For a special holiday treat, we’re going to explore two tales of salmonella disease detectives — the first about Mary Mallon (“Typhoid Mary”) and the birth of a new genre; and the second about a mysterious salmonella outbreak at Massachusetts General Hospital solved with the assistance of a very jolly patient. Along the way, we’ll talk about clinical epidemiology, the long-lasting influence of Berton Roueché, and the joys of being an internist!
You can sign up for the Digital Education conference at cmeregistration.hms.harvard.edu/digitaleducation.
- Buckle GC, Fischer Walker CL, and Black RE, Typhoid fever and paratyphoid fever: Systematic review to estimate global morbidity and mortality for 2010.J Glob Health. 2012 Jun; 2(1): 010401.
- Marineli F et al, Mary Mallon (1869-1938) and the history of typhoid fever.Ann Gastroenterol. 2013; 26(2): 132–134.
- Soper GA, The Curious Career of Typhoid Mary, read on May 10, 1939 before the Section of Historical and Cultural Medicine. Retrieved from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1911442/pdf/bullnyacadmed00595-0063.pdf
- Norrington B, Cochineal: A Little Insect Goes a Long Way, UCSB Geography.
- Roueche B, The Santa Claus Culture, The New Yorker, Aug 27, 1971.
- Lang DJ et al, Carmine as a Source of Nosocomial Salmonellosis, NEJM. Apr 13, 1967.
You can buy Medical Detectives here: https://www.amazon.com/Medical-Detectives-Collection-Award-Winning-Investigative/dp/0452265886
Based on editorial copy and may not be the same as what is broadcast.
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. This episode is called Santa’s Salmonella, and it’s my first-ever (and probably only) holiday episode! But really, it’s about the most appetizing thing that you could think of as you’re preparing for your winter (or really, given that I know you guys listen from all over the world, also your summer, or New Year, or almost-Lunar New Year) holiday with delicious treats — and that of course is everyone’s favorite GI pathogen, salmonella! In this special holiday episode, we’re going to gather ‘round the hearth and listen to heart-warming story about disease detectives tracking down two different salmonella outbreaks 50 years apart, as well as the importance of storytelling in medicine with Berton Roueche, and how much of an influence he’s had on me and this podcast. So let’s mix up some eggnog — nothing like some raw eggs, throw another yule log in the fire, and let’s get salmonella-ing!
I always like to start my Christmas day with a little microbiology review. (that’s actually not that far from the truth — I think I’ve worked Christmas for the last 7 years). So I want to welcome you all to my family tradition by going over salmonella. Salmonella is a gram negative rod; when I was a student, I always assumed that the name was given because of the salmon-colored rose spots that happen during typhoid fever, but no — it’s named after Daniel Salmon, who was the first Doctor of Veterinary Medicine in the US, and was the director of the Department of Animal Industry for the US government when the bacteria was discovered as the cause of hog cholera . The original name was salmonella choleraesuis (now salmonella enterica), and Theobald Smith, the discoverer, more or less randomly decided to name it after his boss.
Honestly, it kinda worked out, because salmonella is actually a pretty okay name as far as gastrointestinal pathogens go. It’s transmitted through the fecal-oral route, and usually requires a high number of organisms to be ingested to cause infection (and the reason we know this, as I was perusing the literature, is that people — often prisoners — were made to ingest increasing amounts of salmonella as part of vaccine trials). Most infections cause only a mild gastroenteritis, indistinguishable from other infectious diarrheas, with fevers lasting 2-3 days, and diarrhea up to a week. A certain serovar of salmonella enterica, which used to be called salmonella typhi, causes typhoid fever, also known as enteric fever. Salmonella disseminates through the body’s lymphatic system, leading to a syndrome of hepatosplenomegaly, generalized sepsis with high fever, and the aforementioned rose spots (which appear in about a third of patients). The mortality is considerably high — 10-30% in the pre-antibiotic era, but now only 1-4% with appropriate treatment.
Nontypoidal salmonella is still incredibly common across the world; in the US the CDC estimates 1.35 million infections, 26,500 hospitalizations, and 420 deaths a year. Salmonella is generally spread from poultry, eggs, and milk, but pets are also a major source, especially lizards. As I’m writing this, there are currently multi-state outbreaks associated with pet Bearded Dragons and pet hedgehogs. So if you have those — consider this a public service announcement.
There has long been interest in typhoid fever from a public health perspective; long before a microorganism was identified, or even before germ theory existed, it was clear that the disease was spread to others through food. And because of that, one of the first real examples of the “disease detective” narrative — and of how seriously the authorities took it — comes from salmonella, with the case of Mary Mallon, then and now better known as Typhoid Mary.
This “infectious narrative” comes from George Soper, a sanitary engineer who identified the first asymptomatic carrier in history, and wrote extensively about his experiences after the fact, which is what I’m referencing here. Mary Mallon was an Irish immigrant to the US, and was working as a cook for a wealthy banker at a rented summer residence on Long Island when a typhoid fever outbreak sickened 6 of the 11 people staying at the house. Several local doctors made detailed examinations but were unable to figure out where typhoid fever had come from. This is when Soper was brought in. Soper is quick to mention that the family actually did not necessarily subscribe to germ theory for typhoid fever; they felt that foul odor and water were the cause — the persistence of miasma into the 20th century. Soper, however, was up-to-date with the latest literature, and had successfully cultured salmonella in his investigations in the past. He started by confirming the “essentials” of the outbreak — when everyone became sick, what they ate and when, who they were in contact with, etc. The next was to investigate all possible sources. Before Soper had arrived, a meal of freshwater clams had been the most likely culprit, though certainly not definitively so. With his detailed questions, however, the timing didn’t work — the clams didn’t fit the bill. He then turned his attention to the well water on the property, the cesspool, the manure on the lawns, even investigating the sanitary facilities at the neighboring house — but there was nothing suspicious, and the cultures he took were negative.
The trail would have gone cold here if it were Soper had not been an expert in typhoid fever. From the literature, as well as his personal experience, he knew that convalescent patients could carry the bacteria in their stool for several weeks after recovering. Could someone have been recovering from typhoid fever and unknowingly infected the others? He went to his details again, and noted that all of the infected people had fallen ill shortly after Mary Mallon’s arrival. With this insight, he again went back to the menus, and found the culprit dish — an ice-cream with fresh peaches cut up and frozen inside, which does honestly sound quite lovely. If Mary had had typhoid fever several weeks prior, she could be the source.
Unfortunately, by this time, Mary had already returned to New York City, and Soper had no way of contacting her directly. However, through the family he had her employment agency. They unfortunately had no address or way to contact her, but he did have the addresses of previous families who had employed her, so he started making house calls to track her down. And what he found shocked him — in all, he found 8 families she had worked for, and all of them had outbreaks of gastroenteritis, twenty two people in total. Seven of them had typhoid fever, and some had died.
All of this happened in the two years she had been with her employment agency — which would mean that Mary was not just a convalescent patient, but an asymptomatic carrier, which had never been described before. Because the implications for public health were so great, Soper felt that he had to confirm his suspicions, and I’m just going to quote the man’s recollections himself, because the situation is so ridiculous. So remember, he didn’t have her home address, so he burst in on her while she working in a fancy house on Park Avenue in New York City:
“I had my first talk with Mary in the kitchen of this house. I suppose it was an unusual kind of interview, particularly when the place is taken into consideration. I was as diplomatic as possible, but I had to say I suspected her of making people sick and that I wanted specimens of her urine, feces and blood. It did not take Mary long to react to this suggestion. She seized a carving fork and advanced in my direction. I passed rapidly down the long narrow hall, through the tall iron gate, out through the area and so to the sidewalk. I felt rather lucky to escape.”
So it turns out that bursting in on a person’s place of employment, accusing them of killing the people who they work for, and then asking for a stool sample right then and there, is not an effective strategy in investigating disease outbreaks. Who knew. After failing that spectacularly, Soper dispatched the state health department, who sent the police to forcibly take a sample. The story of her apprehension is similarly bonkers. I have Soper’s piece in the show notes if you want to read it. Let’s just say she led the police on a five hour chase through the snowy city, and even after she was caught she fought the entire way to the hospital.
The stool culture showed, of course, bacillus typhosus — salmonella enterica as we would now call it. Multiple cultures were taken, all of which showed salmonella.
The past year, there had been ~3000 cases of typhoid fever in New York City; 10% of those had died. From the perspective of the city health department, Mary was a threat to public health. Especially after the chase and the arrest, the health department was not inclined to trust her, and sent her to North Brother Island at Riverside Hospital, where she was forcibly quarantined, eventually in a cottage. She was treated with a number of theoretical therapies — hexamethylenamin, urotropin, brewer’s yeast, all sorts of laxatives with the hope it would clear the salmonella. She was offered to have her gallbladder out, with the thought that gallstones might be causing the infection, which she declined. But despite all these therapies, 120 of 163 stool samples continued to show salmonella. She was released two years later but quickly started working again as a cook under an assumed name, since Typhoid Mary had gone viral, or I guess bacterial, in the press. Unfortunately but expectedly, disaster followed her, and at the Sloane Maternity in Manhattan, she sickened 25 doctors, nurses and staff, leading to two deaths. She was again forcibly quarantined on North Brother Island where she lived an additional 28 years until her death. Soper actually visited her several times during her final confinement, and in his opinion felt she was at peace — her life as Typhoid Mary had been extraordinarily difficult and stressful. I’m skeptical of that, but we do know that she never again fought her confinement in the courts. In the end, Mary sickened 53 people with typhoid fever and caused three deaths — which Soper himself points out was not even a particularly large toll by the time he was writing in 1939 when many further asymptomatic carriers had been found. Regardless — because of her large footprint in media, her essential “life sentence” despite not committing a crime, and for being the first asymptotic spreader of typhoid fever (and the greatest example of the dangers of asymptomatic spread of all infectious diseases), Mary Mallon’s story became the de facto
So now I want to jump fifty years in time to 1966, traveling a little north to a city I call home — Boston, MA, in the Massachusetts General Hospital, where a mysterious outbreak of a rare subtype of salmonella, called cubana, was raging.
There are over 2600 serotypes of salmonella — though in the 1960s, these were actually considered different species. I don’t want to get too much in the weeds with the debates on salmonella taxonomy, but many of these subspecies, now called serovars, were named after where they were discovered, and salmonella cubana, as you can imagine, was first encountered in Cuba in 1946. Up until this point, it was extraordinarily rare outside of Cuba. In 1970, of the 24,000 salmonella cases in the US, only 166 were from cubana.
Just like Soper with Mary Mallon, the responsibility for tracking down this outbreak fell into the lap of the first person to realize what was going on, which in this case was David Lang, a pediatrician at MGH. This outbreak was clustered on the pediatric ward; four infants had become ill with salmonella cubana, and one of them — a boy with congenital kidney disease — had died. Most concerningly, it appeared that none of them were admitted with any symptoms of a gastroenteritis, and the earliest stool sample showing salmonella had been taken after the child was in the hospital for a week. That meant the infection was coming from inside the hospital — a nosocomial infection, nosos, of course, meaning sickness in Greek as in nosology, and komein meaning to take care of.
As Lang sat down to consider this cluster, it was clear that these four patients were part of a larger outbreak stretching back almost a year. Throughout 1965, from March to October, there had been a mysterious rash of salmonella cubana noted in cultures from the MGH microbiology lab — five females and two males, ranging from 1 to 70 years old. But these cases were spread all over the hospital and not clustered in a single unit like the current situation. In fact, the most curious thing was that 4 of the 7 had actually been diagnosed with samlonellosis on admission — they were admitted to GI symptoms, and then later diagnosed with salmonella cubana. The fifth patient had a positive culture on hospital day 2. So five of these 7 patients seemed to have been infected outside of the hospital. At the time this was odd for sure, but not compelling evidence of a hospital outbreak. But with this new outbreak on the pediatric ward, these previous cases certainly seemed to be part of a larger nosocomial outbreak.
All of that could wait, though, since Lang wanted to first ensure the safety of the children still on the ward. He ordered stool cultures from all the patients from the latest patient’s room (this was the 1960s, so that was six patients), as well from all the nurses and food handlers who had been on the pediatric unit during outbreak period. While waiting for the results, Lang shut down the pediatric units to anything except emergencies or the critically ill. In the meantime, any patient with loose stools was placed on strict isolation and contact precautions. All patients and staff traveling off the unit had to be in gowns, and the entire unit went through a deep clean.
In a few days, the results of his studies started to trickle in. Two of the children from the room where the boy had died came back positive for S. cubana; the rest of the patients and the staff were negative. Lang felt the cause must be close — there were now six patients, all in the pediatric ward, all with Salmonella cubana. It would only take tried and true epidemiological methods to track down the source.
But then the whole case was thrown a curveball. A few days later, a 72 year-old man in a completely different building developed symptoms of salmonella during his hospital stay, with the cultures showing salmonella cubana. Lang was now in charge of investigating a hospital-wide outbreak.
What could possibly cause salmonella infections over the entire hospital? The first thought was pretty obvious — just like Mary Mallon a half century before, food was most likely source. And if the kitchen was implicated, that could mean that MGH was on the verge of a massive outbreak. But there were a couple of problems with this theory — at this point, infections had been going on for six months. If the kitchen were implicated, there would have already been a massive outbreak, rather than the 10 or so scattered cases. And then, as Lang learned, there wasn’t a single kitchen — in the 1960s, just as today, MGH is comprised of multiple buildings, and each had its own kitchen. But cases had been in multiple buildings.
But then Dr. Lang realized that there was one kitchen that served all buildings in the hospital — the special-diet kitchen. The special diet kitchen produced meals for the sickest patients, which at the time meant, for example, low-carbohydrate for diabetes, low protein for “nephritis,” or soft foods for those with diarrhea. Dr. Lang and his team inspected the special diet kitchen and collected stool cultures from all the staff members, as well as taking swabs from throughout the kitchen. None of them grew salmonella.
The trail had gone cold. MGH had a smoldering salmonella outbreak, and with no obvious source, there was no end in sight.
But then Lang became aware of one extra infection in a very, very unusual patient.
I bet you’ve been wondering what the Christmas connection was. Well, at some point in the early 1950s, a strange tradition had started on the surgical service at MGH. Back then, all admissions from the emergency room were split by a single medicine resident and surgical resident. The exact details are lost in the sands of time — which essentially means not published in the single grand rounds on the topic — but on one late and apparently not particularly busy night in December, a surgical resident decided to admit a fictitious patient to the male surgical ward — which makes me very nostalgic, because it Botswana, wards were divided by male and female. The patient, of course, was Santa Claus! From that time on, this entire admission was an opportunity to have some fun, For example, admitting orders might read:
Activity: OOB to sleigh TID, back of sleigh at 30 degrees, or sled rest, no flying until cleared for takeoff.
Condition: Critical but jolly
Medications: Egg to noggin’ PRN insomnia, decadron the half with boughs of holly q6h
Labs: Deerlantin level
Nursing: HO2, 2L nasal cannula, check his blood sugar twice, find out if he’s being naughty or nice, holy foley! Monitor the urine output!
Misc: Duke the Claus the Bells of Holly, Call “Ho-!HH!-Ho!” for visions of sugar plums, Patient insists on room with a chimney, Allow the patient to have a silent night.
A couple things here for my younger American listeners — back in the day, and still through much of the world, we had to write a long series of admission orders when a patient came into the hospital. The mnemonic I used was ADC VAN DIMLS, and merely explaining this is making me feel old.
By the 1960s, the tradition was in full swing — the ER nurses were responsible for providing the basic suit, and the entire outfit would be customized by the nurses on the surgical ward. Nurses used food coloring to make his IV fluids red and green; fake consults were placed to all sorts of services. Take, for example, a radiology read requested for Santa in 1960:
By fluoro and film and every device
The lungs were congested with everything nice
Tinsel and ribbons and candy canes galore
Bespoke of the journey that lay afore
The heart was generous as noted last year
But full size not expected until the 25th near
An added exam was requested this year — a roentgen of that famous white beard
Low kV technique clearly reveals the bushy
Yule strands filling the field
For an impression was gained this year — as in many
His real diagnosis is good willy aplenty
The films are elusive, but the meaning is clear
The Yuletide message of kindness and cheer
And as part of general Christmas cheer, the residents would send fake specimens to the laboratory; an example is actually shown in a photo in the paper of an agar plate that the bacteriologist inoculated with SOMETHING that I’m a little scared to speculate, spelling out Ho Ho Ho and Merry Christmas Too; the request was for “Saturday Night Fever of gastrointestinal etiology.” The preliminary culture results said:
Santa, poor dear, has severe diarrhea
Caused by some unknown bacteria
We’re doing our best in this laboratory
To get Santa Out of his lavatory
And the final culture grew out: Yuletidia multijocularis, with a recommendation to try high dose clausicillin.
So back to salmonella. Prior to the pediatric outbreak, there had actually been an eighth patient who had been diagnosed with the disease, which Lang had largely overlooked. In December of 1965, just like for the past 15 years, the house officers and nurses on the surgical wards had assigned a bed to Santa, and one night they decided to sneak down some fake specimens to the lab, presumably with some absolutely absurd order, to see if the lab technicians would either play along with the joke and respond with something cheery, or get fooled. It was fun either way. They had mixed two fake samples — one of fake blood, and the other fake urine, to send to the microbiology lab. Whoever was processing the blood presumably let out a deep sigh, cursed the residents, and threw the sample out in the trash. But the urine was processed by a good natured technician on Christmas Eve, probably with some joking message sent back to the floor. It was a holiday weekend, so no one checked on the culture until Monday — but when the technician came back, the agar plate was growing salmonella, which would later be confirmed as salmonella cubana. Lawrence Kunz, the chief bacteriologist of MGH, was immediately called in. Kunz called the floor and tracked down one of the nurses who was in on the joke. She had helped make the specimen — it was made of water, intravenous multivitamins (a banana bag as they’re both affectionately and very weirdly called), a pinch of carmine dye for coloring, and a throat swab taken from another surgical patient, she said. Kunz immediately returned to the floor and recreated their concoction — banana bag, water, carmine dye. He sent this concoction down to the lab, as well as a second throat swab from the patient (which he expected to be negative in any event, since the patient had no gastrointestinal symptoms). The cultures of both were negative for salmonella cubana. Kunz then tracked down the original residents who had made the cuttures. It turns out the nurse had remembered wrong — the carmine dye had been used on the fake blood, which had immediately been thrown away; the urine sample had only contained the vitamins and water.
By this point it was February, and the first signs of the broader outbreak had become clear. Kunz’s first thought was the water — there had been a big salmonella outbreak in Riverside, CA the year before. But there was no evidence of a greater Boston outbreak, and while theoretically there could certainly be contamination inside of MGH, each building had a completely separate hook up to the municipal water supply. That left the vitamin preparation. Kunz remembered a paper he had read almost a year earlier on culturing salmonella from yeast. At the time, Vitamin B, which was a component of the banana bag, was produced from yeast. Could this possibly have been the source? Aware that the concentration would have to be extremely low (because after all, it hadn’t grown in the sample he had created), he passed 50 vials of the vitamin preparation through a micropore filter. But even this superconcentration of vitamins was negative
Kunz and Lang had reached dead end after dead end. The Massachusetts Department of Public Health had been contacted per protocol, and had done a full inspection of the hospital, finding nothing wrong. It seemed like the outbreak might have burnt out on its own, no matter where it was coming from. And that wouldn’t be without precedent — a source was never identified in the Riverside Salmonella outbreak either.
But just six days after they dared to hope the outbreak was over, salmonella roared back with a vengeance. Five more cases were identified in March, and four of those were clearly hospital acquired. The patients ranged from an infant, a teenager, and a 79 year-old man, and they were found in different buildings. They couldn’t be more different.
Lang was back at the drawing board. He thought that perhaps an asymptomatic carrier — again, directly recalling Mary Mallon — might be responsible — an x-ray tech, patient transporter, a consulting surgeon, any sort of hospital personnel who moved throughout the hospital. For months, Lang drew up detailed lists of these staff members to try to see if he could link anyone to the new cases, but again, there was no discernible pattern.
By chance, Lang read the monthly CDC Salmonella bulletin, which reported an outbreak in powdered milk. Powdered milk was commonly used as a nutritional additive in the hospital, and Lang excitedly looked at all the charts again — virtually every patient with nosocomial salmonella had received the milk. The manufacturer that provided MGH’s powdered milk was different than the one implicated in the CDC’s bulletin, but Kunz tested dozens of samples anyway. And again, all were negative.
Cases struck again in June and July, bringing the total number up to 21. It had now been a slow-moving outbreak of over a year, and they were still no closer to the answer. Lang and Kunz were about to ask for the help of the CDC in Atlanta. Running out of hope, Lang turned his attention to the newest patient — a four month old boy who was admitted with failure to thrive and who developed diarrhea during his stay in the hospital, which of course turned out to be salmonella. Fortunately, he had responded to treatment and was gaining weight. He examined the order book — fortunately, because he was a child with failure to thrive, the orders were relatively simple; the hospital was controlling everything that was going in and out of his body. He was getting vital signs every four hours; there was a dietary warning telling of a milk allergy, a diet order only for “strained foods and meats,” an order for D-xylose, which is a sugar used in gut transit studies, and a milk substitute called Nutramigen. There were only two medications: multivitamins and oral iron. All of these, of course, had been extensively investigated previously. Lang then turned the page to the tests. As part of his failure to thrive work up, the child had received a urinalysis, and four different stool tests: the first for fat and fibres, the second for trypsinenzyme activity, the third a routine stool culture (ordered likely as part of the unit-wide work up for salmonella that had turned up his case), and the fourth was a carmine transit-time test,
And it was at this point that Dr. Lang thought about Santa Claus. Carmine dye had been the ingredient that the surgical residents had used to color Santa’s blood, and which Kunz had erroneously thought was part of the urine culture that grew salmonella. Dr. Lang went immediately to the pharmacy. Where did carmine come from? The chief pharmacist looked at the purchasing records. The powder was bought in bulk from an aniline dye company in New Jersey, and then put into capsules at MGH. That seemed to be another dead end — aniline dyes, of course, were completely synthetic and would be extraordinarily unlikely to be contaminated with salmonella. Tiny aside — at some point I need to dig deeply into aniline dyes, since their discovery was largely responsible for the birth of the pharmaceutical industry. But an assistant pharmacist, armed with the Merck manual knew better. The dye may have been purchased from an aniline dye company, but it was certainly not synthetic — carmine dye came from cochineal.
This is not the first time I’ve stanned dyes on Bedside Rounds — and for good reason! So allow me to go off on a little tangent now. Cochineal is a tiny scaled insect that live in a parasitic relationship with cactuses throughout South and Central America. Their desiccated, ground-up bodies make a brilliant red dye — carmine. A lot of dried up bodies — 80 to 100 thousand have to be harvested in order to make a kilogram of carmine. Cochineal had been harvested by the Aztecs since at least the 10th century, and when the Spanish conquistadors invaded, became a prized export, second only to silver. The Spanish kept a strict monopoly, and intentionally spread disinformation that their dye came from a grain. It’s brilliant red quickly made it one of the most popular textile dyes; think of the brilliant “red coats” of the British officers. It was only when Leuwenhoek examined it under his microscope that he saw that it clearly came from an insect, and even after the secret was out, it remained the color of choice until synthetic alazarin was invented in the middle of the 19th century, destroying the cochineal market for textile dyes more or less overnight. But carmine found a second as a food and cosmetic coloring, as many of the red synthetic dyes proved to be toxic or carcinogenic.
Ground up insects from central America — that was enough for Lang. He took all the carmine from the floor and planted it on agar plates himself; they were plated on August 1st, and by August 3rd, salmonella had grown. After months of searching, this was the first real break in the case! But of course, at this point there were 21 cases total, and 14 of those were considered to be nosocomial infections. The breakthrough meant nothing unless the carmine dye could be linked to the other cases. Lang looked over their charts again — carmine had been prescribed in 10 of the previous cases. As for the remaining four, one of those was the nurse who had clearly caught it from her pediatric patients.
His next question — was the carmine becoming contaminated inside of MGH? Lang tested the sealed containers that were still in the pharmacy, and he found salmonella there too. The other hospitals in Boston used the same distributor, so why hadn’t they had similar nosocomial outbreaks? Dr. Kunz called the other bacteriologists around town and soon had the answer. Salmonella was grown on a culture medium, but the subtype was confirmed serologically. The other hospitals confirmed only the most common serologic subspecies; only MGH had the ability to distinguish cubana. And in fact, when the State Diagnostic Library performed surveillance cultures in other hospitals later in the month, they found cubana contamination.. The CDC and FDA’s investigation turned up cases of tainted carmine dye all over the United States — the same manufacturer had sold it to 28 food companies, 10 spice companies, 33 pharmaceutical companies, seven cosmetic companies, and 56 other companies, contaminating products as diverse as candy, chewing gum, preservatives, seasonings, meat, ice cream, and tomato extracts. Even worse — because carmine was a natural, rather than artificial, dye, it did NOT need to be listed on any labels — consumers would have had no idea. Ultimately, the contamination was linked to six more cases in California, one in Oregon, three in Ohio, and six more in England. Why didn’t it cause a wider outbreak despite so many potential sources? It’s probably a combination of the high dose given to medical patients, as well as the fact that salmonella cubana thrived with a weakened host, and weakened GI tract, which made sick hospitalized patients the ideal substrate.
Which still left the mysterious case of Santa’s salmonella. Kunz had attempted to recreate the Santa culture with carmine, but it was negative, whereas the original “positive” urine culture had presumably been inoculated when the same throat swab was used to mix both the urine and the blood. Had the carmine been positive then, the case would have been solved a whole six months earlier. Ultimately, Kunz had to accept that he had probably used another supply of carmine from before the contamination.
The CDC sent investigators to Peru to look into the cochineal industry, and actually sent Dr. Lang and Kunz to the Canary Islands as well. And while the Peru investigation did turn up salmonella newport and enteritidis — both common straints — no cubana was found on any of the insects. Ultimately, the investigation narrowed down the contamination to machinery and “slurries” in a single processing plant (called only Company A in the CDC report), and after performing appropriate decontamination, no more samples of contaminated carmine were produced. And in fact, CDC monitoring data showed that nationally, salmonella cubana levels returned to their pre-1965 lows after these modifications — suggesting that in fact, Santa had been responsible for helping to stop a nation-wide outbreak. And now to this day, the FDA requires all carmine to be pasteurized, and to be listed as an additive in food products — another legacy of Santa.
So I hope you enjoyed this story. Really, that’s it — there’s no deeper message about, say, epistemology, or the vagaries of nosology and diagnosis, or the ethical conundrums built into modern medicine. It’s just a nostalgic episode about a very influential writer in my life, and for this podcast. And go figure that stories about salmonella outbreaks would make me nostalgic.
So let me explain how I came about this story. I work for the hospitalist group at Beth Israel Deaconess Medical Center, and in pre-COVID times, we always ate lunch together. And sometime last year, my boss, Joe Li, told me the MGH Santa story. It sounded a little, well, too good to be true, but when I looked it up — I found a pleasant surprise. Well, two pleasant surprises, the first being that he, in fact, was correct. And second, while there were some dry pieces in NEJM by Lang and Kunz, there was also a fantastically colorful article in the New Yorker on the subject published in 1971 by none other than Berton Roueche. If you’re a medical professional of a certain age — or a nerdy podcast host a bit younger for that matter — Roueche needs no introduction. But for everyone else, Roueche was the storied New Yorker staff writer who chronicled medical mysteries in his Annals of Medicine column for the magazine, starting with his famous case of Eleven Blue Men. Eleven Blue Men is wonderful; it describes the mysterious case of eleven homeless men who had shown up with cyanosis at a local hospital, and which health authorities tracked down to the use of sodium nitrate instead of sodium chloride at a local cafeteria. He continued to be a staff writer right up until his death at the age of 83. Roueche is important in his own right — his started at a time of great change in public health, when the field transitioned from large-scale interventions to prevent mass outbreaks of infectious disease — think building sewage systems, clearing swamps, ensuring safe water — to dealing with outbreaks and clusters of infectious disease. It’s no coincidence that his popularity coincided with the founding of the both the Centers for Disease Control and the Epidemic Intelligence Service. He did much to popularize this new style of “disease detective” (and to ensure that fewer people threatened public health workers with carving forks) — his stories intentionally evoked classic mystery novels, especially Arthur Conan Doyle — even in the Santa Claus Culture, the piece this episode was based off of, Lang and Kunz come across like Sherlock Holmes and Watson.
But Roueche has also been very influential for ME personally. When I was a fourth year medical student at Tulane, I took a class called “Advanced Internal Medicine” with Chad Miller and Ben Rothwell, and one of our assignments was to read the book Eleven Blue Men, in which Roueche goes full Sherlock and rewrote a number of his classic pieces from the first person perspective of a detective novel. When I first read the book, I clearly wasn’t 100% with the purpose of the assignment, since I was like, “these are awesome stories, but it’s clearly all fiction.” There was no way that clinical epidemiology and medicine was actually that exciting. And that model has been incredibly influential to me. First for the podcast; as I’ve worked on crafting Bedside Rounds over the years, I’ve tried to create a similar “exploratory” narrative model where exploring quite academic — if important — subjects is full of wonder and mystery. And then as a doctor — this is probably more just me than Roueche, but I try to remind myself that being an internist really is living a detective novel.
So that’s it for the show! Honestly, I just wanted you all, dear listeners, to experience the joy of Berton Roueche, and a fun connection with Santa Claus this holiday season, no matter what holiday you celebrate. There’s only one Roueche book still in print — Medical Detectives; this story comes from Medical Detectives Part 2, which is very hard to get a hold of. I’ll have a link on Amazon if you want to buy it — I highly recommend it. And if you have ever find a copy of 11 Blue Men, grab it as quickly as you can.
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, and I think I’ve been pretty well-behaved lately.
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.