08Chapter 6. Diabetes as a Treatment for Madness: Manfred Sakel’s Comas

In 1927 Berlin, a young neurologist named Manfred Sakel (1900–1957), working at the private Lichtenfeld Clinic, found himself facing a difficult problem. He had been put in charge of a group of heroin addicts, many of whom developed severe withdrawal states when the drug was taken away — agitation, aggression, bouts of extreme restlessness. The usual sedatives simply weren’t up to the job.

Sakel bet on a new weapon. A tool that had only entered medicine’s arsenal a few years prior. Insulin. Discovered in Toronto in 1921, by the mid-1920s it had already brought diabetes to heel. Its mechanism was simple: it crashed blood glucose levels. For diabetics, this meant salvation. But what if a healthy person received an injection? What lurked beyond therapeutic thresholds? Administered in sufficient doses, insulin could drop sugar levels so violently that the brain would shut down. First came fine tremors. Shaking. Then icy sweat broke out. Reason dissolved into delirious confusion. Then—nothing. A deep hypoglycemic coma.

Sakel observed that small doses of insulin reduced agitation and induced sedation in his heroin-addicted patients. He started increasing the dosage—until the patient would slip into a genuine hypoglycemic coma: a deep unconscious state marked by muscle rigidity, automatisms, and sometimes convulsions. Then, an intravenous glucose drip was administered, and the patient gradually regained consciousness.

Sakel wrote that after such doses, his addicts became “calm, compliant, more amenable to treatment.” This observation simmered for years. Then in 1933, after moving to Vienna and working at a psychiatric clinic, he took the decisive step: applying this method not to addicts, but to schizophrenia patients.

This is how it began. A patient received insulin—dose after dose, step by step—until their blood sugar plunged so low they slipped into deep coma. Minutes. Sometimes an hour. Sometimes two. All this time they hovered between life and d••th, while doctors stood by and waited. Then—glucose. Through a tube or IV. And the patient returned. Opened their eyes. Regained consciousness. But one coma wasn’t treatment. It was just the beginning. A full course meant thirty, forty, fifty comas. Every other day. Each one a separate d••th to return from. The entire course stretched over two months. Two months between life and d••th. Two months on the edge.

What happened during the coma? The brain, deprived of glucose, ceased functioning properly. Automatic movements emerged, sometimes convulsions, sometimes epileptic seizures. The patient lay utterly helpless. Any mistake—too deep a coma, delayed glucose—could mean irreversible brain damage or d••th.

The mortality rate from insulin coma therapy, according to various reports, ranged from one to five percent. That was per course. In other words, for every twentieth to hundredth patient who underwent a course, the treatment ended in death.

But Sakel’s reports contained astonishing statistics. He claimed that up to 88 percent of his schizophrenia patients showed improvement, and many achieved full remission. He called his method “glucohypoglycemic therapy.” The medical community was stunned.

In 1937, Switzerland hosted an international congress on insulin coma therapy. By then, the method had spread to dozens of countries: Harvard, Mayo Clinic, Salpêtrière, Bekhterev Institute in Leningrad. “Insulin wards” appeared across the developed world, where hundreds endured weeks-long coma regimens.

Sakel emigrated to the United States, worked in New York clinics, and became one of the stars of American psychiatry. He was tipped for a Nobel Prize — much as Wagner-Jauregg had won one for malarial therapy.

But.

However, when researchers in the mid-20th century began conducting rigorous comparative trials of insulin coma therapy—comparing it to control groups receiving standard care without induced comas—the results painted a completely different picture.

London, 1953. The study meant to settle the matter. Harold Bourne and colleagues randomly assigned patients to two groups. One received insulin coma therapy. The other got sedative barbiturates—no coma, just sleep. They compared results. Then came the unforeseen. The outcomes were nearly identical. Improvements that Sakel and followers attributed specifically to comas occurred just as often in schizophrenia patients who never entered coma at all. So was the coma irrelevant? Then what—or who—was doing the healing?

This meant something simple and terrifying. Sakel’s 88% success rate was a methodological artifact. Sakel worked without control groups. He compared his patients not with others treated by conventional methods in parallel, but with his own subjective impression of “what would have happened had I not treated them.” Such comparisons cannot be made objectively. Many patients with acute schizophrenia improve within weeks due to the spontaneous course of the illness — especially during first episodes — even without any treatment. Those who deteriorated were categorized in Sakel’s reports as “non-responders,” while those who improved were counted as successes.

This is a classic methodological error that many modern pseudo-therapies still fall victim to: gather impressions from clients, tally “improvements,” lack a control group—and then declare that the method is effective.

By the mid-1950s, skepticism was growing in global psychiatry. Then came a turning point in 1952, when French doctors Jean Delay and Pierre Deniker published the first results using chlorpromazine — the first true antipsychotic. Within a few years, everything changed. Insulin coma therapy rapidly lost ground. By the late 1960s, it had vanished from most Western clinics.

In the USSR, as with sulfosin therapy, things were different. Insulin coma treatment persisted for decades. When I trained at the Bekhterev Institute between 2001 and 2003, there were still wards administering insulin therapy — though by then, they used sub-coma doses, inducing hypoglycemia without full unconsciousness. This was considered a “gentler method.”

One thought still haunts me whenever I remember those insulin wards.

What made patients “better” after undergoing a course of thirty to fifty comas? Definitely not the coma itself—that much is clear today. So what was it?

First: regression. A patient who spent five weeks in a semi-comatose state, dehydrated and emaciated, emerged mentally and physically depleted. They became passive, stopped protesting, making demands, or disturbing staff. This passivity could easily be mistaken for “improvement,” especially in an environment where psychiatric staff labeled patients as “agitated” or “anxious.”

Second: attention. In insulin wards, patients received far more attention than in regular units. Nurses checked their pulse, blood pressure, and glucose levels multiple times a day. Staff were always nearby, talking to them. One of the most powerful therapeutic factors in any medical field is simply having someone pay attention to you.

Third: Spontaneous Remission. Schizophrenia is a condition characterized by episodic progression. Many patients naturally improve during the first few months of the acute phase. If this coincided with the administration of coma therapy, the improvement was attributed to the treatment.

Together, all of this produced the “88% improvement” statistic, which struck those who witnessed it firsthand as nothing short of a miracle. The myth took hold. The method was given time to reach a stage where, in certain clinics, it had effectively replaced the whole of psychiatry.

But we must honestly acknowledge another truth. Within the confines of psychiatry as it existed at that time, Sakel’s insulin coma therapy was neither madness nor a crime. It represented an attempt to provide relief in circumstances where other methods had failed and no better alternatives existed.

Sakel was an educated physician. He was not a charlatan in the way modern “family constellation” practitioners or “ThetaHealing masters” are. He genuinely believed in his method. He published his results in peer-reviewed journals. He engaged in scientific debates. He listened to critics and responded to them.

Herein lies the abyss. The chasm that forever separates him from modern esotericists. Yes, his results were an artifact. A methodological illusion. But he remained within science. He played by its rules — albeit with a fatal flaw in his methods. And precisely because of this, science was able to dismantle his system. Test it. Refute it. Discard it. A follower of Hubbard or the creator of a commercial “unique method” would never allow their brainchild to face the crucible of placebo-controlled trials. It would annihilate their business. It would obliterate their entire commercial structure. Sakel — allowed it. Sakel — lost. And history accepted the outcome.

This distinction is crucial. Scientific psychiatry makes mistakes and corrects them. Pseudoscientific systems don’t correct themselves, because they never make mistakes—within their own framework, they are never wrong.

But none of this absolves the doctors of my generation in my country of one great failure: why did insulin coma therapy linger in post-Soviet psychiatry for so long after the rest of the world had abandoned it?

The answer is the same as with sulfosin. Cultural inertia. Textbooks written in Soviet times were never updated. Seasoned specialists taught newcomers only what they themselves knew. Professional debate? Nonexistent. A young doctor couldn’t raise uncomfortable questions — hospital walls absorbed nothing but silent compliance. And there was a simple administrative reality. This approach required no new medicines, no ward renovations, no staff retraining. Insulin was available. Glucose was stocked. Syringes lay ready on tables. Beds stood lined in rows. An inexpensive method—a crucial factor in an underfunded healthcare system.

But saving money at the system level came at a colossal cost to every individual patient on whom these methods were used.

Today, Sakel’s insulin coma therapy is nearly obsolete. Its historical purpose was singular: to prove schizophrenia wasn’t a death sentence. Before it, most psychiatrists viewed schizophrenia as an irrevocably worsening catastrophe. Afterwards—even if only as an illusion—it seemed dramatic improvement was possible.

The illusion shattered when true antipsychotics entered the scene. Their arrival instantly rendered insulin comas pointless. Yet that mirage had already served its covert purpose. It gave psychiatry hope. It proved attempts could be made, searches undertaken, optimism warranted. Without that seismic emotional shift, would the generation of researchers who revolutionized psychiatry in the 1950s have emerged? Unlikely. People seldom pour energy into hopeless causes.

A small but important caveat here. Not every mistaken practice in medical history is purely harmful. Some serve as transitional stages, preparing the ground for what comes next. Insulin coma therapy is just such a case.

After the formula. A grain of truth: after a hypoglycemic coma, patients looked calmer. An overgeneralization: “coma cures schizophrenia.” The method: dozens of comas per course—no control group. The industry: insulin wards around the world, Sakel’s global fame, “88% improvement.” The victims: one in fifty—dead; the rest—those whose “improvement” was exhaustion.

Other practices of the same era were far more dramatic. In the next chapter, we will encounter the most notorious—and the darkest—offshoot of this age of exploration. Lobotomy—a method that would earn a Nobel Prize and leave behind the most harrowing legacy of that period.