Diabetes Insipidus: History and Discovery

Diabetes insipidus (DI) and diabetes mellitus share one dramatic symptom — the body passes enormous volumes of urine and the patient is tormented by thirst — yet they are entirely different diseases. The old physicians could tell them apart with a method that sounds startling today: they tasted the urine. In diabetes mellitus it was sweet, like honey; in diabetes insipidus it was “insipid,” meaning flat, dilute, and tasteless. The story below traces how that bedside taste-test led, over three centuries, to the modern understanding that diabetes insipidus is a disorder of a single water-balance hormone — antidiuretic hormone, also called vasopressin — and how the disease has, very recently, been given an entirely new name to stop it being confused with sugar diabetes ever again.

Two Diseases, One Confusing Name
Thomas Willis and the Taste of Urine (1670s)
Naming “Diabetes Insipidus”: Cullen and Frank (Late 1700s)
Tracing the Cause to the Pituitary Gland
Pituitary Extracts as the First Treatment (1913)
Du Vigneaud and the Synthesis of Vasopressin (1954)
Central vs. Nephrogenic: Two Roads to the Same Symptom
From Hormone to Gene: AVPR2 and Aquaporin-2
A New Name: Arginine Vasopressin Deficiency and Resistance (2022)
Research Papers and References
Connections

Two Diseases, One Confusing Name

The word diabetes is far older than any understanding of blood sugar or hormones. It comes from a Greek root meaning “to pass through” or “siphon,” and it is usually credited to the physician Aretaeus of Cappadocia, who in the second century described a dreadful wasting illness in which the flesh and limbs seemed to “melt down” and pour out of the body as urine. To the ancient observer the defining feature was not sweetness or salt but sheer volume: a person who could not stop urinating and could not stop drinking, as if water ran straight through them. For many centuries that single picture — massive urine output (polyuria) with unquenchable thirst (polydipsia) — covered what we now know to be at least two completely separate diseases.

This shared name is the root of a confusion that has lasted, remarkably, into the present day. Diabetes mellitus is a disorder of blood sugar, caused by problems with the hormone insulin; its excess urine is a side effect of the body trying to flush out surplus glucose. Diabetes insipidus has nothing to do with sugar or insulin at all. It is a disorder of water, caused by a problem with a different hormone — antidiuretic hormone (ADH), also called vasopressin — whose job is to tell the kidneys how much water to keep. The two diseases produce the same flood of urine for entirely different reasons, and the central task of this history is to show how physicians slowly learned to pull them apart.

The first reliable tool for separating them was, by modern standards, extraordinary: the physician dipped a finger and tasted the patient’s urine. Sweet urine meant one disease; flat, watery, tasteless urine meant the other. Crude as it was, this taste-test was a genuine diagnostic breakthrough, and it is where the modern story properly begins.

Back to Table of Contents

Thomas Willis and the Taste of Urine (1670s)

The pivotal figure is the English physician Thomas Willis (1621–1675), an Oxford anatomist whose name survives today in the “circle of Willis,” the ring of arteries at the base of the brain. Around 1674, in a work whose memorable English title was The Diabetes, or Pissing Evil, Willis recorded that the urine of certain patients with this wasting, thirst-ridden disease was “wonderfully sweet, as if imbued with honey or sugar.” He was not the first human being ever to notice this — Ayurvedic physicians in ancient India had described “honey urine” (madhumeha), attracting ants, more than a thousand years earlier, and the Persian physician Avicenna had noted the sweetness too — but Willis is generally regarded as the first physician in the modern European medical tradition to tie the sweet taste of the urine specifically to the disease.

Willis’s great practical contribution was to recognise that not all of these copious-urine patients had sweet urine. Some passed equally enormous quantities of urine that was not sweet — it was bland and watery. By using taste at the bedside, he could sort the patients into two camps. It is important to be precise about what Willis did and did not do, because sources sometimes blur this: he clearly distinguished the sweet form of the disease from a non-sweet form, and the Latin word mellitus (“honeyed,” from mel, honey) became attached to the sweet variety. He did not, however, understand why the urine was sweet — he did not know it contained sugar — and the now-familiar paired term diabetes insipidus for the tasteless form was not yet in use in his writing. That formal naming came a century later.

Even so, Willis’s taste-based split was the conceptual seed of everything that followed. For the first time, a single overwhelming symptom — the flood of urine — had been shown to hide two different diseases, distinguishable at the bedside. A hundred years afterward, the English physician Matthew Dobson would finally prove that the sweetness in the “mellitus” form really was sugar, by evaporating the urine and tasting the sweet residue left behind — but that thread belongs to the history of diabetes mellitus. Our concern is the other branch: the tasteless, insipid disease.

Back to Table of Contents

Naming “Diabetes Insipidus”: Cullen and Frank (Late 1700s)

The disease got its modern name in the late eighteenth century, and the credit is genuinely shared rather than belonging to one person — a point worth stating plainly, since popular accounts often pick a single “discoverer.” The Scottish physician and chemist William Cullen (1710–1790), a towering teacher at Glasgow and Edinburgh, is widely credited with drawing the clear distinction in 1769 between sweet diabetes and a diabetes whose urine was insipid to the taste. Cullen explicitly contrasted the two forms, applying mellitus to the sweet disease (following Willis) and reserving the idea of an insipid diabetes for the tasteless one. In this sense the conceptual pairing — mellitus versus insipidus — is firmly associated with Cullen’s era and teaching.

The formal Latin term diabetes insipidus as a fixed disease label is frequently attributed to the influential German physician and public-health pioneer Johann Peter Frank (1745–1821), in the 1790s (often dated to 1794). Historical sources differ on exactly who first set the two words side by side in print, and rather than force a false certainty this page presents both attributions honestly: Cullen for the clear taste-based distinction around 1769, and Frank for crystallising the formal term around 1794. The Latin insipidus simply means “without flavour” or “tasteless” — the linguistic opposite of mellitus, “honeyed.” The everyday English word insipid, meaning dull or flavourless, comes from exactly the same root, which is why the name, once explained, is easy to remember.

So by the close of the 1700s, physicians possessed a clean two-part scheme. Diabetes mellitus: too much urine, and it is sweet. Diabetes insipidus: too much urine, and it is tasteless. What no one yet knew was the cause of the insipid form — whether it sprang from the kidneys, the nerves, the blood, or somewhere else entirely. Answering that question would take the better part of the next century and a half, and the trail would lead, surprisingly, to a small gland hanging beneath the brain.

Back to Table of Contents

Tracing the Cause to the Pituitary Gland

For most of the nineteenth century the cause of diabetes insipidus remained a genuine mystery. Physicians could describe it vividly — patients drinking and urinating gallons a day, sometimes more than ten litres — but they could not point to the responsible organ. The crucial clue came from a structure that earlier anatomists had dismissed as a vestige: the pituitary gland, a pea-sized body dangling from the base of the brain, just behind the eyes. The breakthroughs were laid down brick by brick, and it is worth crediting the right people for the right steps so that no single “first” is overstated.

The foundation was laid in 1895, when George Oliver and Edward Albert Schäfer (Sharpey-Schafer) in England showed that an extract of the pituitary gland, injected into an animal, sharply raised its blood pressure — the so-called pressor effect. This proved the once-ignored gland secreted something biologically powerful. In 1901, work associated with Magnus and Schäfer demonstrated that extract of the gland’s posterior lobe also had a striking antidiuretic action — it made experimental animals produce less urine. Here, for the first time, was a substance from the pituitary that did the exact opposite of what happens in diabetes insipidus, where urine pours out uncontrollably. The implication was almost impossible to miss: perhaps the disease was caused by a lack of this pituitary substance.

That suspicion hardened around 1912, when clinical observers linked damage to the pituitary region — from tumours, injury, or disease — with the appearance of diabetes insipidus in patients. The pieces were converging: a gland that, in extract form, powerfully reduced urine output; and patients who developed runaway urine output when that same gland was damaged. The obvious next experiment was to give the extract to a patient with the disease and see whether the flood could be stemmed.

Back to Table of Contents

Pituitary Extracts as the First Treatment (1913)

That experiment succeeded in 1913, the year diabetes insipidus first became a treatable disease. Working independently, the Italian investigator A. Farini (with B. Ceccaroni) and the German physician R. von den Velden reported that injecting posterior pituitary extract into patients with diabetes insipidus dramatically reduced their urine output and relieved their desperate thirst. For the first time, a sufferer of this disease could be given something that actually worked — not a cure for the underlying defect, but a genuine replacement for the missing hormonal signal that told the kidneys to conserve water.

This was a profound moment for patients. Before 1913, the diabetes-insipidus patient faced a life ruled by the bathroom and the water jug, with nothing a doctor could offer beyond sympathy. After 1913, the disease joined the small but growing list of conditions that early twentieth-century medicine could control by replacing a substance the body had stopped making — a strategy that, a few years later, would transform diabetes mellitus through the discovery of insulin (1921–1922). The early pituitary extracts were crude and short-acting, and over the following decades they were refined into more practical forms, including a posterior-pituitary powder that patients could insufflate (snuff) into the nose for absorption. These were imperfect, but they kept people alive and comfortable.

The success of the 1913 extracts also settled the scientific argument. If giving back pituitary extract fixed the disease, then the disease was, in essence, a deficiency of something the pituitary normally supplied. The hunt was now on to identify that “something” — to extract it pure, work out its chemical structure, and ideally build it in the laboratory so that patients need not depend on ground-up animal glands. That chemical quest would be crowned, four decades later, by one of the most celebrated achievements of twentieth-century biochemistry.

Back to Table of Contents

Du Vigneaud and the Synthesis of Vasopressin (1954)

The hormone behind the antidiuretic effect was eventually purified and named vasopressin — a name that captures its blood-vessel-constricting (vaso-pressor) action, the very effect Oliver and Schäfer had first seen in 1895. It is the same substance physiologists call antidiuretic hormone (ADH), the two names referring to one molecule viewed through two of its actions: tightening blood vessels, and ordering the kidneys to hold on to water. It is produced in the hypothalamus of the brain and released from the posterior pituitary, and it is the master regulator of the body’s water balance.

The towering figure here is the American biochemist Vincent du Vigneaud (1901–1978) of Cornell University Medical College. Following what he called “the sulfur trail” — a career-long study of sulfur-containing biological compounds — du Vigneaud and his team worked out the exact amino-acid structures of the two posterior-pituitary hormones, vasopressin and the closely related oxytocin, and then achieved their total chemical synthesis. In 1953 they synthesised oxytocin, and in 1954 vasopressin, building these small peptide hormones from scratch in the laboratory. This was the first time anyone had chemically synthesised a polypeptide hormone, proving that such biologically active molecules — previously obtainable only by extraction from glands — could be manufactured and would behave exactly like the natural article. For this landmark work du Vigneaud was awarded the 1955 Nobel Prize in Chemistry, “for his work on biochemically important sulphur compounds, especially for the first synthesis of a polypeptide hormone.”

For diabetes insipidus this changed everything. A deficient hormone whose structure is known and which can be made in a factory is a deficiency that can be reliably and cleanly corrected. Du Vigneaud’s synthesis opened the door to laboratory-made vasopressin and, ultimately, to the engineered analogue desmopressin (DDAVP) — a modified vasopressin tailored to keep the water-conserving action while shedding most of the blood-pressure effect, and to last far longer in the body. Desmopressin, available as a nasal spray, tablet, or injection, remains the cornerstone treatment for central diabetes insipidus to this day, a direct descendant of the molecule du Vigneaud first built in 1954.

Back to Table of Contents

Central vs. Nephrogenic: Two Roads to the Same Symptom

As the hormone story matured through the twentieth century, physicians realised that diabetes insipidus itself comes in two fundamentally different forms — a distinction every bit as important as the original split from diabetes mellitus. The crucial question is: is the problem a shortage of vasopressin, or a failure of the kidneys to obey it?

In central diabetes insipidus (sometimes called cranial or neurogenic DI), the brain and pituitary fail to make or release enough vasopressin. The hormonal “hold the water” signal is simply not being sent. This is the form that pituitary damage causes, and the form the 1913 extracts and modern desmopressin treat — because if the body is short of the hormone, supplying it from outside fixes the problem. In nephrogenic diabetes insipidus, by contrast, the brain makes plenty of vasopressin, but the kidneys cannot respond to it — the signal is sent loud and clear but the receiving tissue is deaf to it. Giving more hormone does little good, because the fault lies downstream in the kidney itself. This form can be inherited, or caused by certain drugs (lithium being a classic example) or by metabolic disturbances.

The brilliantly simple test that separates the two was built directly on du Vigneaud’s synthetic hormone, and it remains conceptually central to diagnosis. First the patient undergoes a carefully supervised water-deprivation test; then they are given vasopressin (historically the preparation Pitressin, later desmopressin). The logic is elegant: if the injected hormone makes the kidneys concentrate the urine and the flood slows, the kidneys clearly can respond — so the original problem was a hormone shortage, meaning central DI. If, however, the patient keeps producing dilute urine even after receiving the hormone, the kidneys are not responding to it at all — meaning nephrogenic DI. In one bedside sequence, the disease declares which of its two forms it is. (Modern laboratories increasingly add a blood test for copeptin, a stable fragment released alongside vasopressin, to make this distinction more precisely, but the underlying central-versus-nephrogenic logic is unchanged.)

Back to Table of Contents

From Hormone to Gene: AVPR2 and Aquaporin-2

The final layer of understanding arrived in the late twentieth century, when molecular biology explained why some kidneys are deaf to vasopressin in the first place — the deepest cause of inherited nephrogenic diabetes insipidus. The discoveries clustered in the 1990s and pinpointed the two molecular components of the kidney’s vasopressin-response machinery.

The first is the vasopressin V2 receptor, the docking site on kidney cells that vasopressin must bind to in order to issue its “conserve water” order. Its gene, AVPR2, sits on the X chromosome, and in 1992 researchers showed that mutations in AVPR2 cause the most common, X-linked form of hereditary nephrogenic DI — which is why it classically affects boys. When the receptor is faulty, vasopressin arrives at the kidney and finds no working lock to open; the signal cannot get in. The second component is the molecular “water gate” the receptor is supposed to open: a water-channel protein called aquaporin-2, which the kidney inserts into its cell membranes to let water be reabsorbed. In 1994, Deen and colleagues showed that mutations in the aquaporin-2 gene (AQP2) cause a rarer, non–X-linked form of nephrogenic DI. Here the receptor may work, but the water channel it controls is broken, so water still cannot be pulled back and is lost as a torrent of dilute urine.

With these discoveries, the entire chain finally lay exposed end to end: vasopressin is made in the brain, released by the pituitary, travels to the kidney, binds the V2 receptor, and triggers aquaporin-2 channels to reclaim water. A failure at any link — too little hormone (central DI), a broken receptor or a broken channel (nephrogenic DI) — produces the same ancient symptom Aretaeus described nearly two thousand years ago: a body that cannot hold its water. The tasteless flood that once could only be identified by a physician’s tongue was now traceable to specific molecules and specific genes.

Back to Table of Contents

A New Name: Arginine Vasopressin Deficiency and Resistance (2022)

History has a striking final chapter, and it returns to the very confusion this page opened with. For three and a half centuries, the name diabetes insipidus has caused trouble precisely because it contains the word diabetes — a word the public, and even many clinicians, instantly associate with sugar diabetes. Patients with diabetes insipidus are repeatedly mistaken for diabetes-mellitus patients; in some documented hospital cases this confusion led to the wrong management and to vasopressin treatment being dangerously withheld, with serious harm. In a survey of affected patients, about 85% wanted the name changed, largely because of frightening experiences with health professionals who did not understand their condition.

In response, a coalition of the world’s leading hormone-medicine bodies acted. In 2022, the Working Group for Renaming Diabetes Insipidus issued a position statement proposing that the disease be renamed to reflect its real cause, the vasopressin (AVP) system, rather than its superficial resemblance to sugar diabetes. The proposed new terms are arginine vasopressin deficiency (AVP-D) for the central form (too little hormone) and arginine vasopressin resistance (AVP-R) for the nephrogenic form (kidneys resistant to the hormone). The proposal was endorsed by a remarkable line-up of professional societies, including the Endocrine Society, the European Society of Endocrinology, the Pituitary Society, the Society for Endocrinology, the European Society for Paediatric Endocrinology, the Endocrine Society of Australia, the Brazilian Society of Endocrinology and Metabolism, and the Japan Endocrine Society.

It is worth being clear about the present status, since adoption is a process rather than an instant switch: the older name diabetes insipidus remains in very wide everyday use among patients, clinicians, and existing medical literature, and you will encounter it everywhere — including on this site — while the new AVP-D / AVP-R terminology is gradually taken up. The two names refer to exactly the same conditions. In a fitting twist, the change finally completes the work Thomas Willis began in the 1670s: where he separated the two “diabetes” diseases by taste, the modern endocrine community is separating them by name, so that the tasteless water-balance disease is no longer shackled to the sugar disease it was never truly related to.

Back to Table of Contents

Research Papers and References

The references below combine peer-reviewed historical reviews and primary modern papers on the discovery of vasopressin and the genetics of diabetes insipidus with curated PubMed topic-search links. Historical primary texts — Thomas Willis’s The Diabetes, or Pissing Evil (1674) and the eighteenth-century writings of Cullen and Frank — are named in the article as historical sources rather than as modern citations. Each link opens at its source (PubMed, PMC, or the Nobel Foundation) in a new tab.

Lindholm J. Diabetes insipidus: historical aspects. Pituitary. 2004;7(1):33-38. — doi:10.1023/B:PITU.0000044633.52516.e1
Valenti G, Tamma G. History of diabetes insipidus. Giornale Italiano di Nefrologia. 2016;33(S66). — PubMed 26913870
Eknoyan G. A history of diabetes insipidus: paving the road to internal water balance. American Journal of Kidney Diseases. 2010;56(6):1175-1183. — doi:10.1053/j.ajkd.2010.08.012
du Vigneaud V, Ressler C, Swan JM, Roberts CW, Katsoyannis PG, Gordon S. The synthesis of an octapeptide amide with the hormonal activity of oxytocin. Journal of the American Chemical Society. 1953;75(19):4879-4880. — doi:10.1021/ja01115a553
du Vigneaud V, Gish DT, Katsoyannis PG. A synthetic preparation possessing biological properties associated with arginine-vasopressin. Journal of the American Chemical Society. 1954;76(18):4751-4752. — doi:10.1021/ja01647a089
The Nobel Prize in Chemistry 1955 — Vincent du Vigneaud (official prize record and biography). — NobelPrize.org — du Vigneaud 1955
Arima H, Cheetham T, Christ-Crain M, et al.; Working Group for Renaming Diabetes Insipidus. Changing the name of diabetes insipidus: a position statement of the Working Group for Renaming Diabetes Insipidus. Endocrine Connections / J Clin Endocrinol Metab / Eur J Endocrinol (joint). 2022. — PMC9578068
Christ-Crain M, Bichet DG, Fenske WK, et al. Diabetes insipidus. Nature Reviews Disease Primers. 2019;5(1):54. — doi:10.1038/s41572-019-0103-2
Bichet DG. Genetics and diagnosis of central and nephrogenic diabetes insipidus. — PMC5159460
Deen PM, Verdijk MA, Knoers NV, et al. Requirement of human renal water channel aquaporin-2 for vasopressin-dependent concentration of urine. Science. 1994;264(5155):92-95. — doi:10.1126/science.8140421
Pituitary posterior-lobe extract in the treatment of diabetes insipidus (historical, early antidiuretic therapy, 1913 era). — PubMed: pituitary extract history of DI treatment
Mutations in the vasopressin V2 receptor gene (AVPR2) in X-linked nephrogenic diabetes insipidus. — PubMed: AVPR2 and nephrogenic DI
Central versus nephrogenic diabetes insipidus — diagnosis and the water-deprivation / vasopressin test. — PubMed: central vs nephrogenic DI diagnosis
Thomas Willis and the early history of diabetes — sweet versus insipid urine. — PubMed: Thomas Willis and the history of diabetes

External Authoritative Resources

Back to Table of Contents

Connections

Back to Table of Contents