Aminorex

Alexander M Walker, MD, DrPH

Frau Z focused on each breath. It was February 1967, in the cardiology waiting room of the University Clinic in Bern, Switzerland. Frau Z was a 35 year old mother of five who could not do even small chores without gasping for air. Her fingers were thick and dusky and her ankles were swollen. All this had come on in a matter of months. (1)

Frau Z’s illness, as it would turn out, was decidedly out of the ordinary, and its cause was widespread. She was the harbinger of an epidemic that has run its course. Fortunately, the physicians and epidemiologists who were in the midst of the crisis recounted it over many years, so that it is still possible to reconstruct who discovered what and why, and to learn about decisions very much like ones we face today.

Hans Peter Gurtner, the clinic’s chief cardiologist, directed Frau Z’s work-up, an evaluation that would determine the course of his own professional life. He began with a physical examination and questions to his new patient. An electrocardiogram followed, as did a chest x-ray, and cardiac catheterization.

First, the physical exam: seen through living tissue, blood without oxygen is blue. Mixtures of tissue and unoxygenated blood run the spectrum from the robin’s egg color of a vein close to the surface, to azure, to slate gray, to blue black.. The medical term "cyanosis" describes collectively all the unhealthy shades manifest by tissue that darkens for lack of for oxygen. Frau Z had moderate cyanosis. Together with her shortness of breath, the dusky color of her fingers told Gurtner that her lungs were not getting enough air in contact with enough blood.

Swelling in the ankles can come from too much fluid in the body, weak liver function, malnutrition, or heart failure. Frau Z was not overhydrated or undernourished ( in fact, she was considerably overweight). She had no history of hepatitis or alcohol abuse, the most common causes of liver disease. If the swelling were from heart failure, it would have to be that her right ventricle, the chamber of the heart that pumps blood into the lungs, was failing. When the right ventricle goes, the circulation was literally backs up.

Through their stethoscopes, cardiologists heard a loud dub in the lub-dub, lub-dub of Frau Z’s heart sounds, particularly just up from her left breast The point overlies the heart valve that keeps blood from flowing backwards from the lungs each time the heart relaxes. When the leaflets of the valve snap together hard, there is an emphatic dub, and it is a sign of elevated pressure in the pulmonary artery. To designate this hypertension in the arteries that lead to the lungs, physicians use the term "pulmonary hypertension."

"Listen to your patient," admonished William Osler, the great American physician of the nineteenth century. "He is telling you the diagnosis." Dr. Gurtner and his colleagues interrogated Frau Z about predisposing conditions. There were the accepted causes of pulmonary hypertension: heart failure itself, congenital or acquired abnormalities of the heart, and various forms of pulmonary embolus, a clot formed elsewhere that breaks loose in the circulation and travels back to the fine arteries of the lung. The common causes of pulmonary embolus are surgery in the lower body, inflammation of the veins, and pregnancy. Frau Z had none of these when her symptoms began.

An electrocardiogram, or ECG, was needed because the most common cause of pulmonary hypertension is the failure of a damaged left ventricle to pump blood away from the lung. In such cases, the back-up causes blood to pool in the lung, and the pulmonary hypertension comes from the right ventricle pushing blood in where little can get out. Frau Z’s ECG gave no evidence of damage to the muscle of the left ventricle. Instead, it revealed stronger signals than normal over the right heart. The muscle had hypertrophied. It had "bulked up," almost surely for the same reason that a weight lifter does: overwork. This was the same part of the heart whose putative failure had caused the swelling in Frau Z’s ankles.

The chest x-ray showed no lung disease. There was an enlarged heart. The vessels carrying blood from the right heart into the lung were distended.

Cardiac catheterization in 1967 involved snaking a flexible tube, or catheter, from the large vein in the crook of a patient’s arm up into the vena cava, which is the great vein that returns blood to the heart. You can also start from the largest veins in the neck or in the groin, and work your way down or up to the heart, as need be. All the while x-ray images tell the cardiologist exactly where the catheter lies. When it reaches the heart the cardiologist bends the tip of the catheter, and passes it in sequence into the right atrium (the antechamber that feeds blood into the right ventricle), the right ventricle itself, and finally into the pulmonary artery. An attached pressure gauge registers the intensity and temporal profile of the blood’s pressure at the catheter tip.

Frau Z’s pulmonary artery pressure hit 69 millimeters of mercury with the ventricle contracting. The pressure would have been strong enough to lift a column of liquid mercury about two and a half inches up a glass tube. Normal is less than 25 millimeters, roughly an inch.

Frau Z had pulmonary hypertension. What the heart sounds suggested the pressure gauge now proved. Investigation of usual causes was fruitless. There was no heart disease that might cause the hypertension, there was no destruction of lung tissue; there was no backflow of blood from the left side of the heart though a congenital defect. Frau Z’s case was one of "primary" disease. Her lungs’ blood vessels were resisting the flow of blood, her right ventricle was overworked and decompensating, and her body was starved for oxygen.

Ayerza’s Black Hearts

Alfred Fishman is the Measey Professor of Medicine at the University of Pennsylvania. When I first spoke to him he told me of his early correspondence with Dr. Gurtner. Fishman himself has recounted the early history of primary pulmonary hypertension, which I paraphrase below. (Fishman 1997 - 2)

The nineteenth century witnessed an escalation of medical technique that parallels in some ways the technological explosion of our own time. Dominance in the earlier wave came about through increasing scrutiny of the life course, symptoms, and signs of patients, and through comparing these to the morbid anatomy, which was discoverable at death. The observations were the springboards for speculation, hypothesis, and the search for confirmation. For academic medicine, which was international and polyglot, it was a war of acuity, interrogation, integration, and exposition. The veterans of those struggles are long dead, and it is hard to read their journals without a sense of loss.

The first description of primary pulmonary hypertension captured a deranged anatomy rather than errors of function. In 1865 Klob (3) reported the autopsy findings of a 59 year old man who had died after developing swelling in his ankles. To his surprise Klob found an impressive narrowing of the finer branches of the pulmonary artery with a localized arteriosclerosis, instead of the usual cardiac pathology. In 1891 Ernst Romberg, a physician in Leipzig, took Klob’s curious report a step further with the clinical description of a 24 year old man whose eventual autopsy yielded findings identical to those reported by Klob.(4) Apart from an undiagnosed "muscular rheumatism" 18 months before, Romberg’s patient had led an unremarkable life. Starting shortly after the bout of muscle pains, he experienced a progressive shortness of breath, "and his formerly healthy countenance became bluish. Indeed after physical exertion, the patient often looked ‘electric blue.’" He came to Romberg already in this advanced state. The young man took to bed, and his color darkened to an "enormous cyanosis". Digitalis, one of the few drugs of that time still in use today, did not help. "The respiratory frequency doubled. The facial expression became anxiety-ridden." He died within weeks. Autopsy showed "extraordinarily widespread, high grade sclerosis of the pulmonary arteries with consequent hypertrophy of the right half of the heart." The right ventricle was so oversized that the normally larger left ventricle looked as if it were a tiny appendage. There were no other defects.

The next important advance was a thoroughgoing description based on eleven cases, by an Argentinean physician, F.C. Arrillaga. (5) "The first sign that attracts the attention in examining the patient is cyanosis; it is very noticeable on the head and the extremities; on the face everything is cyanosed: not just the lips, the nose, the ears, but also the tongue, which is the color of wine still, and all the oral mucous membranes, the surface of the tonsils, the epiglottis, the palatine arch, the posterior pharynx; just the same all the tissue of the face, the eyelids, and the eyes." Arrillaga credited one of his professors, Abel Ayerza, for kindling his interest in these cases. In 1901 the older man had called his colleagues’ attention to a patient with cyanosis and shortness of breath who died of failure of the right ventricle. Because of his dark hue and his cardiac death, Ayerza had called this a "black heart" patient.

The general syndrome of cyanosis associated with heart failure in the absence of heart disease, was being publicized at the time by William Osler. (6) Osler-Vasquez disease, as it was then called, also included an increase in blood hemoglobin and swelling of the liver and spleen. It was really a syndrome, a complex of patient symptoms and doctor-elicited signs. The cyanosis stemmed from too little oxygen in the blood. The bone marrow compensated by increasing the number of red blood cells to augment the oxygen-carrying capacity of the blood. The enlargement of liver and spleen followed on the progressive inability of the heart to pump venous blood back into the lungs.

Arrillaga suggested that the cause of Ayerza’s disease was infection that had spread to the pulmonary artery. Syphilis was chief among these, he thought, though he mentioned malaria as well. The straightforward basis for this claim was that the patients he saw were typically infected with syphilis.

An American venerologist, Aldred S. Warthin upped the descriptive ante in 1919 with the account of 605 day-long hospitalization of a laborer in Ann Arbor. The clinical evaluation, which left the man fit only to become a popcorn salesman, led to a diagnosis of Ayerza’s disease. (6) When the patient had finally succumbed, Warthin’s meticulous description of the autopsy ran to another eight pages. He advanced the thesis that there were pulmonary lesions indicative of syphilitic effects, though the disease was only really demonstrable elsewhere in the body. In a literate discussion that added more heft, Warthin distinguished Ayerza’s disease from the Osler-Vasquez syndrome. Ayerza’s disease, Warthin argued, was specifically the conjunction of the syndrome with a syphilitic infestation of the pulmonary artery. Several years later Arrillaga noted that authorities were coming down on both sides of the issue of syphilis as the cause of the black heart syndrome, but he was able add his own observation of another seven cases. (Arrillaga 1924 -7) The syphilis in these patients supported his earlier speculation. Possibly to account for the cases in which syphilis had not been directly observed, he added a novel twist, that the syphilitic lesion might be attracted to the artery by earlier damage.

As years progressed critics pointed out Arrillaga’s contorted reasoning, as he tried to fit the clinical evidence into a an ill-fitting box.(8) In the end it fell to Oscar Brenner, a pathologist from Birmingham to bring forth a sufficient number of contrary examples to discredit the hypothesis altogether. In a two hundred page, five part review, Brenner summarized all the published cases to date, and added a hundred from his own experience. (9) Syphilis, which had been identified in only one of the hundred cases he knew personally, was not part of the picture. Brenner sniffed at "the undue tendency ... of French and South American authors to exaggerate the frequency of syphilis." The term Ayerza’s disease was to persist in pockets of academic medicine for another 40 years, but Arrillaga’s theory of causation was abandoned.(10)

Brenner proposed no alternative etiology for primary pulmonary hypertension, nor did his successors. Research in the middle of the 20th century turned instead to elucidating the mechanics of the disease, and this depended on the emergence of a technique for clinicians to measure blood pressures inside the beating heart.

Cardiac catheterization was devised in 1929 by a German surgeon, Werner Forssmann. He had the idea of delivering drugs directly to the heart. Forssmann, who was still in training, asked a colleague insert a large bore needle into a vein in the crook of his right arm. Forssmann himself then threaded a heavily oiled urinary catheter through the needle. He had coaxed about 13 inches of the catheter into his arm before his collaborator insisted that they stop. Undeterred, Forssmann went by himself to the operating room the following week. He did not think that he could manage the big needle by himself, so instead he anesthetized the insertion spot, and cut open the skin over the vein. After that it was a simple matter to nick the vein and slip in another catheter. This he was able to run in its whole length of two feet. From the length of the catheter and the sensations in his chest, Forssmann guessed that he had reached his heart. To be sure, he made a "trip from operating room to x-ray department, which is an unusually long distance in our establishment and during which I had to ascend stairs, covering the distance on foot with the probe lying in my heart ... The position of the catheter was proven by x-ray and indeed I observed the tip of the catheter itself in a mirror held in front of the illuminated screen by a nurse." (11)

Forssmann’s discovery lay fallow for some time. In the rigid hierarchy of German academic medicine, he was too young to be taken seriously. His more measured contemporaries may also have been off by his enthusiasm. Forssmann himself went on to be trained in urology, and to marry a urologist. After a wartime detour as a prisoner on the Western Front, he returned to practice with his wife.

The next significant development came in 1945, the year of Forssmann’s release from an Allied camp. André Frédéric Cournand, a French physiologist who had taken up residence in New York in 1931, brought in the idea of a double bore catheter. Cournand set the outlet holes of the two joined tubes a few inches apart, so that pressures could be simultaneously measured at different points in the circulation. With this device Cournand obtained the first known readings of pulmonary artery pressures in man. Forssmann and Cournand, together with Cournand’s mentor at Columbia, Dickinson Richards, shared a Nobel Prize in 1956 for the development of cardiac catheterization. (11, pp. 330-336)

With the advent of cardiac catheterization, it became evident that pulmonary arteriosclerosis was not a necessary component of pulmonary hypertension. It was instead a late stage consequence, in the same way that atherosclerosis is a result of chronic hypertension elsewhere in the circulatory system. The primary disorder was hypertrophy and spasm of the muscled wall of the small arteries in the lung. The constriction could be reversed with intravenous injections of muscle relaxants, but the treatment proved more instructive for physiologists than it was helpful to patients. The hypertrophy and failure of the heart’s right ventricle, just as Romberg had concluded half a century before, were the consequence of an intolerable work load posed by pumping blood into an increasingly resistant lung.

More recent studies have identified prostacycline as a drug that can be given over long periods to treat primary pulmonary hypertension. The procedure involves a portable intravenous infusion pump, and is hardly ideal. When Hans Peter Gurtner faced Frau Z in 1967, however, the only treatment was one informed by the most tenuous clinical logic. Pulmonary hypertension is often the result of traveling blot clots. Clots can be prevented by blood thinning agents, though only at the risk of unwanted and often dangerous internal bleeding. Perhaps, clinicians reasoned, inapparrant clots played a role in "primary" pulmonary hypertension and anticoagulants would help. Besides, there was no other therapy. In fact, there was no effective therapy at all.

The Epidemic

Primary pulmonary hypertension is rare. The last patient preceding Frau Z at the Bern Circulation Laboratory had been there just over a year earlier, in 1965. In the decade since catheterization had become routine, the clinic had typically uncovered a case every year or two.

Dr. Gurtner’s team did not have to wait long for the next patient, who came in April. There were two in July, two in August, and one in September. The eighth patient of 1967 was diagnosed in October, and one of the August patients died of pulmonary hypertension that same month. Two more appeared at the clinic during December, and in January 1968 there were three new cases, and another death. Most of the patients were women.

With the dramatic increase, the doctors began to ask why. Patients were quizzed as to whether they had come into contact with any new substance at home, in the workplace, on their daily travels. (12) "Since women were predominantly affected, we thought at first that some daily household product, a fabric component, or a cosmetic could be responsible," reported M. Gertsch, a colleague of Dr. Gurtner’s. He and the others put together a list of every common exposure that they could imagine: "cooking fumes, cleaning and washing products, furniture polish, floor wax, Air-Fresh, plant sprays, various cosmetics, common medications, and all manner of exogenous factors: pets, parasites, allergens, eating utensils, hobbies, and so forth." (13)

The clinical presentation and the pathologic picture of primary pulmonary hypertension were very much the same during the epidemic period of 1967 and 1968 as they had been before. However, taken as a group, the patients who had been seen since 1967 did differ from earlier patients with primary pulmonary hypertension in several ways, according to Dr. Gurtner: "The progression of their disease was more rapid, a history of pulmonary embolism was much less common, more than half of the newly diagnosed patients were at least a little overweight, and around 80% reported having taken an appetite suppressant before the first appearance of their symptoms." (12)

Prior to 1967 patients with primary pulmonary hypertension had been on the skinny side. Among the eight seen from 1956 through 1966 at the Bern clinic, the mean was eight percent under Swiss norms for weight. In 1967 the average patient was twelve percent over. Many of the patients were very heavy, and of those who were not, a number had only just reduced.

Frau Z’s story was illustrative. At 153 centimeters (five feet one-half inch) tall, she weighed 89 kilograms (196 pounds) . She had taken amphetamine-based diet pills years before. More recently, from mid-March to mid-June of 1966, she had tried a newly introduced regimen called Menocil and had lost 15 kilograms, more than 30 pounds. The weight had unfortunately come back with her growing incapacity. (1)

There was one underweight patient in 1967. A 54 year old woman who had never been heavy was seen for increasing shortness of breath in October, the month of the first death. In the spring of 1966 she had taken a tablet of Menocil each day for three or four months. In April of 1967 her symptoms began. She weighed 61 kilos (134 pounds) at the time, which was normal for her height and age. She embarked on another course of Menocil, in preparation perhaps for the summer, and lost five kilos. (1)

"There was only one overriding, common factor" Dr. Gertsch concluded. "Among the 23 patients whom we investigated from beginning of 1967 to the middle of June 1968, 13 reported that they had undergone a weight reduction program using aminorex fumarate (Menocil®). This observation led us to inquire about Menocil with the private physicians, pharmacists, and health insurance agencies of those who had denied Menocil consumption. In addition, we looked into temporal correlations between the onset of symptoms (usually shortness of breath with exertion) and the first consumption of this appetite suppressant." The interval from first use of Menocil to first symptom tended to be on the order of six months, though in individual cases it ranged from three weeks to two years. Even though the time from symptoms to first diagnosis was accelerated in Menocil patients compared to others with primary pulmonary hypertension, the typical interval was sufficiently long, about eleven months (14) that the search for temporal correlations was difficult. In the end, though, among 67 patients seen from 1967 through mid-February 1970, there would be 60 women, 50 of whom had used aminorex, and 7 men, 3 of whom had used the drug. Eleven died of their disease within a few years. (12)

Amphetamines, "speed," were also part of the picture. In addition to Frau Z, there were two patients had used amphetamine derivatives prior to trying Menocil, without apparent adverse effects at the time. One patient had used amphetamines exclusively, several years before the onset of any problems. (12)

On May 10, 1968, Dr. Gurtner presented the results of the Bern team’s investigations to at the annual meeting of the Swiss Society for Internal Medicine in the resort town of Chur. (1) Whereas the cardiology service had seen eight patients with primary pulmonary hypertension from 1955 through 1966, they had seen 12 cases in 1967, and already another 11 in 1968. The microscopic changes associated with the Swiss cases were similar to what had been seen previously in primary pulmonary hypertension. Nearly half of the muscle walls of the arteries showed internal fibrosis, or scarring. Something was destroying the muscle.

Science and Advocacy

From the first report of the Bern group, it appears that Dr. Gurtner and his colleagues were more comfortable with clinical evaluation and treatment than they were with etiology, or causation. Of the 23 pages of their initial monograph, only three were given over to the question of why there had been a 20 fold increase in the number of cases of primary pulmonary hypertension coming into their clinic. Though they had worked hard, and had pinpointed the cause, the final evaluation was amazingly circumspect. "It remains to be seen," they concluded, "whether this correlation is merely chronological or whether a causal connection can be detected."

Few physicians, then or now, have any training in the methods of drawing general conclusions about the origin of diseases from simple observations. There is no limit to the further documentation one can ask for in observational studies, and observational research almost never comes up to the mark of evoking general acceptance, as often happens with true medical experiments, even inferior ones. Dr. Gurtner and his colleagues were certainly aware that Arrillaga’s case series of syphilitics had mislead a generation of cardiologists. Arrillaga’s error was based in observation of a dozen cases, but the refutation had required a life’s work of Brenner’s clinical data.

These were the doubts that the Bern group harbored, as Dr. Gurtner explained it. (12)

The university team might have been fooled by changes in referral patterns. Against this idea, when they looked at patients whose pulmonary hypertension was due to congenital heart disease or acquired cardiac defects, Dr. Gurtner and his colleagues found no increase anything like that seen for primary pulmonary hypertension. Since the presenting symptoms would be largely the same in the primary pulmonary hypertension and pulmonary hypertension that was due to identifiable causes, it seemed unlikely that there could have been selective referral of the primary pulmonary hypertension patients.

Perhaps the technical capacity of the cardiology clinic had improved during the preceding decade, so that cases of primary pulmonary hypertension were no longer being mistaken for something else. If this were the case, they reasoned, there should have been a decline in the apparent numbers of cases of pulmonary hypertension ascribed to other causes. There was none.

Conceivably a better trained corps of general physicians was recognizing and referring primary pulmonary hypertension to the cardiology clinic. Younger physicians were taking their place in the community, and the cardiology clinic itself was vigorous in its efforts at postgraduate medical education. Unfortunately for this hypothesis, at most a quarter of the patients who came into the clinic did so with primary pulmonary hypertension listed as even a possibility by the referring doctor.

None of the alternative explanations seemed to explain the increase in cases.

The Bern group’s careful public writings did not prevent them from pushing for early action by regulatory authorities. A month after the report in May, Gurtner and his colleagues formally communicated their suspicions to the Swiss Intercantonal Pharmaceutical Control Agency (IKS). The IKS recommended that the drug be withdrawn from the market. (15) Shortly thereafter the agency restricted Menocil prescription to a "no-refill" status. (16) When word of Dr. Gurtner’s report came back to the United States in the summer of 1968, the U.S. manufacturer, McNeil Laboratories, notified the Food and Drug Administration, and suspended its own research program. (17) On August 23 the Swiss Medical Association sent a cautionary note to all its members, and asked them to report any suspicious cases.. The Committee on Safety of Medicines in the United Kingdom declined permission for the drug to be used in British clinical trials in September, and the IKS published a notification that Menocil would be officially removed from the list of approved Swiss medicines. (18,19) In early October the European distributor of Menocil, Cilag Chemie wrote to physicians in Switzerland. Noting the lack of clear scientific evidence incriminating Menocil, implying that Gurtner had observed many other causes of primary pulmonary hypertension in the affected pateints (this was misleading, since predisposing factors for the most part had not been uncovered), and citing work in Austria (which had not been published so far), the company announced that it had decided "as a purely precautionary measure to suspend sales for the time being." (20) There was a simulataneous action in Austria.

The affair moved more slowly in Germany (19). The West German agency charged with drug safety, the Pharmaceutical Commission, had not received any word from their Swiss counterparts, and it was not until Cilag warned the commission in August that there was any consideration of the problem. The Swiss replies to its queries were "too vague" to form the basis for action, according to the woman who headed the German commission, a Dr. Homann. Instead, at the end of October, some three weeks after Menocil sales had been suspended in Switzerland and Austria, the Pharmaceutical Commission and Cilag sent a joint letter to all West German physicians, inquiring about possible Menocil side effects. The letter stressed that "scientifically certain evidence for an etiologic connection has not been adduced..." The joint wording was eerily reminiscent of the company’s letter announcing suspension of sales in Switzerland, but there was no mention in the German letter of withdrawals in either Switzerland or Austria.

At an international conference in Vienna in November, with support from Cilag, researchers from Switzerland, Austria, and West Germany, the three countries in which Menocil was formally on the market, reported a confluence of incriminating results. On November 28, the company announced suspension of sales in Germany, again "as a purely precautionary measure." The announcement went out by ordinary mail, and was overlooked by both pharmacists and physicians until an angry essay in Die Zeit by German epidemiologist Eberhard Greiser called attention to the emerging scandal on December 13. (19) The drug was not sold anywhere after that.

Cilag Chemie procrastinated from the time of Gurtner’s report in May to the withdrawals of Menocil in October and November. The grudging tone of the company’s letter to physicians and the delays in West Germany suggest that senior managers were reluctant to accept a causal link between Menocil and primary pulmonary hypertension. There may well have been a struggle within the company. During the November conference, which Cilag sponsored, Friedrich. Kaindl, chair of the cardiology department in Vienna reported that "for months we had investigated all possible causes, including hair spray, contraceptives, animal hair, permanent wave, and so forth. The person who got the ball rolling was the gentleman who represented Menocil in Austria. He encouraged us to look into the relationship between Menocil and pulmonary hypertension. Mr. Schreitl’s unique stance must be acknowledged unequivocally, because his service was remarkable." (21) Summing up the conference, he alluded to the ambivalence of his academic colleagues about the company’s role in promoting Menocil and Schreitl’s moves to hasten its exit from the market. "Although some people might prefer that I didn’t say this, the distributor performed a real service in making us aware of the possible connection between Menocil and pulmonary hypertension ."

The suspension of Menocil sales invoked what philosophers would call a moral certainty, a conviction quite distinct from both faith and logic, and based only on imperfect evidence. In the tales of gumshoe epidemiology that graced the New Yorker’s pages for four decades, Berton Roueché’s protagonists would sometimes invoke moral certainty to describe themselves in the moment when the suspect poison or bacterium finally declared its role. (22) The year after the withdrawal of Menocil Dr. Kaindl reiterated the core problem. "We note with satisfaction that as a result of the Vienna conference ... the sale of this appetite suppressant was immediately suspended by the distributor everywhere and without exception, an act that deserves the greatest recognition," he said, slightly rewriting history in Cilag’s favor. But he added his reasoning, "Airtight proof had not been presented, for understandable reasons; the company’s action nonetheless eliminated this malady." (23) Kaindl may have wanted to raise up a model of firm response to a probable menace. Other clinicians and scientists continued to invoke the notion that mere observation did not amount to scientific proof that Menocil had caused pulmonary hypertension. We have seen Gurtner’s own hesitation. Prof. Kaindl’s admiration for Mr. Schreitl may have stemmed from the businessman’s ability to act where neither the academics nor his own management could do so, each for reasons of their own histories being incapable of drawing moral certainty from the plain data in front of them.

Quants

Five years after Gurtner’s first report, 23 cardiology centers in Switzerland, Germany, and Austria pooled their patient data to get a fuller picture of the epidemic. (24) There were 582 patients with primary pulmonary hypertension from 1960 through 1972. The rate of diagnosis had not varied through 1966. The proportion of cases with primary pulmonary hypertension suddenly doubled in 1967, doubled again in 1968, and remained high in 1969. The numbers then fell as precipitously as they had risen. The rate dropped by half in 1970 and again in 1971. By 1972 the disease was almost as rare as it had ever been. Essentially all of the fluctuation was ascribable to patients who said that they had used Menocil. Cardiologists and epidemiologists elsewhere in Europe, in Russia, and in the United States, where Menocil had not been on the market, reported no change in the occurrence of primary pulmonary hypertension. (25)

Eberhard Greiser, the epidemiologist who put together this perspective, was the same man who had denounced the delay in the West German withdrawal of Menocil in 1968. He also undertook a number of telling studies in Lower Saxony, where we worked in the Medical College of Hannover with the internist K. Gahl (26). The most important of these was to estimate the actual level of danger faced by patients who took Menocil. Gahl and Greiser set out to identify all the cases of primary pulmonary hypertension among persons who had taken Menocil in the Hannover area. Working with local medical organizations to contact all general practitioners, gynecologists, and internists, they organized a cardiac screening of patients who had taken Menocil and who had become ill in any way between December 1968 and July of the following year. Out of 605 women and 126 men with medical complaints, the physicians found 23 women with primary pulmonary hypertension demonstrated either by cardiac catheterization or at autopsy.

Another nine with suspicious symptoms developed proven disease in the year following the survey. Greiser mapped out the residences of all the cases, and found that they clustered around Hannover. (24) In an infectious disease outbreak this kind of clustering might have pointed to a common source of infection, but Greiser took it as evidence that he had been more successful in finding cases nearby the Medical College than far away, so he focused his research on the locals.

Greiser and Gahl performed a survey of the use of appetite suppressants in Hannover. (24 , 25) 5000 people provided information on their use of weight loss drugs, which the researchers then sought to confirm through the records of the regional health insurance agency. By statistical projection, the survey provided an estimate of how many people had taken appetite suppressants. The survey told Greiser and Gahl what proportion were men and women, and what their ages were. Most importantly, the survey let Greiser and Gahl estimate how many Hannoverians had taken how many tablets. Using the demographics of anorexigen consumption, Greiser could estimate, for all of Hannover, the numbers of citizens at possibly different levels of risk.

The pattern was highly incriminating for Menocil. Primary pulmonary hypertension was most common in middle aged women, but the likelihood of developing the disease was tightly related to the duration of consumption of Menocil. Among women born between 1910 and 1949, Greiser estimated a risk of one in a thousand for those who took one or two of the 40 tablet packs. The risk rose rapidly with increasing consumption, to one in a hundred for women who had taken five to six packs, and to nearly one in ten for those who had used eight packs or more. The question of a "dose-response" relation between Menocil and primary pulmonary hypertension would be crucial for comparing Menocil to later anorexigens.

For all their vital contributions to the Menocil debate, Greiser and Gahl missed an important opportunity to confirm the Hannover findings, and their failure illuminates a curious sidelight of epidemiologic investigation. In addition to the Hannover studies, the two had contrasted the Menocil intake of confirmed cases of primary pulmonary hypertension in the three nation survey to a that of comparison group of closely matched patients from the same cardiology centers. This technique can provide both powerful and accurate estimates of the variation in risk, but Greiser and Gahl chose instead only to report that the patterns of use of the drug in the diseased and healthy groups were "significantly" different. This custom dates back to the 1920’s, when the great English statistician R.A. Fisher was pioneering analytic techniques for agricultural research. Fisher himself was a giant of modern statistical thought, and the conventions he adopted for fertilizer selection crept over into fields in which a yes-or-no decision was irrelevant. Even before the Greiser and Gahl analysis, it was abundantly clear already that Menocil was significantly associated with primary pulmonary hypertension. The question of health concern was not whether, but how strongly, and they missed the chance.

Fisher was a passionate believer in experiment. Random allocation of treatments between study subjects formed the physical basis for much of his mathematics, for which he eventually received a knighthood. A chain smoker himself, Sir Ronald was the last person of any scientific stature to question the purely epidemiologic inference that smoking causes lung cancer. The lack of a randomized trial, in his mind, left too many alternate explanations on the table. He died of lung cancer in 1959.

A Better Mousetrap

A full page advertisement for Menocil in the Deutsche Medizinische Wochenschrift in June of 1968 shows a turned figure, a Bauhaus chess piece with a hint of female form, an elegant Dame illuminated from the side by a stark light. In the foreground, in the shadow and out of focus, stands her pudgy counterpart, a former self.

Aminorex fumarate was discovered in the United States, by McNeil Laboratories, in 1960. (27 , 28) It attracted attention because it suppressed appetite in rats, and it was found to have a similar effect in humans in 1962. Aminorex is chemically related to the naturally occurring stimulant ephedrine, which is found in the Chinese herb ma huang and is a favorite component of "natural" invigorants. It is a cousin of epinephrine, the internal chemical messenger produced by mammals in response to danger. Though not so addictive, aminorex is closely related to the amphetamine derivatives that gave the term "diet pill" its smarmy ring in 1950s. The drug induces a mild euphoria. There were reports of its misuse by groups of young people in Austria, including intravenous injection of tablets crushed and dissolved in mineral water (check this: "Donnauwasser’). (21 , 29)

Long term safety studies of aminorex in rats and dogs at 30 times the recommended human dose had shown excitability, irritability, gastritis, and the occasional ulcer as side effects of therapy, but there had been no evidence of cardiac or pulmonary changes. (Reported cases of human overdose showed very much the same pattern.) (30)

Few of the human trials of aminorex made their way into the scientific literature. Those that did surface were done in Boston by Arthur J. Hadler, who was the chief of the Medical Service at the Veteran’s Administration Outpatient Clinic in 1967. McNeil had supplied aminorex to him as an investigational drug with the trade named Apiquel. Hadler found that aminorex was as effective as other anorexigens in promoting weight loss in veterans, and he concluded that aminorex fumarate might be an effective alternative to amphetamines. (31) Three years later, Dr. Hadler reported on further studies. (32) Clearly sensitized by the growing controversy over adverse effects in Europe, and by McNeil’s own withdrawal of the Apiquel New Drug Application from the FDA, Hadler devoted more than half of the "Results" section of his second report to side effects. There were no instances of primary pulmonary hypertension. The total follow-up time veteran’s studies, 16 weeks, was less than the interval between first Menocil use and disease onset for almost all of the Swiss patients. The record for speedy onset was held by a patient in Lausanne, whose first symptoms appeared after two months. (33) Frau Z’s interval had been four months, and almost all other patients had enjoyed longer symptom-free times. Dr. Hadler’s studies were too short to observe the effect.

Dr. Hadler’s veterans were at least 20 pounds overweight and had lost from eight to twelve pounds on therapy, depending on how many weekly tablets of aminorex they had been given. Dr. Hadler concluded from his experience that "The author has noted no sign or symptoms suggestive of pulmonary hypertension in patients who have taken aminorex or other anorexigenic medication and believes aminorex to be a safe, effective anorexiant with few troublesome side effects." Although he cited the Committee on Safety of Medicines’ withholding permission to test Menocil in the UK, Hadler did not mention Swiss reports, which had been published two years earlier, and it is not clear from his writings that he was aware of the clinical picture that he claimed to have ruled out.

The following year, Dr. Hadler reported on yet another anorexigen, fenfluramine, which was destined to become half of the fen/phen diet. (34) Two of eight patients on high dose fenfluramine completed their scheduled 12 week course, as did seven of eight patients on low dose fenfluramine, almost all with weight loss. Although he noted that his series was too small to establish an optimal dosing schedule, Dr. Hadler concluded in part that "... flenfluramine appears to be a safe anorexigenic drug without CNS [central nervous system] stimulating properties..." that merited further investigation.

A South African study, begun before Gurtner’s first publication and concluded shortly afterward, afforded a longer follow-up than Hadler had achieved with the veterans, and specifically looked into primary pulmonary hypertension because of the Swiss reports. (35) Dr. M.C. Kew of Johannesburg Hospital gave eighteen Bantu and two colored patients four months of aminorex fumarate and four months of a placebo pill. He arranged the treatments in random order, aminorex or placebo first, in what is called a "crossover" trial. At the end of eight months, there was no clinical evidence of pulmonary hypertension in any patients. As in Hadler’s studies, the periods of aminorex consumption were marked by weight loss in almost every patient. The patients who crossed over from aminorex to placebo also afforded Dr. Kew an opportunity to examine rebound effects. Two thirds of the Bantu and Colored who took aminorex first and placebo second gained weight during the placebo portion of the trial.

Hadler and Kew were not alone in their failure to identify cases of pulmonary hypertension in their trials. A subsequent review of all the human experimental data, carried out by scientists working for McNeil, identified only a single case in 4,440 subjects. All of the trials were too short, and the complication, though devastating on a population scale, was just too rare to be caught in the net of clinical research. (17)

Oddball or Prophet?

Frau Z’s case evolved from an oddity into the forerunner of an epidemic. Is it possible that aminorex-induced pulmonary hypertension might itself be a prototype of medical disasters to come?

Menocil had about 80% of the appetite suppressant market in Switzerland. In Germany Menocil accounted for only about 15% of sales. It was therefore ominous that doctors in Cologne were observing essentially the same time course of appearance of new cases that Gurtner and his colleagues had reported in Bern. At the University Clinic in Cologne almost all cases had taken Menocil. (36) In Hamburg the number of cases of primary pulmonary hypertension jumped from 5 in 1964-1966 to 27 in 1967-1969, but only 11 of the later cases reported having used Menocil. (37) This last report was scanty on clinical details, and it may have been that the physicians did not inquire with sufficient care to learn about their patients’ drug histories. Years later Dr. Gurtner himself noted the gradual background rise in cases of pulmonary hypertension not associated with Menocil and the occasional exclusive use of other anorexigens in primary pulmonary hypertension patients. He speculated that the problem may have been one of inadequate inquiry. (38) He acknowledged that the question of whether other appetite suppressants might cause primary pulmonary hypertension was still open, but he pointed to the rapid decline in cases in Austria and West Germany, as well as in Switzerland, after Menocil was removed from the market as evidence that the collective effect of other agents was likely to be small.

Outcome

Although Gurtner quickly summarized his experience for a wider audiences,(39 , 40) the first reports in primary English language journal did not appear until 1971.(41) A British Medical Journal account from Basel described a series of cases, of whom the preponderance had taken Menocil for over a year. A large proportion of cases among long-time users would be most easily explained by an elevated of risk in heavy users; this tended to confirm Greiser's observation in Hannover that risk rose with increasing dose. The presentation aroused little interest, however, except in The Lancet, which was BMJ’s arch rival. An anonymous editorialist noted that aminorex had after all been given to calves at high altitude, to rats, and to beagles without causing pulmonary hypertension. "Thus the vital link between aminorex and the epidemic of pulmonary hypertension remains unproved despite the epidemiologic association." (42) If readers disagreed, their letters were in any case not published, and that was the end of the primary pulmonary hypertension epidemic in mainstream English-language medical literature.

The long term survival of the patients with aminorex-induced primary pulmonary hypertension was not good. Looking back 25 years later, Dr. Gurtner reported that the average survival of the aminorex patients was seven and a half years. (43)Gurtner found that patients who had taken more Menocil had mostly shorter survival, though others reported some reversibility of aminorex-induced primary pulmonary hypertension and occasional good survivals. (44) Some physicians believed that the better prognosis was limited to patients who discontinued therapy early. (45) Since the mechanism of action Menocil was never brought to light, there was never the possibility of a rational therapy to counteract the damage. Anticoagulants were of no help.

There is little doubt that other weight loss drugs can cause pulmonary hypertension. There have been many individual reports of cases, and one formal comparative study of the kind Greiser and Gahl just missed pulling off. Foods can cause primary pulmonary hypertension in animals, but there seem to be no clear human counterparts. There have been epidemics of primary pulmonary hypertension. At the turn of the century, there were reports of relatively high rates among Bengalis, for unknown reasons. (6) An epidemic poisoning in Spain, caused by sale of chemically treated canola oil (rapeseed) for cooking, entailed a number of cases of primary pulmonary hypertension. AIDS victims have an increased risk, for reasons that have yet to be elucidated.

One of the most important questions for future epidemics would be the actual magnitude of the risk associated with Menocil, and the extent to which that risk might attach to other, similar agents. Ideas of cost and benefit evaluation were not current in medicine in the late 1960s, and the data conflict. Eberhard Greiser, the epidemiologist and writer of angry essays, reported risks that ran to almost ten percent for prolonged use. Even two packets of Menocil appeared to bring the danger into the realm of one in a thousand, which would have been itself a thousand times higher than the annual incidence of primary pulmonary hypertension in the absence of a specific cause. Set against this is the fact that the epidemic in all three affected countries touched only several hundred people. Cardiology centers reported jumps of 20 fold in the incidence of primary pulmonary hypertension in 1967 and 1968, which is frightening, but still a far cry from the thousand to hundred thousand fold increase in users predicted by Greiser. Was the answer in a relatively low market penetration of Menocil? Or were Greiser's figures overblown? He and Gahl reported their results, but not much of their methodology, so it is hard to tell.

Why did Gurtner and his colleagues see a causal nexus that others missed? I put that question to Gurtner in 1997. By chance, he had been well prepared for a pulmonary hypertension investigation: he had trained with André Cournand at Bellevue Hospital in the 1950’s, when Cournand won the Nobel Prize with Forssmann and Dickinson. Having been the first to obtain pressure readings in the pulmonary artery, Cournand had naturally kept pulmonary hypertension as a special area of interest, which he had passed on to his students. Gurtner also recalls that "an important clue was given to our team by two residents who had left the Department of Pathology and entered the Section of Cardiology in 1968. They remembered having assisted in the investigation of lung biopsies of patients with pulmonary hypertension who had taken an anorexigen."

Friedrich Kaindl, the Austrian cardiologist whom Mr. Schreitl had finally pressed to investigate Menocil, ruminated on his own earlier failure to track down the culprit. "... how difficult it can be," he said, "to uncover a causal connection when it touches on the realm of the intimate or when it concerns the self esteem of individuals, who in this case had to confess that they lacked the discipline to maintain an ideal weight." (23) Today, when eating disorders themselves become enshrined as celebrity diets, we may have traded in the sense of shame that obesity connoted to an Austrian professor thirty years ago, but Kaindl’s generic warning still stands. For all the legitimate concerns about privacy that shadow the medical detective business that we practice now, it is still undeniably the unknown that gives us greatest pause. What we keep most secret is what can kill us.

After leaving the University Hospital in 1967, Frau Z was treated with digitalis to strengthen her heart muscle, diuretics to keep off fluid, and anticoagulant therapy, on the lingering premise that blot clots might play a role in her pulmonary hypertension. She continued to deteriorate, so that she could do only light kitchen chores sitting down. She was hospitalized several times. Dr. Gurtner’s last knowledge of her was in 1970. An abdominal swelling "the size of a baby’s head" brought her back into the hospital. This is turned out was an enormous pouch full of blood. The anticoagulants had prevented her clotting system from stanching a small internal bruise. The swelling came down during her hospital stay, during which she developed jaundice, and for a time her kidneys failed. She rallied somewhat and was sent home, prognosis "extremely guarded."

References
 

1 .  Gurtner HP, Gertsch M, Salzmann C, Scherrer M, Stucki P, Wyss F.  Häufen sich die primär vasculären Formen des chronischen Cor pulmonale? Journal Suisse de Médicine 1968;41:1579-1589 and 1968;43:1695-1707
2 .  Fishman AP.  A century of primary pulmonary hypertension.  In L.J. Rubin and S. Rich (eds.)  Primary Pulmonary Hypertension  Marcel Dekker, New York, 1997  I have drawn on Dr. Fishman’s account for most of the history recounted here.
3 .  Klob.  Wiener Wochenblatt 1865; 31:45  (quoted in Romberg)
4 .  Romberg E.  Über Sklerose der Lungen Arterie. Deutsches Archives fur klinische Medizin 1891;48:197-206
5 .  Arrillaga FC.  Sclérose de l’artère pulmonaire secondaire à certains états pulmonaires chroniques (cardiaques noirs).  Archives des Maladies du Coeur 1913;6:518-529
6 .  Warthin AS.  A case of Ayerza’s disease: chronic cyanosis, dyspnea, and erythremia, associated with syphilitic arteriosclerosis of the pulmonary arteries.  Transactions of the American Society of Physicians.  1919;34:219-239
7 . Arrillaga FC.  Sclérose de l’artère pulmonaire secondaire (cardiaques noirs).  Bulletins et Mémoires de la Société Médicale des Hôpitaux de Paris.  1924;48(3rd series):292-303
8 .  Brachetto-Brian D.  Concepto anatomo-patologico de los cardiacos negros de Ayerza. Revista de la Sociedad de medicina interna y de la Sociedad de Fisiologia. 1925;1:821-931
9 .  Brenner O.  Pathology of the vessels of the pulmonary circulation.  Archives of  Internal Medicine 1935;56:211-237, 457-497,724-752,976-1014,1190-1241
10 .  Kaindl F.  Pulmonale Hypertonie.  Wiener medizinische Wochenschrift 1970;120:631-634
11 .  Sourkes TL.  Nobel Prize Winners in Medicine and Physiology, 1901-1965.  London, Abelard-Schuman, 1967,  pp. 333-334. The quote is Sourkes’s translation from W. Forssmann, Die Sondierung des rechten Herzens, Klinische Wochenschrift 1929;8:2085-2087
12 .  Gurtner HP.  Pulmonale Hypertonie nach Appetitzüglern. Medizinishe Welt 1972;23:1036-1041
13 .  Gertsch M.  Die Häufigkeit der primär vaskulären pulmonalen Hypertonie am Inselspital Bern.  Zeitschrift für Kreislaufforschung 1970;59:884-886
14 .  Blankart R.  Epidemiologische Untersuchungen zur Häufung der pulmonalen Hypertonie in den Jahren 1967-1969. In R. Blankart (ed) Adipositas - Kreislauf - Anorektika / Obesity - Circulation - Anorexigens.   Hans Huber Publishers, Bern, 1970
15 .  Gurtner HP.  Aminorex and pulmonary hypertension.  A review.  Cor Vasa 1985;27;160-171
16 .  König.  Betrifft Menocil. Tabletten.  Letter to members of the Swiss Medical Association, August 23, 1968
17 .  Peters L, Gourzis JT.  Pharmacological, toxicological and clinical observations with Aminorex in the United States. In R. Blankart (ed) Adipositas - Kreislauf - Anorektika / Obesity - Circulation - Anorexigens.  Hans Huber Publishers, Bern, 1970, pp. 61-77
18 .  Cahal DA.  Appetite suppressants and pulmonary hypertension (letter).  Lancet 1969;i:947
19 . Greiser, E.  Hochdruck in der Lunge.  Die Zeit (Hamburg).  15 December 1968 , p. 30
20 .  Schregenberger O, Hauisch K.  Menocil.  Letter to Swiss physicians, October 10, 1968
21 .  Kaindl F. Diskussion. Wiener Zeitschrift für Innere Medizin und ihre Grenzgebiete 1969;50:505, 511
22 .  Roueché, Berton.  The Medical Detectives.  Truman Talley Books/Plume, New York, 1991.  See “Eleven Blue Men” (1947) and “The Liberace Room” (1960).
23 .  Kaindl F.  Medikamentös bedingte pulmonale Hypertonie; Erfahrungen in Österreich.  Wiener Klinische Wochenschrift 1970;83:373-375
24 .  Greiser E.  Epidemiologische Untersuchungen zum Zusammenhang zwischen Appetitzüglereinnahme und primär vasculärer pulmonaler Hypertonie.  Internist 1973;14:437-442
25 .  Greiser E, Gahl K.  Kausalitätsnachweis der durch Aminorexfumarat induzierten primär vasculären pulomonalen Hypertonie.  Verhandlungen der Deutschen Gesellschaft für Innere Medizin.  1972;78:1073-1076
26 .  Gahl K, Greiser E.  Erfahrungen einer klinisch-epidemiologischen Studie zur Problem der primär vaskulären pulmonalen Hypertonie.  Schweizerische medizinische Wochenschrift 1970;100:2154-7
27 .  Peters L, Gourzis JT.  Pharmacological, toxicological and clinical observations with aminorex in the United States.  In R. Blankart (ed) Adipositas - Kreislauf - Anorektika / Obesity - Circulation - Anorexigens.  Hans Huber Publishers, Bern, 1970
28 .  Poos GI, Carson JR, Rosenau JD, Roszkowski AP, Kelley NM, McGowin J. 2-Amino-5-aryl-2-oxazolines.  Potent new anorexigens.  Journal of Medical Chemistry 1963;6:266-272
29 .  Brücke F.  Diskussion. Wiener Zeitschrift für Innere Medizin und ihre Grenzgebiete 1969;50:501
30 .  Borbély F, Pasi A, Velvart J.  Die akute perorale Vergiftung durch 2-Amino-5-Phenyl-Oxazolin-Fumarat beim Menschen anhand von 30 Beobachtungsfällen.  Archiv für Toxicologie 1970;21:117-124
31 .  Hadler AJ.  Studies of aminorex, a new anorexigenic agent.  J New Drugs (or is the J Clin Pharmacol, which is the way it is cited in Hadler 1970?) 1967;7:296-302
32 .  Hadler AJ.  Further studies of aminorex, a new anorexigenic agent.  Current Therapeutic Research 1970;12:639-644
33 .  Rivier JL, Sangra R, Tsicotis A.  Klinik und Ätiologie. Wiener Zeitschrift für Innere Medizin und ihre Grenzgebiete 1969;50:489-493
34 .  Hadler AJ.  Fenfluramine, a new anorexigenic agent.  Journal of Clinical Pharmacology 1971;52-55
35 .  Kew MC.  Aminorex fumarate: A double-blind trial and examination for signs of pulmonary arterial hypertension.  South African Medical Journal 1970;44:421-423
36 .  Von Smekal P, Standfuss K, Rau G.  Zunahme der primär vaskulären pulmonalen Hypertonie im Zusammenhang mit der Einnahme von  Appetitzüglern?  Zeitschrift für Kreislaufforschung 1970;59:892-897
37 .  Voß H, Harms H.  Epidemiologie und Klinik der primär vaskulären pulmonalen Hypertonie. Zeitschrift für Kreislaufforschung 1970;59:887-891
38 .  Gurtner HP.  Chronische pulmonale Hypertonie vaskulären Ursprungs, plexogene pulmonale Arteriopathie and der Appetitzügler Aminorex:  Nachlese zu einer Epidemie. Schweizerische medizinische Wochenschrift 1985;115:818-827
39 .  Gurtner HP, Scherrer M.  Chronic cor pulmonale due to primary vascular obstruction of the lungs -- is there a true increase in incidence.  Proceedings of the Vth European Congress of Cardiology, Athens (1968)
40 .  Gurtner HP.  Pulmonary hypertension produced by ingestion of substances.  Bulletin de physiopathologie Respiratoire.  1969;5:435:441, 1969
41 .  Follath F, Burkhart F, Schweizer W.  Drug-induced pulmonary hypertension?  British Medical Journal 1971;1:265-266
42 .  An epidemic of pulmonary hypertension.  (editorial) Lancet 1971;ii:252
43 .  Frank H, Gurtner HP, Kneussl M, Lang I, Mlczoch J.  Aminorex-induzierte, plexogene pulmonale Arteriopathie: 25 Jahre danach!  Zeitschrift für Kardiologie 1993;82:568-572
44 .  Mlczoch J, Probst P, Szeless ST, Kaindl F.  Primary pulmonary hypertension:  Follow-up of patients with and without anorectic drug intake.  Cor Vasa 1980;22:251-257
45 .  Loogen F, Worth H, Schwan G, Goeckenjan G, Lösse B, Horstkotte D.  Long-term follow-up of pulmonary hypertension in patients with and without anorectic drug intake.  Cor Vasa 1985;27:111-124
 

This document has been provided through the Drugs and Devices Information Line, http://www.hsph.harvard.edu/Organizations/DDIL/ddil.html