Fig. 7. Frontispiece and page from Sebastiani Pissinii De Diabete Dissertatio. Milan, 1654. The Museum of Diabetes, Turin. www.museodeldiabete.it. Reprinted with kind permission.
“Vannella Moriconi, a very wise and very noble woman, having suffered long from diabetes and being tormented by a ferocious lust for drinking, lacked any strength but without fever, she died almost without realizing it.
Domitilla Arnolfini, a very noble young woman, even though she enjoyed a blossoming of life which she had been used to leading in health, concluded her life in a high fever for several months before and could not avoid dying of diabetes.”
On the other hand, Francesco Arma (1550) of Chivasso, near Turin, physician to Duke Emanuele Filiberto of Savoy, boasted his success on “Domino Joanne Maria de Contino” after “septem grana” of pepper.
Sweet Urine
Tasting urine was an integral part of the medical art, above all because there were no other methods to analyze it, and diabetes was a rare case in which the practice proved useful in making a diagnosis because the urine was abundant and above all sweet. The first to establish a distinction between diabetes “mellitus,” where the urine is as sweet as honey, and the other forms (“potius salsus quam dulcis”) was the Englishman Thomas Willis (1621–1675) in his Pharmaceutice Rationalis sive Diatriba de Medicamentorum Operationibus in Humano Corpore of 1676 (Fig. 8) [17].
Fig. 8. Frontispiece and passage “De Diuresi nimia, eiusque remedio, and speciatim de Diabete …” from Thomas Willis Pharmaceutice Rationalis sive Diatriba de Medicamentorum Operationibus in Humano Corpore. London, 1676. The Museum of Diabetes, Turin. www.museodeldiabete.it. Reprinted with kind permission.
Subsequently, by letting the urine evaporate and challenging the residue with yeast, Matthew Dobson (1732–1784) demonstrated in 1776 that the sweet taste was due to the presence of sugars capable of fermentation [18]. Incidentally, already in the first half of the 500s, Paracelsus had boiled to evaporation the urine of diabetic patients obtaining “salt” as residue. Although he did not recognize this as sugar, he had claimed that thirst and diabetes were always due to the accumulation of such “salt” in the blood.
The first to take advantage of Matthew Dobson’s observations on the presence of sugar in the urine, and apply it to monitor the results of treatment, was John Rollo (ca. 1750–1809). Surgeon-General of the Royal Artillery, Rollo should also be credited for giving the first clear directions to limit carbohydrates in the diet, although for the wrong reason. Convinced that the cause of diabetes was “a primary and peculiar affection” of the stomach in which some morbid change in “the natural powers of digestion and assimilation” led to the formation of sugary material in the organ, mostly from vegetable matter, he recommended a diet rich in old meat and fat [19, 20]. In his Notes of a Diabetic Case of 1797, followed a year later by An Account of Two Cases of the Diabetes Mellitus, Rollo reported on the diet inflicted on his first patient, a fellow army officer and personal friend (Fig. 9). On October 19, 1796, Captain Meredith was started on a definitely unpalatable regimen:
Fig. 9. Frontispiece and first two pages of John Rollo’s Traité du Diabète Sucré. French translation. Paris, 1798. Reprinted with kind permission.
“… The diet to consist of animal food principally and to be thus regulated:
Breakfast. One and a half pints of milk and half a pint of lime-water, mixed together; and bread and butter.
Noon. Plain blood puddings, made of blood and suet only.
Dinner. Game, or old meats, which have been long kept; and as far as the stomach may bear, fat and rancid old meats, as pork. To eat in moderation.
Supper. The same as breakfast.”
Meredith was militarily abiding, and did improve. Others, including a general, were not so lenient. Rollo, a Scotsman, published mostly on matters of military medicine and also in support of Edward Jenner’s vaccination (Medical Report on Cases of Inoculation, 1804).
Finally, in 1815, the French chemist Michel-Eugène Chevreul (1786–1889) identified the sugar as glucose, opening the road to its quantitation for diagnostic and therapeutic purposes [21]. The solution was developed by the Germans Carl August Trommer in 1841 and, with better luck, in 1848 by Hermann Christian von Fehling (1811–1885) [22], who developed the method used to measure glycosuria almost to the present day. These developments were not universally acknowledged, anyway, and leveraging upon rare cases of renal diabetes and other non-diabetic “melliturias,” detractors fought fierce diatribes which endured well into the first half of the 20th century.
Sweet Blood
Already in 1806, the Englishman Willian Hyde Wollaston (1766–1828) had shown the presence of sugar in the blood in a proportion of 1:30 to the amount found in urine and, in 1843, Thomas Watson also developed a method to measure glucose in the blood, the clinical use of which was seriously hampered by the requirement of at least 300 mL for a single assay. The first to show that the liver can release glucose into the blood and, more generally, to propose a model of the human body as a set of collaborating parts maintaining an internal balance (“milieu interieur”) was the Frenchman Claude Bernard (1813–1878), father of experimental medicine. This replaced the then dominant concept of various organs as separate entities not communicating with each other except through the ill-defined “humors.” Bernard theorized in 1855 that diabetes was the result of an overproduction of glucose by the liver, a revolutionary vision for its time and still valid today [23]. He found that ligation of the pancreatic ducts results in degeneration of the gland, an observation that was to carry important consequences. To him we also owe the first reliable method of measuring the concentration of glucose in the blood.
The Pancreas and Diabetes
In 1685, at the dawn of experimental medicine, the Swiss Johann Conrad Brunner (1653–1727), who described the homonymous glands in the duodenum, removed the pancreas from a dog in the course of his studies on the role of pancreatic juices in digestion [24]. To contrast the theories of Frans de la Boë (Sylvius of the mesencephalic aqueduct) and his pupil Reigner