Tosh, Pritish K. MD*†‡; Bulens, Sandra N. MPH†; Nadle, Joelle MPH§; Dumyati, Ghinwa MD∥; Lynfield, Ruth MD¶; Schaffner, William MD#; Ray, Susan M. MD**††; Jain, Seema MD‡‡; Fridkin, Scott K. MD†; Sievert, Dawn M. PhD†
In the late 1990s, community-associated methicillin-resistant Staphylococcus aureus (MRSA) strains were identified in previously healthy individuals and in outbreaks; epidemiologic studies suggested these strains may be more easily transmissible or more likely to result in infection than other S aureus strains.1–11 The emergence of MRSA in community settings was accompanied by anecdotal reports and perceptions that these infections were more severe than corresponding infections caused by MRSA strains traditionally confined to health care settings.12,13 These observations coupled with limited treatment options make patient care challenging despite guideline recommendations for empiric antimicrobial therapy.14–17
Lower respiratory tract infections (LRIs) with S aureus requiring hospitalization in young persons was increasingly reported during the 2003 influenza season.10,18–20 The complications of these community-associated S aureus respiratory infections included necrotizing pneumonia with an associated mortality as high as 51%.19 The role of MRSA in hospitalized LRI is not clear, and much of what is known about this association is from case reports and case series. This includes observations that community-acquired S aureus LRI tend to affect children and young adults (median age, 16–21 years), are predominantly caused by MRSA (73%–88%), have high morbidity (79%–81% receive critical care) and mortality (29%–51%), and occur with or after influenza infection (33%–71% testing positive for influenza).10,18–21 As case series are subject to reporting bias, further studies are needed to explore these observations.
To help understand the relationship between MRSA, influenza, and community-onset S aureus LRI requiring hospitalization in otherwise healthy individuals, we performed a systematic surveillance for these infections in a large, geographically diverse population. The objectives of this study were to (1) describe the demographic and clinical characteristics of patients with community-onset S aureus LRI requiring hospitalization, and (2) evaluate potential associations between influenza infection and community-onset S aureus LRI requiring hospitalization.
MATERIALS AND METHODS
Design and Patients
Surveillance was established through the Active Bacterial Core Surveillance (ABCs) system within the Emerging Infections Program (EIP), an established surveillance network of the Centers for Disease Control and Prevention (CDC) as previously described.22 Active Bacterial Core Surveillance/EIP has performed active population-based laboratory surveillance for invasive MRSA cultured from sterile body sites since 2005. Five of the 9 Active Bacterial Core Surveillance/EIP sites participating in invasive MRSA surveillance participated in this study. Of the 5 sites, 3 sites performed population-based surveillance: San Francisco County, CA; Ramsey and Hennepin Counties, MN; and Monroe County, NY. Sentinel surveillance was performed in 2 large acute care facilities (one urban and one suburban) in the Atlanta, GA metropolitan area, and in one large acute care facility in Davidson County, TN. For the duration of this study, these 5 sites deviated from routine invasive MRSA surveillance by identifying all cases of S aureus (not just MRSA) isolated from blood and respiratory specimens (rather than sterile sites alone). All 5 sites included hospitals with pediatric patients. Human subjects review at the CDC and the other participating surveillance sites determined this to be a nonresearch public health surveillance activity except for the surveillance site in California, where the study protocol was approved by the institutional review boards of the California Committee for the Protection of Human Subjects and the participating health care facilities.
Trained surveillance staff reviewed laboratory test results of the cases. In addition to routine surveillance for bacteremia, for this study, the staff also reviewed all respiratory cultures growing S aureus from all participating laboratories between September 1, 2008 and August 31, 2010 (Fig. 1). Respiratory culture was defined to include samples from sputum, bronchoalveolar lavage fluid, lung tissue, or pleural fluid. Patients eligible for record review (suspect cases) included hospitalized residents of the surveillance areas (or receiving care at a sentinel surveillance hospital), younger than 50 years with a positive blood or respiratory culture for S aureus, which was collected within 2 days of hospital admission. Suspect case patients without recent history of hospitalization or presence of an indwelling medical device on admission underwent complete medical record review. Recent hospitalization was defined as having an overnight stay in a hospital or long-term care/skilled nursing facility in the previous 12 weeks. Indwelling medical device was defined as a central vascular catheter, tracheostomy, gastrostomy, urinary catheter, or any hemodialysis access device at the time of hospital admission.
Abstracted data included demographics, medical comorbidities, clinical data from the first 4 hospital days, discharge diagnoses, and clinical outcomes. Evidence of influenza infection was defined as a positive laboratory test for influenza by polymerase chain reaction or rapid influenza diagnostic tests, receipt of influenza antiviral agents between 7 days before admission and 2 days after admission, or a history of influenza diagnosis within 10 days before the initial S aureus culture. Periods of influenza circulation were defined using national weekly summary reports of positive influenza tests reported to the CDC/Influenza Division by US World Health Organization (WHO) and National Respiratory and Enteric Virus Surveillance System collaborating laboratories.
Cases were classified as a confirmed case when a suspect case had evidence of an infection caused by S aureus listed in the discharge summary. Cases with a diagnosis of pneumonia listed in the discharge summary then underwent further review to identify those with objective evidence of LRI to minimize misclassification of hospitalized patients with exacerbations of chronic conditions mimicking pneumonia (eg, congestive heart failure and chronic obstructive pulmonary disease) with respiratory specimens contaminated with S aureus. Objective evidence of LRI included (at the time of hospital admission) fever, and either (a) dyspnea, hypoxia, rales/bronchial breath sounds, hemoptysis, tachypnea, or (b) radiographic evidence of LRI (radiology report of bronchopneumonia/pneumonia, air space density/opacity, cavitation, consolidation, single lobar infiltrate, interstitial infiltrate, unilateral or bilateral multilobar infiltrate, or new or changed infiltrate), or (c) blood cultures positive for S aureus and pneumonia as the only infectious disease diagnosis listed in the discharge summary. Cases with a pneumonia diagnosis in the discharge summary without objective evidence of LRI at admission were excluded from further analysis.
Correlation of medical comorbidities and clinical outcomes were performed using the Charlson Comorbidity Index score for each case, comparing those with no substantial medical comorbidity (Charlson score, zero) to those with any substantial medical comorbidity (Charlson score ≥1).23 Cases were evaluated for microbiologic characteristics (eg, specimen source, methicillin resistance, and other sites of infection), demographic and clinical characteristics (eg, age, sex, prior hospitalizations, preceding influenza infection, bacteremia, and specific medical comorbidities), and clinical outcomes (eg, need for care in an intensive care unit [ICU], inhospital mortality, and length of hospital stay). Further analyses were performed on the subset of case patients with LRI to compare risk factors and outcomes between those infected with MRSA versus MSSA as well as those with and without influenza infection.
Comparative analyses were performed using the χ2 test, the Fisher exact test, and the Kruskal-Wallis test for comparison of medians. Statistical significance was defined as P < 0.05. All statistical analyses were conducted using SAS software version 9.2 (SAS Institute Inc, Cary, NC).
A total of 708 suspect hospitalized community-onset S aureus case patients aged 50 years or younger were identified; no S aureus infectious syndrome was listed in the discharge summary in 76 patients, and 107 patients were classified as having pneumonia but did not have supporting evidence for LRI and were excluded (Fig. 1). The remaining 525 patients were confirmed hospitalized community-onset S aureus cases; 154 were identified from the San Francisco, CA surveillance site; 95 were identified from the Ramsey and Hennepin Counties, MN surveillance site; 118 were identified from the Monroe County, NY surveillance site; 87 were identified from the Atlanta, GA surveillance site; and 72 were identified from the Nashville, TN surveillance site. Approximately half (45%) of these case patients were infected with MRSA.
Confirmed hospitalized community-onset S aureus case patients aged 50 years or younger had a median age of 38 years, and 67% were men (Table 1). Smoking (29%) and type 2 diabetes mellitus (17%) were the most common medical comorbidities, although 65% had a Charlson score of zero, indicating that they had no significant medical comorbidities. A total of 154 case patients (30%) required ICU care within 7 days of hospital admission, and 41 case patients (8%) died during the hospitalization. The median length of hospital stay was 8 days. Ninety-four (18%) of the confirmed hospitalized community-onset S aureus case patients met our definition for LRI; 25 case patients were identified from the San Francisco, CA surveillance site; 19 were identified from the Ramsey and Hennepin Counties, MN surveillance site; 27 were identified from the Monroe County, NY surveillance site; 11 were identified from the Atlanta, GA surveillance site; and 12 were identified from the Nashville, TN surveillance site. The other 431 case patients had other S aureus infectious syndromes identified in their discharge summary including 103 (24%) with skin or soft tissue infections, 62 (14%) with bone or joint infections, 89 (21%) with blood stream infections without another identified source, 39 (9%) with cardiac infections, 157 (36%) with other infectious syndromes, such as meningitis, peritonitis, internal abscesses, and otitis, and 19 (4%) with 2 or more non-LRI diagnoses.
Confirmed cases occurred throughout the surveillance period with roughly equal distribution by month (Fig. 2), regardless of status as LRI or non-LRI cases. There was no overt clustering around a single time period. Of note, influenza, either 2008–2009 seasonal influenza or influenza A(H1N1)pdm09, was in circulation for 14 months of the 24-month study period (from December 2008 through January 2010). Lower respiratory tract infection cases were as likely to occur during months when influenza was circulating (n = 64) as in other months (20% vs. 15%; P = 0.24).
The proportion of case patients with medical comorbidities was similar between the LRI case patients and non-LRI case patients: 57 (61%) of the LRI case patients and 285 (66%) of the non-LRI case patients had a Charlson score of zero; P = 0.12, meaning that they had no underlying medical comorbidities known at the time of hospitalization. Methicillin-resistant Staphylococcus aureus infection was found in 42 (45%) of 94 LRI case patients and 194 (45%) of 431 non-LRI case patients (P = 1.0).
Characteristics of the LRI cases are summarized in Table 2. Forty percent of the LRI case patients required ICU care and 6% of the LRI case patients died. Although 64 LRI cases occurred during the 14 months of influenza circulation, only 35 case patients (55%) were tested for influenza at hospital admission. Of the 35 LRI case patients who were tested for influenza during hospitalization, 9 (26%) were positive for influenza infection. Nine (14%) of the 64 LRI case patients during periods of influenza circulation received influenza antiviral agents within a week before or 2 days after admission and 2 case patients (3%) had a history of influenza diagnosis within 10 days before hospital admission. Overall, 13 (20%) of the 64 LRI case patients during periods of influenza circulation had evidence of recent or concurrent influenza infection. Concurrent S aureus infections at other sites were rarely seen among the LRI cases, with 3 skin and soft tissue infections (3%) and 5 internal abscesses (5%).
Lower respiratory tract infection cases with evidence of concurrent or recent influenza infection had significantly higher inhospital mortality compared to LRI cases without influenza infection (31% vs. 2%; P = 0.003; Table 3). The LRI case patients with and without influenza infection had similar baseline characteristics including age, sex, hospitalization within the past year, and Charlson score. The LRI case patients with influenza infection had similar frequency of infection with MRSA, need for ICU care, and duration of hospital stay compared with the LRI cases without influenza infection. Of the 4 LRI case patients with evidence of influenza infection requiring ICU care, 3 patients were bacteremic with S aureus, 3 patients had influenza infection confirmed by laboratory test result, and all four died.
Among the LRI case patients, those with MRSA were significantly more likely to have been hospitalized in the prior year compared to those with MSSA (33% vs. 13%; P = 0.03; Table 4). Other demographic and medical comorbidity parameters were similar between the 2 groups. Of the outcome variables examined, there was only a significant difference detected for median length of stay between LRI case patients with MRSA (9 days) compared to those with MSSA (6 days, P = 0.04). There was a trend toward increased mortality among the LRI case patients with MRSA (12%) compared to the LRI case patients with MSSA (2%, P = 0.09), although this did not reach statistical significance. Overall, the LRI case patients with MRSA were as likely to have evidence of concurrent or recent influenza infection as the LRI case patients with MSSA (17% vs. 12%; P = 0.55). When stratifying the LRI case patients by the Charlson score, a higher proportion of the LRI case patients were admitted to the ICU among the case patients with a Charlson score of zero compared to case patients with scores greater than zero (27 [50%] of 57 vs. 9 [25%] of 37; P = 0.03). However, there were no statistically significant differences in the proportion of infections caused by MRSA, proportion of cases with bacteremia, or inhospital mortality between the case patients with a Charlson score of zero and the case patients with a Charlson score greater than zero.
This study was able to systematically identify persons aged 50 years or younger without recent hospitalization in a multisite surveillance effort across several geographically diverse areas with community-onset S aureus LRI requiring hospitalization. Our findings suggest that S aureus LRI is not a rare event among previously healthy persons aged 50 years or younger, occurs without any demonstrable seasonality, and can occur with or without influenza infection. The morbidity of S aureus LRI was high (40% of the case patients required ICU care), and high mortality (6%) was observed especially among those with evidence of recent or concurrent influenza infection (31% mortality). Less than half of the LRI case patients were tested for influenza despite influenza circulating for 14 of the 24 months of surveillance. The percentage of hospitalized community-onset invasive or respiratory S aureus infections caused by MRSA was approximately 50%, which was similar between the LRI and non-LRI cases.
Despite not including persons older than 50 years in this study, the median age of the case patients (38 years) was substantially higher than that reported by 2 case series where the median ages were 16 and 21 years.18,19 In these case series, which included 17 and 51 cases, 73% to 88% of S aureus LRI was caused by MRSA, which is higher than what was observed in our study, where 45% of hospitalized community-onset S aureus LRI were caused by MRSA.18,19 The high visibility of community-associated MRSA or regional differences in relative strain abundance in the community may have biased these initial case reports of S aureus LRI toward an association with MRSA.
Our study found S aureus LRI cases to have substantial morbidity and mortality; however, the morbidity and mortality was less than what had previously been described by case series (79%–81% requiring ICU care and 29%–51% dying), which suggests some bias toward overreporting severe cases in the case series.18,19 This difference in morbidity and mortality may also be related to the higher frequency (33%–71%) of influenza coinfection reported in prior case series.18,19 Consistent with what was reported in prior case series, our study found that those with S aureus LRI with evidence of influenza infection had high mortality (31%).18,19 In our study, only 55% of the hospitalized community-onset S aureus LRI case patients were tested for influenza during times of influenza circulation. The high percentage of influenza and S aureus coinfection (26%) seen among our LRI case patients who were tested, the high mortality associated with coinfection, and the low levels of testing highlight the need for increased influenza testing of patients hospitalized with LRI while influenza is circulating, as recommended by Infectious Diseases Society of America/American Thoracic Society guidelines.16
The increased mortality associated with S aureus LRI cases with influenza infection likely stems from the synergistic nature of these 2 pathogens to produce severe respiratory illness. Although the exact pathogenesis of this synergy is still the subject of substantial ongoing basic and clinical research, influenza infection does seem to increase the vulnerability of the host to severe bacterial disease.24–27 Experimental coinfection in mice with influenza and S aureus was found to result in increased titers of both pathogens and higher mortality compared with infection with either pathogen alone.24
We also found a similar proportion of bacteremia and need for ICU care among the hospitalized community-onset S aureus LRI case patients with MRSA compared to those with MSSA. There was a significantly longer duration of hospital stay among the case patients with MRSA (9 days) compared to those with MSSA (6 days). A trend toward higher mortality (12% for MRSA and 2% for MSSA; P = 0.09) was also seen among the LRI case patients with MRSA; this was not a statistically significant difference but may have been limited by the small sample size. It is not clear from this study whether these findings were due to increased virulence of the MRSA strains independent of antimicrobial resistance, reduced effectiveness of antimicrobials used to treat MRSA, patient factors not captured by this study, or a combination of these factors. Case patients with LRI with MRSA were more likely to have been hospitalized in the prior year than those infected with MSSA. This suggests a few possible explanations including that there may have been hospital-acquired, community-onset MRSA infections included in our study and that illness requiring hospitalization may set up patients for subsequent colonization with MRSA in the hospital and the community. Hospitalized community-onset S aureus LRI case patients with Charlson scores of zero were twice as likely to require ICU care as those with Charlson scores greater than zero, although the reason for this association is unclear.
There were several limitations to this study. The surveillance population in this study was almost entirely urban and was not population-based in two of the study sites, limiting the generalizability of the findings. This study only included hospitalized community-onset S aureus case patients with bacteremia or positive respiratory cultures and was not designed to capture other important community-onset S aureus syndromes such as skin or soft tissue infections, septic arthritis, or epidural abscess with negative blood cultures. Case finding may have allowed inclusion of patients without a definitive lower respiratory infection, as (1) respiratory cultures often grow S aureus as an incidental finding among patients with other pulmonary signs and symptoms and (2) patients were included if they had a discharge diagnosis of pneumonia and blood culture positive for S aureus but did not have supporting clinical or radiographic evidence for LRI. However, the number of confirmed LRI cases without true pneumonia was likely small because (1) patients without clinical or radiographic evidence of LRI were excluded and (2) patients included because of positive blood culture had pneumonia listed in their discharge summary as the only infectious disease diagnosis. S aureus isolates from case patients were not collected to analyze potential associations with strain type (eg, USA300) or virulence factors. Another limitation was the relatively small number of patients in the study, which hindered our ability to detect significant differences in patient outcomes and risks such as a potential association between MRSA LRI and increased mortality.
The results of this study with regard to influenza infection need to be interpreted with the caveat that the sensitivity and specificity of indicators of influenza infection were suboptimal. Testing for influenza was not standardized and was not performed in most of the LRI case patients. The method of influenza testing was also not standardized, and the rapid antigen test (which was used commonly) had suboptimal sensitivity and specificity for detecting influenza A(H1N1)pdm09 infections. Furthermore, with treatment guidelines recommending empiric antiviral therapy in hospitalized or severely ill patients with suspected influenza infection, it is possible that the 4 patients with negative influenza tests who were considered to have evidence of influenza infection due to receipt of antivirals may have been misclassified. A large part of this study was performed during the 2009 pandemic, which may affect the generalizability of the findings to seasonal outbreaks, especially with regard to age distribution and temporal distribution.
In summary, this study demonstrates that in these 5 metropolitan areas, community-onset S aureus LRI requiring hospitalization affects adults (and not just younger adults as suggested by case series) and is not predominantly associated with MRSA and influenza activity does not have a major impact on incidence but does seem to affect severity. We observed high mortality with influenza coinfection but a low level of influenza testing among patients with community-onset S aureus LRI requiring hospitalization, emphasizing the importance of performing influenza testing on patients presenting with LRI during influenza season, as recommended by the Infectious Diseases Society of America/American Thoracic Society guidelines.16 By providing systematically derived epidemiologic evidence, the findings of our study support the observation made by previously published case series that S aureus LRI has high morbidity and mortality.18–20 Furthermore, we provide new epidemiologic evidence that S aureus LRI occurs without any demonstrable seasonality and can occur without influenza infection. Further studies are needed to confirm these observations and to devise prevention strategies.
The authors acknowledge the contributions of the following people who were vital to the work represented in this manuscript: Anna Bramley, Lauren Pasutti, Pam Daily, Art Reingold, Janine Ladson, Wendy Baughman, Monica Farley, Lindsey Lesher, Anita Glennen, Billie Juni, Elliot Stivers, Anita Gellert, Brenda Barnes, Terri McMinn, Katie Gore, and Susan Tymensky.
1. Four pediatric deaths from community-acquired methicillin-resistant Staphylococcus aureus—Minnesota and North Dakota, 1997–1999. MMWR Morb Mortal Wkly Rep. 1999; 48: 707–710.
2. Methicillin-resistant Staphylococcus aureus infections among competitive sports participants—Colorado, Indiana, Pennsylvania, and Los Angeles County, 2000–2003. MMWR Morb Mortal Wkly Rep. 2003; 52: 793–795.
3. Faden H, Ferguson S . Community-acquired methicillin-resistant Staphylococcus aureus and intrafamily spread of pustular disease. Pediatr Infect Dis J. 2001; 20: 554–555.
4. Fridkin SK, Hageman JC, Morrison M, et al. Methicillin-resistant Staphylococcus aureus disease in three communities. N Engl J Med. 2005; 352: 1436–1444.
5. Herold BC, Immergluck LC, Maranan MC, et al. Community-acquired methicillin-resistant Staphylococcus aureus in children with no identified predisposing risk. JAMA. 1998; 279: 593–598.
6. Huijsdens XW, van Santen-Verheuvel MG, Spalburg E, et al. Multiple cases of familial transmission of community-acquired methicillin-resistant Staphylococcus aureus. J Clin Microbiol. 2006; 44: 2994–2996.
7. Kazakova SV, Hageman JC, Matava M, et al. A clone of methicillin-resistant Staphylococcus aureus among professional football players. N Engl J Med. 2005; 352: 468–475.
8. L’Heriteau F, Lucet JC, Scanvic A, et al. Community-acquired methicillin-resistant Staphylococcus aureus and familial transmission. JAMA. 1999; 282: 1038–1039.
9. Naimi TS, LeDell KH, Como-Sabetti K, et al. Comparison of community- and health care–associated methicillin-resistant Staphylococcus aureus infection. JAMA. 2003; 290: 2976–2984.
10. Sattler CA, Mason EO Jr, Kaplan SL . Prospective comparison of risk factors and demographic and clinical characteristics of community-acquired, methicillin-resistant versus methicillin-susceptible Staphylococcus aureus infection in children. Pediatr Infect Dis J. 2002; 21: 910–917.
11. Shahin R, Johnson IL, Jamieson F, et al. Methicillin-resistant Staphylococcus aureus carriage in a child care center following a case of disease. Toronto Child Care Center Study Group. Arch Pediatr Adolesc Med. 1999; 153: 864–868.
12. Gillet Y, Vanhems P, Lina G, et al. Factors predicting mortality in necrotizing community-acquired pneumonia caused by Staphylococcus aureus containing Panton-Valentine leukocidin. Clin Infect Dis. 2007; 45: 315–321.
13. Miller LG, Perdreau-Remington F, Bayer AS, et al. Clinical and epidemiologic characteristics cannot distinguish community-associated methicillin-resistant Staphylococcus aureus infection from methicillin-susceptible S. aureus infection: a prospective investigation. Clin Infect Dis. 2007; 44: 471–482.
14. Daum RS, Seal JB . Evolving antimicrobial chemotherapy for Staphylococcus aureus infections: our backs to the wall. Crit Care Med. 2001; 29: N92–N96.
15. Liu C, Bayer A, Cosgrove SE, et al. Clinical practice guidelines by the infectious diseases society of america for the treatment of methicillin-resistant Staphylococcus aureus infections in adults and children: executive summary. Clin Infect Dis. 52: 285–292.
16. Mandell LA, Wunderink RG, Anzueto A, et al. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis. 2007; 44:(suppl 2): S27–S72.
17. Stevens DL, Bisno AL, Chambers HF, et al. Practice guidelines for the diagnosis and management of skin and soft-tissue infections. Clin Infect Dis. 2005; 41: 1373–1406.
18. Hageman JC, Uyeki TM, Francis JS, et al. Severe community-acquired pneumonia due to Staphylococcus aureus, 2003–04 influenza season. Emerg Infect Dis. 2006; 12: 894–899.
19. Kallen AJ, Brunkard J, Moore Z, et al. Staphylococcus aureus community-acquired pneumonia during the 2006 to 2007 influenza season. Ann Emerg Med. 2009; 53: 358–365.
20. Kallen AJ, Hageman J, Gorwitz R, et al. Characteristics of Staphylococcus aureus community-acquired pneumonia during the 2006–2007 influenza season. Clin Infect Dis. 2007; 45: 1655
21. Roberts JC, Gulino SP, Peak KK, et al. Fatal necrotizing pneumonia due to a Panton-Valentine leukocidin positive community-associated methicillin-sensitive Staphylococcus aureus and Influenza co-infection: a case report. Ann Clin Microbiol Antimicrob. 2008; 7: 5
22. Klevens RM, Morrison MA, Nadle J, et al. Invasive methicillin-resistant Staphylococcus aureus infections in the United States. JAMA. 2007; 298: 1763–1771.
23. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. Journal of Chronic Diseases. 1987; 40: 373–383.
24. Iverson AR, Boyd KL, McAuley JL, et al. Influenza virus primes mice for pneumonia from Staphylococcus aureus. J Infect Dis. 2011; 203: 880–888.
25. McCullers JA, McAuley JL, Browall S, et al. Influenza enhances susceptibility to natural acquisition of and disease due to Streptococcus pneumoniae in ferrets. J Infect Dis. 2010; 202: 1287–1295.
26. Parker D, Prince A . Immunopathogenesis of Staphylococcus aureus pulmonary infection. Semin Immunopathol. 2012; 34: 281–297.
27. Smith AM, Adler FR, McAuley JL, et al. Effect of 1918 PB1-F2 expression on influenza A virus infection kinetics. PLoS Comput Biol. 2011; 7: e1001081