Episode 57: The Second Wave

In August of 1918, a horrific second wave of the Spanish Flu crashed across the world. In this episode, the third of a four-part series exploring hydroxychloroquine and COVID-19, I’ll explore this single moment in time, through the mysterious origins of the Spanish Flu and historiographical controversies, scientific missions to mass burial sites in remote Alaskan villages, the ill-fated journey of the HMS Mantua, debates about how to count victims of a pandemic, and the mystery behind Pfeiffer’s bacillus. Plus a new #AdamAnswers about that annoying yellow on blue powerpoint template so common in the medical field!


In August of 1918, a horrific second wave of the Spanish Flu crashed across the world. In this episode, the third of a four-part series exploring hydroxychloroquine and COVID-19, I’ll explore this single moment in time, through the mysterious origins of the Spanish Flu and historiographical controversies, scientific missions to mass burial sites in remote Alaskan villages, the ill-fated journey of the HMS Mantua, debates about how to count victims of a pandemic, and the mystery behind Pfeiffer’s bacillus. Plus a new #AdamAnswers about that annoying yellow on blue powerpoint template so common in the medical field!



  1. Viboud, C. et al. Age- and Sex-Specific Mortality Associated With the 1918–1919 Influenza Pandemic in Kentucky. J Infect Dis 207, 721–729 (2013).
  2. Oxford, J. S. & Gill, D. A possible European origin of the Spanish influenza and the first attempts to reduce mortality to combat superinfecting bacteria: an opinion from a virologist and a military historian. Hum Vacc Immunother 15, 2009–2012 (2019).
  3. Epps, H. L. V. Influenza: exposing the true killer. J Exp Medicine 203, 803–803 (2006).
  4. Patterson, S. W. & Williams, F. E. PFEIFFER’S BACILLUS AND INFLUENZA. Lancet 200, 806–807 (1922).
  5. Taubenberger, J. K. & Morens, D. M. The 1918 Influenza Pandemic and Its Legacy. Csh Perspect Med a038695 (2019) doi:10.1101/cshperspect.a038695.
  6. Trilla, A., Trilla, G. & Daer, C. The 1918 “Spanish Flu” in Spain. Clin Infect Dis 47, 668–673 (2008).
  7. Taubenberger, J. K. The origin and virulence of the 1918 “Spanish” influenza virus. P Am Philos Soc 150, 86–112 (2006).
  8. Heinz, E. The return of Pfeiffer’s bacillus: Rising incidence of ampicillin resistance in Haemophilus influenzae. Microb Genom 4, (2018).
  9. Barry, J. M. The site of origin of the 1918 influenza pandemic and its public health implications. J Transl Med 2, 3 (2004).
  10. Johnson, N. P. A. S. & Mueller, J. Updating the Accounts: Global Mortality of the 1918-1920 “Spanish” Influenza Pandemic. B Hist Med 76, 105–115 (2002).
  11. Tomkins SM, Colonial Administration in British Africa during the Influenza Epidemic of 1918-19. Canadian Journal of African Studies / Revue Canadienne des Études Africaines. Vol. 28, No. 1 (1994), pp. 60-83 (24 pages)
  12. Qiang Liu et al, The cytokine storm of severe influenza and development of immunomodulatory therapy. Cell Mol Immunol. 2016 Jan; 13(1): 3–10.
  13. Spreeuwenberg et al. Reassessing the Global Mortality Burden of the 1918 Influenza Pandemic.Am J Epidemiol . 2018 Dec 1;187(12):2561-2567. doi: 10.1093/aje/kwy191.
  14. R. F. J. Pfeiffer: Vorläufige Mittheilungen über den Erreger der Influenza. Deutsche medicinische Wochenschrift, Berlin, 1892, 18: 28. Die Aetiologie der Influenza. Zeitschrift für Hygiene und Infektionskrankheiten, 1893, 13: 357-386.


Please note that this transcript is based off the editorial copy and may not exactly match with the audio podcast.

This is Adam Rodman, and you’re listening to Bedside Rounds, a monthly podcast on the weird, wonderful, and intensely human stories that have shaped modern medicine, brought to you in partnership with the American College of Physicians. 


The last two episodes have taken the long view of the drug hydroxychloroquine as a way of explaining a single moment in time — the excitement and hype over the potential of that medication to prevent and treat, perhaps dramatically so, COVID-19. In the first episode I discussed the discovery of cinchona and the development of quinine; in the second I talked about our understanding of influenza and how quinine came to be a major — if not the most important — medication to treat the disease by the time of the 1889 Russian Flu. This episode is called “The Second Wave.” I want to step back from quinine for an episode and take a deep look at a single moment in time — August and September 1918, when it became clear that a horrifying “Second Wave” of the Spanish flu was crashing against the world. It’s the obvious parallel to our “Wuhan” moment, as we watched COVID-19 spread through that city in early 2020. Along the way, we’re going to discuss the mysterious origins of the Spanish Flu, treading deep into historiographical controversies, scientific missions to mass burial sites in remote Alaskan villages, the ill-fated journey of the HMS Mantua, debates about how to count victims of a pandemic, and the mystery behind Pfeiffer’s bacillus, the purported cause of the Spanish Flu If you’re binging all these episodes in a row, you know that’s a lot to cover — so let’s get started!


In this miniseries, we’ve traveled through many different epistemes — that is, frameworks of knowledge that fundamentally shape our approach to reality and our patients.  From the traditional medicine of Quechua healers, to the humoral medicine of the Spanish who took their remedies, to the Enlightenment-era nosologists who attempted to redefine fevers, to the pathological anatomists who extracted quinine and then studied it in all sorts of periodic diseases, to the German school and the beginnings of scientific medicine — we’ve run the gamut of Western medicine in the past few hours. But in this episode, I think it’s helpful to do a little check in with the state of medicine at the turn of the 20th century. This terms of debate in this episode should sound very familiar to you, because I’d argue that we’re still living in the episteme of “scientific medicine” that had come into its own in this period (though as a complete aside, the very nature of epistemes is that they are invisible until they suddenly “rupture” or shift underneath your feet — so we don’t really have a good way of knowing if that’s going on right now). 


By the teens, medicine had largely been re-organized on a scientific basis. The strength of the so-called German school had led to a worldwide movement to ground medical education in the basic sciences, and especially laboratory science.  In the United States, this was cemented by the publication of the Flexner report, which gave us the basic structure of medical education that persists to this day. Germ theory was ascendent and unchallenged — to the point that some diseases were mistakenly thought to be infectious, like I talked about way back in episode 36, about Joseph Goldberger’s mission to prove pellagra was caused by a nutritional deficiency. Laboratory science had given us important insights into treating disease, including our first understanding of the humoral immune system. Diagnostics were part of mainstream medical practice. After Roentgen had discovered x-rays, roentgenology had spread across the world — both with fluoroscopy, and printed on film or paper. By the time the Spanish flu hit, x-ray technology was still not quite able to get effective pictures of the lungs (it was still mostly used for bones), but certainly fluoroscopy was being used to examine large tubercles. Medical residents and house officers — and yes, they actually did live in the hospital — would spin their patients’ urine, looking for casts and protein. They made slides of blood to look for anomalous cells. The ran centrifuges to estimate the hematocrit. While routine blood chemistries were still a decade or so off, microbiology had become quite advanced, to the point we still largely use the same technique today — gram staining, petri dishes, different culture media. 


And unlike the late 19th century, therapeutics were catching up with diagnostics. The pharmaceutical industry as we know it had come into existence, growing out of the dye industry. There were already controversies about drug pricing and industry influence. Many drugs we still use today had been discovered or isolated — aspirin, adrenaline, atropine, digitoxin, heroin. New vaccines had been invented — from Pasteur’s rabies vaccine (giving our modern use of the word) in 1885, there had been either vaccines or antitoxins for diphtheria, tetanus, anthrax, cholera, plague, and the ill-fated tuberculin. And despite all the press that penicillin gets, the first antiinfective had been invented — arsphenamine, or Salvarsan, which cured most syphilis (in the early days, “chemotherapy” was synonymous with antibiotic). Communication technology had vastly shrunk the world. Medical journals and professional societies were still the main way physicians communicated with one another, just as they had been fifty years earlier. But the dominance of the telegraph meant that information traveled almost as quickly as it did today. I’m not entirely suggesting that the physicians who were faced with the Spanish flu were just like us — but the medical world of 1918 was clearly far closer to that of 2020 than 1820 — or even 1889, where we left off last time.


We also need a little grounding in the historical context of the outbreak of the Spanish flu — and that, of course, is that it started during what was then called the Great War, now depressingly called WWI after its sequel. The United States declared war in April of 1917, but for the first year of the war, support was mostly in terms of supplies and money. The US had traditionally kept a relatively small standing army; in 1915 it numbered at 100,000. Teddy Roosevelt and other members of the Preparedness Movement felt the United States should keep a far larger army — after all, at those numbers, the Germans outnumbered the Americans 20-1. While it slowly increased the number of troops in the following years, after the declaration of war, the US would mobilize almost 4 million people. As you can imagine, this required a number of training camps across the country, packed with young men from all over the nation. Even then, it wasn’t until the summer of 1918 that significant numbers of troops started to arrive at the Western front. To put that in perspective, the end of the war was just a few months later, on November 11th, which is still celebrated throughout the world as Veteran’s day, Remembrance day, and Armistice Day.


The origin of the Spanish Flu is still under pretty vigorous debate, so I’m going to start with the low-hanging fruit. I think everyone has heard the story that the name “Spanish flu” became popular because Spain was neutral in the Great War and didn’t have dramatic restrictions on the press, whereas in the UK, the US, France, and Germany, news about a disease spreading in Army camps would naturally be suppressed on national security grounds. The first evidence of the disease in Spain comes from May 22, 1918, about a new, relatively mild influenza that was rapidly spreading. A week later, the king of Spain, Alfonso XIII, became sick, along with several members of his government, at which point the Spanish flu essentially became daily news. The initial Spanish name was the “Soldado de Napoles,” a playful reference to a song from a popular musical that was “equally contagious,” which also references how relatively mild the disease seemed to be — it would be like if we called COVID-19 the “Hamilton disease” because everyone couldn’t stop “catching it.”


The Spanish themselves were under no illusion that the flu had originated in Spain. The newspapers reported that the disease had been carried across the French border by unskilled laborers replacing those Frenchmen killed at the front, and when the Second Wave crashed on Spain later that fall, the name “French Flu” replaced any playfulness. Regardless of what the general public felt, I should also note that physicians and epidemiologists during and immediately after the Spanish flu also did not seriously think that the flu had started in Spain. An American Medical Association report published in the 20s immediately disabuses the reader of that fact, and goes over a list of competing theories. Today this has basically been narrowed down into three competing arguments — and I want to go over and consider them, if only to illustrate the difficulty in using historical sources, both medical and nonmedical, to “know” the past.


The first theory was, unsurprisingly, China, for two reasons — the first is that, in fact, most traceable pandemics have seemed to start in Asia, and China in particular, and secondly, that anti-Chinese bias, especially as carriers of contagion, was in full spring in the West, and given the racism that some Asian Americans have received here in the US, probably hasn’t abated all that much.  In November of 1917, an outbreak of “winter sickness” was noted in the villages of Northern China, spreading incredibly rapidly — traveling 500 kilometers in only six weeks. During this same period, the British and French recruited almost 100,000 Chinese laborers as part of the Chinese Labor Corps; in March of 1918, a ship from Northern China set sail for Vancouver. Three thousand of these workers ended up in medical quarantine in Canada with “flu-like” symptoms; in fact, their rail car was sealed shut to prevent the spread of the disease. 


This sounds like a pretty compelling story — but there’s a major problem. In the early days of this outbreak, a number of American-trained Chinese physicians investigated this outbreak, clustered in northern villages bordering Manchuria, and determined that it was pneumonic plague — the pulmonary manifestation of the bubonic plague, a disease that still pops up from time-to-time in the very same area, always to great fanfare in the media. The Manchurian Plague, in fact, had killed almost 60,000 people just a few years earlier — and had led the physician Wu Lien-teh to utilize face masks for the first time in controlling the outbreak of the disease. Also, it was 1917. The Chinese physicians who investigated the outbreak were familiar with the disease and confirmed their findings with microbial cultures showing Yersinia pestis. And I just need to make an aside — there was an excellent 99 percent invisible about Dr. Wu — I’ll place a link in the shownotes. This was an old theory widely accepted in the 20s, and has found increased currency lately, most notably through the historian Mark Humphries who uncovered those archival documents about the Canadian quarantine of Chinese laborers.


The second theory is probably the oldest, but has been re-popularized by the military historian Oxford — from 1916-1917, two clusters of an incredibly deadly respiratory disease were noted, the first at a British field hospital in France, and the second at the Aldershot army hospital which baffled the authors: “a symptom complex so distinctive as to constitute a definite clinical entity.” The poor soldiers presented with mild respiratory illness and then rapidly progressed to bronchopneumonia. The major clinical finding was “a peculiar dusky heliotrope type of cyanosis of the face, lips, and ears, so characteristic as to hall-mark the nature of the patient’s malady.”  Mortality was as high as 40%, and post-mortem invariably showed bacterial superinfection with pneumococcus. At the time, it was still thought that Pfeiffer’s bacillus was the unique cause of influenza, whereas pneumococcus was more associated with bronchial pneumonia. This led the authors to conclude that the disease was a distinct entity from the flu, but by 1919 they had rethought their hypothesis: “in essentials the influenza pneumococcal purulent bronchitis that we and others described in 1916 and 1917 is fundamentally the same condition as the influenza pneumonia of this present pandemic.”


The third theory is the Kansas hypothesis — also an old theory, and now promoted by John Barry, the historian who wrote The Great Influenza. In January 1918, a strange new influenza was noted in Haskell Country, KA — in particular striking the young and healthy, rapidly progressing to pneumonia, and carrying a high mortality rate. It concerned a local doctor Loring. Miner so much that he reported the disease to the US Public Health Service. As Barry points out, this area of the country was sparsely populated with humans, but had a large population of birds. Camp Funston — now Fort Riley — was located nearby, and on March 4th, 1918 the first case of influenza was reported at Camp Funston — the first definitive documented case of the Spanish flu. The camp held an average 56,000 troops which freely moved throughout other training camps in the US. And on March 18th, Camp Forrest and Greenleaf in Georgia reported an outbreak of influenza, and by April 24 of 36 army training camps had influenza outbreaks. 


So these are the three historical arguments — China, France, and Kansas, all based primarily of primary documentation — in the China hypothesis, of Canadian quarantine records, for France, from medical reporters from doctors published in 1916 and 1917, and for Kansas, from a frontline doctor, as well as contemporary newspapers documenting a mysterious flu in the same period.  There’s an interesting historiographical discussion to be had about how to  weigh these different hypotheses — none of them entirely fit, since from what we know about the disease, it was initially a relatively mild flu, and the Kansas and French diseases appeared quite severe.. But fortunately, we can actually use laboratory science to help narrow down these hypotheses. Jeffery Taubenberger, a virologist at the US Armed Forces Institute of Pathology in 1995 realized that polymerase chain reaction, PCR, technology could be used to reconstitute the genetic material from the Spanish flu. His team identified 10 samples from soldiers who had died of the disease in an archive dating back to the Civil War; two were positive for an H1N1 influenza. Using PCR, they were able to partially sequence the viral nucleic acid. 


This gave a very exciting opportunity — if they could find an appropriate gravesite where a victim of the Spanish Flu might be found, and be appropriately preserved. They found it in Brevig MIssion, a remote village in Alaska not far from Nome. When the Spanish Flu struck the Seward Peninsula in the winter of 1918, 72 of the 80 villagers had been killed. The devastation was so total, and the fear of the bodies of the infected so great, that the Provincial government hired gold miners to thaw the ground and dig a massive trench for the bodies. The frozen soil kept the bodies preserved for 90 years, until a mission recovered the lung tissue from a well-preserved woman buried 7 feet deep. This sample — inactivated of course — allowed Taubenberger’s team to recreate the genome. A few years later, two additional samples were found to be positive from the Royal London Hospital.


Now this is all starting to get way above my head, but analysis of the neuraminidase protein — the N1 — matches very closely circulating avian viruses. This strongly suggests that firstly, it was transferred from birds to humans, and then secondly, it had not circulated in humans for very long, and probably not more than a couple months after the transfer. Second, all three samples — all from the Second Wave and not the less virulent first wave, are more-or-less identical. All of this would certainly make the French hypothesis far less likely, which calls for two years of circulation before it crashed on the world. There’s also some epidemiological evidence that suggests against the French hypothesis — the armies may not have tracked influenza initially, but they definitely tracked bronchopneumonia, and morbidity reports from the major armies showed no increases in bronchopneumonia (not withstanding the French examples) until spring of 1918.


But, of course, does it really matter? It probably shouldn’t be that surprising that we don’t know where the Spanish Flu came from. Mystery is the rule, not the exception. All circulating influenza A today is a successor of the Spanish Flu — which is insane, if you think about it, because the Spanish Flu is technically still killing people. But even in 2009, we still have no idea where the Swine Flu, also H1N1, originally came from. That we can track COVID-19 back to a single city is an exception. In any event, from the perspective of people on the ground, the Spanish flu seemed to spread all over the world, all at once. So Camp Funston had the first definitive case — but the next week, cases were reported at San Quentin Prison, in Detroit, and in South Carolina as well as the aforementioned army bases. By April, it had followed the American Expeditionary Force to France, and then to Spain (where it got its moniker). By June it was in North Africa, Bombay, Calculla, China, New Zealand, and the Philippines. It took only four months to circle the world.


Now here is what makes the Spanish flu so traumatic. At this point, it is still the “Soldado de Napoles” — a relatively mild influenza pandemic that killed only the elderly or the infirm, much like the Russian Flu of 1889. Sure, there were concerns about it affecting soldiers’ ability to fight in the trenches, which is why governments were suppressing news of its spread. But in August of 1918, something horrific happened, and a century later we still don’t know exactly what that was. 


On the 15th of August, the HMS Mantua, an ocean liner refitted as an armed merchant ship, docked in Freetown in Sierra Leone, then part of British West Africa. Multiple crew members and passengers were sick with what appeared to be the regular Spanish flu and were taken to the hospital. We still don’t know what happened — had the virus mutated on the Mantua? Did it mutate while in Sierra Leone? — but the effects were absolutely horrifying. This new influenza appeared to have a death rate almost 10 times higher than previously, and killed primarily young healthy people. A localized outbreak in the dockworkers of Freetown soon overwhelmed the city, killing almost 5% of the population within three weeks. During the same time period, this “second wave” of the Spanish flu was noted in Brest, in France, and in Boston, where it rapidly burned across the world. This version of the Spanish flu killed primarily people aged 20-40, most notably from a secondary pneumonia caused by streptococcus; contemporary autopsy reports also suggest lung findings consistent with what we now call acute respiratory distress syndrome or ARDS, possibly as a result of something called a cytokine storm, where the body’s immune system overreacts to effects of the virus on respiratory epithelial cells, which would explain why young people with healthy immune systems were most at risk. An alternative/complementary hypothesis would be that older adults had antibodies from another H1 influenza from their youth; there’s some interesting epidemiological data, but since the Spanish flu is the oldest flu virus we have pathological samples from, this is conveniently untestable.


There weren’t just two waves — the disease returned in the winter of 1919 as a “third wave,” popped up for another two years, and as I mentioned earlier still technically continues to circulate as influenza A. Estimating the human toll of the pandemic has been very difficult, and epistemologically fraught. In the 1920s, Edwin Oakes Jordan estimated 21.5 million global deaths, essentially by using death certificates in the US to try and estimate a local death rate, and then extrapolating this to the rest of the world. 


In 2002, Johnson and Mueller used regional excess mortality to attempt to calculate a global burden of influenza mortality. Excess mortality has again been in the news lately with COVID-19 — the essential idea is that case definitions are notoriously difficult, especially in the Spanish Flu, where we didn’t have a diagnostic test, but even in 2020 when local testing systems were rapidly overwhelmed with COVID-19, and when there’s an unknown number of asymptomatic or minimally symptomatic cases.


To take two extreme examples — if someone has a massive heart attack and dies, but has a positive swab for COVID-19 with no other symptoms, did they die of the disease?


Or — someone has a heart attack at home, but because of fear of contracting COVID-19, they stay away from the hospital, and die the next day. Their swab is negative. Did they die from the disease?


Case definitions can be difficult to pin down, especially in a rapidly changing pandemic, but vital statistics registers are far more accurate — there’s not much controversy about what constitutes a death — so you can get a more reliable measure of mortality from a pandemic by comparing the number of deaths during the pandemic to a “baseline” mortality figure from the years before — hence excess mortality. Using this methodology, Johnson and Mueller calculated a number that is far higher — 50 million, though allowing for uncertainty they suggested almost 100 million deaths. 


And now this gets to the heart of the difficulty in calculating the burden of the Spanish flu, because statistically there was another mass casualty incident ongoing during this period — the Great War. In 2018, Spreeuwenberg and colleagues used publicly available mortality data sets for 13 countries and British India, and applied a “multilevel regression model to control for distorting factors such as war and the underlying time trend in mortality”. I know you can’t see me, but I’m using air quotes here, because I understand those words, yet really have no idea what that means. And they came up with a far different number — 17.4 million people. No matter which estimate you go with, the burden was huge — the world’s population was 1.8 billion; 17.4 million people was  1% of the world’s population; 100 million is 5.4%. In comparison, about half a million people died in the last influenza pandemic, the 2009 swine flu, and as of today 740,000 official deaths from COVID-19 are counted, though that number is assuredly much higher if/when an excess mortality methodology is used. 


And here’s the crazy thing — we still don’t know WHY the Spanish Flu was so deadly. Taubenberger points out that an analysis of hemagglutinin and neuraminidase do not suggest anything especially unusual that would cause especial virulence. We don’t even know what sort of mutation happened between the first and second wave– presumably some sort of antigenic drift, but all of the samples that have been recovered are from the second wave, not the first. 


The Spanish Flu has been in the news a lot because of comparisons with COVID-19 — more than it’s 100-year anniversary two years ago.  I mean, when I started this series on hydroxychloroquine, you knew that at some point I was going to get here, right? And a lot of the press seems to suggest, well, at least COVID-19 isn’t as bad as the Spanish Flu. But I want to again place my tinfoil hat on and interrogate that idea for a second.First, I’d like to challenge it from a purely pragmatic perspective. We know that a lot of people died from streptococcus pneumonia and respiratory failure in the Spanish Flu. Unfortunately for those victims, sulfa drugs wouldn’t be invented for another decade, and penicillin for 25 more years; mechanical ventilators would wait even longer. If the people of 1918 had had those interventions, I think we can reasonably assume the death rate would have been lower. 


But just how deadly was the Spanish flu to an individual in the 1918 setting? Or putting it another way, what was the case-fatality rate/ratio — the proportion of deaths from a disease compared to the total number of cases? What was an individual’s chance of dying? In 2013, Taubenberger and his team performed a fascinating analysis of death certificates from a single state in the US — Kentucky. They manually inputted every single death certificate from the state between 1911 to 1919, linking them with three outcomes — death from pneumonia and influenza, respiratory deaths, and total deaths — 310,363 death certificates in total. Their analysis found an excess death rate of 0.54% of the entire population over the period of the pandemic — driven almost totally by respiratory deaths and influenza and pneumonia deaths. Just like was reported in the original pandemic, the death maximum was between the ages of 24 and 26, and in under 1 year old; age 9-10 and 56-59 had the lowest deaths. The paper has a number of fascinating observations in it — including advancing the hypothesis that older adults were protected because of an older influenza virus infection — and I recommend you read it if you’re interested. But he has an almost throwaway line in there about estimating the CFR — given estimates on the percentage of the population infected (very high), and the excess mortality, we can begin to estimate a CFR, which Taubenberger guesses was between 0.5-1% in the United States (with estimates in other parts of the world to be considerably higher). Now there is still considerable uncertainty about what the CFR and IFR are of COVID-19, but 0.5-1% is certainty within a range that most people find believable.  So to take a moment to think about this — with 1918 technology, COVID-19 and the Spanish Flu may have had a similar fatality rate, at least within the United States. 


We’ve now set up our moment in time. It’s August 1918 in the United States. Reports of a deadly second wave of Spanish Flu start to pour in throughout the country, we now know it spread from Boston outward. The Surgeon General of the United States, Rupert Blue had cut his teeth fighting plague. It hardly seemed likely that a mild disease such as influenza could cause such a panic. But over the month of August, he performed a telegraphic survey of the state health departments across the country, and what he found disturbed him. The disease had already taken hold across the US, in particular in army camps. A tramp steamer had shown up at Newport News with the entire crew sick. Fort Morgan, in Mobile, had needed to shut down to protect recruits. Clusters of faces were popping up in the civilian populations of New York, Boston, Philadelphia, and New Orleans. 


While Blue mobilized a national plan, he realized he needed to issue practical advice to every frontline physician in the country. Therefore, he issued a special scientific bulletin going over the latest recommendations in fighting the Spanish flu for every “medical man” in the country. And in case anyone missed it, on September 18, 1918, the Associated Press circulated an article that was picked up in almost every newspaper in the country with this information. Quoting directly:


“Infectious agent — the bacillus influenza of Pfeiffer. Mode of transmission — droplet infection plays an important part. Method of control: early recognition of disease, bed isolation during the course of disease, with screened beds. Immunization — vaccines are used with only partial success. Quarantine — none: impractical. The attendant of the case should wear a gauze mask. During epidemics persons should avoid crowded assemblages, street cars, and the like. Education as regards the danger of promiscuous coughing and spitting. Patients, because of the tendency of the development of the bronchopneumonia, should be treated in well-ventilated warm rooms.”


This should sound pretty familiar, because it’s essentially the same working advice for influenza that was present during the last pandemic, the Russian flu, 30 years earlier, with one big exception — Pfeiffer’s bacillus. Remember that by the time of the Russian flu, there was still space to debate whether influenza was caused by an as-yet unidentified infectious microoganism (like Osler argued), or whether it was caused by “local conditions,” essentially a neo-miasmatic theory, like Flint suggested. The staggering number of influenza cases during the Russian Flu, as well as increasingly sophisticated microbiological techniques led to the remarkable discovery of Richard Pfeiffer. In the nose of multiple influenza patients, he saw a small, rod-shaped bacteria, though they were quite difficult to culture. He called his discovery Bacillus influenzae, though it was commonly called Pfeiffer’s bacillus. From the very beginning, it was unclear whether the bacillus caused the flu, or just happened to be around in influenza victims. A similar controversy happened with the pneumococcus — it was originally identified in pneumonia patients and thought to be the cause of the disease pneumonia; very quickly it became apparent that while streptococcus certainly caused SOME pneumonia, it wasn’t present in every case of the disease.


Still, it was generally accepted that Pfeiffer’s bacillus was the cause of influenza. Things took on increasing urgency with the Spanish flu. As the death toll increased, there was a worldwide search to discover a vaccine — sound familiar? — though in this case, that meant anti-sera, similar to the diphtheria antitoxin — and in fact, convalescent plasma would soon be used in this fashion. But if you wanted bacillus influenzae antiserum, you needed to be able to identify infected patients. Because of this, widespread gram staining and culturing took place, which again showed that Pfeiffer’s bacillus was present in many cases — but certainly not every patient. By now, it had become clear that there were agents called viruses that were invisible under microscopes, and that were filterable — that is, they could pass through a filter that blocked all bacteria. In 1921, researchers at the Rockefeller Institute successfully filtered nasal secretions, which nonetheless still caused disease in rabbits. That is to say, they had discovered influenza was a filterable virus. But not understanding the magnitude of their discovery, they instead hypothesized it was an atypical bacterium. Their results never caught on; it didn’t help that during this same time, chocolate agar was developed, which could rapidly and easily grow Pfeiffer’s bacillus. As we now know, H Flu can be found in normal upper airways — and certainly was far more common before the Hib vaccine — and was soon being isolated from Spanish Flu patients all over the world. 


That influenza was caused by a virus wouldn’t be proven until a decade later, when Richard Shope repeated the filter experiments with pigs and swine flu, finding that a filterable agent indeed caused the disease in pigs. He discovered a “pig version” of Pfeiffer’s bacillus, and performed a very clever experiment — given alone, bacillus influenzae suis caused no disease. But when given with the “filterable agent” it caused a disease far worse than the agent alone. Thus, he realized that influenza was caused by a virus, and that Pfeiffer’s bacillus merely caused a superinfection. This experiment was soon repeated in humans, and by 1933, it was commonly accepted that influenza was another mysterious “virus”. 


That was a bit of a tangent. My whole point here is that at the beginning of the deadly second wave of the influenza pandemic, the most “up-to-date” information that the surgeon general mailed to every doctor in the country, as well as telegraphing to every paper just in case they missed anyone, was already rather old fashioned. Influenza just hadn’t been that exciting of a disease.


And I think you can see where this entire episode has been leading. In this announcement, he also gave the most up-to-date advice on treatments: “During the present outbreak in foreign countries the salts of quinine and aspirin have been most generally used during the acute attack.”


Why did he advise this? Well, it was because in the last flu pandemic, the treatment of choice had been quinine and antipyrine, and aspirin had largely replaced antipyrine — more effective and much less disgusting to take. But if you’ve been listening to the miniseries, the causal chain goes back much farther than that — because quinine was thought to work in influenza because it was a periodic disease, similar to malaria, dating from the early 19th century, which in turn was based off the centuries-old notion that certain classes of fevers — tertians and quartans in particular — responded to the bark of the cinchona tree. 


This all sounds very quaint, and a few decades earlier, this would have been the end of it. But medicine was changing, and the idea of efficacy — that you needed to test to see if treatments were actually effective — was starting to take hold in the wider medical world. That’s it for this episode, but next time we’re going to talk quinine, chloroquine, and hydroxychloroquine — how our notions about therapeutic efficacy changed over the 20th century, and how these drugs roared back in the 21st century after almost a century-long pause.


By the way, I know I’ve been teasing Dr. Rahul Ganatra’s presence for three months now — and I’ve actually recorded the episode with him. But I keep getting stuck in fascinating rabbit holes, so he will have to wait for the fourth and (hopefully final) episode.


But wait — it’s time for a #AdamAnswers! This is from, well me! I asked on Twitter, “Where did the godawful yellow-text-on-blue-background default powerpoint template that was all the rage in Med Ed for the past decade come from? I remember as a med student at Tulane being specifically told to do my powerpoints like that…”

So for some context, I was a medical student from 2009 to 2013, and a high percentage of my preclinical lectures were projected with a bright blue background with yellow and white text. I have some examples up on my Twitter feed if you want to see. I honestly didn’t think a lot about it, until I was in an amazing medical humanities class that was taught by Chad Miller and Ben Rothwell. We had a lecture during this class and told us something to the extent of, if you want to be taken seriously as a medical educator, you need to use a blue background with bright yellow text. As a fourth year medical student, I wasn’t the sort of obnoxious person that I am now who questions everything he’s told, so I dutifully switched my powerpoints to yellow on blue for a few months, because of course I wanted to be taken seriously. 


CME is available for this episode if you’re a member of the American College of Physicians at www.acponline.org/BedsideRounds. All of the episodes are online at www.bedsiderounds.org, or on Apple Podcasts, Spotify, Google Podcasts, or the podcast retrieval method of your choice. The facebook page is at /BedsideRounds. The show’s Twitter account is @BedsideRounds. I personally am @AdamRodmanMD on Twitter, which is where you can see the Twitter thread I made about blue backgrounds and yellow text.


All of the sources are in the shownotes, and a transcript is available on the website.


And finally, while I am actually a doctor and I don’t just play one on the internet, this podcast is intended to be purely for entertainment and informational purposes, and should not be construed as medical advice. If you have any medical concerns, please see your primary care practitioner.