Bacteria May Cause H1N1 Death Rate Increase

[Guest guest]

Bacteria May Cause H1N1 Death Rate Increase

By , North American Correspondent, MedPage Today

Published: December 31, 2009

Reviewed by Zalman S. Agus, MD; Emeritus Professor

University of Pennsylvania School of Medicine and

Dorothy Caputo, MA, RN, BC-ADM, CDE, Nurse Planner

A bacterial coinfection with the H1N1 pandemic flu may explain the unusually

high case fatality seen in Argentina earlier this year, researchers said.

Argentina had 137 deaths out of 3,056 cases between May 17 and July 16,

representing a case fatality rate of 4.5%, according to Gustavo Palacios, PhD,

of Columbia University's Mailman School of Public Health, and colleagues.

That death rate was markedly higher than the roughly 0.6% seen in most other

countries and was without an obvious explanation, Palacios and colleagues

reported online in PLoS ONE.

But a molecular examination of nasopharyngeal swab samples from 199 people with

laboratory-confirmed H1N1 infection found that Streptococcus pneumoniae was

strongly correlated with severe disease and death, they said.

The finding suggests that a noninvasive test could have important prognostic

significance, Palacios and colleagues said, and " may provide insights into

clinical management. "

The elevated case fatality rate in Argentina might have been caused by

under-reporting of mild disease, the researchers noted, but the alternative -- a

real increase owing to changes in the virus or the patients -- could have had

serious global implications.

Gene sequencing of 26 virus samples from Argentina -- from patients with both

severe and mild disease -- gave no evidence of genetic changes that might

contribute to virulence, they said.

Nor was there any indication of an increased frequency of risk factors for the

pandemic flu, they said.

But multiplex polymerase chain reaction testing of the nasopharyngeal swab

samples -- looking for evidence of 33 microbial agents simultaneously -- found

that at least one additional pathogen was present in 72% of the 199 samples.

Of the samples, 39 were from people who had severe disease, defined as either

death or hospitalization, while the remaining 160 were from people with mild

disease.

S. pneumoniae was found in 62 samples and was highly correlated with severe

disease: 56.4% of the 39 severe cases versus 25% of the 160 mild cases, a

difference that was significant at P=0.0004, the researchers reported.

The relationship, however, was only significant (at P<0.0001) among people in

what the researchers called the low-risk age group -- those ages 6 through 55.

Among those patients, there were 20 severe cases, of which 13 (or 65%) were

coinfected with S. pneumoniae.

In contrast, among those younger than 6 or older than 55 -- the so-called

high-risk age groups -- half of the severe cases also had S. pneumoniae

infection and half did not, they wrote.

In patients in the low-risk group, the adjusted odds ratio of severe disease in

the presence of S. pneumoniae was 125.5, with a 95% confidence interval from

16.95 to 928.72, they noted.

The most common pathogen observed was Haemophilus influenzae, which was found in

104 patients but was significantly less common in severe cases (at P<0.0001).

" Three practical implications emerge from our study, " said co-author W. Ian

Lipkin, MD, also of Columbia.

The first is that S. pneumoniae is " important in the pathogenesis and prognosis "

of disease associated with the H1N1 pandemic flu, he said.

" Second, " he said, " easily accessible samples such as nasopharyngeal swab

samples may be used as an index to risk of severe disease. "

And finally multiplex diagnostic methods like those used by the researchers " can

enable rapid detection of a broad spectrum of viral and bacterial agents and

inform clinical care, " Lipkin said.

http://www.medpagetoday.com/InfectiousDisease/SwineFlu/17762