Share this article on:


Rowland, Lewis P. MD

Infobytes: Book Review

Dr. Rowland, Editor-in-Chief of Neurology Today, is Professor of Neurology at Columbia University Medical Center in New York City.

Back From the Brink: How Crises Spur Doctors to New Discoveries about the Brain by Edward J. Sylvester

296 Pages • Dana Press • 2004

Back from the Brink: How Crises Spur Doctors to New Discoveries about the Brain by Edward J. Sylvester is a paean to the modern Neuro-ICU. Sylvester, an experienced journalist, enthusiastically describes how the intensivists save a drug addict with multi-organ failure and how they come close to saving a man with a ruptured aneurysm, “close” but not fully recovered.

Sylvester focuses on two leaders, Marek Mirski at Johns Hopkins and Stephan Mayer at the Columbia University Medical Center. Both are overworked by their institutions and are committed to using the ICU as a research center with the goal of improving patient care. Both are in the process of expanding. Both are in the second generation of leaders. Daniel Hanley started the unit at Hopkins and Matthew Fink launched the Columbia ICU. Another unit was started at the Massachusetts General Hospital by Alan Ropper, who was succeeded by Walter Koroshetz.

Sylvester tells stories of status epilepticus, brain edema, cardiac arrest, and an astrocytoma discovered by accident. The stories are not limited to patients but also pertain to faculty. The writing is lively, accurate, informative, and engaging.

Sylvester states that Neuro-ICU made its debut in the mid-80s, which is technically true. However, the Columbia story starts earlier and so also do the others – a tale I can attest to from personal experience that actually began in the early 1950s.

When I was a resident from 1950 to 1953, my attention was drawn to myasthenia gravis. This became the subject of my first clinical paper, a report of 25 postmortem examinations. The fatality rate then was 33 percent, an amazing number because almost no one is allowed to die of myasthenia these days, not even octogenarians with other concomitant life-threatening diseases.

The problem in the old days is starkly defined in retrospect. We did not really know how to care for patients who had trouble breathing from neuromuscular causes. We depended on tank respirators that created external negative pressure to suck air into the lungs. The US had made a huge investment in these devices for the treatment of paralytic poliomyelitis, which came in vast epidemics every summer.



The last polio epidemics in industrialized countries occurred in the mid-50s, and one hit Denmark most severely. Copenhagen soon ran out of tank respirators and had to borrow from other countries in Europe, which were having their own problems. Before long there were none to loan. So the Danes did tracheostomies and medical students were drafted to push Ambu bags day and night around the clock. But they ran out of medical students, too. So someone had a bright idea: the “mechanical medical student.” They rigged up simple electrical pumps to compress the bags and force air into the patient's lungs.

It worked – and that was the start of modern positive-pressure respiration. The advantages over the tank were immediately apparent. Nurses could actually get to the patient directly, not through awkward portholes in the steel walls. Patients could be turned for pulmonary care or comfort. Why hadn't that been done earlier? It was another example of an institutionalized error created by the hubris of authorities; they were convinced that positive pressure in the lungs would impede venous return to the heart. It did, but not enough to prevent the imminent revolution in pulmonary care.

Three years later, in 1956, I was a research fellow in England, learning to use plethysmography for measuring muscle blood flow. I would do for muscle what Seymour Kety and Louis Sokoloff had done for the brain. That metabolic approach did not succeed but I made it a point to visit the first Neuro-ICUs in England, one at the National Hospital at Queen Square under Roger Gilliatt and the other at Oxford under John Spaulding. I was much impressed by the advantages of positive-pressure respiration and also by the special units where nurses and physicians could gain the necessary skills.

When I returned to New York I went to see our Chief, Houston Merritt, to report my new enthusiasms. But not much happened. We went on using tank respirators, not just at Columbia but throughout the country. I concluded that the investment in tanks must have been too great to change. The mortality rate for patients with myasthenia in tank respirators was a whopping 50 percent. It took about a decade before positive-pressure ventilators became available commercially and the tanks gradually disappeared.

Back to Top | Article Outline


In that decade, we gradually developed a team interested in neuromuscular disease. We had a weekly conference and I personally saw every patient in myasthenic crisis. There was almost always at least one in the hospital and, often, two or more. We had positive-pressure respirators but not a special unit. We moved our patients to the neurosurgery recovery room, where the nurses feared that our infected tracheostomies would contaminate the craniotomy wounds. They tried to push us out as soon as they could.

At one point, while still a junior faculty member and with the approval of Houston Merritt, I joined forces with the heads of Urology and Anesthesiology to develop plans for an ICU that would combine renal dialysis and respiratory support. When the three of us went to a meeting of the Medical Board, they did the talking and I, much the junior, listened. The President of the Medical Board was the all-powerful Chairman of Medicine, and I was astonished to hear him berate his senior colleagues for a plan that I also deemed necessary for patient care. But he was the Professor of Medicine and he was training “complete physicians;” our plan, he said, “would fragment care.”

By 1967, I had crawled up the academic ladder and was invited to be Chairman at the University of Pennsylvania. When I left for Penn, Columbia-Presbyterian still had no adult ICUs. Plans for a Coronary Care Unit were just being implemented. At Penn, I found three flourishing ICUs: Medical, Surgical, and Respiratory. The Respiratory Unit was administered by a trio: Pulmology, Anesthesiology, and Neurology. In contrast to my earlier experience, the nurses loved our patients, because they recovered from the acute effects of Guillain-Barré syndrome or myasthenic crisis, with many happy endings that contrasted to the ultimately bleak outlook for pulmonary patients with emphysema.

In 1973, I returned to Columbia as Chairman. The dean asked me what we needed. At the top of my list, I said, was an ICU. “No problem,” he said. “We had promised one to your predecessor five years ago and it is in the works.” The works took another decade; we opened the unit in 1983.

By that time, we had a wonderful candidate to lead us. Matthew Fink had been our Chief Resident. He had also been a Chief Resident in Medicine at Boston University – twin technical skills that would be valuable in his new role, to say nothing of his evident leadership. When he started, the neurosurgeons did not trust neurology residents to care for their patients but, in a few years, they had come to admire Fink and the house staffs shared responsibility.

In 1995, Fink's reputation had grown. He was appointed Chairman of Neurology at the Beth Israel Hospital in New York and was soon President of that hospital system. In 1999, we conducted a national search for his successor and recognized the talents of Fink's own student, Stephan Mayer; he has become a true leader in the field and Neuro-ICUs are now the norm for all teaching hospitals.

Sylvester was right. Neuro-ICUs appeared in the mid-80s, but not without the groundwork of three earlier decades.

©2005 American Academy of Neurology