31.9 (76/238) 40.7 (33/81)
Mechanical ventilation; N(%) 11.1 (528/4769) 11.7 (386/3302) 20.2 (111/549) 21.9 (52/238) 33.3 (27/81)
Any complication; N(%) 52.0 (2476/4763) 53.6 (1769/3299) 57.1 (313/548) 62.5 (148/237) 75.3 (61/81)
New arrhythmia; N(%) 10.7 (264/2476) 10.5 (185/1769) 13.4 (42/313) 12.2 (18/148) 8.2 (5/61)
Myocardial infarction; N(%) 4.3 (106/2476) 4.2 (74/1769) 2.6 (8/313) 2.7 (4/148) 1.6 (1/61)
Unstable angina; N(%) 0.7 (16/2476) 0.7 (13/1769) 0.0 (0/313) 0.0 (0/148) 0.0 (0/61)
Congestive heart failure; N(%) 8.6 (124/1440) 9.0 (90/998) 5.5 (9/165) 4.1 (3/74) 9.1 (3/32)
Frailty Index (day 30f); Mean ± SD 0.2 ±0.1 (1347/ 0.2 ±0.1 (1614/3302) 0.2 ± 0.1 (335/ 0.2 ±0.1 (165/238) 0.2 ±0.1 (56/81)
4769) 549)
a All-cause CAP meeting clinical case definition. b CAP testing negative by sputum culture, blood culture or UAD. c CAP and positive UAD or S. pneumoniae culture (blood/sputum/sterile site). d CAP with blood culture positive for S. pneumoniae.
e Disease without CAP and culture positive for S. pneumoniae from sterile site.
f The Frailty Index is defines by scoring 40 health variables, and a score of 0 signifies the absence of deficits and 1 represents presence of all deficits [19].
Table 4
Multivariate analysis of risk factors for death in adults hospitalized with CAP.
Covariate Category Overall P valuea Category P valuea Odds ratio % (95% CI)
Age 16-49 0.000
50-64 0.188 2.1 (1.3-3.6)
65-74 0.817 2.5 (1.5-4.1)
75-84 0.038 3.2 (1.9-5.2)
>84 0.000 6.0 (3.7-10.0)
Gender, male 0.007 1.4 (1.1-1.7)
Immunocompromised 0.000 1.8 (1.4-2.2)
Body mass index (BMI) 18.5-24.99 0.000
<18.5 0.000 2.2 (1.6-3.1)
25-29.99 0.100 0.8 (0.6-1.0)
30-40 0.700 0.9 (0.7-1.2)
>40 0.030 0.5 (0.2-1.0)
The multivariate analysis was based on 394 cases and 3468 controls.
a Overall P values apply to significance of the covariate whereas categorical values denote differences between members within the covariate categories. The comparator used for categorical comparison is denoted by an asterisk (*).
This proportion is consistent with recent meta-analysis in the UK at 27%, and exceeds the US estimate of 5.8% over the same time period [2,24]. However, the true proportions of pneumococcal CAP and vaccine-preventable serotypes in this study would have been better reflected if all four tests were performed on all 4769 CAP cases.
While the diagnostic yield increased with four tests for S. pneu-moniae detection, only three helped characterize the serotype distribution. While other studies have combined testing with cultures and UADSpn to assess the contribution of S. pneumoniae to CAP, the UADPCV13 used in this study enables better estimates of the proportion of hospitalized CAP caused by PCV13 serotypes [5,15]. Of 144
S. pneumoniae cases identified (in the 621 cases in whom all tests were performed), 81% had a serotypeable result and 41% of these were uniquely identified by UADPCV13. In absence of a similar UAD assay for PPV23 and NVTs, the contributions of these sero-types may be underestimated.
The potential benefits of pneumococcal vaccines on the disease burden in Canadian adults are dependent on the presence of vaccine-preventable serotypes. In this study, the majority of cases of CAPSpn and IPD were cause by PCV13- and PPV23-serotypes, and the most predominant serotypes were 3, 7F, 19A, and 22F. A similar serotype distribution was recently noted in Canadian children and US adults aged >50 [28,29]. PCV13-based childhood
Table 5
Multivariate analysis of risk factors for intensive care unit admission or requirement of mechanical ventilation in adults hospitalized with CAP.
Covariate Category Overall P value3 Category P valuea Odds Ratio% (95% CI)
Age 16-49* 0.000
50-64 0.024 0.9 (0.6-1.2)
65-74 0.103 0.8 (0.6-1.2)
75-84 0.038 0.6 (0.4-0.8)
>84 0.001 0.4 (0.3-0.7)
Gender, male 0.005 1.4(1.1-1.7)
Pneumococcal vaccine 0.006 0.7 (0.6-0.9)
Antibiotics within 4 h of admission 0.033 1.3 (1.0-1.7)
The multivariate analysis was based on 388 cases and 2140 controls.
a Overall P values apply to significance of the covariate whereas categorical values denote differences between members within the covariate categories. The comparator used for categorical comparison is denoted by an asterisk (*).
immunization programs were introduced in all Canadian provinces by 2011, and PPV23 adult programs have been in place for many years. In this study, encompassing the three years following implementation of universal PCV13 in infants, up to 14.3% and 15.8% of all adult hospitalizations for CAP across the SOS Network were caused by PCV13 and PPV23 serotypes, respectively. These values exceed the proportional contributions of total and vaccine-preventable S. pneumoniae to all-cause CAP in the US [30]. While the incidence of vaccine-preventable disease caused by PCV13-serotypes is expected to decline over time if herd immunity afforded by childhood immunization mirrors the successes of the previous PCV7-based programs, PCV7 serotypes continue to be observed in 7.6% of adults hospitalized with CAPSpn even after five to eight years of PCV7-based childhood immunization (Fig. 1). This suggests that even with maximum herd protection, there will still be residual disease that has escaped herd immunity, and supports broader use of PCV13 in adults. Unlike recent trends in IPD [31,32], we did not observe a significant decline in the proportion of CAPSpn cases caused by PCV13 serotypes over the study period (Supplemental Table S3 and S4). Amongst older adults (age >50, age >65, and total cases), no significant differences between years were observed for the proportion of cases of CAPSpn caused by PCV13 serotypes, but lower numbers were noted in 2013 compared to 2011 or 2012 (potentially reflecting the early impact of PCV13 immunization in infants). While the study period presented here, encompassing the three years following implementation of infant PCV13 immunization programs, is too early to expect a significant decline in PCV13-type CAP in adults, it will be critical to continue to monitor this trend among hospitalized adults in order to assess the burden of disease attributable to PCV13 serotypes over time, as the impact of herd protection conferred by pediatric vaccination is optimized.
Funding
This work was supported by the Public Health Agency of Canada (PHAC), the Canadian Institutes of Health Research (CIHR), and by an investigator-initiated research grant to CIRN from Pfizer Canada [Grant numbers CIHR 23338, GSK1106, and PF1001]. No study sponsors were involved in study design, specimen collection, analysis and interpretation of data, scientific writing, or decision to submit the paper for publication.
Conflicts of interests
SAM received research grants from GlaxoSmithKline, Pfizer, Sanofi Pasteur; LV received research grants from GlaxoSmithKline, Pfizer, Optimer, Cubist and Merck, and personal fees from Merck, Optimer and Cubist. No other conflicts were declared.
Authors' contributions
The CIRN SOS principal investigator (SAM) and CIRN SOS site investigators (AM, AM, DW, GB, GS, JJ, JL, LV, MA, ML, MS, ST, WB) were involved in the conception and design of the study; SAM, JL, TH, and ME conducted/supervised the CIRN SOS Network central laboratory; SAM, ME, and AA were involved in CIRN SOS site coordination. LY, DM-C, LL, ME, and JL were involved in data management and statistical support. JL, AL, and HG were involved in method development and molecular testing for the study. IM provided reference serological testing at the NML. JL, ME, TH, and SAM drafted the manuscript and interpreted the data. All authors revised the manuscript critically for important intellectual content and all authors reviewed and approved the final draft of the manuscript.
Key points
With new recommendations being discussed for use of pneumococcal vaccines in adults, the study provides data to demonstrate pneumococcal CAP and IPD are significant causes of morbidity and mortality in hospitalized adults, with most S. pneumoniae serotypes being vaccine-preventable in the three years following introduction of infant PCV13 immunization programs in Canada.
Acknowledgments
The authors would like to thank all CIRN SOS surveillance monitors and staff at the Canadian Center for Vaccinology (CCfV) who were instrumental in the recruitment, collection, and processing of CAP and IPD specimens. In particular, the authors would like to acknowledge Michelle Warhuus.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/jj.vaccine.2017.05. 049.
References
[1] McNeil SA, Qizilbash N, Ye J, et al. A retrospective study of the clinical burden of hospitalized all-cause and pneumococcal pneumonia in Canada. Can Respir J 2015:17154.
[2] Jain S, Self WH, Wunderink RG, et al. Community-Acquired Pneumonia Requiring Hospitalization among U.S. Adults. N Engl J Med 2015;373:415-27.
[3] O'Brien KL, Wolfson LJ, Watt JP, et al. Burden of disease caused by Streptococcus pneumoniae in children younger than 5 years: global estimates. Lancet 2009;374:893-902.
[4] Drijkoningen JJ, Rohde GG. Pneumococcal infection in adults: burden of disease. Clin Microbiol Infect 2014;20:45-51.
J.J. LeBlanc et al./Vaccine xxx (2017) xxx-xxx
[5] Chalmers JD, Campling J, Dicker A, Woodhead M, Madhava H. A systematic review of the burden of vaccine preventable pneumococcal disease in UK adults. BMC Pulm Med 2016;16:77-88.
[6] Holter JC, Müller F, Bj0rang O, et al. Etiology of community-acquired pneumonia and diagnostic yields of microbiological methods: a 3-year prospective study in Norway. BMC Infect Dis 2015;15:64-75.
[7] Lim WS, Baudouin SV, George RC, et al. BTS guidelines for the management of community acquired pneumonia in adults: update 2009. Thorax 2009;64.
[8] World Health Organization (WHO). Estimated Hib and pneumococcal deaths for children under 5 years of age; 2000. <http://www.who.int/ immunization/monitoring_surveillance/burden/estimates/Pneumo_hib_2000/ en/> [accessed April 2, 2017].
[9] Gil-Prieto R, Pascual-Garcia R, Walter S, Älvaro-Meca A, Gil-De-Miguel Ä. Risk of hospitalization due to pneumococcal disease in adults in Spain. The CORIENNE study. Hum Vaccine Immunother 2016;22:1-6.
[10] Public Health Agency of Canada (PHAC). National advisory committee on immunization (NACI). <http://www.phac-aspc.gc.ca/naci-ccni/index-eng.php> [accessed April 2, 2017].
[11] Bettinger JA, Scheifele DW, Kellner JD, et al. The effect of routine vaccination on invasive pneumococcal infections in Canadian children, immunization monitoring program, active 2000-2007. Vaccine 2010;28:2130-6.
[12] Kellner JD, Vanderkooi OG, MacDonald J, Church DL, Tyrrell GJ, Scheifele DW. Changing epidemiology of invasive pneumococcal disease in Canada, 19982007: update from the Calgary-area Streptococcus pneumoniae research (CASPER) study. Clin Infect Dis 2009;49:205-12.
[13] Rudnick W, Liu Z, Shigayeva A, et al. Pneumococcal vaccination programs and the burden of invasive pneumococcal disease in Ontario, Canada, 1995-2011. Vaccine 2013;31:5863-71.
[14] Centers for Disease Control and Prevention (CDC). Direct and indirect effects of routine vaccination of children with 7-valent pneumococcal conjugate vaccine on incidence of invasive pneumococcal disease - United States, 1998-2003. Morb Mortal Wkly Rep 2005;54:893-7.
[15] Bonten MJ, Huijts SM, Bolkenbaas M, et al. Polysaccharide conjugate vaccine against pneumococcal pneumonia in adults. N Engl J Med 2015;372:1114-25.
[16] Tomczyk S, Bennett NM, Stoecker C, et al. Use of 13-valent pneumococcal conjugate vaccine and 23-valent pneumococcal polysaccharide vaccine among adults aged >/=65 years: recommendations of the advisory committee on immunization practices (ACIP). Morb Mortal Wkly Rep 2014;63:822-5.
[17] National Advisory Committee on Immunization (NACI). Update on the use of 13-valent pneumococcal conjugate vaccine (PNEU-C-13) in addition to 23-valent pneumococcal polysaccharide vaccine (PNEU-P-23) in immunocompetent adults 65 years of age and older - Interim Recommendation; June 2016. <http://www.phac-aspc.gc.ca/naci-ccni/index-eng.php> [last accessed April 2, 2017].
[18] McNeil S, Shinde V, Andrew M, et al. Interim estimates of 2013/14 influenza clinical severity and vaccine effectiveness in the prevention of laboratory-
confirmed influenza-related hospitalisation, Canada, February 2014. EuroSurveillance 2014:19.
[19] Searle SD, Mitnitski A, Gahbauer EA, Gill TM, Rockwood K. A standard procedure for creating a frailty index. BMC Geriat 2008;8:24.
[20] Lang AL, McNeil SA, Hatchette TF, Elsherif M, Martin I, LeBlanc JJ. Detection and prediction of Streptococcus pneumoniae serotypes directly from nasopharyngeal swabs using PCR. J Med Micobiol 2015;64:836-44.
[21] Huijts SM, Pride MW, Vos JM, et al. Diagnostic accuracy of a serotype-specific antigen test in community-acquired pneumonia. Eur Resp J 2013;42:1283-90.
[22] Pride MW, Huijts SM, Wu K, et al. Validation of an immunodiagnostic assay for detection of 13 Streptococcus pneumoniae serotype-specific polysaccharides in human urine. Clin Vaccine Immunol 2012;19:1131-41.
[23] Vila-Corcoles A, Ansa X, Ochoa-Gondar O, et al. Pneumococcal pneumonia in adults 60 years or older: Incidence, mortality and prevention. Med Clin (Barc) 2016;146:199-202.
[24] Said MA, Johnson HL, Nonyane BA, et al. Estimating the burden of pneumococcal pneumonia among adults: a systematic review and meta-analysis of diagnostic techniques. PLoS One 2013;8:e60273.
[25] Castiglia P. Recommendations for pneumococcal immunization outside routine childhood immunization programs in Western Europe. Adv Ther 2014;31:1011-44.
[26] Horita N, Miyazawa N, Kojima R, et al. Sensitivity and specificity of the Streptococcus pneumoniae urinary antigen test for unconcentrated urine from adult patients with pneumonia: a meta-analysis. Respirology 2013;18:1177-83.
[27] Bartlett JG. Diagnostic tests for agents of community-acquired pneumonia. Clin Infect Dis 2011;52:S296-304.
[28] Demczuk WH, Martin I, Griffith A, et al. Serotype distribution of invasive Streptococcus pneumoniae in Canada after the introduction of the 13-valent pneumococcal conjugate vaccine, 2010-2012. Can J Microbiol 2013;59:778-88.
[29] Sherwin RL, Gray S, Alexander R, McGovern PC, Graepel J, Pride MW, Purdy J, Paradiso P, File Jr TM. Distribution of 13-valent pneumococcal conjugate vaccine Streptococcus pneumoniae serotypes in US adults aged >/=50 years with community-acquired pneumonia. J Infect Dis 2013;208:1813-20.
[30] Center for Disease Control and Prevention (CDC). Active bacterial core surveillance (ABCs): emerging infections program network. <http://www. cdc.gov/abcs/reports-findings/survreports/spneu-types.html> [accessed September 28, 2016].
[31] Desai S, Policarpio ME, Wong K, Gubbay J, Fediurek J, Deeks S. The epidemiology of invasive pneumococcal disease in older adults from 2007 to 2014 in Ontario, Canada: a population-based study. CMAJ 2016;4:E545-50.
[32] Cabaj JL, Nettel-Aguirre A, MacDonald J, Vanderkooi OG, Kellner JD. Influence of childhood pneumococcal conjugate vaccines on invasive pneumococcal disease in adults with underlying comorbidities in calgary, alberta (20002013). Clin Infect Dis. 2016;62:1521-6.
