Bird flu from the skies? Lessons from 1918
Chandra Wickramasinghe
Photograph of hospital ward with victims of flu in 1918. Over
50 million people died worldwide
As new cases of bird flu continue to turn up at
our doorstep, the appearance of streams of migrating birds in our
autumn skies must fill us with a sense of foreboding. Over flight
paths that extend across thousands of miles, several billion migrating
birds inhale and recycle large volumes of air at a height of about
a kilometre above the ground. If the birds are incubating the dreaded
H5N1 virus, it is possible that vast numbers of viral particles
will be discharged into the atmosphere, some of which would serve
to nucleate raindrops, others which rise in updrafts into the stratosphere
and are carried around the world.
There is a wealth of information to be gleaned concerning the 1917-1919
influenza pandemic if one has the patience to leaf through tomes
of yellowing papers in the dusty archives of libraries. This is
precisely the task that the late Sir Fred Hoyle and I undertook
in 1978, examining sources that included The London Times, Times
of India, The Lancet and US Senate Committee reports to name but
a few. Our researches led to a stark, yet inescapable conclusion:
at the very least some component of the infective agent responsible
for the 1918-1919 outbreaks of a lethat brand of influenza fell
directly through the skies.
Long before the viral cause of influenza was established, Physicians
in the late 19th century had confidently asserted that epidemic
influenza spreads so rapidly across the country that it defies explanation
on the basis of person-to-person spread alone. The distinguished
English Physician Charles Creighton described its spread as a miasma
descending over the land. The viral nature of the causative agent
does not preclude a reservoir being established in the atmosphere,
for instance through the agency of high flying birds.
Whilst we await the prospect of another influenza pandemic with
apprehension, various estimates of the economic cost of such a major
disaster are being aired. One such estimate gives a worldwide death
toll of 7.4 million, a cost to GDP in Britain alone of £95millon,
and close to a billion job losses as companies go bankrupt. Whether
or not one accepts these forecasts, it is clear that any measures
to reduce influenza attack rates by even a few percent would have
huge economic benefits. In the absence of a full knowledge of how
pandemic influenza arises and spreads, lessons from history are
worth looking at more carefully than they have been done so far.
Particularly so as the bird flu virus currently in circulation in
the avian population has been discovered to be very similar in its
gene structure to the virus recovered from victims of the great
pandemic of 1918/1919.
If we had even the slightest clue pointing to pandemic flu being
carried though the air, for instance in fog and mist, it would surely
be foolish to dismiss this possibility out of hand. There are indeed
tantalising hints from the historical record that such a process
might have occurred for the lethal second wave of the pandemic that
ripped across the world in 1918, taking a huge toll of life.
Estimates of the death toll in 1918 vary from a minimum of 30 million
to about twice that number. There are some estimates that suggest
that 20 million deaths occurred in India alone. In parts of Alaska
and in the Pacific Islands over half the total population in some
villages and cities had perished. The rapid spread of influenza
across the frozen wasteland of Alaska in November/December 1918
remains a mystery on the basis of person-to-person transmission.
With a population of 50,000 people very thinly spread over an area
of the size of Europe, and with ground transportation essentially
impossible, the only route of viral transfer must have been through
the air.
There were three waves of pandemic influenza occurring in less
than 12 months. In the first wave, which occurred in the spring
of 1918 the attack rate was 50%, but the mortality rate was not
high. The second wave, which came in the autumn, was also characterised
by high attack rates but very high mortality rates. The peak death
rates from this lethal second wave happened to be in the 20-40 year
age group, precisely the age group that would presumably have spent
more time breathing the external virus-laden air in the course of
their hectic social activities and working habits.
The overall scale of the disaster caused by the pandemic is difficult
to imagine. Populations in many cities and villages were decimated
in a matter of weeks. In the State of Punjab in India, streets were
reported to have been strewn with corpses of victims, and at railway
stations carriages had to be continually cleared of dead or dying
passengers.
On the other hand some places miraculously escaped from the pandemic.
St Helena, an island in the mid-Atlantic, is known to have definitely
escaped, despite all the shipping that had called there. Then there
was a puzzling long delay before the pandemic reached the shores
of Australia. This country seems to have been quite remarkably free
of the disease until early in 1919, despite all the ships that called
there from infected ports, and despite the well-attested attacks
that occurred in mid-ocean. The first influenza death in Australia
occurred at Sydney on 10 February 1919 and was reported in the Times
of London of 20 February 1919.
Again there was an enormous variability in the way ships at sea
were affected. Passenger liners arriving in Australia during the
pandemic recorded attack rates that ranged between four and forty
three percent. And there were similar differences in the attack
rates on crews of ships in the British Navy.
The erratic behaviour of the influenza virus, particularly in the
lethal second wave of 1918, is described graphically in an article
by Dr. Louis Weinstein:
"The lethal second wave, which started at Ford Devens in Ayer,
Massachusetts, on September 12, 1918, involved almost the entire
world over a very short time…..Its epidemiologic behaviour,
was most unusual. Although person-to-person spread occurred in local
areas, the disease appeared on the same day in widely separated
parts of the world on the one hand, but, on the other, took days
to weeks to spread relatively short distances. It was detected in
Boston and Bombay on the same day, but took three weeks before it
reached New York City, despite the fact that there was considerable
travel between the two cities. It was present for the first time
in Joliet in the State of Illinois four weeks after it was first
detected in Chicago, the distance between those areas being only
38 miles….."
The lethal second wave also provided striking evidence of local
patchiness from one American city to another. Death rates from respiratory
disease recorded in the late months of 1918 varied dramatically
between different cities. A striking contrast came from Pittsburg
and Toledo, neighbouring cities with normally almost identical death-rates
and with populations engaged in similar daily occupations. The late
1918 death-rate from respiratory diseases in Pittsburg exceeded
that in Toledo, not by a few percent or a few tens of percent, but
by an enormous 400 percent.
The only reasonable inference to be drawn is that the virus was
airborne with an incidence at ground level that was temporally erratic
and spatially very patchy. It would be unwise to dismiss the historical
evidence that leads to this conclusion as being flawed and inadmissible.
It is true that we have no detailed record all the causative agents
that might have been involved in the 1918 disaster, but the indications
are that a pure avian virus, ominously related to the present H5N1
bird flu, was implicated. Evidence that the human cases of bird
flu recorded over the past 2 years did not show person-to-person
infectivity but had a 50% fatality rate, taken together with similar
historical evidence from 1918, should make us approach an impending
pandemic with a measure of caution as well as humility.
Despite the great strides of progress achieved in the past few
years in the understanding of viruses, the complex patterns of genetic
variability found in isolates of the H5N1 virus is admitted by experts
to defy complete understanding. An external reservoir of the virus
in the high atmosphere, amplified by the exudations of billions
of high-flying migratory birds, cannot be ignored, no matter how
unlikely it might sound. Winter downdrafts could bring down the
amplified virus as nuclei of mist that can directly enter the respiratory
tracts of susceptible humans. A heavy fall-out in any location could
give a semblance of high infectivity which would then be an illusion.
In these circumstances, it would make sense if contingency plans
for the next pandemic include measures to minimise unprotected exposure
to mist and weather, as soon as cases are detected in any locality.
The use of face masks could possibly reduce attack rates, as well
as a general reduction of non-essential travel. It might also be
profitable to explore the feasibility of deploying modern techniques
of molecular biology to indentify viruses in the environment (air
and rainwater samples), with a view to preparing vaccines ahead
of major infective outbreaks. Such measures should of course to
be considered in addition to the other precautions currently in
train.
Accepting the possibility of a vertical atmospheric incidence of
the pandemic virus, to the extent of taking cost-effective measures
such as I have discussed, is fully consistent with the currently
accepted Rio declaration of a "Precautionary Principle"
for dealing with serious threats to the environment and to human
health. The Rio declaration of 1992 states that "in order to
protect the environment, the precautionary approach shall be widely
applied by States according to their capability. Where there are
threats of serious or irreversible damage, lack of full scientific
certainty shall not be used as a reason for postponing cost-effective
measures to prevent environmental degradation".
The price to be paid for neglecting this principle could be too
high.
Relevant further reading
For an original exposition of the Hoyle-Wickramasinghe
theory of Diseases from Space see Archive Paper
published in Viruses from Space (Univ Coll Cardiff Press
1986)
Wickramasinghe, C., The Independent, Letter to the Editor,
11 October 2005Taubenberger, J.K. et al, 2005, Nature, 437,
889
Weinstein, L., 1976, New England J. Med, 294, 1058
Hoyle, F. and Wickramasinghe, N.C., 1990, J. Roy,Soc.Med,
83, 258
© Chandra Wickramasinghe
|
