13
Peptic Ulcers: A Discovery in an Empty Lab
The greatest obstacle to knowledge is the illusion of knowledge.
—Daniel Boorstin, historian
Dr. Marshall and Dr. Warren
For some time, Dr. Robin Warren, a staff pathologist at the Royal Perth Hospital, had been observing some mysterious bacteria found near the site of stomach and duodenal ulcers in the biopsies of patients with these disorders. He had found the same organism in patients with chronic inflammation of the stomach lining (gastritis), but they were not found in patients with other diseases of the stomach. Warren was unsure of the significance of his findings, since they violated the orthodoxies of gastroenterology and pathology, which held that no bacteria could survive the acidity of the stomach, and that any bacteria seen in the stomach were the result postmortem change. Dr. Warren’s findings could not be taken seriously because of the preceding reasons.
Warren was not the first to see these bacteria, which had unusual shapes that were suggestive of a corkscrew. A search of the literature uncovered numerous other papers beginning in 1889 describing the presence of these bacteria, but none had generated any hypothesis as to their significance. Organized medicine often adopts a hostile attitude whenever its traditional teachings are attacked. This was certainly the case of Warren’s discovery. In 1940, Dr. A. Stone Friedberg published a paper showing these bacteria in fresh stomach specimens taken at operation. His professor discouraged any further work, saying, “Maybe, you made a mistake.” Had Friedberg been able to pursue this work, it might have led to a much earlier use of antibiotics as treatment for ulcers. Generations of medical students had been taught that an excess of stomach acid generated by stress, cigarette smoking, and dietary factors caused peptic ulcers.
In 1977, Dr. Barry Marshall, a well-trained gastroenterologist, joined the staff of the Royal Perth Hospital. Marshall was looking for a research project when he heard about Dr. Robin Warren’s observation of bacteria in the stomach. Marshall was interested, and he visited Warren to learn the details. In their discussions, Marshall heard about Warren’s inability to convince others of the validity of his many observations, including some in which the bacteria were found in fresh samples taken by biopsy of living patients.
Marshall was convinced that he and Warren should join in a project designed to study the occurrence of bacteria in patients with various disorders of the stomach. In the work, Marshall would send Warren biopsies from patients with ulcer disease or gastritis that Warren would evaluate as well as samples from other patients having other disorders of the stomach. Marshall would collect clinical information on these patients in order to see if there was any correlation between the findings on the biopsies and the symptoms expressed by the patients.
Over the next several years, Marshall obtained one hundred consecutive biopsies from patients with gastric complaints. The fresh biopsy material was sent to Warren for culture and microscopic examination. Most of the patients had symptoms of peptic ulcer or pain. In collecting clinical information, Marshall asked each biopsied patient to complete a detailed clinical protocol that listed every possible symptom that they might have had. The results were totally unexpected. The presence of bacteria did not correlate with any specific symptom, except perhaps for bad breath and burping. The biopsy reports showed that the corkscrew-shaped bacteria were present in every case of duodenal ulcers and had a significant association with gastric ulcers.
Koch’s Postulates
But the fundamental question remained: Do these organisms cause ulcers? Marshall set out to test this question. In doing so, he intended to follow the famous postulates laid down by the great microbiologist Robert Koch in 1890 in his studies of the bacteria that cause tuberculosis. The postulates spell out four requirements that must be met to establish that a given microbe is the cause of a given disease. Koch’s postulates are:
1. The microorganism must be found in abundance in all individuals suffering from the disease, but should not be found in healthy animals.
2. The microorganism must be isolated and cultured from a diseased individual.
3. The cultured microorganism should cause disease when introduced into a healthy, disease-free individual.
4. The microorganism must be isolated from the diseased experimental host and must be shown to be identical to the original specific causative agent.
Warren had already shown that the bacteria were present in large numbers in patients with gastritis or ulcer, and of equal importance, they were absent in patients who had other, unrelated diseases of the stomach. Marshall set out to test Koch’s second postulate, namely to grow the suspected organism in pure culture. He made many attempts to culture the organisms, but all failed. It was frustrating, since many clusters of the organisms could be easily seen under the microscope, but they were resistant to growth in many different culture media.
Easter Holiday
On April 8, 1982, the Thursday before the start of a five-day Easter holiday vacation, another attempt was made to culture the organism. Like all of the previously failed cultures, this one was treated in the same way that all of the other routine diagnostic cultures taken from other patients in the hospital had been handled, namely that they were incubated for forty-eight hours, removed from the incubator, and evaluated by a technician, whose report was sent back to the doctors and then discarded.
This routine was being followed on this particular weekend. The cultures that were planted on Thursday were to be read by the technicians on the following Saturday. But in their haste to get off on the holiday, they left Marshall’s cultures untouched in the incubator. When they returned on the following Tuesday, the cultures were examined. To their great surprise, these cultures showed luxuriant growth! It was easy to show that these organisms were identical to those seen on the biopsy specimens. Koch’s second postulate had been satisfied, thanks to the Easter vacation!
Marshall presented these findings at a 1983 Brussels conference that was chaired by Professor Martin B. Skirrow of the Gloucestershire Royal Hospital and a renowned authority. Skirrow was impressed with the work of young Dr. Marshall, who created a sensation when he declared that that people who were free of the corkscrew bacteria would be free of peptic ulcers. It was a brash statement considering that the hypothesis had not been critically tested.
Marshall and Warren sent a paper describing their joint work to the leading medical journal, the Lancet, in 1984. Although the editors were willing to publish it, they were unable to find any referees for the peer review, which is the requirement of the journal. Learning of these difficulties in getting the work published, Skirrow first repeated and confirmed the results in his own laboratory and then informed the editors of the Lancet of his results. Shortly afterward the paper was published.
In their Lancet paper, Marshall and Warren described their success in growing and isolating the unidentified bacillus from biopsies of patients with gastritis or ulcer. These bacteria were able to survive the extreme acidity of the stomach by penetrating the thick mucus lining of the stomach walls where they grew in the grooves between cells. Since the deeper mucus layers are slightly alkaline, the bacteria thrived in a neutral environment protected from gastric acid. The paper also contained a discussion of the inadequacy of acid-suppressing H2-receptor antagonists (the so-called group of H2-RA drugs) that were universally being used to prevent ulcer recurrences. Their paper also contained the hypothesis, which Dr. Marshall would go to great lengths to prove, that these bacteria could be the cause of chronic gastritis and peptic ulcer disease.
The Bold Self-Experiment
The next task was to infect a disease-free host to see if the disease could be reproduced. But here were no known animal models from which to choose. Marshall thought that piglets might be a trial animal model. He inoculated them with a culture of organisms, but none developed the pathological picture of the disease. After seriously pondering the idea of using himself as the experimental host, Marshall decided that he had no other choice. He would be the “guinea pig.”
On the morning of the experiment, Marshall skipped breakfast but took 400 milligrams of cimetidine (an H2-RA drug), believing that the organisms would have a better chance of surviving if his level of stomach acid was reduced. Two hours later, he gulped down a cloudy brown liquid that contained a billion Helicobacter pylori organisms. For the next three days, he had no symptoms, and he continued his usual activities. But on the third day after a small evening meal, the food felt like a lump of lead in his stomach. On the morning of each of the fifth to the eighth days, he awoke very nauseated and vomited a slimy liquid that seemed to contain no acid. He began to feel very fatigued and to sleep poorly. At that point, his wife told him that he had “a putrid breath.” His colleagues had also noticed that during the previous week, but they were too polite to tell him that he had foul-smelling breath.
After ten days, a biopsy of his stomach showed teeming clusters of Helicobacter, and at that point, Marshall decided to terminate the experiment. The experiment had succeeded. Helicobacter was a proven pathogen! The two biopsies taken on day ten were not enough to define the pathology completely, and he scheduled another biopsy four days later. By then, the vomiting had stopped, and he was symptom-free. The next biopsy, taken on the fourteenth day, showed a picture of healing gastritis. At that point, he began the treatment with an antibiotic, which was continued for a week.
No Helicobacter were seen in any of the eight subsequent samples. Cultures, histology, and electron micrographs were all bacteria-free. The Helicobacter had been eradicated without any additional treatment. Whatever happened to cause the organism to disappear continues to be a mystery. But Marshall had satisfied the conditions of Koch’s third postulate, and since the organisms recovered from his own biopsy during the experiment were identical with those seen in his patients, he had met the conditions of the fourth postulate as well. Marshall and Warren had decisively proven that Helicobacter pylori were the cause of gastritis and peptic ulcer disease. In other words, these disorders were really infectious diseases and were treatable with antibiotics.
Natural History of Helicobacter pylori
In pondering the results, Marshall was intrigued by his observations that the vomiting that he had experienced contained no acid. In searching the literature, he discovered some descriptions of gastritis in an account by Sir William Osler in his 1910 Textbook of Medicine, in which he describes a similar illness in children in which he had observed a lack of gastric acid. Suddenly the whole natural history of the infection became clear. The reason that ulcer patients could not recall an acute infection with Helicobacter was because it had most probably occurred when they were small children, afflicted with some transient illness that was associated with vomiting due to the infection.
Family members might infect these children or others, probably through the oral-intestinal contamination route. Once planted, the Helicobacter settled into a lifelong asymptomatic phase, which sometimes was punctuated by the appearance of clinical ulcer disease in adulthood. Because the bacteria were not permanently affected by any of the then customary ulcer therapies, ulcer disease became a lifelong problem, with a pattern of relapses.
Well-controlled double-blind studies from medical centers throughout the world have shown clearly that antibiotic treatment of ulcer patients for a two-week period results in eradication of the Helicobacter and healing of the ulcer. Furthermore, the rate of relapse, so common with all previous therapy programs, is nearly zero. One study included heavy cigarette smokers since they were thought to have added risk for ulcers presumably due to stress. But in fact, the level of cigarette smoking remained unchanged in the treated patients. Assessment of the mental status of the treated group showed that their sleep patterns, their sense of optimism, and their feeling of well-being all improved with the eradication of the bacteria. These findings suggest that the so-called “ulcer personality” may simply reflect a diminished state of health related to chronic infection of the stomach.
In 1984, Dutch scientists discovered that Helicobacter produce large amounts of the enzyme urease. This enzyme catalyzes the breakdown of urea into ammonia and bicarbonate. The latter probably protects the organism against the acid of the stomach, but in doing so it forms carbon dioxide, which is exhaled through the lungs:
Urea + Urease ==> Ammonia & Bicarbonate
Bicarbonate ==> Carbon Dioxide and Water
Marshall saw the urease reaction as a means for developing a simple, rapid diagnostic test for Helicobacter. This test consists of having the patient swallow a capsule containing radioactive-labeled C14 urea and ten minutes later blow up a two-liter balloon to catch the C14 carbon dioxide. The presence of C14, labeled carbon dioxide, in the breath indicates that Helicobacter is present. One modification has been to use the non-radioactive carbon isotope C13 labeled urea. The test has proven to be quite effective in confirming the eradication of Helicobacter after treatment.
Recognition was slow in coming, but it eventually came. Investigators in Australia, Europe, and the United States began to report confirmatory studies. In 1994, the National Institutes of Health convened a consensus panel of experts on “Helicobacter pylori in Peptic Ulcer Disease.” The publication of this meeting fully supported Marshall and Warren’s theory, although it noted that most people infected with H. pylori do not develop duodenal or gastric ulcers, implying that some other factors are at work. The consensus report recommended that patients with H. pylori ulcers be treated with antibiotics to eradicate the infection, as well as using acid-suppressing agents to relieve the symptoms during the treatment periods.
In 1995, Barry Marshall received the Albert Lasker Clinical Medical Research Award “for his visionary discovery that Helicobacter pylori cause peptic ulcer disease.” It went on to say “flying against the winds of orthodoxy, Dr. Marshall put the infection theory of ulcers on the map of medical discoveries. The gratitude of untold millions of ulcer patients is embodied in this Lasker Award.” The Lasker Award is one of the most prestigious awards for scientists conducting medical research. In many instances, it has been a reliable stepping-stone to the Nobel Prize.
That was ten years in coming, but it came in 2005, when Barry Marshall and Robin Warren jointly received the Nobel Prize in Physiology or Medicine. The citation recognized that they challenged the prevailing dogmas with tenacity and prepared minds. By using technologies generally available (fiber endoscopy, silver staining of histological sections, and culture techniques), they made an irrefutable case that the bacterium Helicobacter pylori caused peptic ulcer disease.
Comment
The results of the studies of Marshall and Warren revolutionized the diagnosis and treatment of peptic ulcer disease. Nearly all of the basic findings underpinning their success took place during the early 1980s, yet it took more than a decade before they were fully recognized and appreciated. Dr. Warren’s repeated observations of Helicobacter were made possible because of the invention of the flexible gastroscope that allowed gastroenterologists to easily take snippets of the stomach lining for microscopic study without undue discomfort of the patient. This was the first time that he could study “fresh” immediate samples from the lining of the stomach or duodenum taken from patients. Despite receiving no recognition of the significance of his work, Warren had persisted.
It was Marshall’s curiosity that brought the two together for a long-lasting partnership that had as its foundation the idea of bacteria causing peptic ulcer, which was at the time universally regarded as outrageous. Everyone knew that ulcer disease was caused by a specific “ulcer” lifestyle, the important components of which were stress, cigarette smoking, alcohol, drugs, and perhaps even some genetic factors. Furthermore, orthodox medical authorities “knew” that no bacteria could possibly live in the acid environment of the stomach.
Thanks to Marshall’s risky self-experiment, the role of Helicobacter in producing ulcer disease was firmly established. Well-controlled studies conducted by Marshall with the essential assistance of Warren were convincing in demonstrating the central role of these bacteria in the genesis of ulcer disease. Looking back, all of the work seems very straightforward. But basically it relied on several unpredictable events. There was the chance visit of Marshall to Warren’s laboratory in quest of a research project. It was another stroke of good fortune that Professor Skirrow took up work to confirm the results of Marshall and Warren and then to persuade the editors of the Lancet to publish their paper. Then there was the happy circumstance of a long Easter holiday that allowed the Helicobacter cultures to grow undisturbed for five days rather than being discarded after two days.
Given the convincing results of the work, what were the sources of opposition to its acceptance? Marshall mentions an economic motive on the part of the pharmaceutical industry. At the time, companies making so-called H2-RA drugs were enjoying sales of $3 billion per year and reaping handsome profits. If these drugs would vanish and be replaced by antibiotics, it might be a serious blow to their financial well-being.
Then there were the gastroenterologists, some of whom may have been receiving “research grants” from grateful drug companies. Furthermore, a significant source of their income was derived from fees for endoscopy procedures and from repeated office visits for adjustments of therapy and for the management of relapses. These income streams for their practices would be markedly reduced, given the simple urease test and the nearly zero rate of relapses. Finally, the gastroenterologist faced potential competition for ulcer patients from general practitioners, who were now armed with antibiotics and the C14 or C13 urease test and could now competently manage their patients with ulcer disease without endoscopy.
Armed with new diagnostic tests for Helicobacter, epidemiologists looking at populations on all continents began to see that the bacteria were so widely spread that that it may be the world’s most common chronic infection. But even more fascinating were the reports that a decrease in the prevalence of H. pylori was strongly associated with a declining incidence of gastric cancer, which in many parts of the world ranks as number two in terms of cancer deaths.
Cohen, S. “Helicobacter Pylori.” The Merck Manual of Diagnosis and Therapy, 18th ed. Hoboken: John Wiley & Sons, 2006.
Marshal, B. J. “Helicobacter Connections.” Nobel Lecture, Nobel Prize in Physiology or Medicine, 2005, 250–277.
Marshall, B. J. et al. Helicobacter Pylori in Peptic Ulceration and Gastritis. Cambridge, MA: Blackwell Science, 1991.
Warren, J. R. “Helicobacter—The Ease and Difficulty of a New Discovery.” Nobel Lecture, Nobel Prize in Physiology or Medicine, 2005, 292–305.