Mar 29 2010
One of the themes of SBM has been, since the very beginning, how the paradigm of evidence-based medicine discounts plausibility (or, perhaps more appropriately, implausibility) when evaluating whether or not a given therapy works. One of our favorite examples is homeopathy, a therapy that is so implausible on a strictly scientific basis that, for it to work, huge swaths of well-established science supported by equally huge amounts of experimental and observational evidence would have to be found to be all in serious error. While such an occurrence is not per se impossible, it is incredibly unlikely. Moreover, for scientists actually to start to doubt our understanding of chemistry, biochemistry, pharmacology, and physics to the point of thinking that our understanding of them is in such serious error that homeopathy is a valid description of reality, it would take a lot more than a bunch of low-quality or equivocal studies that show no effect due to homeopathy detectably greater than placebo.
On Friday, Kim Atwood undertook an excellent discussion of this very issue. What really caught my attention, though, was how he educated me about a bit of medical history of which I had been completely unaware. Specifically, Kim discussed the strange case of John Lykoudis, a physician in Greece who may have discovered the etiology of peptic ulcer disease (PUD) due to H. pylori more than a quarter century before Barry Marshall and Robin Warren discovered the bacterial etiology of PUD in 1984. One reason that this story intrigued me is the same reason that it intrigued Kimball. Lykoudis’ story very much resembles that of many quacks, in particular Nicholas Gonzalez, in that he claimed results far better than what medicine could produce at the time, fought relentlessly to try to prove his ideas to the medical authorities in Greece at the time, and ultimately failed to do so. Despite his failure, however, he had a very large and loyal following of patients who fervently believed in his methods. The twist on a familiar story, however, is that Lykoudis may very well have been right and have discovered a real, effective treatment long before his time.
Kimball is right to point out that discoveries require context. Medicine can be prone to dogma. Of that, there is no doubt. Sometimes, physicians resist challenges to prevailing medical dogma. Of that, too, there is no doubt. However, reading the story of John Lykoudis, I couldn’t help but wonder the full context of his activities and efforts to convince the Greek medical authorities of the 1950s and 1960s that he was on to something. It also allows me to indulge myself in a bit of my surgical training, the bulk of which, ironically enough, occurred during the time period when the discoveries of Warren and Marshall were first revolutionizing the treatment of PUD and becoming increasingly accepted in the 1990s. When evaluating a story like that of Dr. Lykoudis and why he was unable to convince the medical profession of 50 years ago that his findings had merit, it’s very important to understand (1) the prevailing notion at the time of PUD etiology and, even more importantly, (2) how PUD was diagnosed and treated in the 1950s and 1960s.
Throughout most of the 20th century, PUD was thought to be caused by stress and dietary factors. The primary treatment for disease that had not yet developed complications was bed rest and prescription of bland diets. Towards the mid-20th century it became increasingly appreciated that gastric acid is a major factor in the etiology of PUD, and increasingly antacids, then later H2 receptor antagonists like cimetidine and ranitidine, and then still later proton pump inhibitors like omeprazole. Cimetidine and ranitidine were the mainstays of PUD treatment when I was a medical student and a resident. For complicated PUD that either was not adequately managed by drugs, sometimes surgery would be necessary. Endless were the discussions and arguments among surgeons fo the time what operation was best for PUD management, a Bilroth I or Bilroth II, vagotomy and pyloroplasty, and, at the time I was finishing my residency, the highly selective vagotomy. Then, of course, there were the discussions of what to do in the case of a bleeding ulcer that couldn’t be stopped using endoscopy or what operation to perform in the case of a perforated ulcer.
One must also remember that in the 1950s and 1960s, the diagnosis of PUD was much different than it was even in Marshall and Warren’t time in the 1980s. That’s because in the 1970s and 1980s fiberoptic endoscopy became the dominant method of diagnosing PUD. Not only did it allow for direct visualization of the ulcer, but it also allowed for potential therapy without laparotomy. More relevant for purposes of this discussion, endoscopy allows isolation of the H. pylori that causes PUD. Prior to that, the methods of diagnosis were not as accurate. In general, PUD would be diagnosed by history, physical, and then an upper GI X-ray series, in which the patient would swallow barium-containing slurries, after which fluoroscopy would visualize the stomach and duodenum.
Consequently, if you imagine yourself back in the 1950s and 1960s, the time when Dr. Lykoudis practiced, the diagnosis of PUD was less certain than it was in the 1980s. Unless it was severe, it might not even show up on upper GI series, given the limitations of the imaging technology of the time. It wasn’t always possible to distinguish between gastritis and true PUD. Consequently, a lot of diagnosis and treatment of PUD in pre-endoscopy times ended up being a lot more empiric than it is now. The diagnosis was arrived at clinically, and it was symptomatic relief that was used as the main measure of treatment success.
So what would it have taken to demonstrate, given the technology of the time, that Dr. Lykoudis’ treatment was effective? It would have taken pretty much what it would take today, minus the use of endoscopy. Consequently, we could envision a clinical trial in which standard of care at the time (bland diet, antacids, etc.) was tested against the standard of care plus Dr. Lykoudis’ antibiotic regimen. The design would be simple: Two groups, one receiving standard of care plus placebo, one receiving standard of care plus antibiotics. Alternatively, although it probably wouldn’t be considered ethically acceptable now, back in the 1960s, it probably would have been considered ethical to do an alternative study, directly comparing the then standard of care to Dr. Lykoudis’ antibiotic regimen. Whichever design were chosen, before entry in the study, each patient would have to undergo upper GI to document the presence of PUD and measure its severity. They would also have to undergo upper GI at the end of the study to document healing. Throughout the study, pain scores would be measured, and patient outcomes tracked to see what proportion of patients end up requiring surgery for their PUD and which proportion can go back to eating a regular diet. Unfortunately, Dr. Lykoudis didn’t have data anywhere near that level of rigor.
On the other hand, what I’m describing above is the equivalent of what we would call today a phase III clinical trial. In general, phase III trials aren’t started without preliminary data. But what sort of preliminary data are usually needed to provide adequate scientific justification for a phase III trial? Today, we require preclinical evidence in the form of cell culture and animal studies, as well as lesser levels of clinical evidence, such as smaller trials like phase I or phase II trials. The preclinical data provide scientific justification and plausibility, and the human data bolster that. Of course, we can’t apply today’s standards to the science of 50 years ago. Back then, the methodology and ethics of randomized clinical trials were not as advanced and well-worked out as they are today.
So what happens if we look at Dr. Lykoudis’s story and whether he might have had enough data to justify a large phase III-like clinical trial in the time in which he was practicing? By today’s standards, he probably did not. As Kimball pointed out, although plausibility does not mean understanding the mechanism, it does mean that there should be a potentially plausible mechanism, as there does not exist, for example, for homeopathy. The concept that there might be a bacterial cause of PUD is not, on the surface, incredibly implausible. But was it plausible enough to justify a clinical trial of the sort that Dr. Lykoudis proposed? What would it take? This is what Dr. Lykoudis proposed:
In 1967, Lykoudis succeeded in getting the attention of the Prime Minister’s office. His correspondence with the Minister of Health on 21 August, 1967, a sad document indeed, is revealing. He registers his frustration that medications with apparently no effect on PUD were approved, whereas Elgaco was repeatedly rejected. He proposes, in essence, a phase III trial: 100 PUD patients to be treated at a State hospital by the eminent professors, 50 with conventional treatment and 50 with Elgaco. ‘Their refusal to approve it is understandable, but their refusal to test it is not!’ he writes.
A not unreasonable assertion. Unfortunately, as Kimball pointed out, the technology to isolate and culture H. pylori didn’t exist at the time. Absent that, providing strong evidence for a bacterial etiology for PUD would require an obvious and strong response to antibiotics in the form of unambiguous symptom relief and healing documented on followup upper GI imaging studies. Even if Lykoudis had had that, he would have been unable to culture the organism responsible for PUD, which would have left scientists in a quandary. Response to antibiotics is storngly suggestive of an infectious etiology, but, absent an organism, one can never determine for sure whether it is in fact a bacteria causing a disease or the antibiotic has an activity other than its ability to kill bugs, an additional activity. For example, erythromycin increases GI motility. Thus, in the context of the time, it’s not surprising that Lykoudis’ ideas were considered highly implausible, and it would have required very strong evidence to make the idea seem plausible.
Steve Novella made a cogent observation:
But taken at face value, I think the real lesson is that process is more important than whether or not one turn’s out to be correct. Science is about process.
The problem with Lykoudis is that his behavior was indistinguishable from the myriad quacks and charlatans that existed then and exist today. That in hind sight one turned out to be on the right track is not all that surprising, and their contemporaries should not be faulted for their inability to predict the future.
The question is – what did Lykoudis do to convince the scientific community of his claims. Did he perform carefully controlled double-blind placebo-controlled trials? Did he attempt to enlist the help of a microbiologist to try to isolate the organism? Or did he just expect people to take his word for it?
What did he do to deserve being taken seriously? Being right in the hindsight of history is not enough.
And I think that this is the key point. Science is a process. It is by its very nature constrained by what is known at any given time in history. In the context of Dr. Lykoudis’s time and given what was known then, it is not surprising that his idea would have encountered heavy resistance from the scientific orthodoxy of the time. Another issue to consider is regional variation in physicians’ attitudes. One example I like to use to illustrate this is the reaction of European physicians to the ideas of Ignaz Semmelweis. Semmelweis, as most readers will recall, first demonstrated that the high rate of puerperal fever in the obstetrics ward run by physicians was due to physicians not washing their hands after doing autopsies, going straight from the morgue to the delivery room. Semmelweis’ findings were far more favorably received in England, for example, than they were on the continent. It is possible, although by no means assured, that Lykoudis’ ideas might have been better received if he had lived in a different part of the world.
Here’s one final consideration. Science is performed by human beings. Although it is a process designed to overcome human biases, communicating the results of scientific research is subject to the same idiosyncracies to which any human communication is subject. Anyone who’s ever been to a scientific conference knows that. It is quite possible to be right and, to put it bluntly, to piss off the very scientists that need to be convinced so much that they harden their positions protecting the scientific consensus. One example is Semmelweis himself. Sherwin Nuland, in his book The Doctors’ Plague: Germs, Childbed Fever and the Strange Story of Ignac Semmelweis, suggested that if Semmelweis had communicated his findings more effectively and managed not to antagonize the medical establishment so thoroughly he might not have been marginalized and dismissed in the manner that he was. At the very least, he may not have been met with as much hostility. Galileo, although apparently not intentionally, alienated the Pope at the time by putting his words in the mouth of a character named Simplicio defending the Aristotelian Geocentric view in Dialogue Concerning the Two Chief World Systems. Reading between the lines, one can see echoes of this sort of antagonism in Lykoudis’s story. Indeed, Michael Phillips of St. Vincent’s Medical Center wrote in an letter to The Lancet:
I propose a less dramatic metaphor: medicine is a marketplace of ideas, with sellers and buyers. Sellers (innovators with new ideas) advertise their intellectual property to potential buyers (other physicians). This buying public is highly sophisticated and sceptical. Quite correctly, physicians will only accept the highest quality new ideas because the lives of their patients are at risk. So physicians buy only when they see the hallmark of quality: publication in a respected peer-reviewed journal.
That in a nutshell explains the tragedy of John Lykoudis. He had a wonderful intellectual product: the insight that peptic ulcer disease is infectious, supported by the evidence that it can be cured with antibiotics. But he lacked the ability (or the training) to sell this insight to his colleagues. He was a retail trader who treated individual patients instead of targeting the wholesale market of other physicians.
Of course, this is the problem. Most physicians are not scientists, and many physicians are very prone to being swayed by anecdotal evidence. Without well-designed clinical trials based on the best basic and translational science available, way to validate or refute anecdotal data. I find Lykoudis’s story to be a cautionary tale. Whether he was correct and thus the true “Galileo” of H. pylori, rather than Warren and Marshall or whether he was just another crank, his story demonstrates that we scientists should be very careful to guard against excessive smugness. As has been repeated by many skeptics in many variants over the years, it is not sufficient to claim the mantle of Galileo as a persecuted martyr for science. You must also be right. Even though it is not clear whether, taken in the context of the time, Lykoudis was a crank or a misunderstood physician who was ahead of his time, Warren and Marshall’s vindication of his ideas that PUD is bacterial in etiology reminds us that not all who claim the mantle of Galileo are necessarily cranks. The vast majority usually are, but on very rare occasions we do see a real Galileo.
10 Responses to “The case of John Lykoudis and peptic ulcer disease revisited: Crank or visionary?”