The Laws of Medicine

To Thomas Bayes (1702–1761), who saw uncertainty with such certainty

“Are you planning to follow a career in Magical Laws, Miss Granger?” asked Scrimgeour.

“No, I'm not,” retorted Hermione. “I'm hoping to do some good in the world!”

J. K. Rowling

The learned men of former ages employed a great part of their time and thoughts searching out the hidden causes of distemper, were curious in imagining the secret workmanship of nature and . . . putting all these fancies together, fashioned to themselves systems and hypotheses [that] diverted their enquiries from the true and advantageous knowledge of things.

John Locke

Y
ears ago, as a medical student in Boston, I watched a senior surgeon operate on a woman. The surgeon, call him Dr. Castle, was a legend among the surgical residents. About six feet tall, with an imposing, formal manner that made the trainees quake in their clogs, he spoke in a slow, nasal tone that carried the distinct drawl of the South. There was something tensile in his build—more steel wire than iron girder—as if his physique had been built to illustrate the difference between stamina and strength. He began rounds at five every morning, then moved down to the operating theaters in the basement by six fifteen, and worked through the day into the early evening. He spent the weekends sailing near Scituate in a one-mast sloop that he had nicknamed
The Knife
.

The residents worshipped Castle not only for the precision of his technique, but also because of the quality of his teaching. Other surgeons may have been kinder, gentler instructors, but the key to Castle's teaching method was supreme self-confidence. He was so technically adept at surgery—so masterful at his craft—that he allowed the students to do most of the operating, knowing that he could anticipate their mistakes or correct them swiftly after. If a resident nicked an artery during an operation, a lesser surgeon might step in nervously to seal the bleeding vessel. Castle would step back and fold his arms, look quizzically at the resident, and wait for him or her to react. If the stitch came too late, Castle's hand would reach out, with the speed and precision of a falcon's talon, to pinch off the bleeding vessel, and he would
stitch it himself, shaking his head, as if mumbling to himself, “Too little, too late.” I have never seen senior residents in surgery, grown men and women with six or eight years of operating experience, so deflated by the swaying of a human head.

The case that morning was a woman in her fifties with a modest-size tumor in her lower intestine. We were scheduled to begin at six fifteen, as usual, but the resident assigned to the case had called in sick. A new resident was paged urgently from the wards, and he came quickly into the operating room, tugging his gloves on. Castle walked up to the CAT scans hung on the fluorescent lightbox, studied them silently for a while, then moved his head ever so slightly, signaling the first incision. There was a reverential moment as he stretched out his right hand and the nurse handed him the scalpel. The surgery began without incident.

About half an hour later, the operation was still under perfect control. Some surgeons liked to blast music in the operating room—rock and roll and Brahms were common choices—but Castle preferred silence. The resident was working fast and doing well. The only advice that Castle had offered was to increase the size of the incision to fully expose the inner abdomen. “If you can't name it, you can't cut it,” he said.

But then the case took a quick turn. As the resident reached down to cut the tumor out of the body, the blood vessels surrounding it began to leak. At first, there was only a trickle, and then a few more spurts. In a few minutes about a teaspoon of blood had run into in the surgical field, obscuring the view. The carefully exposed tissues were submerged in
a crimson flood. Castle stood by the side, his hands folded, watching.

The resident was clearly flustered. I watched a pool of sweat forming over his brow, mirroring the pool of blood in front of him. “Does this patient have a known bleeding disorder?” he asked, his desperation mounting. “Was she on a blood thinner?” Usually he would have studied the chart the night before and known all the answers—but he had hurriedly been assigned to the case.

“What if you didn't know?” said Castle. “What if I told you that I didn't know?” His hands had already reached into the woman's abdomen and closed the vessels shut. The patient was safe, but the resident looked devastated.

But then, it was as if a tiny bolt of knowledge had moved, like an electric arc, between Castle and his resident. The resident modified his approach. He walked over, past the surgical drapes above the woman's head, to confer with the anesthesiologist. He confirmed that the anesthesia was adequate and the patient was safely sedated. Then he returned to the surgical field and blotted out the remnant blood with some gauze. Now, he began cutting around the blood vessels when he could, charting their course with the tip of his Babcock forceps, or separating them with his fingers with exquisite delicacy, as if polishing the strings of a Stradivarius. Each time he neared a blood vessel, he turned the blade of the scalpel to its flat side and dissected with his hands, or moved farther out, leaving the vessel untouched. It took significantly longer, but there was no further bleeding. An hour later, with Castle nodding approvingly, the resident closed the incision. The tumor was out.

We walked out of the operating room in silence. “You might want to go and check her chart now,” Castle said. There was a note of tenderness in his characteristic nasal twang. “It's easy to make perfect decisions with perfect information. Medicine asks you to make perfect decisions with imperfect information.”

....

This book is about information, imperfection, uncertainty, and the future of medicine. When I began medical school in the fall of 1995, the curriculum seemed perfectly congruent to the requirements of the discipline: I studied cell biology, anatomy, physiology, pathology, and pharmacology. By the end of the four years, I could list the five branches of the facial nerve, the chemical reactions that metabolize proteins in cells, and parts of the human body that I did not even know I possessed. I felt poised to begin practicing real medicine.

But as I advanced through my training—becoming an intern, then a resident, a fellow in oncology, and then an attending doctor treating patients with cancer—I found that a crucial piece of my education was missing. Yes, I needed the principles of cell biology to understand why, say, a platelet transfusion lasts only two weeks in most patients (platelets live in the body for only about two weeks). Anatomy helped me recall why a man had woken up from a surgical procedure with his entire lower body paralyzed (an unusual artery that supplies the lower spinal cord had become blocked by a clot, resulting in a “stroke” of the spinal cord, not the brain). An equation from pharmacology reminded me why one antibiotic was dosed four times a day while its close molecular cousin was given only once a day (the two chemicals decay at different rates in the body).

But all this information could, I soon realized, be looked up in a book or found by a single click on the Web. The information that was missing was what to
do
with information—especially when the data was imperfect, incomplete, or uncertain. Was it appropriate to treat a forty-year-old woman with acute leukemia
with an aggressive bone-marrow transplant if her health was declining rapidly? At first glance, textbooks and published clinical trials gave you an answer. In this instance standard wisdom held that patients with declining health and performance should not be given a transplant. But what if that answer did not apply to
this
woman, with
this
history, in
this
particular crisis? What if the leukemia itself was causing the rapid decline? If she asked about her prognosis, I could certainly quote a survival rate pulled from a trial—but what if she was an outlier?

My medical education had taught me plenty of facts, but little about the spaces that live between facts. I could write a thesis on the physiology of vision. But I had no way to look through the fabric of confabulation spun by a man with severe lung disease who was prescribed “home oxygen,” but gave a false address out of embarrassment because he had no “home.” (The next morning, I got an irate phone call from the company that had attempted delivery of three canisters—to a Boston storefront that sold auto parts.)

I had never expected medicine to be such a lawless, uncertain world. I wondered if the compulsive naming of parts, diseases, and chemical reactions—frenulum, otitis, glycolysis—was a mechanism invented by doctors to defend themselves against a largely unknowable sphere of knowledge. The profusion of facts obscured a deeper and more significant problem: the reconciliation between knowledge (certain, fixed, perfect, concrete) and clinical wisdom (uncertain, fluid, imperfect, abstract).

This book began as a means for me to discover tools that might guide me through a reconciliation between these two
spheres of knowledge. The “laws of medicine,” as I describe them in this book, are really laws of uncertainty, imprecision, and incompleteness. They apply equally to all disciplines of knowledge where these forces come into play. They are laws of imperfection.

The stories in this book are of real people and cases, but I have changed names and identities and altered some contexts and diagnoses. The conversations were not recorded verbatim, but have been paraphrased from my memory. Some situations, tests, and trials have also been changed to maintain the anonymity of patients and doctors.

In
Harry Potter
, that philosophical treatise disguised as a children's book, a teacher of wizardry asks Hermione Granger, the young witch-in-training, whether she wishes to learn the Magical Laws to pursue a career in magic. “No,” says Granger. She wishes to learn the laws so that she can do some good in the world. For Granger, magical laws do not exist to perpetuate magic. They exist as tools to interpret the world.

....

I
n the winter of 2000, during the first year of my medical residency, I lived in a one-room apartment facing a park, a few steps from the train station at Harvard Square.

Lived
is a euphemism. I was on call every third night at the hospital—awake the whole night, admitting patients to the medical wards, writing notes, performing procedures, or caring for the acutely ill in intensive care units. The next day—
postcall
—was usually spent in a dull haze on my futon, catching up on lost sleep. The third day we named
flex
, for “flexible.” Rounds were usually done by six in the evening—and the four or five hours of heady wakefulness that remained were among the most precious and private of all my possessions. I ran a three-mile circuit around the frozen Charles River as if my life depended on it, made coffee on a sputtering Keurig, and stared vacantly at the snowdrifts through my window, ruminating on the cases that I had seen that week. By the end of the first six months, I had witnessed more than a dozen deaths, including that of a young man, no older than I, who died of organ failure while awaiting a heart transplant.