What caused previously healthy children to die during the 2009 H1N1 influenza pandemic? Yesterday, the journal Pediatrics published the results of a study I conducted with the Pediatric Acute Lung Injury and Sepsis Investigator’s (PALISI) Network. Results have been widely reported, by the New York Times, the Washington Post, USA Today and TIME, among many others, provoking a lot of reader commentary, questions and, I fear, some misconceptions.
Our study collected data on 838 children with 2009 H1N1 infection admitted across 35 pediatric intensive care units (ICUs) in the U.S. Most of these children were severely ill, the majority requiring mechanical ventilator support for respiratory failure, and 9 percent died. Many (70 percent) had underlying illnesses like asthma or neurologic conditions that increased their risk. But among those who were previously healthy, the chief risk factor for death was co-infection with methicillin-resistant Staphylococcus aureus, or MRSA. It increased the risk of mortality 8-fold.
So what do these results mean? First and foremost, everyone eligible should get the flu vaccine. The 2009 pandemic H1N1 influenza A virus continues to circulate worldwide, and is included in this year’s vaccine along with the influenza A and B seasonal strains predicted to predominate. Influenza vaccine isn’t perfect, but it’s highly effective in stimulating antibody production and preventing severe infections.
Yet many families decide not to get their children vaccinated against the flu. Reasons include “not enough time – too much hassle,” “not worried – the pandemic flu strain wasn’t as bad as everybody expected,” and fear of rare vaccine-related complications. Some readers commenting on the news reports took our results as evidence that flu vaccination doesn’t work.
While our study couldn’t ascertain the children’s flu vaccination status for certain, we do know that almost 65 percent were admitted during September-October-November of 2009, when the 2009 H1N1 vaccine was just becoming widely available. So most probably hadn’t received it. They were left with no immunity to 2009 H1N1, leading to high rates of hospitalizations, school closures and deaths.
Now what about MRSA? Some commentators suggested that these critically ill children were really dying of MRSA, not flu. The truth is that influenza has long been known to be the most fatal when it strikes in combination with bacterial organisms. Organisms like Streptococcus pneumoniae (a.k.a pneumococcus) and Staphylococcus aureus often reside in people’s nasal passages as colonizing bacteria, usually causing no harm. But in the presence of an influenza infection, these organisms can become invasive and overcome host defenses. In the infamous influenza pandemic of 1918, for example, a very high proportion of the deaths was believed to be caused by a fatal combination of influenza and pneumococcal infections.
Many prior studies of influenza and bacterial co-infection are based on incomplete voluntary reporting methods or autopsy studies. In contrast, we rigorously screened our pediatric ICU patients to document cases of 2009 H1N1 infection and extracted data directly from medical records, interviewing the children’s doctors when facts needed confirmation. A third of the children who were critically ill with influenza had a clinically suspected or laboratory-confirmed bacterial co-infection that was thought to be present on admission. Staphylococcus aureus was cultured in the respiratory secretions of 8.5 percent of our 838 patients, but pneumococcus in fewer than 2 percent.
Alarmingly, almost half of the Staphyloccus aureus superinfections were with MRSA. That’s a very high rate of respiratory co-infection with a “superbug” that is common in the elderly but usually not seen at high rates in children. Surprisingly, the rate of MRSA co-infection of the lung was higher in the previously healthy children (6 percent) than in children with chronic medical conditions who are more likely to be hospitalized, a known risk factor for MRSA (3.2 percent).
Of the 18 previously healthy children in our study who died, 8 also had S. aureus, and 6 of them had MRSA. The 9 children who survived influenza-MRSA were extremely ill, most requiring weeks of mechanical ventilator support; two were on heart-lung bypass due to lung failure.
Our study adds to growing reports that Staphylococcus aureus, especially MRSA, is a rising cause of severe, often fatal infections in children. And while pneumococcus appears to be causing fewer severe influenza co-infections in U.S. children, likely because of pneumococcal immunization, there is no vaccine for S. aureus.
In 2003, after an especially severe strain of H3N2 influenza A caused a rise in pediatric deaths, the CDC initiated a registry with mandated reporting to track them. Lyn Finelli and colleagues from the CDC reported in 2008 that bacterial co-infections were involved in 6 percent of reported pediatric flu deaths in 2004-2005, 15 percent in 2005-2006, and 34 percent in 2006-2007. Almost all this increase was due to S. aureus — and most of these children hadn’t been vaccinated against flu.
The bottom line is that rates of MRSA infections are increasing in U.S. children and pose a special risk during flu season. Unfortunately, testing for MRSA isn’t a solution. While there is a regimen that can sometimes decolonize the nose, widespread or repeated attempts at decolonization pose the danger of making MRSA more resistant and even harder to get rid of.
That leaves preventing influenza infection as the best method of preventing this highly fatal flu-MRSA combination. And that makes vaccination against flu our #1 defense.
Adrienne Randolph, MD, MSc, of the Division of Critical Care Medicine at Children’s Hospital Boston, is founder and chair of the Pediatric Acute Lung Injury and Sepsis Investigator’s (PALISI) Network. She also directs the RSV and Asthma Research Study Center at Children’s. Randolph previously discussed flu vaccination in our sister blog, Thriving.