02 February 2010

Once again, flu and bacterial co-infection

With the H1N1 pandemic trending down, it may seem that the question of how much bacterial co-infection affects the outcome of flu is less important than it was. But though the pandemic is subsiding — for ever, for this season, or just until a third wave, who can say — researchers are just now getting enough good data to be able to make solid observations about what happened during the past 10 months.

Case in point: Writing in the journal Public Library of Science (PLoS) ONE, a team of researchers from Australia has pinpointed the incidence of MRSA co-infection during flu in two hospitals in Perth last summer, which was the Australian winter and the height of their flu season. Of 252 patients admitted for H1N1 infection, 3 were identified during treatment as having MRSA pneumonia. They survived, but two other patients who died were found to have MRSA pneumonia during post-mortem exams.
There were 3 female and 2 males, aged between 34 and 79 years... Two patients lived at the same long-term care facility, whilst the other patients lived independently in the community. Four of the 5 patients had conditions that may have increased their risk of pneumonia, including quadriplegia (two patients) asthma (one patient), cirrhosis (one patient) and diabetes mellitus (one patient). Two of the 5 cases (patients 3 and 4) had known MRSA infection/colonization prior to the onset of their illness (with the same cMRSA clone that subsequently caused their co-infection).
 There are some interesting points embedded here. First, incidence: In the Australian patients, MRSA pneumonia was much more common. The Perth researchers found 5 MRSA cases out of 252 flu patients. When the CDC analyzed the occurrence of MRSA pneumonia in flu last summer, it found only 1 case out of 272. Second, treatment: None of the 5 patients got antibiotics that would have affected MRSA — even though two of them were already known to be MRSA carriers. The possibility of MRSA pneumonia subsequent to flu seems not to have occurred to the health professionals taking care of them.

And third, the pathogen: The 5 Australian cases were caused by 3 community MRSA strains that are common in Australia — but only one of the 3 made PVL, the toxin that has so frequently been associated with MRSA pneumonia. That is interesting, and troubling at the same time. At this point, the association of PVL and necrotizing pneumonia has become practically taken for granted; and yet here are two strains that did not make PVL and yet caused severe and fatal pneumonia. It may be an indication that the inflammation that flu causes in the lung can open the door to more severe damage even when PVL is not present; it's certainly an indication that the absence of PVL does not signal a mild or not-dangerous strain.

The cite is: Murray RJ, Robinson JO, White JN, et al. 2010 Community-Acquired Pneumonia Due to Pandemic A(H1N1)2009 Influenzavirus and Methicillin Resistant Staphylococcus aureus Co-Infection. PLoS ONE 5(1): e8705. doi:10.1371/journal.pone.0008705.

Simultaneously, a new paper in the American Journal of Pathology seeks to clarify how often and in what circumstances bacterial superinfection becomes a risk during flu. Using a range of mice — both healthy ones, and "knockout" mice bred to be without particular immune-system components — researchers from San Diego confirmed that infections with flu and with Haemophilus influenzae can be lethal when the flu infection precedes the bacterial one. That was true even for infections that, if experienced separately, would not have been lethal; it was the synergy of the two infections, flu first followed by the bacterial infection, that caused the high mortality rate. The results may not be directly applicable to human medicine (Do you all know the old flu-research saying, "Mice lie and ferrets mislead?"), but they are an important indicator both of the seriousness of bacterial infection after flu, and also of the potential vulnerability of even healthy beings to that one-two punch.

The cite is: Lee LN, Dias P, Han D, et al.: A mouse model of lethal synergism between influenza virus and Haemophilus influenzae. Am J Pathol 176: 800-811.

No comments: