Issue: March 2012
March 01, 2012
11 min read

Secondary bacterial infections common with viruses

Researchers discuss secondary pneumococcus after influenza and other lethal viral–bacterial interactions.

Issue: March 2012
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

One hundred years ago, physicians treating infectious diseases would have been acutely aware of the danger of secondary bacterial infections after viral infection. As the 20th century wore on, a combination of increased specialization in health care research and the widespread use of penicillin created gaps in the understanding of the synergy between virus and bacteria. However, recent developments have brought viral–bacterial interactions back to the forefront of research in infectious diseases.

Jonathan A. McCullers, MD, of the department of infectious diseases at St. Jude Children’s Research Hospital in Memphis, Tenn., told Infectious Diseases in Children that up until the 1930s and 1940s, secondary bacterial infections after influenza infection were appreciated as a major public health concern that caused substantial morbidity and mortality.

“The concept of coinfection was lost for several decades due to the breakdown into a one-pathogen, one-host model,” McCullers said. “But this is not how infections work in the real world.” 

During the past 10 years, the infectious diseases specialty has done a complete turnaround and is now looking more broadly at everything from host–immune response to multiple pathogens, he said.

Dennis W. Metzger, PhD
Dennis W. Metzger, PhD, director of the Center for Immunology and Microbial Disease at Albany Medical College, said interferon-gamma plays a role in the treatment of bacterial infection following influenza.
Photo courtesy of Metzger DW

“Now everybody wants to study this,” McCullers said. “Major public health agencies are funding studies on viral interactions, and data are starting to trickle out that indicate what clinicians have known for some time: Secondary pneumonia following influenza infection is a big problem.”

Although most of the focus in the area of viral–bacterial interactions is on secondary infections after influenza, research also is being conducted on topics ranging from otitis media to respiratory syncytial virus and rhinovirus. The data that are emerging are a unique combination of new information and old information re-examined, and as the paths of study are redefined after a long hiatus, researchers often find themselves starting at the ground level.

History in the present

Although research has been moving away from specialization for nearly a decade, it was similarities between the 1918 and the 2009 influenza A H1N1 pandemics that brought coinfections back to the front of the discussion. 

In a recent review article in Antiviral Therapy, McCullers noted data sets that highlighted disproportionate trends in hospitalizations among younger patients who acquired 2009 H1N1 influenza. Further data indicated that as many as 50% of severe or fatal H1N1 infections were complicated by a secondary bacterial infection.

Keith P. Klugman
Keith P. Klugman

Keith P. Klugman, MB, BCh, PhD, William H. Foege Chair of Global Health at the Rollins School of Public Health at Emory University in Atlanta, told Infectious Diseases in Children that there was a doubling of mortality in US children as a result of the 2009 pandemic.

“Sixty percent of deaths were in kids who had underlying problems,” Klugman said. “But in otherwise healthy kids, the overwhelming number of deaths occurred in children who had secondary bacterial infections.”

McCullers and Dennis W. Metzger, PhD, professor and Theobald Smith Alumni Chair and director of the Center for Immunology and Microbial Disease at Albany Medical College, cited similar data from a recent CDC study of pathological specimens from the 1918 influenza pandemic.

“Over 90% of the people who died in that pandemic died from secondary bacterial infections,” Metzger said.

Klugman also noted similar figures from the 1918 pandemic and said there are parallels to be drawn between 1918 and 2009.

“A lot of basic science is currently being conducted on mortality rates linked to bacterial infections arising from viral respiratory infections,” Klugman said. “We are seeing a lot of staphylococcus and pneumococcus.”

“At the time, the explanation was that the virus causes so much tissue damage that it creates an environment conducive to a foothold for bacterial infection,” Metzger said. “We are now creating models to re-evaluate that hypothesis.”

Host-immune response

Metzger and Keer Sun, PhD, of Albany Medical College, conducted research in mice suggesting that interferon-gamma produced during T-cell responses to influenza infection can have an inhibitory effect on bacterial clearance from the lung by alveolar macrophages.

“This suppression of phagocytosis correlates with lung [interferon-gamma] abundance, but not viral burden, and leads to enhanced susceptibility to secondary pneumococcal infection, which can be prevented by [interferon-gamma] neutralization after influenza infection,” Metzger and Sun wrote.

Further results indicated that secondary pneumococcal infection can be prevented by interferon-gamma neutralization after influenza infection. Metzger and Sun concluded that direct inoculation of interferon-gamma can mimic influenza infection, which in turn causes a down-regulation in the expression of the class A scavenger macrophage receptor with collagenous structure (MARCO) on alveolar macrophages.

“Thus, [interferon-gamma], although probably facilitating induction of specific anti-influenza adaptive immunity, suppresses innate protection against extracellular bacterial pathogens in the lung,” they wrote.

Metzger said administering influenza alone to the mice was not lethal. It was only when they were given a bacterial infection a week later that fatalities occurred.

“It is important to note that the time of greatest susceptibility was the time when the virus was clearing,” he said. “There were high levels of virus early, but they were not susceptible to bacterial infection at that point.”

Similar findings were observed by Kash and colleagues, who infected mice with sublethal doses of 2009 seasonal H1N1 and pandemic H1N1, then followed this with infection with Streptococcus pneumoniae 48 hours later. Infection with pandemic disease plus S. pneumonia resulted in severe disease and a 100% fatality rate.

The researchers analyzed host response to infection, including mice that were coinfected with fatal results. Mice that died experienced significant loss of responses linked to lung repair. Also, enhanced bacterial replication was observed in the lungs of mice that had died as a result of coinfection.

“This study reveals that the extent of lung damage during viral infection influences the severity of secondary bacterial infections and may help explain some differences in mortality during influenza pandemics,” the researchers wrote. 

John F. Alcorn, PhD, assistant professor in the Division of Pulmonology, department of pediatrics, at Children’s Hospital of Pittsburgh of UPMC, and colleagues wrote that mutations in signal transducer and activator of transcription 3 (STAT3), which is required for Th17 immunity, were discovered in patients with hyper-immunoglobulin E syndrome, who often present with skin and lung Staphylococcus aureus infections. They hypothesized that Th17 cells may have a role in S. aureus pneumonia. Their research indicated that mice lacking either interleukin-17R or IL-22 had a decreased ability to clear S. aureus compared with wild-type mice.

John F. Alcorn, PhD
John F. Alcorn

“Little emphasis has been placed on T-cell–dependent immunity against S. aureus infections, despite the increasing incidence of methicillin-resistant S. aureus in the context of community-acquired pneumonia and as a secondary infection,” Alcorn said.

During the recent 2009 influenza pandemic, there were several reported cases of mortality associated with secondary S. aureus infection.

Alcorn’s group administered influenza A PR/8/34/H1N1 followed by

S. aureus to mice, and this resulted in increased inflammation and decreased ability to clear both the virus and the bacteria. Moreover, greater type I and II interferon production was observed in the lungs of mice with coinfection compared with those that had the virus alone. Also, coinfection with influenza A was linked to substantial decreases in IL-17, IL-22 and IL-23 production after S. aureus infection. The decrease in IL-17, IL-22 and IL-23 induced by S. aureus infection required type I interferon production in mice infected with influenza A, but was independent of type II interferon in these mice.

Over-expression of IL-23 in coinfected mice “rescued the induction of IL-17 and IL-22 and markedly improved bacterial clearance,” Alcorn and colleagues wrote. “These data indicate a novel mechanism by which influenza A-induced type I [interferons] inhibit Th17 immunity and increase susceptibility to secondary bacterial pneumonia.”

Regarding the study, Alcorn said: “Since the Th17 pathway is involved in host defense against a wide variety of pathogens, this mechanism of immunosuppression by viral infection may have broad implications in numerous respiratory diseases.”

The role of immunization

As researchers continue to re-examine the mechanisms by which these combination infections occur, similar investigations are taking place with regard to immunization. Toni Darville, MD, of the division of infectious diseases at the University of Pittsburgh Medical Center and an Infectious Diseases in Children Editorial Board member, said vaccinating children against a specific viral infection has a broader positive effect than just protection against the specific virus. 

“We have seen that giving influenza vaccination is linked to reductions in many secondary infections, including acute otitis media,” Darville said.

Toni Darville, MD
Toni Darville

Klugman and Metzger highlighted the cross-protective effect of influenza virus immunization against multiple strains of influenza. Metzger and Sun conducted some basic research in this area by examining the ability of live-attenuated influenza vaccine (FluMist, MedImmune) to mediate protection against H1N1 infection, which originated in swine. Although mice that were not vaccinated were found to be susceptible to H1N1 Cal/04/09 infection, those that received the vaccine were less likely to experience morbidity or mortality. Experimental depletion of CD4 and CD8 cell counts indicated that protective immunity was dependent on vaccine-induced CD4 cells but not CD8 cells. Serum or mucosal Abs played a minimal role in cross-protection in passive protection studies, according to the results.

The study findings indicated that mice that were not vaccinated experienced intensive recruitment of phagocytes as a result of H1N1 infection, but these mice experienced severe suppression of pulmonary innate defense against secondary pneumococcal infection. A correlation was observed between increased susceptibility and increased production of interferon-gamma produced during the recovery stage of infection with H1N1. This increased production of interferon-gamma was suppressed in vaccinated mice.

When type II interferon signaling was absent, susceptibility to secondary bacterial infection was decreased. However, susceptibility was not decreased in the absence of type I interferon signaling.

Metzger and Sun concluded that seasonal immunization with the live influenza vaccine not only promoted resistance to pandemic H1N1 influenza infection but also restored innate immunity against complicating secondary bacterial infections. They also said some clinicians have argued for a combination vaccine that protects against viral and bacterial infections.

“This may not be a bad idea, but you still can’t overcome the macrophage,” Metzger said.

“We need to do everything we can to reduce these secondary bacterial infections because this translates into fewer antimicrobial prescriptions,” Darville said.

New drug development

Although antibiotic overuse is always a concern, the discussion has recently moved toward whether more comprehensive use of antiviral medications is warranted.

“This is a polarizing argument,” McCullers said. “We currently reserve them for the sickest patients. But if you use them in everybody, you can also prevent bacterial pneumonia. Antivirals should help everybody.” 

He said the data on antiviral medications are not solid, and the CDC is being careful in its approach to the issue.

Jonathan A. McCullers, MD
Jonathan A. McCullers

“We need a firm guideline for antiviral use,” he said. “Some doctors never give them, while others give them for every patient. We clearly need to develop options so we don’t just think of them as something to be used in severe cases.”

Darville highlighted the possibility of bacterial infection in viral otitis media, and McCullers said human metapneumovirus, rhinovirus, respiratory syncytial virus and parainfluenza all can lead to secondary bacterial infections. However, the data on these combinations are not clear, and subsequently, effective antiviral treatments are still in the distance.

“Fluid and swelling in the ear and nose in otitis media creates an environment amenable to bacteria,” Darville said. “These are common clinical scenarios, and more effective treatments would be beneficial.”

Preventing infections

Until antiviral therapies become more prevalent, identifying infections before they occur may be the most effective way of combating the problem. William I. Krief, MD, of the departments of pediatrics and emergency medicine at Long Island Jewish Hospital/Steven and Alexandra Cohen Children’s Medical Center in New Hyde Park, N.Y., and colleagues aimed to determine the risks for serious bacterial infections in febrile infants with influenza infections and then compare these risks with those in febrile infants who did not have influenza.

The multicenter, prospective, cross-sectional study was conducted at five participating EDs during three consecutive influenza seasons (October to March) from 1998 to 2001. Eligible infants were febrile and aged 60 days or younger. Blood, urine, cerebrospinal fluid and stool cultures were taken from these infants.

The researchers defined urinary tract infection (UTI) by single-pathogen growth of > 5 x 104 colony-forming units per milliliter (CFU/mL) or > 104 CFU/mL in association with a positive urinalysis, according to the results. They said bacteremia, bacterial meningitis and bacterial enteritis were defined by growth of a known bacterial pathogen, and a serious bacterial infection was defined as any of the previously mentioned infections.

There were 1,091 infants initially enrolled. Of 844 (77.4%) infants who were tested for influenza, 123 (14.3%) tested positive. The serious bacterial infection status was determined in 809 (95.9%) infants. The serious bacterial infection rate was 11.7% in those 809 infants.

Lower rates of serious bacterial infections (2.5%) and UTIs (2.4%) were reported in infants with influenza infections compared with those who tested negative for influenza. No cases of bacteremia, meningitis or enteritis were observed in the children who tested positive for influenza. However, no statistically significant differences were observed between the positive and negative groups for these individual infections.

Krief told Infectious Diseases in Children that febrile infants are at high risk for serious bacterial infections, with rates of 6% to 12%.

“Due to this high risk, current clinical standard of care is to obtain blood, urine and [cerebrospinal fluid] for evaluation of bacterial infections,” he said. “What our study and others have shown is that febrile infants with proven viral infections are at significantly lower risk for these bacterial infections. However, the risk of UTIs remained clinically significant.”

Krief said febrile infants with a virus are at lower risk for UTI, but the risk is still high enough to warrant evaluation for UTI.

“For the clinician, identifying infants at low risk for serious bacterial infections — those with viruses like [respiratory syncytial virus] and influenza — may lead to a reduction of laboratory testing and antibiotic use,” he said. Krief noted that the rate of social bacterial infection was lower in his group’s patient population. “However, this is probably because these were primary influenza infections, and it takes time for secondary infections to evolve.”

Whether treatment moves toward increased immunization, more antivirals, better antibiotics or earlier identification, it is clear that the paradigm must change, according to the experts who spoke with Infectious Diseases in Children.

“Most pediatricians will say that the virus damages the lung tissue,” Metzger said. “This has been the model for the last 90 years, but it has not gotten us where we need to go. We need to develop more effective counter-measures because this problem isn’t going anywhere.” – by Rob Volansky


  • Gonzalez-Galan V. Clin Microbiol Infect. 2011;17:1895-1899.
  • Kash JC. MBio. 2011;doi:10.1128/mBio.00172-11.
  • Krief WI. Pediatrics. 2009;124:30-39.
  • Kudva A. J Immunol. 2011;186:1666-1674.
  • Perez-Padilla R. N Engl J Med. 2009;361:680-689.
  • Pierce VM. J Clin Microbiol. 2011;doi:10.1128/?JCM.05996-11.
  • Sun K. J Immunol. 2011;186:987-993.
  • Sun K. Nat Med. 2008;14:558-564.


  • Dr. Alcorn reports receiving NIH NHLBI funding for co-infection work. Dr. Klugman reports consulting for manufacturers who produce both influenza and pneumococcal vaccines. Drs. Darville, McCullers and Krief report no financial disclosures.

What is the role of antiviral therapy in treating children with influenza and preventing secondary bacterial infections?

Ravi Jhaveri, MD
Ravi Jhaveri

Studies are lacking, but it appears that antivirals to prevent influenza indirectly prevent secondary bacterial infections. 

It is known that direct damage to respiratory epithelium during influenza is a major factor in bacterial superinfection. It is also known that viral proteins serve to facilitate adherence of bacteria to epithelial cells. It would seem obvious that by treating influenza with antiviral agents, one would make an impact in preventing secondary bacterial infections. This has been demonstrated in animals (McCullers JA. J Infect Dis. 2004;190:519-526), but actual patient outcomes data are lacking. It is important to acknowledge the studies that have demonstrated a modest benefit of treating influenza with antivirals. A 1-day reduction in overall symptoms with overall reduced severity occurs if the medication is initiated within 24 to 48 hours of symptom onset.

Regarding whether current antiviral practices should increase or decrease: If a child presents in the appropriate window when treatment has been shown to be beneficial, we should start therapy. I would say that given this, treatment should increase. However, I would also emphasize that we should increase all our efforts to prevent influenza prior to treatment, which means increased rates of vaccination, more hand washing and better diagnostics. If we can prevent the cases of influenza, the secondary bacterial infections will also be prevented.

As for adherence to recommendations, I would stick with the current guidelines. We emphasize appropriate testing to confirm the diagnosis and then prioritize patients for treatment based on their risk of severe disease. We need to do what we are doing better.

Ravi Jhaveri, MD, is Assistant Professor of Pediatrics and Assistant Professor of Molecular Genetics and Microbiology at Duke University School of Medicine in Durham, N.C. Disclosure: Dr. Jhaveri reports receiving research funding from Merck and mentoring a trainee that has a training grant from BMS.


Stephen C. Eppes, MD
Stephen C. Eppes

Antivirals are effective treatment for influenza, but we must be cautious of viral resistance.

What we know about antivirals is that they reduce the duration of illness, they decrease the amount of viral shedding, and they reduce clinical symptoms. There are data that show that oseltamivir (Tamiflu, Roche) can prevent secondary infections, such as otitis media or lower respiratory complications, and can also reduce the duration of hospitalization in children. Some studies indicate that pneumonia can be prevented by oseltamivir, but there are fewer data in children than in adults. 

Regarding the question of the appropriateness of antiviral use: It is well accepted by the CDC and AAP that any child with influenza requiring hospitalization should receive treatment without necessarily awaiting virologic confirmation. In addition, children with underlying medical conditions and those younger than 2 years of age (who have higher rates of hospitalizations and complications) should be treated. More liberal use of antivirals may prevent some secondary bacterial infections, but we have to consider that widespread use of these drugs may increase viral resistance.

We should be adhering to current recommendations for treating influenza, but one point that deserves mention is the recommendation for children younger than 1 year. In 2009, there was an emergency use authorization (EUA) because there was no FDA approval for that age group. That EUA has now expired, but I would still heartily endorse using oseltamivir for children younger than 1 year. That said, for children younger than 3 months, for whom there are no good safety and pharmacokinetic data, I would recommend consulting with an expert.

Stephen C. Eppes, MD, is Chief of the Division of Infectious Diseases at the Alfred I. duPont Hospital for Children in Wilmington, Del., and Professor of Pediatrics at Jefferson Medical College in Philadelphia. Disclosure: Dr. Eppes reports no relevant financial disclosures.