Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Impact of Preceding Flu-Like Illness on the Serotype Distribution of Pneumococcal Pneumonia

  • Joon Young Song ,

    Affiliations Department of Pathology, School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, United States of America, Division of Infectious Diseases, Department of Internal Medicine, Korea University College of medicine, Seoul, Republic of Korea

  • Moon H. Nahm,

    Affiliations Department of Pathology, School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, United States of America, Department of Microbiology, School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, United States of America

  • Hee Jin Cheong,

    Affiliation Division of Infectious Diseases, Department of Internal Medicine, Korea University College of medicine, Seoul, Republic of Korea

  • Woo Joo Kim

    Affiliation Division of Infectious Diseases, Department of Internal Medicine, Korea University College of medicine, Seoul, Republic of Korea

Impact of Preceding Flu-Like Illness on the Serotype Distribution of Pneumococcal Pneumonia

  • Joon Young Song, 
  • Moon H. Nahm, 
  • Hee Jin Cheong, 
  • Woo Joo Kim



Even though the pathogenicity and invasiveness of pneumococcus largely depend on capsular types, the impact of serotypes on post-viral pneumococcal pneumonia is unknown.

Methods and Findings

This study was performed to evaluate the impact of capsular serotypes on the development of pneumococcal pneumonia after preceding respiratory viral infections. Patients with a diagnosis of pneumococcal pneumonia were identified. Pneumonia patients were divided into two groups (post-viral pneumococcal pneumonia versus primary pneumococcal pneumonia), and then their pneumococcal serotypes were compared. Nine hundred and nineteen patients with pneumococcal pneumonia were identified during the study period, including 327 (35.6%) cases with post-viral pneumococcal pneumonia and 592 (64.4%) cases with primary pneumococcal pneumonia. Overall, serotypes 3 and 19A were the most prevalent, followed by serotypes 19F, 6A, and 11A/11E. Although relatively uncommon (33 cases, 3.6%), infrequently colonizing invasive serotypes (4, 5, 7F/7A, 8, 9V/9A, 12F, and 18C) were significantly associated with preceding respiratory viral infections (69.7%, P<0.01). Multivariate analysis revealed several statistically significant risk factors for post-viral pneumococcal pneumonia: immunodeficiency (OR 1.66; 95% CI, 1.10–2.53), chronic lung diseases (OR 1.43; 95% CI, 1.09–1.93) and ICI serotypes (OR 4.66; 95% CI, 2.07–10.47).


Infrequently colonizing invasive serotypes would be more likely to cause pneumococcal pneumonia after preceding respiratory viral illness, particularly in patients with immunodeficiency or chronic lung diseases.


Pneumonia is the leading cause of death in both children and adults worldwide, being estimated to affect approximately 450 million people in the world annually and to result in about 4 million deaths per year (7% of the world's yearly death) [1], [2]. A preceding viral infection significantly increases both the chance for bacterial pneumonia and the chance for death from pneumonia [3]. Many observational studies have found Streptococcus pneumoniae (pneumococcus) to be the predominant bacterial etiology of community-acquired pneumonia (CAP) in both adults and children [4].

Although it is a well-known pathogen, pneumococcus is an opportunistic pathogen and normally resides in the nasopharynx (NP) of a large fraction of a population. To survive in the NP as well as to be capable of causing invasive diseases, pneumococci express a variety of virulence factors that influence pneumococcal interactions with host cells and other bacterial species. These virulence factors include pneumolysin, pneumococcal surface protein A, pneumococcal surface protein C, pneumococcal surface adhesin A, and capsular polysaccharide. Capsular polysaccharide, which shields pneumococci from the host immune system, may be the most important of these virulence factors. It can increase virulence by more than a million fold in experimental invasive infections [5], and can also assist pneumococcal colonization within the NP [6], [7]. Pneumococci are known to express more than 90 serologically and biochemically distinct capsule types (serotypes), and various epidemiologic studies have found serotypes to be correlated with the propensity for a high rate of nasopharyngeal carriage or for invasive diseases [8].

Like other types of bacterial pneumonia, pneumococcal pneumonia is especially common following viral infections [9], [10]. Zhou et al.[10] reported the significant association of invasive pneumococcal pneumonia with the activities of influenza and respiratory syncytial virus. Pneumococci were obtained from 23.5% of lung cultures in autopsy cases during the 1918–1919 influenza pandemic [11], and from 10% of fatal cases during the 2009 influenza A/H1N1 pandemic [12]. Since secondary bacterial infection significantly increases the mortality associated with viral infections, many studies have investigated the synergistic mechanisms between viral and bacterial infections using various (animal) model systems. One group of studies found that viral infections lead to over-expression of pneumococcal binding receptors, impaired alveolar macrophage phagocytosis and neutrophil dysfunction [9], [13], [14]. These findings suggest that the host becomes susceptible to pneumococcal invasion into deeper tissues and develops pneumonia by micro-aspiration of pneumococci that are already colonizing the NP. Other studies found that viral infections make the host susceptible to pneumococci from other individuals, and they increase pneumococcal transmission among susceptible individuals [15], [16]. We hypothesized that infrequently colonizing invasive serotypes may cause post-viral pneumococcal pneumonia with enhanced transmission by preceding respiratory viral infection, or that frequently colonizing weakly invasive serotypes may cause post-viral pneumococcal pneumonia with successful tissue invasion in the susceptible host after preceding respiratory viral infection.

Materials and Methods

Collection of clinical data and pneumococcal isolates

Medical records from January 1, 2007 through December 31, 2011 were examined to select patient records with a discharge diagnosis of CAP at Korea University Guro Hospital (KUGH), a 1000-bed teaching hospital in Seoul, Korea. The clinical, radiological, and microbiological findings of all the selected records were re-evaluated to determine whether the patients fulfilled the following clinical and radiological criteria of CAP: (a) an acute pulmonary infiltrate evident on chest radiographs and consistent with pneumonia within 48 h after admission; (b) confirmatory findings on clinical examination; and (c) acquisition of the infection outside a hospital [17]. Patients with healthcare-associated pneumonia or hospital-acquired pneumonia were excluded [18].

The patients with CAP were determined to have pneumococcal pneumonia if their blood samples or adequate lower respiratory specimens yielded bacterial isolates that were optochin sensitive and had alpha hemolytic colonies in the clinical laboratory [19], [20], [21]. Adequate lower respiratory specimens included trans-bronchial aspirates, broncho-alveolar lavage (BAL) specimens, and sputum specimens with the predominant presence of gram-positive diplococci on a Gram stain of high-quality (>25 WBCs and <10 squamous epithelial cells/low-power field). All such bacterial isolates were presumptively identified as “pneumococci” and routinely stored at -80°C.

Two infectious disease doctors reviewed the medical records, and selected cases meeting the criteria of community-acquired pneumococcal pneumonia. Clinical data from the patients were obtained using a structured case report form, which included demographics, underlying diseases, time (month) of pneumonia development, the presence of bloodstream infection, the presence of a flu-like illness (FLI) in the recent 7 days before pneumonia development, results of multiplex respiratory viral PCR/culture, and 30-day case fatalities. This study was approved by the ethics committee of KUGH (IRB No. 2013-01-0037) and was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice. The institutional review board waived written informed consent. Patient records/information was anonymized and de-identified prior to analysis.


Patients were defined as having post-viral pneumococcal CAP if they had FLI documented in their medical records 0–7 days before pneumonia development and were given respiratory viral tests (PCR or culture) at the time of admission. In KUGH, medical doctors have been regularly educated to perform respiratory viral tests, and document about respiratory symptoms with structured case report form if pneumonic patients have recent FLI. FLI was defined as sudden onset of fever (≥38°C) accompanied by ≥1 respiratory symptoms: cough, sore throat, or nasal symptoms. All the other cases were defined as having primary pneumococcal CAP. Influenza epidemic periods were defined based on Korean Influenza & Respiratory Virus Scheme (KINRESS), which reports an influenza index (influenza-like illness cases among 1000 patients) weekly, and declares influenza epidemic if the influenza-like illness index is higher than the upper limit of the mean influenza-like illness index ±2 standard deviations of the non-epidemic periods of the previous 3 years [22], [23]. Influenza epidemic period is defined to end if the influenza index decline below the upper limit above mentioned for four consecutive weeks.

Serotyping of pneumococcal isolates

All the “pneumococcal” isolates were recovered from the clinical laboratory and were re-identified in the research laboratory by colony morphology, optochin susceptibility, and bile solubility testing. Following the re-identification, all the isolates with appropriate properties were serotyped with the multibead serotyping assay method described previously [24], which included multibead assay with monoclonal antibodies (reaction A) and multibead assay with multiplex PCR (reaction B and C). The assay has been designed to identify the pneumococcal capsular PSs of all 93 serotypes and non-typeable (NT) pneumococci. NT pneumococci were classified as Group I or Group II, and Group II pneumococci were further divided into null capsule clade (NCC) 1, 2 and 3 [25].

Based on the literature review of carriage rates, carriage duration, and invasive disease potential[8], [26], [27], [28], [29], [30], each serotype was classified into four groups: infrequently colonizing but invasive (ICI) serotypes (1, 4, 5, 7F, 8, 9V/9A, 12F, and 18C), frequently colonizing and invasive (FCI) serotypes (3, 14, and 19A), frequently colonizing but weakly invasive (FCWI) serotypes (6A, 6B, 11A/11E, 15F/15A, 15B, 15C, 16F, 19F, 23F, and 35B), and serotypes that have not yet been classified (6C, 6D, 7B/7C/40, 9N, 10A/39, 10B, 13, 17F/17A, 20, 22F/22A, 23A, 24F/24A/24B, 25F/25A/38, 28F/28A, 31, 33F/33A/37, 34, 35F/47F, 35A/35C/42, 36, 41F/41A, 45, and NT).

Statistical analysis

We performed case-control analysis to compare the demographics, clinical characteristics, and serotype distributions of post-viral pneumococcal pneumonia and primary pneumococcal pneumonia. We also investigated the association between serotype distribution and influenza epidemics. Data were analyzed using SPSS version 12.0 (SPSS Inc., Chicago, IL, USA). For categorical data, univariate analysis was performed using either the chi-square test or Fisher's exact test. The Mann-Whitney U test was used to compare ages between two groups and was expressed as a median (interquartile range, IQR). P<0.05 was considered to be statistically significant. Multivariate analysis was carried out to assess independent risk factors of post-viral pneumococcal pneumonia using a logistic regression model.


Demographic and clinical characteristics: post-viral pneumococcal pneumonia versus primary pneumococcal pneumonia

Although the review of medical records identified 930 pneumococcal CAP cases, 11 cases were excluded because the research laboratory found that the associated bacterial isolates were non-pneumococcal streptococci. The remaining bacterial isolates from 919 patients were positive for lytA on polymerase chain reaction (reaction B). Urinary antigen test (BinaxNOW S. pneumoniae assay) was taken in 835 cases, and 697 among 835 cases (83.5%) showed positive results. The included 919 cases had a median age of 63 (interquartile range, 44–72) years and were comprised of more male (66.4%) than female patients (Table 1). Bacteremic pneumonia was found in 19 (2.1%) patients, and 100 (10.9%) patients died within 30 days of developing pneumonia.

Table 1. Comparison of patient characteristics and clinical outcomes of post-viral pneumococcal pneumonia and primary pneumococcal pneumonia.

The majority of cases (592 patients, 64.4%) had primary pneumococcal pneumonia (Table 1), but a small but significant number of cases (327 patients, 35.6%) had post-viral pneumococcal pneumonia. In patients with post-viral pneumococcal pneumonia, cough, rhinorrhea/nasal stuffiness and sore throat were observed in 64.5% (211 patients), 26.9% (88 patients) and 58.1% (190 patients), respectively. Patients with post-viral pneumococcal pneumonia were older than those with primary pneumococcal pneumonia (P<0.01); more than half (53.5%) of the patients with post-viral pneumococcal pneumonia were 65 years old or older. Underlying medical diseases were more common in patients with post-viral pneumococcal pneumonia compared to those with primary pneumococcal pneumonia (74.6% versus 60.8%, P<0.01); immunodeficiency (P<0.01), chronic lung diseases (P<0.01), and chronic liver diseases (P = 0.05) were observed most often among patients with post-viral pneumococcal pneumonia. Four hundred and ninety-three of the cases (53.6%) occurred during influenza epidemic periods without a significant difference between post-viral and primary pneumococcal pneumonia (P = 0.84). The two groups had similar rates of bacteremia and 30-day mortality.

Respiratory viruses were actually isolated from 44 (13.5%) of the post-viral pneumococcal pneumonia patients (Table 2).

Table 2. Serotype distribution of pneumococcal pneumonia based on laboratory-confirmed viral infections.

Comparison of serotype distributions: post-viral pneumococcal pneumonia versus primary pneumococcal pneumonia

When pneumococcal serotypes are grouped according to categories, most pneumococcal CAP cases were associated with pneumococci of FCI (272 cases, 29.6%), FCWI (402 cases, 43.7%) and unclassified (212 cases, 23.1%) serotypes, but a small number of cases (33 cases) were due to ICI serotype pneumococci (Table 3). When the two patient groups were compared (Table 2 and Fig. 1A), the distribution of FCI, FCWI, and unclassified serotypes did not differ between them, but ICI serotypes were observed more often among patients with preceding respiratory viral infections than among those with primary pneumococcal pneumonia (23 cases versus 10 cases, P<0.01). In addition, we compared 44 PCR-confirmed viral cases versus 592 FLI-negative primary pneumococcal pneumonia cases. Likewise, ICI serotypes were significantly associated with preceding respiratory viral infections (P<0.01): ICI serotypes (60.0%, 15 among 25 cases), FCI serotypes (6.3%, 12 among 192 cases), FCWI serotypes (3.9%, 11 among 280 cases) and unclassified serotypes (4.3%, 6 among 139 cases). In the age-stratified analyses, ICI serotypes were rare among the pediatric patients, and statistically significant difference was not evaluable. When we restrict the cases to the elderly aged ≥50 years, the results were the same as those from all study subjects (data not shown).

Figure 1. Serotype distribution of Streptococcus pneumoniae was analyzed regarding preceding respiratory viral infections and influenza epidemic periods.

(A) Serotype distribution of Streptococcus pneumoniae in terms of preceding respiratory viral infections among patients with pneumonia. ICI serotypes (4, 5, 7F/7A, 8, 9V/9A, 12F and 18C) were more likely to cause post-viral pneumococcal pneumonia compared to other serotypes (*P<0.05). (B) Serotype distribution of Streptococcus pneumoniae in terms of influenza epidemic periods among patients with pneumonia. ICI serotypes were more likely to cause pneumococcal pneumonia during influenza epidemic periods, while serotype 6B was more common during non-epidemic periods (*P<0.05). ICI, infrequently colonizing but invasive serotypes; FCI, frequently colonizing and invasive serotypes; FCWI, frequently colonizing but weakly invasive serotypes; NT, non-typeable.

Table 3. Comparison of serotype distribution of post-viral pneumococcal pneumonia and primary pneumococcal pneumonia.

When individual pneumococcal capsule types were compared between the two patient groups, 7F/7A was more common among post-viral pneumonia patients than among primary pneumococcal CAP patients (P = 0.05) (Table 3). Serotypes 3 and 19A were the most common serotypes found in both patient populations, and viral etiology could not be associated with specific serotypes of pneumococci (Table 2). Taken together, it was difficult to identify a serotype that is more common in one group than another. Forty-five cases (4.9%) were caused by NT pneumococci, but most of them (88.9%) belonged to Group I, which has defective cps (Table 3). The remaining five NT isolates belonged to Group II: two isolates of NCC1, one isolate of NCC2, and two isolates of NCC3. Contrary to NT carriage isolates, which usually lack the capacity to produce a capsule (Group II) [31], most of the NT pneumococci from pneumonia cases belonged to Group I in the present study.

To estimate the vaccine coverage rate, the pneumococcal serotypes were classified according to the serotypes included in the three different pneumococcal vaccines. The overall coverage rates of 7-valent pneumococcal conjugate vaccine (PCV7), PCV13 and 23-valent pneumococcal polysaccharide vaccine (PPV23) were 24.8%, 60.7% and 63.3% respectively (Table S1 in File S1). The PCV7 would provide protection against 24.5% (80) of the post-viral pneumococcal pneumonia cases compared with 25.0% (148) of the primary pneumococcal pneumonia cases (P = 0.87). Similarly, PCV13 would provide protection against 59.0% (193) of the post-viral pneumococcal pneumonia cases versus 61.7% (365) of the primary pneumococcal pneumonia cases (P = 0.44), and the PPV23 would provide protection against 62.4% (204) of the post-viral pneumococcal pneumonia cases versus 63.9% (378) of the primary pneumococcal pneumonia cases (P = 0.67). Thus, the two patient groups would be comparably protected by all the vaccines.

Risk factors for post-viral pneumococcal pneumonia

The results of multivariate logistic regression analysis for risk factors associated with post-viral pneumococcal pneumonia are summarized in Table 4. Immunodeficiency (OR 1.66; 95% CI, 1.10–2.53), chronic lung diseases (OR 1.43; 95% CI, 1.09–1.93) and ICI serotypes (OR 4.66; 95% CI, 2.07–10.47) were considered as independent risk factors for post-viral pneumococcal pneumonia.

Table 4. Multivariate logistic regression analysis for risk factors of post-viral pneumococcal pneumonia.

Serotype distribution of pneumonia cases in relation to influenza epidemic periods: epidemic periods versus non-epidemic periods

Considering that influenza is the most common cause of respiratory viral infections that precede cases of pneumonia, we additionally compared the distribution of the pneumococcal serotypes of pneumonia cases between influenza epidemic periods and non-epidemic periods (Fig. 1B and Table S2 in File S1). Pneumococcal pneumonia of ICI serotypes occurred predominantly during influenza epidemic periods (75.8%, P = 0.01), whereas other serotypes (FCI, FCWI and unclassified) were evenly distributed between epidemic and non-epidemic periods.


Bacterial pneumonia often follows respiratory viral infections and can be lethal in some patients. The mechanisms involved in the synergism between viral infections and bacterial pneumonia have been extensively investigated in animal models and have produced evidence for increased invasiveness as well as transmission following respiratory viral infections. Increased transmissibility has been associated with prolonged shedding and wider spreading of pneumococci in a strain- or serotype-dependent manner after respiratory viral infections [16]. In the mouse model, influenza-neutralizing monoclonal antibodies inhibited pneumococcal transmission [15]. Increased invasiveness has been associated with the paradoxical suppression of the host immune system or with enhanced bacterial adherence. Alveolar macrophage function is impaired by dysregulated cytokine responses, while neutrophil function is suppressed as the amount of neutrophil-activating chemokines diminishes [9]. Meanwhile, pneumococcal adherence is facilitated by the up-regulation of pneumococcal binding receptors: fibronectin, glycoproteins of basal progenitor cells, polymeric immunoglobulin receptor (pIgR), and platelet-activating factor receptor (PAFR) [9]. In addition, pneumococcal binding to pIgR and PAFR may facilitate the development of bacteremic pneumonia [32], [33].

If increased invasiveness is dominant in human, we should have observed increases in pneumonia due to the relative non-invasive capsule types that commonly colonize the NP. Instead, we found that ICI serotypes (4, 5, 7F/7A, 8, 9V/9A, 12F, and 18C) occurred more frequently among cases of post-viral pneumococcal pneumonia. This observation strongly suggests that increased transmissibility may be the critical factor in humans. However, our findings do not imply that changes that increase invasion are irrelevant. Such changes may have assisted infections by ICI serotypes.

In addition to serotype distribution based on preceding respiratory viral infections, we compared serotype distribution between influenza epidemic periods and non-epidemic periods considering that influenza viruses circulate in Korea during confined periods, usually between November to April [22], [23]. As expected, ICI serotypes were more prevalent during influenza epidemic periods than during non-epidemic periods.

The serotypes associated with (epidemic) outbreaks of pneumococcal pneumonia have been reported to be serotypes 1, 4, 5, 7F, 8, 9, and 12F [34], [35], [36], [37]. An interesting observation is that these serotypes are almost identical to the serotypes associated with pneumonia cases following respiratory viral infections. Previous studies of outbreaks of pneumococcal pneumonia have not provided information on any respiratory viral infections that may have preceded the outbreaks. Thus, epidemic pneumococcal pneumonia outbreaks might be related to concurrent respiratory viral infections. Consequently, surveillance of respiratory viral infections in relation to pneumococcal outbreaks would be warranted and should be performed with highly sensitive methods (PCR or real-time PCR) considering the limited viral shedding periods [38].

To identify the population(s) at greatest risk of pneumococcal pneumonia after preceding respiratory viral infections, we investigated post-viral pneumococcal pneumonia patients for risk factors. Immunodeficiency and chronic lung diseases increased the risk for post-viral pneumococcal pneumonia by 66% and 43%, respectively. Increased pneumococcal pneumonia in immune-compromised patients may be the result of defective T regulatory cell and immunomodulatory responses [39]. In addition, it is well known that viral shedding persists longer in the immunocompromised patients than in healthy adults [40], [41]. Prolonged viral shedding and susceptible time periods might contribute to the development of concomitant or secondary pneumococcal pneumonia. As for chronic lung diseases, acute exacerbation of asthma and chronic obstructive pulmonary disease (COPD) are triggered by respiratory viral infections, which may increase the risk of secondary bacterial infection. Impaired interferon and Th1 responses result in uncontrolled viral replication and exaggerated inflammatory responses in asthmatic patients [42], and systemic corticosteroids for asthma or COPD exacerbations may be associated with slower viral clearance [43]. However, patients with immunodeficiency or chronic lung diseases are also susceptible to flu, so prospective cohort studies are required to clarify if these are real risk factors for pneumococcal pneumonia after respiratory viral infections.

30-day mortality did not differ between the two groups (post-viral pneumococcal pneumonia versus primary pneumococcal pneumonia). Given the older age and co-morbidities among the post-viral pneumonia cases, this group would be expected to show higher mortality. Interestingly, the ICI serotypes are associated with less mortality than serotypes 3 and 19A [44]. This might account for similar mortality in the two groups.

Pneumococcal vaccination is a part of the influenza pandemic preparedness plan. In our study, serotypes 3 and 19A were the most common, followed by serotypes 19F and 6A. Since serotypes 3, 19A, and 6A are included in PCV13, PCV13 could be much better than PCV7 and be useful as a part of the preparedness plan. However, the immunogenicity of PCV13 appears to be low with serotype 3 [45], [46], and PCV13 does not include serotypes 8 and 12F, which are represented in PPV23. Actually in the study by Sanz et al. [47], the introduction of pediatric conjugate vaccine (PCV7) led to overall decrease in invasive pneumococcal diseases, but those by non-vaccine serotypes (particularly serotype 8) increased markedly in HIV-infected patients. Thus, adults aged ≥19 years with immunocompromising conditions should receive a dose of PCV13 first, followed by a dose of PPV23 with ≥8 week interval [48]. In addition, due to its broader serotype coverage, PPV23 may be preferable to PCV13 for the elderly aged 65 years or more (Table S1 in File S1) [49].

This study has several limitations. First, this study had a retrospective design. To increase its accuracy, we combined clinical FLI criteria and respiratory viral tests in defining preceding respiratory viral infections. Considering that influenza patients with pneumonia are less likely to have nasal symptoms compared to those without pneumonia [50], we did not limit the post-viral pneumonia group to those who have nasal symptoms. Because respiratory viral tests were performed at the time of pneumonia diagnosis, the isolation rate of respiratory viruses (13.4%) was relatively low in the present study. Patients are at great risk for pneumococcal pneumonia at 5–7 days after influenza infection [51], [52], but influenza viral shedding is known to persist less than 7 days in general [38]. Moreover, influenza viruses may bind better to α2-3-linked receptors rather than α2-6-linked receptors in pneumonic patients, resulting in low viral isolation from upper respiratory specimens. In human, sialic acid (SA)-α2,6Gal is dominant on epithelial cells in nasal mucosa, while SA-α2,3Gal is usually expressed in the respiratory bronchiole and alveolus [53]. The low rate of laboratory confirmation for respiratory viral infections was the main limitation of this study, so prospective large-scale studies are required. In addition to respiratory viral test, serological test for influenza need to be taken considering low rate of viral isolation. Secondly, the classification of serotype is not widely used. Although we classified the groups based on comprehensive review of international and local data, some controversies might exist. Third, the yields of blood culture were quite low (2.1%). Given that this study was performed in the referral hospital, prior antibiotic use might have reduced the sensitivity of blood cultures, resulting in an underestimation of bacteremic pneumonia. Consequently, although a recent study estimated that about 4.5%–6.0% of invasive pneumococcal pneumonia can be attributed to influenza [54], we could not ascertain whether invasive pneumococcal pneumonia is more likely to concur following respiratory viral infections or not. To better detect bacteremic pneumonia, molecular diagnostic methods need to be considered. Finally, some uncontrolled confounding factors might exist, including vaccinations (influenza and pneumococcus), climate and socioeconomic levels.

In conclusion, serotypes 3 and 19A were the most prevalent among patients with pneumococcal pneumonia. Current pneumococcal vaccines (PPV23 and PCV13) should be effective against these serotypes. However, ICI serotypes were more likely to cause pneumococcal pneumonia after preceding respiratory viral illness, particularly in patients with either immunodeficiency or chronic lung diseases. ICI serotypes seemed to be transmitted more frequently secondary to increased colonization/carriage following epidemics of respiratory viruses. It is not clear why ICI serotypes would be more transmissible compared to others. Further clinical and experimental studies are warranted to clarify the pathogenesis.

Supporting Information

File S1.

Tables S1 and S2. Table S1. The coverage rates of pneumococcal vaccines, stratified by age group. Table S2. Serotype distribution of pneumococcal pneumonia: influenza epidemic periods versus non-epidemic periods.



We deeply appreciate Dr. M.C. Mcellistrem at the University of Pittsburgh for careful reading of the manuscript and helpful comments and suggestions.

Author Contributions

Conceived and designed the experiments: JYS MHN HJC WJK. Performed the experiments: JYS MHN. Analyzed the data: JYS MHN. Contributed reagents/materials/analysis tools: JYS MHN HJC WJK. Wrote the paper: JYS MHN.


  1. 1. Kabra SK, Lodha R, Pandey RM (2010) Antibiotics for community-acquired pneumonia in children. Cochrane Database Syst Rev: CD004874.
  2. 2. Ruuskanen O, Lahti E, Jennings LC, Murdoch DR (2011) Viral pneumonia. Lancet 377: 1264–1275.
  3. 3. Rothberg MB, Haessler SD, Brown RB (2008) Complications of viral influenza. Am J Med 121: 258–264.
  4. 4. Niederman MS (2004) Review of treatment guidelines for community-acquired pneumonia. Am J Med 117 Suppl 3A51S–57S.
  5. 5. Avery OT, Dubos R (1931) The Protective Action of a Specific Enzyme against Type Iii Pneumococcus Infection in Mice. J Exp Med 54: 73–89.
  6. 6. Magee AD, Yother J (2001) Requirement for capsule in colonization by Streptococcus pneumoniae. Infect Immun 69: 3755–3761.
  7. 7. Nelson AL, Roche AM, Gould JM, Chim K, Ratner AJ, et al. (2007) Capsule enhances pneumococcal colonization by limiting mucus-mediated clearance. Infect Immun 75: 83–90.
  8. 8. Song JY, Nahm MH, Moseley MA (2013) Clinical Implications of Pneumococcal Serotypes: Invasive Disease Potential, Clinical Presentations, and Antibiotic Resistance. J Korean Med Sci 28: 4–15.
  9. 9. Ballinger MN, Standiford TJ (2010) Postinfluenza bacterial pneumonia: host defenses gone awry. J Interferon Cytokine Res 30: 643–652.
  10. 10. Zhou H, Haber M, Ray S, Farley MM, Panozzo CA, et al. (2012) Invasive pneumococcal pneumonia and respiratory virus co-infections. Emerg Infect Dis 18: 294–297.
  11. 11. Morens DM, Taubenberger JK, Fauci AS (2008) Predominant role of bacterial pneumonia as a cause of death in pandemic influenza: implications for pandemic influenza preparedness. J Infect Dis 198: 962–970.
  12. 12. Shieh WJ, Blau DM, Denison AM, Deleon-Carnes M, Adem P, et al. (2010) 2009 pandemic influenza A (H1N1): pathology and pathogenesis of 100 fatal cases in the United States. Am J Pathol 177: 166–175.
  13. 13. Sun K, Metzger DW (2008) Inhibition of pulmonary antibacterial defense by interferon-gamma during recovery from influenza infection. Nat Med 14: 558–564.
  14. 14. Didierlaurent A, Goulding J, Patel S, Snelgrove R, Low L, et al. (2008) Sustained desensitization to bacterial Toll-like receptor ligands after resolution of respiratory influenza infection. J Exp Med 205: 323–329.
  15. 15. Diavatopoulos DA, Short KR, Price JT, Wilksch JJ, Brown LE, et al. (2010) Influenza A virus facilitates Streptococcus pneumoniae transmission and disease. Faseb J 24: 1789–1798.
  16. 16. McCullers JA, McAuley JL, Browall S, Iverson AR, Boyd KL, et al. (2010) Influenza enhances susceptibility to natural acquisition of and disease due to Streptococcus pneumoniae in ferrets. J Infect Dis 202: 1287–1295.
  17. 17. Bodi M, Rodriguez A, Sole-Violan J, Gilavert MC, Garnacho J, et al. (2005) Antibiotic prescription for community-acquired pneumonia in the intensive care unit: impact of adherence to Infectious Diseases Society of America guidelines on survival. Clin Infect Dis 41: 1709–1716.
  18. 18. American Thoracic Society, Infectious Diseases Society of America (2005) Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med 171: 388–416.
  19. 19. Blaschke AJ (2011) Interpreting assays for the detection of Streptococcus pneumoniae. Clin Infect Dis 52 Suppl 4S331–337.
  20. 20. Werno AM, Murdoch DR (2008) Medical microbiology: laboratory diagnosis of invasive pneumococcal disease. Clin Infect Dis 46: 926–932.
  21. 21. Song JY, Eun BW, Nahm MH (2013) Diagnosis of Pneumococcal Pneumonia: Current Pitfalls and the Way Forward. Infect Chemother 45: 351–366.
  22. 22. Song JY, Cheong HJ, Choi SH, Baek JH, Han SB, et al. (2013) Hospital-based influenza surveillance in Korea: hospital-based influenza morbidity and mortality study group. J Med Virol 85: 910–917.
  23. 23. Lee JS, Shin KC, Na BK, Lee JY, Kang C, et al. (2007) Influenza surveillance in Korea: establishment and first results of an epidemiological and virological surveillance scheme. Epidemiol Infect 135: 1117–1123.
  24. 24. Yu J, Lin J, Kim KH, Benjamin WH Jr, Nahm MH (2011) Development of a multiplexed and automated serotyping assay for Streptococcus pneumoniae. Clin Vaccine Immunol 18: 1900–1907.
  25. 25. Park IH, Kim KH, Andrade AL, Briles DE, McDaniel LS, et al.. (2012) Nontypeable Pneumococci Can Be Divided into Multiple cps Types, Including One Type Expressing the Novel Gene pspK. MBio 3..
  26. 26. Yildirim I, Hanage WP, Lipsitch M, Shea KM, Stevenson A, et al. (2010) Serotype specific invasive capacity and persistent reduction in invasive pneumococcal disease. Vaccine 29: 283–288.
  27. 27. Sleeman KL, Griffiths D, Shackley F, Diggle L, Gupta S, et al. (2006) Capsular serotype-specific attack rates and duration of carriage of Streptococcus pneumoniae in a population of children. J Infect Dis 194: 682–688.
  28. 28. Shouval DS, Greenberg D, Givon-Lavi N, Porat N, Dagan R (2006) Site-specific disease potential of individual Streptococcus pneumoniae serotypes in pediatric invasive disease, acute otitis media and acute conjunctivitis. Pediatr Infect Dis J 25: 602–607.
  29. 29. Kronenberg A, Zucs P, Droz S, Muhlemann K (2006) Distribution and invasiveness of Streptococcus pneumoniae serotypes in Switzerland, a country with low antibiotic selection pressure, from 2001 to 2004. J Clin Microbiol 44: 2032–2038.
  30. 30. Cho EY, Kang HM, Lee J, Kang JH, Choi EH, et al. (2012) Changes in Serotype Distribution and Antibiotic Resistance of Nasopharyngeal Isolates of Streptococcus pneumoniae from Children in Korea, after Optional Use of the 7-Valent Conjugate Vaccine. J Korean Med Sci 27: 716–722.
  31. 31. Scott JR, Hinds J, Gould KA, Millar EV, Reid R, et al. (2012) Nontypeable pneumococcal isolates among navajo and white mountain apache communities: are these really a cause of invasive disease? J Infect Dis 206: 73–80.
  32. 32. Cundell DR, Gerard NP, Gerard C, Idanpaan-Heikkila I, Tuomanen EI (1995) Streptococcus pneumoniae anchor to activated human cells by the receptor for platelet-activating factor. Nature 377: 435–438.
  33. 33. Kaetzel CS (2005) The polymeric immunoglobulin receptor: bridging innate and adaptive immune responses at mucosal surfaces. Immunol Rev 206: 83–99.
  34. 34. Vanderkooi OG, Church DL, MacDonald J, Zucol F, Kellner JD (2011) Community-based outbreaks in vulnerable populations of invasive infections caused by Streptococcus pneumoniae serotypes 5 and 8 in Calgary, Canada. PLoS ONE 6: e28547.
  35. 35. Vainio A, Lyytikäinen O, Sihvonen R, Kaijalainen T, Teirilä L, et al. (2009) An outbreak of pneumonia associated with S. pneumoniae at a military training facility in Finland in 2006. APMIS 117: 488–491.
  36. 36. Rakov AV, Ubukata K, Robinson DA (2011) Population structure of hyperinvasive serotype 12F, clonal complex 218 Streptococcus pneumoniae revealed by multilocus boxB sequence typing. Infect Genet Evol 11: 1929–1939.
  37. 37. Ihekweazu C, Basarab M, Wilson D, Oliver I, Dance D, et al. (2010) Outbreaks of serious pneumococcal disease in closed settings in the post-antibiotic era: a systematic review. J Infect 61: 21–27.
  38. 38. Carrat F, Vergu E, Ferguson NM, Lemaitre M, Cauchemez S, et al. (2008) Time lines of infection and disease in human influenza: a review of volunteer challenge studies. Am J Epidemiol 167: 775–785.
  39. 39. Neill DR, Fernandes VE, Wisby L, Haynes AR, Ferreira DM, et al. (2012) T regulatory cells control susceptibility to invasive pneumococcal pneumonia in mice. PLoS Pathog 8: e1002660.
  40. 40. Giannella M, Alonso M, Garcia de Viedma D, Lopez Roa P, Catalán P, et al. (2011) Prolonged viral shedding in pandemic influenza A(H1N1): clinical significance and viral load analysis in hospitalized patients. Clin Microbiol Infect 17: 1160–1165.
  41. 41. Pinsky BA, Mix S, Rowe J, Ikemoto S, Baron EJ (2010) Long-term shedding of influenza A virus in stool of immunocompromised child. Emerg Infect Dis 16: 1165–1167.
  42. 42. Singanayagam A, Joshi PV, Mallia P, Johnston SL (2012) Viruses exacerbating chronic pulmonary disease: the role of immune modulation. BMC Med 10: 27.
  43. 43. Lee N, Chan PK, Hui DS, Rainer TH, Wong E, et al. (2009) Viral loads and duration of viral shedding in adult patients hospitalized with influenza. J Infect Dis 200: 492–500.
  44. 44. Harboe ZB, Thomsen RW, Riis A, Valentiner-Branth P, Christensen JJ, et al. (2009) Pneumococcal serotypes and mortality following invasive pneumococcal disease: a population-based cohort study. PLoS Med 6: e1000081.
  45. 45. Kieninger DM, Kueper K, Steul K, Juergens C, Ahlers N, et al. (2010) Safety, tolerability, and immunologic noninferiority of a 13–valent pneumococcal conjugate vaccine compared to a 7-valent pneumococcal conjugate vaccine given with routine pediatric vaccinations in Germany. Vaccine 28: 4192–4203.
  46. 46. Yeh SH, Gurtman A, Hurley DC, Block SL, Schwartz RH, et al. (2010) Immunogenicity and safety of 13-valent pneumococcal conjugate vaccine in infants and toddlers. Pediatrics 126: e493–505.
  47. 47. Sanz JC, Cercenado E, Marin M, Ramos B, Ardanuy C, et al. (2011) Multidrug-resistant pneumococci (serotype 8) causing invasive disease in HIV+ patients. Clin Microbiol Infect 17: 1094–1098.
  48. 48. Centers for Disease Control and Prevention (2012) Use of 13-valent pneumococcal conjugate vaccine and 23-valent pneumococcal polysaccharide vaccine for adults with immunocompromising conditions: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 61: 816–819.
  49. 49. Musher DM (2012) Editorial commentary: should 13-valent protein-conjugate pneumococcal vaccine be used routinely in adults? Clin Infect Dis 55: 265–267.
  50. 50. Song JY, Cheong HJ, Heo JY, Noh JY, Yong HS, et al. (2011) Clinical, laboratory and radiologic characteristics of 2009 pandemic influenza A/H1N1 pneumonia: primary influenza pneumonia versus concomitant/secondary bacterial pneumonia. Influenza Other Respi Viruses 5: e535–543.
  51. 51. Shrestha S, Foxman B, Dawid S, Aiello AE, Davis BM, et al. (2013) Time and dose-dependent risk of pneumococcal pneumonia following influenza: a model for within-host interaction between influenza and Streptococcus pneumoniae. J R Soc Interface 10: 20130233.
  52. 52. Nugent KM, Pesanti EL (1983) Tracheal function during influenza infections. Infect Immun 42: 1102–1108.
  53. 53. Shinya K, Ebina M, Yamada S, Ono M, Kasai N, et al. (2006) Avian flu: influenza virus receptors in the human airway. Nature 440: 435–436.
  54. 54. Walter ND, Taylor TH, Shay DK, Thompson WW, Brammer L, et al. (2010) Influenza circulation and the burden of invasive pneumococcal pneumonia during a non-pandemic period in the United States. Clin Infec Dis 50: 175–183.