The Discovery of Insulin

by Michael Bliss

  Allen’s work undercut the view that diabetes was mostly a problem of carbohydrate metabolism. It was not just the carbohydrates, but the proteins and fats as well, that the diabetic’s body was having trouble with, Allen argued. All kinds of food tended to over-burden the system. Diets which involved cutting back sharply on carbohydrates and then increasing the proteins or fats to compensate, achieved nothing – or, worse, caused a higher rate of acidosis and death in coma because of their fat content. The answer was to continue to cut back on carbohydrates, but to cut back on everything else, too, so that the diabetic’s total calorie intake was reduced. If over-nourishment or normal nourishment produced diabetic symptoms, notably glycosuria, then the trick was to find the degree of under-nourishment that would enable a diabetic to live sugar- and symptom-free. Any previous diabetic diets that had actually been effective, Allen claimed -high-fat, oatmeal cure, or whatever – had been characterized by a low total calorie count. There was no way a diabetic could save his carbohydrates and eat his calories too.

  An outsider with no advanced degrees, Allen had trouble getting a job until the Rockefeller Institute in New York, impressed by his book, offered him a junior position in 1914. The appointment gave him access to a small ward of diabetic patients, and turned out to be a marvellous opportunity to begin applying his theories to humans.

  After four years’ clinical work, Allen and his associates published their results in 1919 in a second massive volume, Total Dietary Regulation in the Treatment of Diabetes, which ran to 646 pages plus charts. Almost half the book consists of exhaustive case records of seventy-six of the one hundred patients Allen had treated.

  His methods were tried on all sorts of diabetics, mild and severe, recently diagnosed and terminally comatose, old and young, educated and ignorant, well-to-do and desperately poor. The therapy was almost always the same: When a diabetic was admitted to hospital, he or she was put on a fast (liquids only) until the glycosuria and, in the severe cases, the acidosis disappeared. Then there would be a gradual building up of diet, measuring by carbohydrate tolerance, but with strict weighing of all foods, to see how much the patient could take before becoming glycosurie. When sugar appeared in the urine, the limit had been reached. A fast day would clear the urine again and the diet would be fixed at a total calorie intake just under this tested tolerance.

  This quick description of the Allen method might go unremarked by readers unfamiliar with serious diabetes and in an age when most of us have to diet occasionally. At the time he introduced what came to be called the “starvation treatment” of diabetes, Allen was advocating serious dieting in a country where being well-fed was still a sign of good health. More ironically, he was advocating serious dieting to patients two of whose complaints were their terrific hunger and their rapid weight loss. They came to the doctor to be treated for these symptoms and the doctor seemed to be telling them that they had to be hungry more often, that they had to lose even more weight.

  The ironies, the Hobson’s choices, the catch-22’s of the treatment were staggering. An adult diabetic, weak, emaciated, wasted to perhaps ninety pounds, would be brought into hospital and ordered to fast. If the patient or the patient’s family complained that he or she was too weak to fast, Dr. Allen replied that fasting would help the patient build up strength. If the patient complained about being hungry, Allen said that the fasting would help ease the hunger. Suppose the method didn’t seem to work and the symptoms seemed to get worse. The answer, Allen insisted, was more rigorous under-nourishment: longer fasting, a maintenance diet even lower in calories. To top it off, Allen and others were also urging diabetics to take as much physical exercise as possible, claiming it would help them burn more food and increase in strength.

  Where was the limit to the dieting? Where would you stop? In fact there was no limit. In the most severe cases the choice came to this: death by diabetes or death by what was often called “inanition.”

  “The plain meaning of this term,” Allen wrote, “is that the diabetes was so severe that death resulted… from starvation due to inability to acquire tolerance for any living diet.” “The best safeguard against inanition,” he added, “consists in sufficiently thorough undernutrition at the outset.” In those situations where the awful choice between death from diabetes and death from starvation could not be avoided, “comparative observations of patients dying under extreme inanition and those dying with active diabetic symptoms produced by lax diets or by violations of diet have convinced us that suffering is distinctly less under the former program.”24

  To illustrate, consider Rockefeller case 60, a forty-three-year-old housewife who came into the hospital on New Year’s Day, 1916, having lost 60 pounds in the few months since the onset of her diabetes. She weighed 36 kilograms or 79 pounds on admission, and was so weak that even Allen hesitated to go ahead with severe fasting:

  The experiment was tried of feeding more liberally for a short time in the attempt to restore some strength, so as to get a fresh start for further fasting….the attempt caused only harm instead of benefit, as always in genuinely severe cases. The question thereafter was whether the glycosuria could be controlled without starving the patient to death….Though the food was thus pushed to the utmost limit of tolerance, it was not possible to prevent gradual loss of weight.

  She was utterly faithful in following her diet, which during hospital stays averaged 750 calories a day and about 1,000 calories when she was at home. When Allen last saw her, in April 1917, her weight was down to 60 pounds and falling. “Perhaps better results might have been obtained by cutting down the weight to perhaps 30 K (66 pounds) at the outset,” he mused in the conclusion to the discussion of her case. “The question remains whether the pancreatic function is absolutely too low to sustain life, or whether by sufficiently rigid measures downward progress can be halted even at this time.” The answer was given in a footnote added in the final revision of the manuscript: “Largely on account of her residence in a city too far away to permit personal supervision and encouragement, this patient finally broke diet, and after a rapid course of glycosuria and acidosis, died in Feb. 1918.”25

  Many of Allen’s patients broke diet out of hospital, some sooner than others. Case 1 embraced Christian Science four months after her release, began eating everything at will, and died in a few more months. Case 51, a seven-year-old Polish-American schoolboy, was able to sneak food at home unknown to his parents, and died from it; “the essential cause of trouble lay in the home conditions of an uneducated Polish laboring family.” Case 18 was a sixteen-year-old errand boy who adhered to his diet fairly well until summertime when he had a feast of cherries. After that he became uncontrollable and went downhill.26

  Even inside the hospital the staff had to be constantly on the alert to stop the pilfering of “forbidden food.” The most extreme example was case 4, a twelve-year-old boy whose diabetes had already caused blindness when he was admitted. No matter how carefully he was treated, his urine tests on some days would show sugar. It could not be accounted for from his diet. The staff could not understand what was happening:

  It had seemed that a blind boy isolated in a hospital room and so weak that he could scarcely leave his bed would not be able to obtain food surreptitiously when only trustworthy persons were admitted. It turned out that his supposed helplessness was the very thing that gave him opportunities which other persons lacked…. Among unusual things eaten were tooth-paste and bird-seed, the latter being obtained from the cage of a canary which he had asked for….These facts were obtained by confession after long and plausible denials. The experience illustrates what great care is necessary if records of diabetic patients are to be vouched for as correct.

  The gods had their revenge. Thinking the glycosuria was caused by too high a normal diet, the staff cut the boy’s normal food supply further and further. It was too late when they realized their mistake. He weighed less than 40 pounds when he died from starvation.27

  Allen was a stern,
cold, tireless scientist, utterly convinced of the validity of his approach. His therapy for diabetes seemed immensely hard-hearted in the extreme cases, and met much resistance from diabetics, their families, other physicians, and other workers at the Rockefeller Institute. Allen defended himself with iron logic. Yes, the method was severe; yes, many patients could not or would not follow it faithfully; yes, in the worst cases it led to death from starvation; yes, all it could do was prolong the lives of diabetics, in some cases for a few years, in severe cases perhaps only a few months. But what was the alternative? All of Allen’s experimental and clinical evidence showed that total dietary regulation was the only way of prolonging the lives of diabetics. Nobody had a better way. Besides, he claimed, his diet was not impossible to follow: because they were better balanced Allen’s diets were often more tolerable than the destructive high-fat alternatives. Most of his disobedient patients had actually been on the more liberal of the series of diets, “and were the sort of persons who would not abide by any restrictions no matter how slight.” Some of his most undernourished patients had borne their diet in the most faithful way.28

  Generally, diabetics on the diet did feel better than those who broke it, the “simple hunger” from careful fasting or dieting being less tormenting than the sick hunger, or polyphagia, of diabetes. Allen’s “faithful” patients, even those under an obvious sentence of death, regained a degree of strength and comfort and the ability to enjoy life. “Though always hungry, excessively emaciated, and lacking strength for any real exertion,” he wrote of case 60, “some of the noteworthy features are her constant cheerfulness, freedom from infection, and comfort in all other respects. She is able to be up and about, carries on light household duties, and – the point of most importance to her – attends to the bringing up of her child.” By her faith and determination, case 60 had won for herself about two extra years of life.29

  In the final analysis, the only argument against the thorough treatment was the cruelty of prolonging a patient’s suffering. “Euthenasia is no more justified in diabetes than in numerous other diseases,” Allen argued. “Diabetics who overeat for the deliberate purpose of killing themselves are uncommon.”30 Allen was proud that he was not only keeping diabetics alive longer, but was pioneering in methods of keeping starving people alive; some of his patients were living in stages of inanition not thought possible.

  Frederick Allen’s determination to apply his methods ruthlessly (to prove his theories absolutely he wanted to be able to control his patients as thoroughly as laboratory animals were controlled) were probably responsible for a decision at the Rockefeller Institute to take away his control of the diabetes clinic. Instead of being the triumph of medical research it appears, the 1919 volume, Total Dietary Regulation in the Treatment of Diabetes, actually veils a bitter controversy about the treatment of those hundred cases. The book was later denounced by Allen as an inconclusive, failed study. The diabetologist did not believe there were many shades of grey in medical research, or in life generally. He left the Rockefeller Institute intensely frustrated, served in the army diabetes service during the war, and in 1919 launched a daring bid for personal and professional independence by purchasing the Morristown, New Jersey, mansion formerly owned by Otto Kahn. There he founded the Physiatric Institute, intended to be a prestigious centre for treatment of Americans suffering from diabetes, high blood pressure, and Bright’s disease. The fees paid by rich patients for the luxurious facilities in one department of the Institute supported more plebeian facilities in other departments as well as the ongoing research work. One of Allen’s rules was that all in-coming patients had to promise their sincere co-operation in the prescribed treatment. As the Institute flourished in 1919–20, Allen worked frantically to pay off his debts and build the resources to support the grand research plan he felt had been frustrated at the Rockefeller.31

  The total dietary approach to diabetes was the best therapy available at that time. How widely it was actually used is difficult to estimate. In medical schools and among up-to-date practitioners Allen’s methods seem to have been universally adopted. But Allen and the other diabetologists often wrote scornfully of the ignorance with which doctors treated diabetes – at worst with opium and over-feeding, at best by handing out printed, out-of-date diets. And even they were better than the patent medicine men who offered nostrums such as Bauer’s Antidiabeticum, and the religious people who offered prayer, faith, and Christian Science. At his own professional level, as well, Allen was under attack from several researchers, especially Woodyatt in Chicago, who worked out elaborate theoretical critiques of “starvation” and new justifications for high-fat and fairly high calorie diets.32 In these clinics, too, the thorough treatment of diabetes was expensive and complicated, involving prolonged hospital stay, careful preparation and weighing of individually tailored diets, elaborate daily tests, and special nursing for children. In prosperous North America diabetes was becoming something of a specialist’s disease, with special diabetic wards being set up at hospitals and physicians building whole practices on nothing but the treatment of diabetes.

  Other than Allen, the most prominent American specialist in diabetes was Dr. Elliott P. Joslin. A New Englander, a graduate of Yale and Harvard, and student of Naunyn at Strasbourg, Joslin gradually narrowed his medical practice in Boston to diabetes. He was a prolific writer, particularly at the semi-popular level aimed at physicians and the diabetics themselves, and a warm enthusiast. In his writing Joslin tried to put the best face on the diabetic’s situation, stressing that it was “the best of the chronic diseases,” clean and seldom unsightly, not contagious, often painless, and usually susceptible to treatment.33

  A friend of Allen and a strong supporter of under-nutrition, Joslin tended to be optimistic about the therapy. He was almost certainly over-optimistic, possibly deliberately so to bolster his patients’ morale and his own.34 It was hard to keep up your spirits to face each day of urging sick people to keep starving. A nurse at the Physiatric Institute remembered how horrifying it was to watch the starving children lying in their beds. “It would have been unendurable,” she wrote, “if only there had not been so many others.”35

  Because he tempered his own rock-hard puritanism with warmth and charm and a sense of hope, Joslin may have had more success with his patients than the forbidding Dr. Allen. He was particularly popular with children, some of whom were brought to him because no one else would treat them. When von Noorden came to Boston, Joslin remembered, he shuddered and turned away when shown one of Joslin’s skeleton-like diabetic girls. A quarter of a century after the discovery of insulin the doctors were reminded of these pre-insulin diabetics when they saw the pictures of the survivors of Belsen and Buchenwald.36

  VI

  Despite the record of failure, and despite the pessimism of men like Allen, Carlson, and Macleod, attempts to find an effective pancreatic extract continued, “because of the strong theoretical inducements,” Allen noted.37 The most interesting and important of these new attempts involved experiments measuring the effect of pancreatic extracts on blood sugar. High blood sugar, or hyperglycemia, had been recognized for many years as a sine qua non of the diabetic condition. Measurements of blood sugar had not usually been involved in diabetes therapy or research, however, because they were very difficult. The chemical tests required to estimate the amount of sugar in the blood called for a lot of blood, usually twenty cc. or more. It was difficult and possibly dangerous to take many of such large blood samples from either humans or animals. As well, methods for testing the sample were time-consuming and so crude that the margins of error in estimating the percentage of blood sugar were very high. It was much more practical, safer, and perhaps more accurate to test the diabetic condition through urine samples alone.

  But accurate blood sugar readings would obviously be a useful research tool, supplying a far more reliable guide to diabetes than urine tests. All of the problems and complications and alternative interpretations of gl
ycosuria created by the possibility of kidney disorder could be avoided. If good testing procedures (the lack of which was probably central in E.L. Scott’s failure) could be developed, it would be much easier to check short-term fluctuations of blood sugar than to measure, say, the hourly inflow of sugar into the urine. The single most important development in diabetes research, next to Allen’s diets, was the rapid improvement between about 1910 and 1920 in techniques for measuring blood sugar. In 1910 a blood sugar test still required 20 cc. or more of blood; by 1920 it could be done with as little as 0.2 cc.38 The use of blood sugar estimations was soon reflected in the research.

  A young American, Israel Kleiner, became interested in pancreatic extracts and blood sugar while working with S.J. Meltzer at the Rockefeller Institute during the time of Allen’s researches. Pioneering studies were being done there on the speed with which injections of sugar normally disappeared from circulation (that is, were assimilated by the system). By contrast, in diabetic animals much of the sugar continued to circulate. But when an emulsion of pancreas was mixed and injected along with the sugar solution, the diabetic animal handled it almost normally. Observing this, Kleiner and Meltzer began experiments to see how pancreatic extracts would affect the ability of depancreatized dogs to deal with their system’s own excess sugar.

  They reported very promising preliminary findings in 1915, but their work was interrupted by the war. In 1919 Kleiner returned to it, running many more experiments. Late in 1919 he published his findings in the Journal of Biological Chemistry. Of all publications before the work at Toronto, it was the most convincing.