Tuesday, April 11, 2017

Spanish flu anxiety: exaggerated?

There has been recent speculation, according to a journalist, by a leading Australian expert on influenza, that a repeat of the Spanish flu is "very possible", even though this is not mentioned in the paper which this newspaper article is based on.
For well over a decade I have considered it is alarmist to raise fears about a recurrence of Spanish flu. I published about this in a peer reviewed paper (one reviewer being a molecular biologist who is a Fellow of the Royal Society) in Infectious Diseases of Poverty in 2012, and later - more briefly - in a report for the Special Programme for Tropical Diseases Research, which I edited. I have been particularly influenced by Prof Paul Ewald, author of “Evolution of Infectious Disease”. When I read "Pandemic skeptics warn against crying wolf", published in Science in 2005 (which included comments by Ewald), it resonated with ideas I was then independently forming. A key point was that if a virus developed enhanced human to human transmission, it could be at the cost of its human lethality. Of course, that is not automatic. A more recent paper, published in the Proceedings of the National Academy of Sciences (USA), also argued that the case fatality rate among humans infected with highly pathogenic avian influenza may be orders of magnitude lower than suspected, if based on surveys of people exposed to lower viral titre (see endnote 1 for more detail).

This paper did, however, comment that avian flu, if "serially passaged" through ferrets could develop airborne transmissability while retaining virulence (ie harm). More on ferrets, later.

The mortality of the Spanish flu was unprecedented, in the 20th century. Conservatively, it killed 25-50 million people in a global population of about 1500 million (1.5 billion). It killed far more than the direct toll of World War I. This mortality rate is about 50 times that of each of the other major 20th century influenza epidemics (1957 and 1968). 

In comparison, the global mortality rate of the 2009 swine flu pandemic was about 1/1000th of the Spanish flu, even if we accept a human death toll in 2009 of 200,000 (ten times that of earlier estimates).  

I am not saying there is no need to fear influenza, but I am saying that a pandemic on a scale such as at the end of WW1 is currently unlikely.

In the following I attempt to put the core arguments in simple terms.

Key reason 1: Viral dynamics

Most fundamentally, evolutionary dynamics that operated at the close of the Great War to produce this human pandemic have not (yet) been repeated. I believe influenza in modern intensive farms (concentrated animal feeding operations or CAFOs) (see endnote 2) does develop strains that are highly contagious to farmed animals (eg to birds/pigs) and sometimes quickly lethal (eg to birds/pigs) – but they are unlikely to evolve both characteristics against humans - unless humans are perhaps living in extremely crowded conditions, and experiencing serial passage of the virus, which, with each cycle, hones its capacity to harm humans. This is subtly different to the theorising of most influenza experts, which is more that a "natural killer" will emerge from a chicken farm, a pig farm, or a live market and be instantly ready to wreak havoc. The difference is that the serial passaging in humans, even if it occurs, is unlikely to  breed a killer such as evolved at the end of the Great War. It thus also follows that the repeated mass slaughter of domestic animals in an attempt to stop a pandemic are cruel and squander resources.

Evolutionary theory informs us that pathogens (eg viruses, bacteria) are unlikely to kill the host rapidly if there is a penalty; they would "prefer" (or, evolution will "favour" them) to be longer living, and slower killing, with more chances for reproduction, by spreading to another animal.In fact, from an evolutionary perspective, nether sickness nor death are necessary, though sickness may be useful as symptoms such as coughing and sneezing enhance the rate of infection.

The following paragraph rephrases this key message, as it is so important.

Pathogens (like all species) evolve to reproduce themselves – killing the host (whether a bird, a pig or a person) is not their “motivation” but of course sometimes happens. However, the rapid death of a host (ie a short time between infection and death) is uncommon, and generally penalised - because a pathogen that rapidly kills its host generally reduces its chance of being successfully transmitted to another host. But rapid mortality, and especially, rapid transmission, can evolve if there is little or no penalty incurred by the pathogen as a result, such as in a densely populated chicken or pig farm. In fact, pathogens that are rapidly contagious may have an evolutionary advantage in extreme crowding even if they kill their host. This is because hosts will die before they live long enough to contract the slower acting form.

Suppose there are three bird populations, X, Y and Z. Population X is small, with little contact with other birds; population Y is of intermediate density, but a bird in population Z has contact with 100 birds every hour.

Suppose (to make the visualisation exercise simpler) that all the birds in contact with the sick bird (the “index case”) lack immunity.

Now imagine three influenza viruses, A, B and C. They each kill 50%, and they each infect 50% of birds they have contact with. (So, a bird in contact with an infectious bird has a 25% chance of death from infection). But these viruses vary in their time to become infective and the speed with which they kill, with variant C being the most rapid acting.

In population Z (the most crowded one) variant C will have an advantage, but in population X it will be at a disadvantage. See also: Lebarbenchon et al: Persistence of highly pathogenic avian influenza viruses in natural ecosystems (published in Emerging Infectious Diseases in 2010). It follows that large bird (or pig) farms may evolve highly pathogenic forms of influenza. But the high pathogenicity is to the bird, or pig, not the human. It is to the species with which the pathogen has co-evolved.

This does not mean the pathogen that has evolved for one species has zero cross reactivity with others. Indeed, it is thought that the H1N1 virus which caused Spanish flu was also toxic to mice, though I am not sure how toxic. (This is reported in Oxford et al, 2005). If my theory is correct then the mortality rate in mice should not be extremely high). Ferrets are also likely to be more vulnerable; their immune system makes them a good model for flu.

It thus follows that intensive farming of ferrets should be avoided.
However, in no such case, in theory, it seems to me, will a virus evolve that is very likely to kill a very high number of humans, especially rapidly; without additional "ingredients" involving serial passage through large numbers of humans, preferably in very crowded conditions. However, at the end of WW1, conditions to breed such a virus, harmful to humans, were unusually favourable, as John Oxford and colleagues have argued in work I think is hugely under-appreciated. 

As they state: in one article called: "A hypothesis: the conjunction of soldiers, gas, pigs, ducks, geese and horses in Northern France during the Great War provided the conditions for the emergence of the “Spanish” influenza pandemic of 1918–1919".

"The Etaples camp had the necessary mixture of factors for emergence of pandemic influenza including overcrowding (with 100,000 soldiers daily changing), live pigs, and nearby live geese, duck and chicken markets, horses and an additional factor 24 gases (some of them mutagenic) used in large 100 ton quantities to contaminate soldiers and the landscape. The final trigger for the ensuing pandemic was the return of millions of soldiers to their homelands around the entire world in the autumn of 1918."

And, "an immense traffic of young soldiers (more than one million of them by September 1917) moving up towards the front; in the hospitals, sick and wounded men and women undergoing treatment up to 23,000 of them at any given time. They depict, in addition, a degree of dangerous overcrowding. A score or so of fit young men in each and every tent, plus the sick and wounded tightly packed together, not only in the tented wards, but laid our in palliasses in the corridors, mess and recreation rooms."  See images. Also, see Vera Brittain's book "Testament of Youth" and the film.

Oxford et al summarise: "Overcrowding would have allowed rapid “passage” of influenza in literally millions of young soldiers and provided the opportunity for an avian influenza virus zoonosis and adaptation to humans."

In other words, if Oxford and his colleagues are correct, the appalling conditions on the Western Front in the Great War shaped the evolution of a virus well adapted to killing people, similar to how intensely crowded livestock farms can promote pathogens well adapted to causing disease in birds and pigs.
If this hypothesis is right then we should not worry too much about variants of influenza in China or elsewhere emerging from intensive farms or live markets, though such conditions might still produce a pandemic - but crucially, not on a "Spanish flu" scale (see endnote 3)

Also, note, some think Spanish flu evolved in the US, and then may have been serially passaged through humans in Haskell County, Kansas, an isolated and sparsely populated county in the southwest corner of the state, in January 1918, before intense serial passage in humans in Camp Funston, Kansas, which held an average of over 56,000 troops. The mystery, of course, is why it would start in Haskell County, which at that time lacked intensive animal farming and had a small human population. Although there is no record, I wonder if it could have in fact been brought there from Camp Funston, before returning there (and then on to Europe from Camp Funston.) Very severe outbreaks of respiratory illnesses also occurred in Europe before 1918, earlier in the war.

The conventional view

The conventional view seems to be it is only a matter of time before chance produces a combination of genes which mimics the human killing power of the H1N1 variant that spread around the world in 1918. I think this is rather lazy thinking. However there are large resources devoted to pandemic preparedness, and a "status quo" against which this counterview is almost invisible. Perhaps this is also in part because influenza threatens the global middle class, unlike tuberculosis or many other threats to human health. I don't think Oxford's hypothesis has been very influential, and my work even less so.

There is also a misperception that if pathogens cross species they are automatically extremely harmful. They can be extremely harmful to the new species, as occurred with Ebola, but this does not mean they all will be (bird flu being a good example if Palese and Wang are correct - see endnote 1). Nor does such species crossing mean they quick or easy transmission in that new species.
(Ebola is highly contagious via direct contact, including sexual contact. Speculation of airborne transmission of Ebola has not been confirmed.)

Key reason 2: Better treatment of bacterial infections 

The high mortality of Spanish flu was, to a considerable degree, due to bacterial pneumonia which complicated the viral illness. In 1918 and 1919 there were no effective anti-bacterial antibiotics. This is vastly different to today. For this reason alone, discussion of a repeat of the post WW1 pandemic is significantly overstated.

A caveat

In a "breakdown" world, perhaps in association with a large famine due to climate change and conflict, conditions could evolve for a variant of influenza with high human to human transmission and with high mortality. This is a point I suggested in my 2012 paper. Perhaps a terrible refugee camp with widespread undernutrition could provide these conditions. But such a camp might need birds and perhaps pigs in large numbers, and that is inconsistent with severe famine.

A final point

The flu epidemics of 1957 and 1968 were not trivial compared to most modern epidemics. They happened without the WWI "milieu". They could recur. But these epidemics, as I stressed at the beginning, were trivial compared to the so-called "Spanish" flu. And every epidemic in the last 20 years has been minor (in terms of human deaths) even compared to the 1957/1968 epidemics.

Endnote 1: Palese and Wang, PNAS 2012    (emphasis added)

"The frequency and certainty with which this staggering fatality rate is reported is troubling when one considers how the numbers are generated; in order for a case to be confirmed by WHO, a person must have an acute, febrile respiratory illness (temperature >38 °C/100.3 °F) with known H5 exposure in the 7 days preceding symptom onset and have molecular confirmation of H5 infection by a WHO-approved laboratory (e.g., virus isolation, PCR, serology) (29). This definition does not allow for asymptomatic infections and essentially requires that a person actively seek medical help at a hospital that is equipped to draw samples and ship them to an approved laboratory. Given that rural populations in developing countries are most commonly affected by H5 viruses, it seems unlikely that even a small fraction of the total number of infected cases has been accounted for under the WHO surveillance system. Also, the fatal cases that have been reported are most likely caused by mega-doses of H5 virus inhaled by the patients, who are living in very close contact with infected poultry. Transmission involving small doses of virus (as observed under regular aerosol transmission conditions) may not result in overt disease. Of the 10 largest studies of which we are
aware (N ≥ 500) (30–39), which document seroevidence in humans for H5 infections, two studies report no seropositivity and the other eight report rates ranging from 0.2% to 5.6%; the studies are mostly conducted in rural areas where H5 infections infections have been previously documented (30–39). Even if only a low percentage of the rural population is asymptomatically/ subclinically infected, the case fatality rate that is offered by the WHO—and that is driving this controversy—is likely orders of magnitude too high. This suggests that more work should be done to determine an accurate case fatality rate for H5 infections, which takes into account mild and asymptomatic cases."

Endnote 2: There is an excellent video about intensive farming and ethics by Marianne Thieme, a human-rights lawyer who is leader of the "Party for the Animals" in the Netherlands.

Endnote 3: While Oxford may or may not share my views about a Spanish flu recurrence being exagerated, he is concerned about the possibility of a modern pandemic - probably more than I am, eg see