The Pandemic Century

by The Pandemic Century- One Hundred Years of Panic, Hysteria

  By October 6, more than 2,000 people a day in New York were being quarantined and the panic was palpable. From several districts came reports that patients were holding nurses captive in their homes because they were so frightened. Then nurses and doctors also starting falling sick. By now the flu had reached San Francisco and was also raging in cities in the Midwest and South. The flu erupted in Chicago in mid-September, most likely introduced by sailors from the nearby Great Lakes Naval Station. With a capacity for 45,000 men, the station was the largest naval training facility in the world, and, like Devens, a breeding ground for respiratory disease. As flu and pneumonia gripped Chicago, citizens were advised to avoid crowds and other public gatherings and to cover their mouths when sneezing. The most visible signs of the contagion were the gauze face masks worn by policemen and tram attendants. The trend quickly caught on, prompting a prominent Illinois physician to warn that homemade masks were inadequate because they were “made from gauze with meshes too large to catch and strain out the bacilli from the fine spray issuing from the mouths of victims.” This was of special concern in hospitals and other confined spaces as the spray was thought to be infectious at distances of up to twenty feet. Instead, he persuaded the Chicago Herald Examiner to publish a cut-out-and-keep guide on its front page for the proper procedure for making a gauze mask with a narrow mesh. Unfortunately, these masks made little difference as influenza virus particles are many times smaller than the smallest bacteria, and by mid-October Chicago was already reporting 40,000 cases. The hardest hit city of all, however, was Philadelphia.

  By 1918 Philadelphia had grown considerably since its Quaker beginnings as the capital of the Pennsylvania colony and the place where the founding fathers signed the Declaration of Independence. Ringed by steel mills and with its huge shipyards overlooking the Delaware River, Philadelphia was an industrial powerhouse. The needs of war (naval vessels, aircraft, munitions) brought tens of thousands of additional workers flocking to the city, and as its population swelled to nearly two million, so living conditions in Philadelphia became increasingly intolerable. In cramped rooming houses and overcrowded tenements, the virus found ample fodder and steadily increased in virulence, killing people rapidly and indiscriminately. At a time when authorities in other cities were advising people to avoid large public gatherings, the epidemic was almost certainly exacerbated by the decision of Philadelphia’s mayor to proceed with a Liberty Loan Drive on September 28. The drive brought thousands of people crowding into the downtown area, and within two weeks Philadelphia had recorded more than 2,600 flu deaths. By the third week of October deaths had soared to over 4,500. As bodies piled up in morgues for lack of undertakers, the stench became overpowering and the city resorted to digging mass graves—something that had not been seen since the yellow fever epidemics of the late eighteenth century. The sight of rotting bodies became so commonplace that adults made little effort to shield children from the horrors. The fear of influenza was now palpable, and with fear came panic. But this panic was not the press’ fault. “Panic is the worst thing that can happen to an individual or a community,” warned the Philadelphia Inquirer in an editorial at the height of the fall wave. “Panic is exaggerated fear and fear is the most deadly word in any language.” The remedy, it suggested, was to expel fearful thoughts by an act of will. “Do not dwell on the influenza. Do not even discuss it. . . . Terror is a big ally of the influenza.” But once seen, the sight of a cyanotic, influenza-ridden body was not easily forgotten, either in Philadelphia or other places the flu visited. The sight of “big strong men, heliotrope blue and breathing 50 to the minute” was unforgettable, observed Dr. Herbert French, a pathologist based at Guy’s Hospital in London and a physician to Her Majesty’s Household. But the worst case by far was the type that became “totally unconscious hours or even days before the end, restless in his coma, with head thrown back, mouth half open, a ghastly sallow pallor of the cyanosed face, purple lips and ears.” It was “a dreadful sight,” he concluded.

  THE 1918 INFLUENZA pandemic was a shot heard around the world. The scenes described by French were not confined only to London and other large European and American cities but were the same everywhere. In Cape Town, observed one eyewitness, the fall wave “made orphans of between two to three thousand children.” One such orphan who was co-opted into burial duties reported: “I carry the coffin, holding my nose . . . no longer were church bells tolling for the dead . . . there was no sexton to ring the bells.” It was the same in Bombay (Mumbai), where the disease arrived courtesy of a container ship in May. Deaths peaked in the first week of October, the same time as Boston. By the end of the year, the flu had killed an estimated one million people in this populous Indian city. All told, the pandemic claimed the lives of 18.5 million people across the Indian subcontinent, according to the latest estimates, and perhaps as many as 100 million worldwide. With the exception of Australia, where strict maritime quarantines delayed the onset of flu until the winter of 1919, virtually the entire globe suffered the pandemic at the same time. Only American Samoa, St. Helena, and a handful of islands in the South Atlantic escaped the plague. It was truly a shared global disaster.

  It is difficult to imagine deaths of this order of magnitude, much less process them. The scale is too vast. “When one has fought a war, one hardly knows any more what a dead person is,” remarks Camus. “And if a dead man has no significance unless one has seen him dead, a hundred million bodies spread through history are just a mist drifting through the imagination.” However, if there is little point in trying to imagine death on this scale, there is much to be gained from examining variations in mortality rates observed in different geographical locations and ecological and immunological settings. When influenza reached New Zealand, for instance, the local Maori population died at seven times the rate of British settlers. Similarly wide variations in mortality rates were observed between indigenous and European-descended peoples in Fiji and other South Pacific islands (one of the most striking discrepancies was observed in Guam, where the pandemic killed 5 percent of the local population but just one sailor at the US naval base on the island). While the case fatality rate for “white” South Africans was 2.6 percent, for “blacks, Indians, and Coloureds” it was nearly 6 percent. For those who toiled underground in the Kimberley diamond mines the mortality rate was even worse—22 percent. Similar variations were observed at Devens and other large army training camps, with recent arrivals suffering far worse clinical outcomes than men of a similar age who had been at the camps for four months or longer. On the AEF transports, sailors permanently assigned to the ships fared far better than soldiers who had just embarked, even though both were attacked by influenza in more or less equal numbers.

  But perhaps the most striking aspect of the Spanish flu pandemic was the mortality pattern observed in young adults. In a normal flu season, curves of mortality by age at death are typically U-shaped, reflecting high mortality in the very young (children under three) and the elderly (seventy-five and over), with low mortality at all ages in between. This is because infants and the aged tend to be those with the weakest immune systems. By contrast, the 1918–1919 pandemic and the succeeding winter recurrences in 1919 and 1920 produced a W-shaped curve, with a third mortality peak in adults aged 20–40 years. Moreover, adults in these age ranges were responsible for half the total influenza deaths, including the majority of excess respiratory deaths. This abnormal mortality pattern was observed both in cities and rural areas, in major European metropolises and distant outposts of empire. In other words, it was the same everywhere.

  Why this should have been the case has never been satisfactorily explained. Nor, despite the advances in influenza virology and immunology and a better understanding of the pathophysiology of flu, are scientists today in a much better position to say whether the Spanish flu pandemic was a one-off occurrence—a never-to-be-repeated epidemiological disaster—or whether it could happen again. By reviewing what has been learned about the 1918
virus, and the likely identity of previous pandemic viruses, it is possible to rule out some hypotheses and rule others in. However, perhaps the biggest clue to the epidemiological patterns and unusual lung pathologies observed in 1918 comes from the ecology of large army camps and the contemporary accounts of medics who observed the ravages wrought by influenza in them at first hand.

  INFLUENZA, WE NOW KNOW, is a member of the family Orthomyxoviridae, and comes in three types—A, B, and C—named in the order of their discovery. Type C rarely causes disease in humans. Type B can cause epidemics, but the course of infection is milder and the spread of the virus tends to be slower. By contrast, type A is associated with explosive spread and high rates of morbidity and mortality, making it the leading cause of epidemics and pandemics. Like all influenza viruses, type A influenzas are RNA viruses and must infect a living cell in order to replicate. Generally, they do this by attacking the epithelial cells that line the respiratory tract from the nose and through the windpipe to the lungs.

  Although in 1933 scientists had demonstrated that influenza was a virus that could be transferred from ferrets to man (the breakthrough was made by a team headed by Sir Patrick Laidlaw at the Farm Laboratory, in Mill Hill, north London, part of the UK’s National Institute for Medical Research), it was not until the 1940s and the invention of the electron microscope that researchers were able to see the influenza virion for the first time. It measured approximately 100 nanometers (0.10 micrometers), making it slightly smaller than the rabies virus but larger than rhinovirus, the cause of the common cold. Magnified, it resembled nothing so much as the surface of a dandelion bristling with tiny spikes and mushroom-like spines. The spikes are made of a protein called hemagglutinin (HA) that derives its name from its ability to agglutinate to red blood cells. When a person inhales an air droplet containing the virus, it is these spikes that stick to the receptors on the surface of the epithelial cells in the respiratory tract, much as a prickly seed case catches on the fibers of clothing in tall grass. The square-headed mushroom-like protrusions, fewer in number, consist of a powerful enzyme, neuraminidase (NA). It is the combination of these proteins and enzymes that enables the virus to invade epithelial cells and evade the body’s immune defenses. These permutations of proteins and enzymes give each virus a signature shape, making for easy classification. In all, scientists have identified sixteen types of hemagglutinin and nine types of neuraminidase in mammals and birds (besides ferrets, type A flu viruses commonly infect pigs, whales, seals, horses, and wild waterfowl), but so far only three types of each have been found to readily infect humans. These are labeled H1, H2, and H3, and N1 and N2, respectively. The Spanish flu is an H1N1.

  Unlike DNA, RNA does not possess an accurate proofreading mechanism. During replication, when the virus invades and colonizes animal cells, the RNA makes small copying errors, resulting in genetic mutations to the H and N molecules on its surface. In the Darwinian world of the virus, some of these copies can confer a competitive advantage, allowing the viruses to escape the antibodies designed to neutralize them and enabling them to spread more efficiently via coughs and sneezes to the wider environment ready to infect the next person. This process of gradual mutation is known as “antigenic drift.” Type A viruses can also spontaneously “swap” or exchange genetic material. This process is typically thought to occur in intermediary hosts such as pigs, which can be infected with swine and human type A strains simultaneously, and is known as “antigenic shift.” In this case, the result is the emergence of a completely new subtype that codes for proteins that may be new to the immune system and for which human populations may possess few or no antibodies. It is these strains that historically have been the cause of pandemics. However, it is thought that the virus responsible for the 1918 pandemic may have emerged in yet another way.

  In the 1990s scientists at the Armed Forces Institute of Pathology in Bethesda, Maryland, led by molecular pathologist Jeffery Taubenberger, succeeded in retrieving fragments of the Spanish flu virus from lung autopsy specimens stored in the institute’s archives. Further genetic viral material came from a woman who had died of influenza in 1918 in Alaska and had been buried in permafrost, which preserved her lungs from decay. Using this material, Taubenberger’s group was eventually able to sequence the virus’s entire genome. Published in 2005, the results came as something of a surprise because none of the eight genes came from a strain that had previously infected humans, as one would have expected if the Spanish flu had been the result of antigenic shift. Furthermore, large portions of the genetic code matched sequences only found in wild birds. This suggested that the virus may have begun as a bird-adapted strain that, with just a handful of mutations, made the leap to humans. Alternatively, the pandemic strain may have begun life as an H1 which reassorted with an avian virus shortly before 1918. By zoos, it was recognized that mallards and teals were an important reservoir of avian influenza viruses in the wild, and the idea that birds might be the source of novel genes in pandemic viruses was gaining currency. Taubenberger’s sequencing studies also coincided with growing concern about an avian virus that was then infecting chicken flocks across Southeast Asia. The virus, known as H5N1, had first emerged in Hong Kong in 1997, where it infected eighteen people and caused six deaths, before reemerging for a second time in 2002. Since then the virus had spread from Asia to Europe and Africa, sparking hundreds of human cases and forcing authorities to cull millions of chickens. Alarmingly, the H5N1 virus was able to replicate in the human respiratory tract, and the mortality rate averaged 60 percent. However, it did not transmit easily from person to person. Nevertheless, its emergence demonstrated that people could be directly infected with a wholly avian influenza virus, meaning it was no longer necessary to invoke pigs as intermediary hosts in the generation of pandemic strains. Theoretically, such reassortments, or mixing, of avian and mammalian flu strains could also occur in humans. The question was, could something like this have happened in 1918? The short answer is that no one knows, but the possibility cannot be ruled out.

  The precise genetic identities of pandemic strains prior to 1900 are lost to history, but in the twentieth century there have been three major shifts. The first was the H1N1 Spanish flu virus that emerged in 1918, or possibly a little earlier (by comparing older and more recent strains of the virus and running molecular clocks backward in time, evolutionary biologists suggest the virus may have acquired its avian genes somewhere between 1913 and 1917). This was the prevailing strain until 1957, when it was replaced by a new viral strain, typed H2N2. Known as the “Asian flu,” the H2N2 seems to have been generated by a reassortment of descendants of the 1918 virus with an avian influenza strain derived from Eurasian wild waterfowl. It spread rapidly around the globe, displacing descendants of H1N1 Spanish flu and killing an estimated two million people. In 1968 there was a third shift, when an H3N2 suddenly emerged in Hong Kong, also apparently as a result of the acquisition of novel proteins from Eurasian wild waterfowl. Known, unsurprisingly, as the “Hong Kong” flu, this virus is estimated to have killed one million people globally, and at the time of writing remains the leading cause of morbidity and mortality from influenza.

  To complete the picture of pandemic viruses in the modern period, we also need to include the Russian flu. Like the 1918 Spanish flu, this was a true worldwide pandemic. Originating in the Eurasian “steppes”—a vast expanse of grassland that encompassed parts of Russia plus Tsarist-controlled Uzbekistan and Kazakhstan—it spread rapidly along international rail and shipping routes and is conservatively thought to have killed one million in the period between 1889 and 1892. Unfortunately, scientists have been unable to recover fragments of the virus, so its precise genetic identity is unknown. However, serology tests on elderly people who were examined for antibodies at the time of the 1968 Hong Kong flu suggest that, like that virus, it was caused by an H3. This may be an important clue, as those most at risk of dying in 1918 were born in or around 1890, meaning they belonged to a birth cohort
whose first exposure to a flu virus would almost certainly have been to the Russian flu. We will return to this in a moment, but first it is necessary to consider the nature of the pneumonias that killed people in 1918.

  As noted earlier, broadly speaking these pneumonias can be divided into two types—lobar and bronchial. However, it is also important to note that in a previrological era, these distinctions rested on clinical observations and histological examinations of lung tissue and that the two types were often closely related, with the clinical-pathological syndromes sometimes overlapping. The most common type by far appears to have been an acute aggressive bronchopneumonia. In this type, pathological changes were most obvious in the bronchi, and pathogenic bacteria could usually be cultured at autopsy from different parts of the lung. Close to 90 percent of the pneumonias fell into this category. In the second type, the outstanding features were pulmonary hemorrhage and edema with extensive damage to one or more of the lobes, and pathogenic bacteria were less frequently or rarely recovered. In this type, the infection appears to have triggered an acute inflammation of the pulmonary alveoli resulting in cell death (necrosis) and the deposit of damaged cells and fluids in the alveolar air spaces—the microscopic sacs that absorb oxygen in the lung. These features were found whenever victims died within a few days of the onset of illness, as well as in 70 percent of cases in which pneumonia developed after influenza. And they were nearly always found in deaths involving healthy young soldiers or civilians. However, it must be reiterated that this type accounted for only a small percentage of deaths overall. The later-onset bronchopneumonias and mixed infections, in which bacteria could be readily cultured after death, were the ones encountered most frequently. Indeed, it is these bacterial fellow travelers of flu—or what pathologists at the time called “secondary invaders”—that many experts believe best explain the majority of deaths seen at camps like Devens and the variations in mortality observed between recruits of the same age from rural and urban areas.