by The Pandemic Century- One Hundred Years of Panic, Hysteria
It did not take long for the infection to spread from the 42nd infantry to adjacent barracks, and when it did, the flu was nothing like the “mild” spring wave. It was explosive. By September 10 more than five hundred men had been admitted to the base hospital at Devens. Within four days, those numbers had tripled, and on September 15 a further 705 were admitted. The next three days were the worst, however. On September 16 medical orderlies had to find beds for a further 1,189 men and the following day beds for 2,200 more. The pneumonia cases began to mount soon afterward, but they were nothing like the bronchopneumonias associated with measles. They resembled more severe versions of the lobar pneumonias that had developed in some of the flu cases at Camp Funston in the spring. “These men start with what appears to be an ordinary attack of LaGrippe or Influenza, and when brought to the Hosp. they very rapidly develop the most vicious type of Pneumonia that has ever been seen,” recalled a Scottish physician named Roy who was present when pneumonia ripped through the wards. “Two hours after admission they have the Mahogany spots over the cheek bones, and a few hours later you can begin to see the Cyanosis extending from their ears and spreading all over the face, until it is hard to distinguish the coloured men from the white. . . . One could stand it to see, one, two or twenty men die, but to see these poor devils dropping like flies . . . is horrible.”
As the writer John Barry noted in his book The Great Influenza, in 1918 these cyanoses were so extreme that victims’ entire bodies would take on a dark purple hue, sparking “rumours that the disease was not influenza, but the Black Death.” British Army medical officers, many, like Welch and Vaughan, experienced civilian physicians and pathologists who had taken military commissions at the outset of war, were similarly impressed by these cyanotic cases and, struck by the resemblance with the cyanoses seen at Etaples and Aldershot in the winter of 1917, commissioned an artist from the Royal Academy to paint patients in the last throes of illness. The artist labeled the final stage “heliotrope cyanosis” after the deep blue flowers of the same name.
As concerns about measles and pneumonia had grown over the summer, the surgeon general’s office in Washington had kept Welch, Vaughan, and Cole busy. They were sent to make an inspection of Camp Wheeler, near Macon, Georgia, and other camps in the South. On leaving Macon in early September, Welch had suggested they stop at the Mountain Meadows Inn, a fashionable retreat in Asheville, North Carolina. A portly man famous for his love of cigars and gourmet dining, Welch was now in his late sixties and, except for a strip of white around the ears, almost completely bald. To offset the absence of hair on top, he sported a fashionable goatee and moustache, which were also white. To some this gave him the appearance of an elder statesman—an impression underscored by his reputation for being an aloof and distracted teacher. But that was the older Welch. In his youth his imagination had been fired by reports from Germany of the advances being made in the understanding of disease processes using the microscope and new laboratory methods, and in 1876 he had set sail for Leipzig to work with Carl Ludwig, then the foremost experimental pathologist in the world. From Ludwig, Welch learned that “the most important lesson for a microscopist [was] not to be satisfied with loose thinking and half proofs . . . but to observe closely and carefully facts.” The experience made an indelible impression and, on his return to the United States, Welch set about conveying the principles and techniques he had acquired in Europe to a new generation of American medical students, first at Bellevue Medical College, New York, and later at Johns Hopkins, the university that, more than any other American institution, is credited with creating a new paradigm for medical education in the United States. There, to contemporaries such as William Osler and William Steward Halstead, Welch was considered a bon vivant whose favorite pastimes were swimming, carnival rides, and five-dessert dinners in Atlantic City. But for all that they might tease the confirmed bachelor by referring to him as “Popsy,” they also recognized that few could equal Welch’s skills as an anatomist. When Welch cared to, he could also awe his students with his intellect and knowledge of art and culture. As Simon Flexner, who went on to write a biography of his former teacher, recalled, Welch’s technique was initially to ignore his students and leave them to their own devices in the laboratory. But on rare evenings when he invited promising students to dine with him, “a spell fell over the room as the quiet voice talked on, and the young men, some of them already a little round-shouldered from too much peering into the microscope . . . resolved to go to art galleries, to hear music, to read the masterpieces of literature about which Welch discoursed so excitingly.”
Welch and his colleagues used their stay in North Carolina to go over what they had learned during their tour of the South. The consensus was that a better understanding of the immunity of newly drafted men held the key to understanding the measles and pneumonia outbreaks. The Meadows Inn “is a delightful, restful, quiet place,” Welch observed on September 19. It would be the last respite the group would enjoy for some time.
Two days later they were back in Washington, DC, but no sooner had they alighted at Union Station than they were informed that Devens had been struck by Spanish influenza and they were to proceed immediately to Ayer. The scene that confronted them there was shocking and difficult to comprehend. By now the base hospital was overflowing with patients and care was almost nonexistent. More than 6,000 men were crammed into the 800-bed facility, with cots installed in every nook, crevice, and cranny. Nurses and doctors had so exhausted themselves caring for the sick that many also now lay ill or dying, having failed, as one observer put it, to “buck the game.” Everywhere Welch and Vaughan looked there were men coughing up blood. In many instances, crimson fluids poured from nostrils and ears. Even eight years later the images were still etched in Vaughan’s memory. “I see hundreds of young, stalwart men in the uniform of their country coming into the wards of the hospital in groups of ten or more,” he wrote in 1926. “They are placed on the cots until every bed is full and yet others crowd in. The faces soon wear a bluish cast; a distressing cough brings up the bloodstained sputum. In the morning the dead bodies are stacked about the morgue like cord wood . . . such are the grewsome [sic] pictures exhibited by the revolving memory cylinders in the brain of an old epidemiologist.”
The scene that greeted them in the autopsy room, once they had stepped over the cadavers blocking the entrance, was possibly even more gruesome. Before them, on the autopsy table, lay the corpse of a young man. According to Cole, when they tried to move him, bloody fluids poured from his nose. Nevertheless, Welch decided it was imperative to take a closer look at his lungs. What he saw astonished the veteran pathologist. As Cole recalled: “When the chest was opened and the blue swollen lungs were removed and opened, and Dr. Welch saw the wet, foamy surfaces with real consolidation, he turned and said, ‘This must be some new kind of infection or plague’ . . . it shocked me to find that the situation, momentarily at least, was too much even for Dr. Welch.”
By the end of October one-third of the camp’s population, some 15,000 soldiers, had contracted influenza and 787 had died of the pneumonic complications of the disease. Two-thirds of these pneumonias were of the lobar variety. Such pneumonias tended to have a very rapid onset and terminated in either massive pulmonary hemorrhage or pulmonary edema. The devastation was far more extensive than is usually seen in lobar pneumonias, with damage to the epithelial cells that line the respiratory tract but little evidence of bacterial action. The other type was more akin to an acute aggressive bronchopneumonia and was characterized by more localized changes, from which pathogenic bacteria could usually be cultured at autopsy.
The first kind of pneumonia was unlike anything pathologists had observed before in either lobar or bronchopneumonias, fully justifying Welch’s description of it as some new kind of plague. But while Welch’s intuition may have been correct, he was not yet ready to abandon old certainties. Perhaps it was the fault of his formative years in Leipzig, followed by his battles to ge
t the American medical profession to embrace the new German laboratory methods, that made him reluctant to challenge the conclusions reached by Pfeiffer as to the etiological role of his bacillus, even when his gut instincts as a pathologist told him that this was something both new and terrifying. Or perhaps it was the fact that by now American scientists trained in the same bacteriological techniques were finding B. influenzae in influenza patients with similarly gruesome lung pathologies. Foremost among these were William H. Park, the chief of the laboratory division of the New York City Health Department, and his deputy Anna Williams, both highly respected medical researchers. Mindful of the importance of observing “closely and carefully” and “not to be satisfied with . . . half proofs,” Welch approached Burt Wolbach, the chief pathologist of Brigham Hospital, Boston, and asked him to conduct further autopsies to see if all cases of this influenza shared the same peculiar lung pathology he had seen at Devens. Next he called the surgeon general’s office to give a detailed description of the disease and urge that “immediate provision be made in every camp for the rapid expansion of hospital space.” The third person he approached was Oswald Avery at the Rockefeller Institute.
A methodical medical researcher, famous for his austere lifestyle, Avery lived for the laboratory. Working with Cole, he had already perfected techniques for identifying the four main subtypes of pneumococcus responsible for lobar pneumonia using specific serums. Next he had gone on to study how efficiently each type killed mice and in what dosages—experiments that led him to conclude that virulence was a function of the ability of the polysaccharide capsule of the pneumococcus to resist ingestion by white blood cells, the immune system’s first line of defense against invasive bacteria.
One of the challenges of culturing Bacillus influenzae is that it is a fastidious organism that grows only within a very narrow temperature range and which depends heavily on oxygen, meaning it is usually found only on the surface of culture mediums. Because it tends to grow singly or in pairs, and its colonies are translucent and lacking in structure, it is also very easy to miss when looking through the field of an optical microscope. Pfeiffer had realized that a substrate of hemoglobin greatly facilitated growth of the bacillus and promoted his blood agar culture as necessary for establishing it (Pfeiffer recommended pigeon’s blood; other researchers used rabbit’s blood). Once a bacteriologist had obtained colonies of the bacillus, the next step was to stain it with an appropriate dye, wash it with alcohol, then stain it again with a contrasting dye (Gram-positive bacteria retain crystal violet stains, whereas B. influenzae and other Gram-negative bacteria, such as mycobacteria, require red counterstains). Such stains could also be applied directly to slides smeared with sputum from influenza cases. However, a more precise and conclusive method was to prepare pure cultures of the bacillus by inoculating mice with sputum from flu patients and then growing the bacteria from fluids taken from the mice and reintroduced to the blood agar media.
Like other researchers, Avery at first found it difficult to grow Pfeiffer’s bacillus from the sputum and bronchial expectorations of flu victims, so, to increase his chances, he refined his methods, adding acids to his agar culture medium and substituting defibrinated blood for untreated blood (other researchers heated the blood or filtered and dried it to separate the hemoglobin from the fibrin). Gradually, as Avery perfected his techniques, he was able to find the bacillus more and more frequently, until he was able to tell Welch it was present in twenty-two of thirty dead soldiers examined at Devens. Wolbach’s results were even more definitive: he had found the bacillus in every case he examined at Brigham Hospital. That was enough for Welch, Cole, and Vaughan. “It is established that the influenza at Camp Devens is caused by the bacillus of Pfeiffer,” they wired the surgeon general on September 27.
IN FACT, influenza is a viral infection. B. influenza is merely a fellow traveler. Like other bacteria commonly found in the mouths, throats, and lungs of influenza patients, it is not the primary cause of the disease, though it may play a role in secondary infections. In the fall of 1918 no one knew this, though some researchers had begun to suspect it. Instead, failure to cultivate B. influenzae reflected badly on researchers, not the theory of bacterial causation. Indeed, so dominant was the scientific view that influenza was a bacterial infection that, rather than doubt Pfeiffer’s claim, scientists chose to doubt their instruments and methods. If the bacillus could not be cultivated on the first attempt, they needed to improve their culture medium, refine their dyes, and try again.
Anomalies are a common occurrence in science. No two experiments are ever exactly alike, but by refining methods and sharing tools and technologies, scientists are broadly able to reproduce each other’s observations and findings, thereby arriving at a consensus that this or that interpretation of the world is correct. That is how knowledge emerges and a particular paradigm comes to be adopted. However, there is no such thing as absolute certainty in science. Paradigms are constantly being refined by new observations and, if enough anomalies are found, faith in the paradigm may be undermined and a new one may come to supplant it. Indeed, the best scientists welcome anomalies and uncertainty, as this is the way science progresses.
When Pfeiffer first advanced his claim for the etiological role of his bacillus, the science of bacteriology and the germ-theory paradigm (one germ, one disease) was in the ascendancy. With the invention of improved achromatic lenses and better culture-staining techniques, by the late 1880s Robert Koch and Louis Pasteur had brought a series of hitherto hard-to-detect germs into view. These included not only such landmark bacteria as the bacilli of fowl cholera and tuberculosis, but streptococcus and staphylococcus. In short order, their discoveries paved the way for the development of serums and bacterial vaccines against diseases such as cholera, typhoid, and plague, and by the eve of World War I, Avery and Cole were using the same methods to develop vaccines for pneumococcal pneumonias.
When Pfeiffer made his announcement in 1892, it raised hopes that it would not be long before bacteriology had also delivered a vaccine for influenza. But from the beginning, Pfeiffer’s claim was dogged by doubts and anomalous observations. The first problem was that Pfeiffer had failed to find B. influenzae in the majority of clinical cases he had examined in Berlin during the Russian influenza epidemic. Second, as noted previously, he had been unable to reproduce the disease in monkeys inoculated with pure cultures of the bacillus (Pfeiffer does not specify what type of monkey he used, but his failure may have been because many monkeys are a poor refractory species for human influenzas). Soon afterwards, Edward Klein, a Vienna-trained histologist and author of the leading British textbook on bacteriology, succeeded in isolating the bacillus from a series of patients admitted to hospitals in London during the same epidemic of Russian flu. However, Klein also noted finding “crowds” of other bacteria in sputum cultures and observed that as the condition of influenza patients improved, it became progressively more difficult to find Pfeiffer’s bacillus in the colonies on the agar plating medium used to grow bacteria. Finally, Klein noted that B. influenzae had also been isolated from patients suffering diseases other than influenza.
After 1892, the Russian influenza epidemic abated and it was no longer possible to conduct bacteriological exams of influenza patients. Now and then there would be a resurgence of Russian flu, however, and investigators would attempt to culture the bacillus from the sputum and lung secretions of convalescents. Sometimes these efforts succeeded, but just as often they did not. For instance, in 1906 David J. Davis, from the Memorial Institute for Infectious Disease in Chicago, reported being able to isolate the bacillus in only three of seventeen cases of influenza. By contrast he had found the bacillus in all but five of sixty-one cases of whooping cough. The following year, W. D’Este Emery, clinical pathologist at King’s College London, noted that B. infuenzae grew more readily in culture in the presence of other respiratory bacteria and seemed to be more virulent for animals in the presence of killed streptococci, leading him
to speculate that Pfeiffer’s bacillus might, for the most part, be a “harmless saprophyte” and that it required other respiratory pathogens to make it pathogenic.
With the emergence of Spanish flu in 1918, researchers were able to resume their investigations. Again, the results were mixed, and again the anomalies cast doubt on Pfeiffer’s claim. By the summer, concerns had reached such a pitch that a special meeting was convened at the Munich Medical Union. Summarizing the debate, The Lancet wrote that “Pfeiffer’s bacillus has been found but exceptionally,” and that if any bacteria had a claim to be the cause of influenza it should be the far more common streptococci and pneumococci. Britain’s Royal College of Physicians concurred, arguing that there was “insufficient evidence” for Pfeiffer’s claim, though it was happy to allow that the bacillus played an important secondary role in fatal respiratory complications of influenza. In other words, the etiological role of B. influenzae might be open to question, but the bacterial paradigm was not. However, this paradigm was now facing a serious challenge from another quarter.
If Koch was the German father of bacteriology, then Louis Pasteur was its French parent or, as one writer puts it, microbiology’s “lynchpin.” In his first biological paper, published in 1857 at the age of 35, Pasteur, then a relatively unknown French chemist working in Lille, boldly formulated what he called the germ theory of fermentation—namely, that each particular type of fermentation is caused by a specific kind of microbe. In the same paper he suggested that this theory could be generalized into a specific microbial etiology of disease and, later, a general biological principle captured by his phrase, “Life is the germ, and the germ is life.” However, in his own lifetime Pasteur’s fame rested on a famous set of public experiments conducted two decades later, in which he isolated the bacteria of anthrax and chicken cholera and, using basic chemical techniques (heat or exposure to oxygen), weakened the microbes to the point where they lost their virulence. Next, he demonstrated that these weakened strains could confer protection to animals challenged with fully virulent versions of the same bacteria. In so doing, Pasteur opened up a whole new branch of microbiology: the study of immunology. Pasteur realized that weak or attenuated microbes stimulated the host (sheep in the case of anthrax; chickens in the case of cholera) to produce substances (antibodies) that protected them against challenge with more virulent, disease-causing microbes. Eight years later, in 1885, Pasteur conducted an even more astounding microbiological experiment by applying the same principles to the rabies virus. Taking the spinal cord from a rabid dog, he injected the diseased material into a rabbit, and, when the rabbit fell ill, repeated the procedure with another rabbit. By passaging the virus in rabbits every few days, he was able to heighten its virulence for rabbits, but reduce its virulence for dogs. Next, he went a stage further and removed the spinal cord of a dead rabbit and dried it for fourteen days. This new attenuated virus no longer caused disease in dogs at all. Instead, it immunized them against challenge with fully virulent rabies. Next, Pasteur staged a daring public demonstration by administering his vaccine to a nine-year-old boy, Joseph Meister, who had been bitten in fourteen places by a rabid dog. Meister made a rapid recovery, prompting banner headlines. Other than smallpox, this was the first successful immunization with a virus vaccine, and within a few months Pasteur was inundated with requests from victims of rabid animal attacks from Smolensk to New Jersey. However, perhaps the most remarkable aspect of Pasteur’s breakthrough in retrospect is that he developed the vaccine without being able to see the rabies virus or having much idea what a virus was. The reason is that rabies, like other viruses, is too small to be seen through an optical microscope (it measures 150 nanometers, or 0.15 micrometers, and requires magnifications ten thousand times greater than were available in Pasteur’s day). But although Pasteur could not visualize the virus or cultivate it in the laboratory, he could intuit its existence by excluding microbes that he could grow and see, i.e., bacteria. Indeed, in 1892, the same year that Pfeiffer had claimed that a bacillus was the cause of influenza, the Russian botanist Dmitry Ivanovski had shown that tobacco mosaic disease was caused by an unseen agent that passed through porcelain filters with pores too small to admit bacteria. By the turn of the century, these filters, known as Chamberland filters after their inventor Charles Chamberland, were being manufactured and used in research laboratories in Europe and elsewhere, leading to the identification of a variety of “filter passing” agents, including the agents of foot and mouth disease of cattle, bovine pleuropneumonia, rabbit myxomatosis, and African horse sickness. Then, in 1902, a commission headed by US Army Surgeon Walter Reed identified the first filter-passing human disease, yellow fever. At the Pasteur Institute in Paris, these agents were referred to as “virus filtrants”—“filter-passing viruses.”