Community-acquired pneumonia (CAP) is one of the most common causes of morbidity and hospitalization in childhood and an important contributor to antibiotic use. Management guidelines can help reduce antibiotic misuse and therapeutic resistance.1,2 However, current guidelines are based on literature from the past decades. Epidemiological factors, such as improved and extended Childhood Immunizations Programs, better diagnostic methods and the importance of reduced antibiotic misuse, make this a critical time for increased knowledge on the etiology of CAP.3 Especially, the widespread implementation in the developed world of routine vaccination against Streptococcus pneumoniae (introduced in the Norwegian Childhood Immunizations Program in 2006) may have altered CAP etiology and incidence.1,4,5 Increased use of molecular biological diagnostic techniques and the discovery of several new viral pathogens in recent years mean that viruses are more often identified as the cause of pneumonia, and it is likely that the incidence of viral pneumonia has been underestimated in the past.6 The few studies from the developed world on etiology in CAP in later years, especially after routine pneumococcal vaccination is introduced, show contradicting results.4,7 Hence, further studies on the etiology of CAP are warranted.
The main objective of this study was to identify the microbiological etiology of CAP in previously healthy children and adolescents younger than 18 years in a population with high pneumococcal conjugate vaccine coverage. Secondary objectives were to determine the proportions of single and mixed infections and determine differences in etiology by age groups.
MATERIALS AND METHODS
Study Design, Population and Ethics
This prospective, observational study was conducted at the Department of Pediatric and Adolescent Medicine, Akershus University Hospital, Norway, from January 1, 2012 until January 1, 2014. The hospital’s catchment area has a population of 480,000, of which 115,000 are younger than 18 years. All patients younger than 18 years, where chest radiograph was performed because of suspected acute lower respiratory tract infection (ALRI), were considered for inclusion. The study was performed in a routine clinical practice, including children and adolescents who were treated ambulatory or in hospital. Chest radiograph and diagnostic tests were taken at inclusion. Medical history and clinical characteristics were recorded. Final inclusion criteria were (1) fever or history of fever; (2) one or more signs of ALRI: tachypnea, chest retractions and cough and (3) a chest radiograph consistent with pneumonia. Exclusion criteria were (1) acquired pneumonia while in hospital or travelling abroad and (2) children with severe motor impairment, innate or iatrogenic immunodeficiency, cystic fibrosis or other chronic disease that predisposes for pneumonia. Presentation with wheeze was not a criterion for exclusion, as it also can be present in pediatric pneumonia.1,8
Vaccination status at the time of inclusion was retrieved from the Norwegian Immunization Registry (registration of childhood routine immunizations is mandatory, and there is a reported coverage >92% for all childhood routine immunizations).9 In addition, all participants were asked if they had received routine immunizations when appropriate.
Guardians or patients older than 16 years of age signed a written, informed consent. The Regional Ethics Committee and the local Data Protection Officer approved the study.
Two study radiologists experienced in pediatric radiology independently examined all chest radiographs after enrollment and were blinded for clinical data. Localized or interstitial infiltrates were regarded as pneumonia but not sole perihilar changes.10 Findings were described in a standardized scoring sheet.11 In case of disagreement between the 2 radiologists, they were asked to independently review the radiographs for a final description. To increase specificity, only findings identified by both radiologists were labeled positive, as interrater variability can be substantial in pediatric chest radiography, especially in nonalveolar findings.10
Blood cultures, serum and nasopharyngeal specimens were taken from all participants at inclusion. Pleural fluid was taken if clinically indicated. Details of diagnostic tests are presented in Table 1.
- Polymerase chain reaction (PCR): Respiratory syncytial virus (RSV) A and B; parainfluenza virus 1, 2, 3 and 4; influenza virus A and B; human metapneumovirus; Mycoplasma pneumoniae and Chlamydophila pneumoniae were analyzed as previously reported.12 Rhinovirus/enterovirus (Rhino&EV/Cc r-gene 71-042), human bocavirus and adenovirus (AdV/hBoV r-gene 71-043) were analyzed using a commercial test (Argene, Biomérieux, Marcy l’Etoile, France).
- Serological tests: Complement fixation test (Virion\Serion, Würzburg, Germany) was used for RSV A/B, adenovirus, influenza A/B and parainfluenza 1–3 and was analyzed if both acute and reconvalescent serum was obtained. Chlamydophila pneumonia (MedacGmbh, Wedel, Germany) and M. pneumoniae (BioRad, Hercules, CA) were tested in acute and reconvalescent serum. Serological evidence for infection with S. pneumoniae was assessed using 2 methods: (A) in-house enzyme-linked immunosorbent assay (ELISA) for IgG against pneumolysin, in detail described elsewhere.13,14 Briefly, ELISA plates were coated with 1 µg/mL of pneumolysin (a gift from TJ Mitchell, Glasgow Biomedical Research Centre, United Kingdom), and 2-fold dilutions of sera were analyzed. A reference serum with high levels of IgG antibodies against pneumolysin was used to draw the standard curve, and results were given in arbitrary units. (B) Flow cytometry was used to analyze binding of serum antibodies to live pneumococci: a live, noncapsular pneumococcus strain FP23 was cultured overnight and then incubated with patient serum, followed by incubation with fluorescing (fluorescein isothiocyanate) anti-human IgG. Geometric mean fluorescence intensity was recorded and interpolated on a standard curve generated from a serum with high levels of antibodies against this strain.15 For both the ELISA against pneumolysin and the flow cytometry, paired sera were always analyzed on the same plate. A control serum was included in each experiment and gave a coefficient of variation less than 20% for the ELISA against pneumolysin and less than 15% for the flow cytometry.
- Bacterial culture: Blood and pleural fluid was cultured according to standard procedure.
- Antigen tests: Antigen test for S. pneumoniae (Binax NOW S. pneumoniae Antigen Card, Alere, MA) was performed on pleural fluid, as studies have shown a high positive predictive value for S. pneumoniae.16 The test is developed for urine, but we did not include testing in urine as a diagnostic test due to poor specificity in children.17,18
Statistical analyses were calculated using IBM SPSS Statistics version 20. Significance levels were 2-sided and set at P < 0.05. Normally distributed data are presented as means with a 95% confidence interval as a measure of variation, whereas skewed data are presented as median with interquartile range (IQR). Categorical data were analyzed with χ2 test or Fisher exact test. Continuous data were tested with analysis of variance (normally distributed) or Mann–Whitney U test /Kruskal–Wallis test (not normally distributed). Skewed data were log-transformed before linear regression analysis.
Two hundred and sixty-five cases met the inclusion criteria (Fig. 1). Table 2 describes demographic characteristics and the antibiotic treatment received. Two hundred and twenty five (84.9%) had received one or more doses of pneumococcal conjugate vaccine at the time of inclusion. No children whose parents refused routine immunizations were identified. Four of the 418 recruited cases (Fig. 1) were excluded due to unacknowledged chronic disease that predisposes for pneumonia at the time of inclusion (diaphragmatic hernia, cerebral palsy, bronchopulmonary dysplasia and 1 case of interstitial lung disease).
A pathogen was detected in 84.2% of the cases, with only viral infections in 63.4%, atypical bacteria in 7.9% (mixed or single) and other bacteria in 12.8% (mixed or single). Details on etiological agents and obtained microbiological specimens by age are presented in Table 3. When considering only the 203 of 265 cases in which both paired sera and PCR from nasopharyngeal specimen were available, the proportions of bacterial and viral cases were relatively similar compared with those found in all 265 cases (bacterial, single or mixed with virus: 20.7% vs. 24.2%; only viral: 63.4% vs. 62.6%). Specified viral and bacterial findings are presented in Figure 2. For the viral causes where both nasopharyngeal PCR and serology were used (RSV A/B, parainfluenza 1–3, influenza A/B and adenovirus), diagnosis was established by positive serology and positive PCR in 94 cases, by serology alone in 46 cases and by PCR alone in 54 cases. For the atypical causes, 11 cases where positive in both serology and PCR, 10 in serology alone and 1 in PCR alone.
Four of the 265 patients had a positive blood culture: 1 positive for S. pneumoniae, 2 for group A Streptococcus pyogenes and 1 for Haemophilus influenzae. Pleural fluid was aspirated in 7 of 15 cases with radiological signs of parapneumonic effusion and culture positive in 2: 1 positive for group A S. pyogenes and 1 for H. influenzae (also positive in blood culture). One case was antigen test positive for S. pneumoniae in pleural fluid (blood and pleural fluid culture negative).
The median number of days between acute-phase and follow-up sera was 55 days (IQR, 46–69 days). A linear regression analysis comparing log-transformed anti-FP23 and anti-pneumolysin titers (both acute and follow-up) showed a strong correlation (r2 = 0.47; P < 0.001; B = 0.62 with 95% confidence interval: 0.56–0.68).
In addition, to collecting nasopharyngeal specimens at inclusion, 134 of 265 patients (50.6%) consented to a second nasopharyngeal specimen at the time follow-up sera were obtained. In 4 of these 134 specimens, the same virus was found by PCR in the acute and follow-up sample. Neither M. pneumoniae nor C. pneumoniae were found in follow-up samples. In 65 cases (24.5% of the total of 265 cases) more than 1 viral causative agent, and in one case both C. pneumoniae and M. pneumoniae were found.
There was no significant correlation between decision to give antibiotics and positive bacterial etiology: P = 0.19 for atypical bacteria (M. pneumoniae and C. pneumoniae) and P = 1.00 for other bacteria (Fisher´s exact test). As seen in Table 2, approximately 40–50% of the patients younger than 5 years and 15% older than 5 years did not receive a full course of antibiotics. All patients improved regardless of antibiotic treatment.
We studied 265 cases with radiologically proven CAP, to elucidate the etiology of CAP in a pediatric cohort with 85% pneumococcal vaccine coverage. With our comprehensive microbiological investigation, etiology was found in 84.2%, similar to a new multicenter study7 but higher than most former CAP etiology studies from developed countries in the past 3 decades.4,8,19–26 Our main findings were that the majority of cases were diagnosed as viral infections without bacterial copathogens (63.4%) and 11.3% with pneumococcal infection.
In our study, the estimated proportion of CAP attributable to S. pneumoniae and other nonatypical bacteria was low (12.8%), in contrast to former CAP etiology studies estimating it to be 20–40%.8,20,21,23 In line with our findings, a recent, large US multicenter study reports even lower proportions of bacterial CAP with S. pneumoniae accounting for only 4%,7 a substantial fall in pneumococcal CAP if compared with another US CAP etiology study from 2004.8 In contrast, a British pre-pneumococcal and post- pneumococcal vaccination study found similar proportions of pneumococcal infections before and after the introduction of the vaccination (14.7% vs. 17.4%).4 In contrast to our study, pneumococcal diagnostics in these studies were limited to bacteremic cases (blood culture or PCR in blood). In this study, blood culture was positive in only 4 of 216 cases where it was obtained, in line with previous reports.19,23,27,28 Possibly as a consequence of this study’s low bacterial proportion, a lower proportion of mixed viral–bacterial infections (9.4%) was found compared with most previous CAP etiology studies1,8,21–24 but in line with the recent US multicenter study.7 The majority of the pneumococcal infections in our study were mixed with viruses, which may suggest that pneumococcal infections are secondary to viral disease. The proportion of cases with an atypical bacterial cause was lower than in most CAP etiology studies.4,8,20,21,24 Of note, the present study was carried out immediately after a large mycoplasma epidemic in Northern Europe, including Norway.29 It is, therefore, likely that the low incidence of mycoplasma can be ascribed to substantial immunity in the population preventing further spread of the infection.
In recent years, studies indicate that the incidence of viral pneumonia has been underestimated.6 Our findings support this, with 75.5% of CAP cases attributable to one or more viral pathogens, a similar proportion as found by a study looking for the prevalence of viral infections in children with CAP.30 Of our CAP cases, 63.4% were viral pneumonias without evidence of bacterial copathogens. Most other publications report viral etiology in 30–60% of pediatric pneumonia and about half of these were coinfected with bacteria.4,8,21,22,24 We found multiple viral infections in a higher number (24.5%) than in most CAP etiology studies but in line with studies looking specifically at viruses in pediatric CAP.30,31 The role of the relatively newly discovered human bocavirus in ALRI is controversial, although a recent study suggests a causative role.32 In this study, we found bocavirus in 24 cases of CAP, but only in 4 patients as the sole pathogen. Therefore, our findings cannot contribute to resolving this controversy.
This study is one of a few investigations of both hospitalized and ambulatory patients, including children treated solely by primary care physicians. This strengthens our study by avoiding the bias toward the sickest patients inherent in inpatient studies. Caution should be taken, however, when making a direct comparison to the many CAP etiology studies of hospitalized patients. Causative pathogens were found significantly more frequently in the 2 youngest (<2 and 2–5 years) versus the oldest age group (>5 years; 89.8%, 83.7% and 67.5%, respectively), a difference that may be due to greater sensitivity in diagnostic tests for viral causes. Although viral etiology was significantly more common in children aged 5 years or younger with RSV as the most frequent finding, mycoplasma infections were most common in the oldest age group (40%). The need for routine administration of antibiotics for mycoplasma infection in children is an open question as discussed in a new Cochrane systematic review.33 Pneumococcal pneumonia was relatively uncommon in all ages with no significant differences between the age groups, questioning the need for routine administration of antibiotics as recommended (at least to children older than 2 years) in current guidelines.1,2
When interpreting the data, several limitations should be considered; first, the absence of a microbiological standard procedure makes validation of diagnostic tests difficult. The majority of CAP cases in children are non-bacteremic.1 Transthoracic lung aspiration with its ethical and practical issue when attempted determines the etiological agent in only two thirds of patients.34 Induced sputum seems to increase bacterial yield and is implemented in a large multicenter CAP etiology study,35 but retrieving representative material in pediatric patients is demanding and differentiating true bacterial pathogens from commensals makes it questionable as a standard procedure.36 Because of this absence of a standard procedure, pneumococcal serological methods are insufficiently validated. A 2008 literature review of pneumococcal serological methods concludes that their sensitivity and specificity are high enough to detect pneumococcal disease in children.37 In the novel flow cytometric analysis using a noncapsular pneumococcus, we measured antibody responses to other pneumococcal antigens than the capsule. This gave us a solid and general method for detecting IgG against most pneumococci, which was strongly correlated to the widely used test for anti-pneumolysin antibodies. To assure coverage of all serotypes, we chose these general pneumococcal serological methods, in contrast to a serotype specific pneumococcal ELISA recently used by another research group.38,39
Second, an asymptomatic control group may have helped us to identify specific viruses or bacteria as incidental findings and not true etiological agents. However, the pathogens tested for in our study is consistent with pathogens tested for in studies using a control group, among them the large US multicenter study where only rhinovirus was found in noteworthy proportions in the controls.7 We compensated for the lack of a control group by setting a low cycle threshold value for positive viral PCR (cycle threshold < 35) to increase the likelihood that a viral PCR finding represents the current infection and not remnants of a previous infection. To investigate if nasopharyngeal findings during disease could represent incidental viral carriage,40 a second nasopharyngeal sample for PCR analysis at the follow-up visit was accepted by about half the patients/guardians. The follow-up sample revealed an identical virus in acute and follow-up samples in only 4 cases.
Third, our relatively strict case definition, including signs of acute infection and radiographic changes excluding perihilar changes, may have lead to exclusion of true CAP cases. This case definition was chosen to minimize the effect of substantial overlap in clinical and radiological features between pneumonia and bronchiolitis, especially because we also included the youngest age groups in our study. The overrepresentation of the youngest children may itself be a limitation, but on the other hand likely reflects the true disease burden in pediatric CAP. Nevertheless, caution must be taken in the interpretation of the results in the oldest age group due to the limited number of cases.
Fourth, we did not include PCR of blood for the diagnosis of pneumococcal invasive disease, which may have reduced our bacterial yield. Several pediatric CAP etiology studies in the last decade have included this with large variations in the findings.4,7,8,39,41,42 A 2010 meta-analysis questions both the sensitivity and specificity of PCR in blood as proof of invasive pneumococcal disease in children.43 Because pneumococcal bacteremia is expected to be infrequent where pneumococcal vaccination rate is high, the cases we might have missed were likely few,44–46 reflected in the low proportion of pneumococcal, bacteremic pneumonia found by the recent multicenter study using pneumococcal PCR in blood.7
Fifth, missing specimens, although in relative low proportions in this study, and the lack of diagnostic tests for all potential pathogens may have influenced our findings.
Finally, although our findings are comparable with those recently found in North America7 and we believe our population is representative for Northern Europe, we advise caution when extrapolating our results to other sociodemographically similar groups.
In conclusion, we found a high proportion of viral etiology and a low proportion of bacterial etiology in this prospective study of radiologically confirmed CAP in a pediatric population with a high vaccination rate for pneumococci. We hope that our results may contribute to improved CAP management guidelines and eventually reduce antibiotic overuse in areas where widespread pneumococcal vaccination is provided. The low pneumococcal burden may reflect the success of widespread pneumococcal vaccination, and the high viral burden points to the development of viral vaccines and therapy for future reduction in the substantial morbidity due to pneumonia in the pediatric population.
We thank all nurses and doctors on call for assistance in enrollment of patients and data collection and all the staff at the participating microbiological laboratories at the Akershus University Hospital. We are grateful for the help from Morten Linbæk and Jan Emil Kristoffersen in recruiting patients from GPs; Tove Karin Herstad, Anne-Cathrine Kristoffersen and Didrik F. Vestrheim at Norwegian Institute of Public Health for the help on pneumococcal diagnostics; Hege Vangstein Aamot, May Tove Furuseth, Hege Smith Tunsjoe and Tonje Sonerud for molecular biological analyses and organizing the Biobank at Akershus University Hospital and Ragnvald Grotli for radiological examinations.
Authors’ Contribution: All authors provided substantial contributions to the study’s conception and design and acquisition, analysis or interpretation of the data. A.S.B., C.I., H.O.F. and B.N. executed the clinical part of the work. T.M.L. was responsible for the microbiological laboratory analyses. A.A. and I.A. planned and performed all pneumococcal serological methods. R.B. was responsible for planning and performing the radiological examinations. A.S.B. drafted the initial manuscript, and all other authors revised it critically for important intellectual content. All authors approved the final manuscript as submitted, and agreed to be accountable for all aspects of the work.
1. Harris M, Clark J, Coote N, et al. British Thoracic Society guidelines for the management of community acquired pneumonia in children: update 2011. Thorax. 2011;66(Suppl 2):ii1–23
2. Bradley JS, Byington CL, Shah SS, et al. The management of community-acquired pneumonia in infants and children older than 3 months of age: clinical practice guidelines by the Pediatric Infectious Diseases Society and the Infectious Diseases Society of America. Clin Infect Dis. 2011;53:e25–e76
3. Clark JE. Determining the microbiological cause of a chest infection. Arch Dis Child. 2015;100:193–197
4. Elemraid MA, Sails AD, Eltringham GJ, et al.North East of England Paediatric Respiratory Infection Study Group. Aetiology of paediatric pneumonia after the introduction of pneumococcal conjugate vaccine. Eur Respir J. 2013;42:1595–1603
5. Angoulvant F, Levy C, Grimprel E, et al. Early impact of 13-valent pneumococcal conjugate vaccine on community-acquired pneumonia in children. Clin Infect Dis. 2014;58:918–924
6. Ruuskanen O, Lahti E, Jennings LC, et al. Viral pneumonia. Lancet. 2011;377:1264–1275
7. Jain S, Williams DJ, Arnold SR, et al. Community-acquired pneumonia requiring hospitalization among U.S. children. N Engl J Med. 2015;372:835–845
8. Michelow IC, Olsen K, Lozano J, et al. Epidemiology and clinical characteristics of community-acquired pneumonia in hospitalized children. Pediatrics. 2004;113:701–707
9. Trogstad L, Ung G, Hagerup-Jenssen M, et al. The Norwegian immunisation register–SYSVAK. Euro Surveill. 2012;17
10. Cherian T, Mulholland EK, Carlin JB, et al. Standardized interpretation of paediatric chest radiographs for the diagnosis of pneumonia in epidemiological studies. Bull World Health Organ. 2005;83:353–359
11. Hurlen P, Borthne A, Dahl FA, et al. Does PACS improve diagnostic accuracy in chest radiograph interpretations in clinical practice? Eur J Radiol. 2012;81:173–177
12. Tunsjo HS, Berg AS, Inchley CS, et al. Comparison of nasopharyngeal aspirate with flocked swab for PCR-detection of respiratory viruses in children. APMIS. 2015;123:473–477
13. Kanclerski K, Blomquist S, Granström M, et al. Serum antibodies to pneumolysin in patients with pneumonia. J Clin Microbiol. 1988;26:96–100
14. Claesson BA, Trollfors B, Brolin I, et al. Etiology of community-acquired pneumonia in children based on antibody responses to bacterial and viral antigens. Pediatr Infect Dis J. 1989;8:856–862
15. Kolberg J, Aase A, Naess LM, et al. Human antibody responses to pneumococcal surface protein A and capsular polysaccharides during acute and convalescent stages of invasive disease in adult patients. Pathog Dis. 2014;70:40–50
16. Strachan RE, Cornelius A, Gilbert GL, et al.Australian Research Network in Empyema (ARNiE). A bedside assay to detect Streptococcus pneumoniae
in children with empyema. Pediatr Pulmonol. 2011;46:179–183
17. Hamer DH, Egas J, Estrella B, et al. Assessment of the Binax NOW Streptococcus pneumoniae
urinary antigen test in children with nasopharyngeal pneumococcal carriage. Clin Infect Dis. 2002;34:1025–1028
18. Dowell SF, Garman RL, Liu G, et al. Evaluation of Binax NOW, an assay for the detection of pneumococcal antigen in urine samples, performed among pediatric patients. Clin Infect Dis. 2001;32:824–825
19. Drummond P, Clark J, Wheeler J, et al. Community acquired pneumonia–a prospective UK study. Arch Dis Child. 2000;83:408–412
20. Heiskanen-Kosma T, Korppi M, Jokinen C, et al. Etiology of childhood pneumonia: serologic results of a prospective, population-based study. Pediatr Infect Dis J. 1998;17:986–991
21. Don M, Fasoli L, Paldanius M, et al. Aetiology of community-acquired pneumonia: serological results of a paediatric survey. Scand J Infect Dis. 2005;37:806–812
22. Tsolia MN, Psarras S, Bossios A, et al. Etiology of community-acquired pneumonia in hospitalized school-age children: evidence for high prevalence of viral infections. Clin Infect Dis. 2004;39:681–686
23. Juvén T, Mertsola J, Waris M, et al. Etiology of community-acquired pneumonia in 254 hospitalized children. Pediatr Infect Dis J. 2000;19:293–298
24. Cevey-Macherel M, Galetto-Lacour A, Gervaix A, et al. Etiology of community-acquired pneumonia in hospitalized children based on WHO clinical guidelines. Eur J Pediatr. 2009;168:1429–1436
25. Wubbel L, Muniz L, Ahmed A, et al. Etiology and treatment of community-acquired pneumonia in ambulatory children. Pediatr Infect Dis J. 1999;18:98–104
26. Hamano-Hasegawa K, Morozumi M, Nakayama E, et al.Acute Respiratory Diseases Study Group. Comprehensive detection of causative pathogens using real-time PCR to diagnose pediatric community-acquired pneumonia. J Infect Chemother. 2008;14:424–432
27. Shah SS, Dugan MH, Bell LM, et al. Blood cultures in the emergency department evaluation of childhood pneumonia. Pediatr Infect Dis J. 2011;30:475–479
28. McCulloh RJ, Koster MP, Yin DE, et al. Evaluating the use of blood cultures in the management of children hospitalized for community-acquired pneumonia. PLoS One. 2015;10:e0117462
29. Blystad H, Anestad G, Vestrheim DF, et al. Increased incidence of Mycoplasma pneumoniae
infection in Norway 2011. Euro Surveill. 2012;17
30. Esposito S, Daleno C, Prunotto G, et al. Impact of viral infections in children with community-acquired pneumonia: results of a study of 17 respiratory viruses. Influenza Other Respir Viruses. 2013;7:18–26
31. Cilla G, Oñate E, Perez-Yarza EG, et al. Viruses in community-acquired pneumonia in children aged less than 3 years old: High rate of viral coinfection. J Med Virol. 2008;80:1843–1849
32. Christensen A, Døllner H, Skanke LH, et al. Detection of spliced mRNA from human bocavirus 1 in clinical samples from children with respiratory tract infections. Emerg Infect Dis. 2013;19:574–580
33. Gardiner SJ, Gavranich JB, Chang AB. Antibiotics for community-acquired lower respiratory tract infections secondary to Mycoplasma pneumoniae
in children. Cochrane Database Syst Rev. 2015;1:CD004875
34. Vuori-Holopainen E, Salo E, Saxén H, et al. Etiological diagnosis of childhood pneumonia by use of transthoracic needle aspiration and modern microbiological methods. Clin Infect Dis. 2002;34:583–590
35. Murdoch DR, O’Brien KL, Driscoll AJ, et al. Laboratory methods for determining pneumonia etiology in children. Clin Infect Dis. 2012;54(Suppl 2):S146–S152
36. Lahti E, Peltola V, Waris M, et al. Induced sputum in the diagnosis of childhood community-acquired pneumonia. Thorax. 2009;64:252–257
37. Korppi M, Leinonen M, Ruuskanen O. Pneumococcal serology in children’s respiratory infections. Eur J Clin Microbiol Infect Dis. 2008;27:167–175
38. Tuerlinckx D, Smet J, De Schutter I, et al. Evaluation of a WHO-validated serotype-specific serological assay for the diagnosis of pneumococcal etiology in children with community-acquired pneumonia. Pediatr Infect Dis J. 2013;32:e277–e284
39. De Schutter I, Vergison A, Tuerlinckx D, et al. Pneumococcal aetiology and serotype distribution in paediatric community-acquired pneumonia. PLoS One. 2014;9:e89013
40. Jartti T, Jartti L, Peltola V, et al. Identification of respiratory viruses in asymptomatic subjects: asymptomatic respiratory viral infections. Pediatr Infect Dis J. 2008;27:1103–1107
41. Resti M, Moriondo M, Cortimiglia M, et al.Italian Group for the Study of Invasive Pneumococcal Disease. Community-acquired bacteremic pneumococcal pneumonia in children: diagnosis and serotyping by real-time polymerase chain reaction using blood samples. Clin Infect Dis. 2010;51:1042–1049
42. Nascimento-Carvalho CM, Ribeiro CT, Cardoso MR, et al. The role of respiratory viral infections among children hospitalized for community-acquired pneumonia in a developing country. Pediatr Infect Dis J. 2008;27:939–941
43. Avni T, Mansur N, Leibovici L, et al. PCR using blood for diagnosis of invasive pneumococcal disease: systematic review and meta-analysis. J Clin Microbiol. 2010;48:489–496
44. Vestrheim DF, Løvoll O, Aaberge IS, et al. Effectiveness of a 2 + 1 dose schedule pneumococcal conjugate vaccination programme on invasive pneumococcal disease among children in Norway. Vaccine. 2008;26:3277–3281
45. van Deursen AM, van Mens SP, Sanders EA, et al.Invasive Pneumococcal Disease Sentinel Surveillance Laboratory Group. Invasive pneumococcal disease and 7-valent pneumococcal conjugate vaccine, the Netherlands. Emerg Infect Dis. 2012;18:1729–1737
46. Black S, Shinefield H, Fireman B, et al. Efficacy, safety and immunogenicity of heptavalent pneumococcal conjugate vaccine in children. Northern California Kaiser Permanente Vaccine Study Center Group. Pediatr Infect Dis J. 2000;19:187–195