This article is more than 5 years old. Information may no longer be current.
The 1918-1919 pandemic influenza revisited: a new perspective based on old information
Within the last six weeks, two articles have been published that have forced infectious disease investigators and the public health community to reconsider their thinking about the cause of the remarkable mortality recorded during the 1918-1919 influenza pandemic, and in addition to reconsider recommendations made for pandemic preparedness.
In one, a group of investigators at the National Institute of Allergy and Infectious Diseases undertook a retrospective autopsy study of people who died during that pandemic. They examined lung tissue specimens from 48 autopsies, and reviewed bacteriologic and pathologic data from 109 published autopsy series from the pandemic. They found that the substantial majority of deaths in that pandemic resulted directly from secondary bacterial pneumonia, including pneumococci, streptococci, Bacillus influenzae, and in some series, staphylococci. Bacillus influenzae is, of course, the same organism we know today as Haemophilus influenzae. The other striking finding was a desquamative tracheobronchitis and bronchiolitis, extending throughout much or all of the tracheobronchial tree, with necrosis and sloughing of the bronchial epithelium. Some clinical observations recorded at that time included the coughing up of “casts” of tracheal or bronchial epithelium.
Fundamentally, this is the same pathologic lesion of influenza that we still see today, although certainly much less severe. Similarly, it is much the same group of bacterial pathogens we see today causing secondary bacterial pneumonia complicating seasonal influenza. The major difference is the intense severity of the disease in 1918-1919, and that is still largely unexplained.
Not found
Not found was much histological evidence of an out-of-control cytokine storm, with a huge outpouring of inflammatory fluid into alveoli, thus essentially drowning the patient in his or her own secretions. This was the lesion seen in mice experimentally infected with the reconstructed 1918 influenza virus and also seen in some of the human cases of H5N1 influenza in Southeast Asia. It should be noted however, that the NIAID data are based on hospital autopsy series; by definition, individuals who died before hospital admission were not represented here. It is possible, therefore, that a number of such cases with early death could have been missed.
That this was likely not a large number of cases is suggested by the data in the second paper, from the U.S. Armed Forces Health Surveillance Center and the Australian Army Malaria Institute. These investigators studied the distribution of influenza illness from the day of onset to day of death in as many different populations as they could find that had the requisite data, and then constructed time to mortality distributions. Peak mortality usually occurred between days seven and 10. Fewer than 25% of deaths occurred before day seven, and in no case was there anything resembling an early peak of mortality which might suggest early death due primarily to influenza virus pneumonia. The majority of deaths appeared to occur between days seven and 21 – exactly in the range that might have been expected for complicating bacterial pneumonias.
None of the information in these two reports is new; it has been there for almost a century! They do provide a perspective of which we seem to have lost sight over the decades. Rather like rediscovering the wheel!
Important implications
Authors of both of these reports point out that their findings have important implications for pandemic preparedness today. U.S. preparedness policy, and indeed that of almost all other countries, has been focused on preventing or modifying influenza virus infection itself. Thus, vaccine development and anti-viral drugs (eg, neuraminidase inhibitors) have been the major efforts, and a great deal of stockpiling has already taken place. Clearly it is equally necessary to stockpile antibiotics effective against primarily community-acquired organisms causing post-influenza pneumonia today, including both MSSA and MRSA. Much more consideration needs to be given to the possible role of pneumococcal and possibly other bacterial vaccines as part of pandemic preparedness.
Several other items related to this year’s expected seasonal flu outbreak are worth mention. As readers will recall, all three component strains included in seasonal flu vaccine were changed this year, leading to concerns that there might be production delays and consequent delayed availability of vaccine. Fortunately, that proved not to be the case, and several manufacturers are already shipping vaccine to distribution centers. No manufacturer is reporting any significant delays, and the vaccine supply is expected to be excellent. The major unknown at this time is the extent of use within the healthy pediatric population.
A somewhat unique approach to influenza vaccination is being undertaken this election year. The “Vote and Vax” program is sponsored largely by the Robert Wood Johnson Foundation, through SPARC (Sickness Prevention Achieved through Regional Collaboration). The latter is a nonprofit organization that offers technical expertise and guidance to local health jurisdictions in setting up and operating election day vaccine clinics. Although the program has existed since the late 1990s, Johnson Foundation support resulted in significant expansion in 2006. That year 127 such clinics were operated; this year the goal is to establish 1,000 election clinics across the United States.
This interesting initiative deserves support and encouragement; it is yet another way of reaching the elderly. The major downside is that it’s potentially “doable” only every other year and could likely function best only in presidential elections. Increasing use of absentee balloting and increasing use of mail ballots – at least in some states – also tend to diminish utilization of election clinics. Nonetheless, it’s an idea worth pursuing further.
Finally, a brief note in follow-up of my comments last month on “the smoking jalapeno”: The FDA has now lifted its ban on the use of jalapeno and Serrano peppers grown in Mexico; the suspect farm(s) are no longer producing or supplying the market. FDA and CDC jointly have declared the epidemic “over.” We’ll likely never know precisely how the contamination occurred.
For more information:
- Morens DM, Taubenberger JK, Fauci AS. Predominant Role of Bacterial Pneumonia as a Cause of Death in Pandemic Influenza: Implications for Pandemic Influenza Preparedness. J Infect Dis. 2008:198;DOI 10.1086/591708.
- Brundage JF, Shanks,GD. Deaths from Bacterial Pneumonia during 1918-19 Influenza Pandemic. Emerg Infect Dis. 2008;14:1193-99.