How Doctors Think (22 page)

I did not sign the paper. I felt dazed. I had called in numerous favors to get to Dr. C, and he had flown by without pausing to share any of his reputed brilliance. Pam continued to query the resident. If Dr. C decided the cysts in the bones accounted for the pain and swelling, what kind of grafts would he insert?

"No, we don't do bone grafts here," the resident said. "Here, we fuse the bones."

Pam and I looked at each other knowingly. We had both trained at the Massachusetts General Hospital, where at times in a complex situation the rationale given trainees for choosing one therapy over another was that an eminent MGH physician "did it that way." It was received wisdom, handed down as if from the heavens. When we left Boston and went to the University of California at Los Angeles, we found another kind of received wisdom. Again, a distinguished physician on the UCLA staff had his personal approach to the same complex clinical problem, but to our surprise it could be very different from the one at the MGH. Nevertheless, the UCLA strategy was spoken of just as reverently, as if it too had descended from celestial heights.

"I really would like to hear what Dr. C thinks is going on," I said to the resident. "I don't know if he realizes that we are both doctors."

The resident said that he would try to bring Dr. C back to the room. Twenty minutes later, Dr. C returned. "Nice to meet you," he said, not exactly calmly, but not at the Gatling-gun pace of the earlier visit. Dr. C began to list the names of doctors he knew who might have crossed paths with me during my career in Boston and Los Angeles. It turned out, not unexpectedly, that we knew several in common.

Pam asked what he thought was the leading diagnosis. "Chondrocalcinosis," Dr. C replied. Chondrocalcinosis is also called pseudogout. Calcium crystals form deposits in the cartilage, making the normally pliable tissue stiff and inflamed. The crystals also float in the fluid of the joint space.

"Wouldn't you see the calcium deposits on the x-rays?" Pam asked.

"There are cases where the x-rays are negative," Dr. C replied.

"And the bone cysts?"

Dr. C said again that he would "figure them out" during the arthroscopy.

Dr. C started to fidget in his chair and then took my left hand and grasped it as a sign of concluding the contact. "My resident will set up the procedure. Don't worry."

But I was worried. Pam was worried. And both of us were disheartened. We had come with great expectations, and everything about Dr. C deflated them. Many years before, when I had serious back pain from a sports injury, the surgeons said they would explore my spine and "figure it out." Out of frustration, I had impulsively opted for the procedure. They ended up fusing the vertebrae. It left me debilitated. In hindsight, I blamed myself more than the surgeons: I had pressed them for a solution when, in fact, none was apparent because the cause of the pain was obscure. That catastrophe had chastened me. It felt like déjà vu with Dr. C.

But Dr. C was world-renowned, the featured speaker at international conferences. So I went to a standard medical textbook and reviewed the chapter on chondrocalcinosis. Nothing in any of my prior tests suggested chondrocalcinosis. If the x-rays don't show the calcium deposits, the easiest way to find the crystals is to withdraw some fluid from the joint with a small needle—not by performing an arthroscopy. And treatment of chondrocalcinosis involved anti-inflammatory drugs like naproxen or steroid injections into the joint—both of which I had already tried, to no effect. Dr. Light echoed these thoughts. Chondrocalcinosis made no sense. "If you think someone has chondrocalcinosis, they don't need arthroscopy. They need a strong anti-inflammatory medication like indomethacin." Dr. C had offered a diagnosis that, while not invented, like Dr. A's hyperreactive synovium, was nevertheless inventive. I decided to do nothing.

Nearly a year passed. I did not use my right hand much. Instead of writing, I switched to using a dictating machine. I avoided the computer entirely. Occasionally, something trivial, like swimming a few extra laps, followed by writing three or four checks to pay bills, would cause a flare-up. My hand would turn red, swell, and become excruciatingly painful. I would apply ice, support it with a splint, and after a few days the inflammation would subside.

A new young hand surgeon whom I will call Dr. D came to Boston, and the word among the senior staff was that he was a hotshot. I was curious and scheduled an appointment with him. Dr. D had a warm, affable manner and listened intently as I recounted the sequence of traumatic events to the hand and the episodic flare-ups. He surprised me by examining not only my right hand but also my left, and then he said that he wanted x-rays of both hands, not only when they were stationary, but when I flexed them as if gripping something tightly. This was the first time anyone had ever paid attention to the left wrist or tried to picture the bones of my hands during a maneuver.

"As I suspected," Dr. D said without a whiff of arrogance in his tone. He put the x-rays on a light box and showed me how the space between the scaphoid and lunate bones widened when my right hand was in a gripping position; the left hand showed no widening of the joint.

"I think the ligament between the scaphoid and lunate is partially torn, or at least not functioning well," he said. The reason I had pain with even minor stress on my right hand was that the lax or torn ligament caused friction between the bones. He went on to explain that there also could be channels from the cysts into the joint, so that they functioned like lakes with thin canals: as the fluid in the cysts came under pressure, the liquid would be squeezed out through the canals into the joint. This set off the inflammation.

Dr. D's scenario made sense to me, but the MRI had shown neither problems with the ligament nor channels from the cysts. Dr. D replied that despite the MRI, he would bet the ligament was abnormal and that connections existed between the cysts and the joints. Doctors relied too much on such sophisticated scans, he continued, so sometimes you had to discount their findings if they were out of sync with the clinical picture. Filling the cysts with bone grafts would probably not do much good in the long run without repairing the ligament, because the loose joint would continue to generate friction, causing pain. Dr. D proposed taking bone grafts from my hip, filling in the cysts, and repairing the ligament. As to the other abnormalities on the MRI scan that Dr. B wanted to fix—the tendon to the pinkie and the tiny cyst in the other bone—Dr. D was reluctant to operate on them. He said the wear and tear on the hands of a man in his fifties who banged on a computer and played sports and was something of a klutz in elevators could cause such findings on an MRI, but trying to fix them might do more harm than good.

Dr. D seemed sober and independent in his thinking, not bowing to technology when it clashed with a patient's history and physical exam. But was he right? I decided to assume for a moment that he was, and asked him how many times he had performed the kind of operation he was proposing. He paused and said, "Once." Then he elaborated, saying that he had done it several times with supervision, but only a single time by himself. He was just at the beginning of his career.

"That's the struggle when patients are having pain," Dr. Light said after hearing Dr. D's thoughts and contrasting them with Dr. B's. "You can see many things on an MRI, but nothing that's clearly responsible for the symptoms. So you begin to go around and around. The hateful part of MRIs—I mean they can be a wonderful technology—but they find abnormalities in everybody. More often than not, I am stuck trying to figure out whether the MRI abnormality is responsible for the pain. That is the really hard part."

The key, Dr. Light continued, is for "everything to add up—the patient's symptoms, the findings on physical examination, what appears meaningful on the MRI or other x-rays. It has to come together and form a coherent picture." In effect he was describing pattern recognition, and saying that if a clear pattern is not apparent, the surgeon is in a quandary. "Picking up a scalpel and cutting can be just the wrong thing." This, though, was what Drs. A, B, and C, without recognizing a coherent and consistent pattern, were set to do.

"I have made the diagnosis Dr. D made, of dynamic scaphoid-lunate instability," Light told me, affixing the technical term to the problem of a lax ligament that causes the bones to shift out of alignment. "The patient usually comes in with a stack of x-rays, like you did. And then I tell him that I want to get a grip view, an x-ray when he is crunching his hand, and he says, 'But you guys have taken every possible x-ray already.' Then you look at the joint space between the scaphoid and lunate bones, and you can drive a truck through it. The point is, you have to think of it."

Why did it take three years to "think of it"?

Light said that no one had ever really taught him how to "think of it." Instead, he learned to observe senior surgeons closely, often one-on-one in the OR, and then began to imitate those who worked with "clarity and effectiveness." He also observed surgeons who did not seem especially clear in their judgments or effective in the OR. He tried to figure out what made the difference. "It is still very much an art, a guild, where you are an apprentice and work with a master craftsman," he said.

Light added that although the conventional wisdom states that surgeons must have "great hands," that successful surgery requires manual dexterity, in fact it is more about deft decision-making. "Of course, if you are a total klutz, you have a problem in the OR," Light allowed. And having good eye-hand coordination is helpful. But Light referred to an article titled "Less Than Ten" by Dr. Paul Brown, an ex–military surgeon practicing in Hartford, Connecticut. Brown reported on surgeons who themselves suffered injuries to their hands, like losing part of or an entire finger. "Of course, there are certain very technically demanding procedures, like sewing together a small blood vessel, that require exquisite dexterity," Light said, but short of this, as Brown's article showed, there is a surprising degree of latitude. Most surgeons learn dexterity through repeated practice. Where they differ most, Light said, is not in technique, the kind of stitch they prefer, or the particular instrument they like to use in a particular setting, but in how they conceptualize a patient's problem and understand what surgery can and cannot do to remedy it. The surgeon's brain is more important than his hands.

Terry Light trained at Yale–New Haven Hospital, and during his internship worked closely with Dr. Richard Selzer, renowned not only as a surgeon but also as a writer. Selzer showed the young Terry Light that a surgeon has to have a high level of confidence to operate, or, as Selzer had written, the "audacity to take a knife to another human being." A certain bravado goes with being a surgeon, Light admitted.

I told Light that I had begun to learn about the types of cognitive shortcuts we use as doctors, and how at times that "certain bravado" affects cognition. Together, Light and I assessed the pitfalls in the thinking of the hand surgeons I had consulted. Dr. A showed what is called "commission bias." This is the tendency toward action rather than inaction. Such an error is more likely to happen with a doctor who is overconfident, whose ego is inflated, but it can also occur when a physician is desperate and gives in to the urge to "do something." The error, not infrequently, is sparked by pressure from a patient, and it takes considerable effort for a doctor to resist. "Don't just do something, stand there," Dr. Linda Lewis, one of my mentors, once said when I was unsure of a diagnosis. This was one of the rare instances, I told Terry Light, when a senior physician had explicitly cautioned me about what can be categorized as a cognitive mistake. It was a firm, no-nonsense injunction from Dr. Lewis, culled from her decades of clinical experience, handed down as master crafts-woman to apprentice. Lewis explained that inaction is not at all what is expected from a physician, nor what a physician expects from himself. But sometimes it is the best course.

Dr. B made a different cognitive error, called "satisfaction of search" by some and "search satisficing" by others. This is the tendency to stop searching for a diagnosis once you find something. It has an analogy in everyday life. Say you are getting ready to leave the house for work and time is short to catch your train. Perhaps you were out late the night before, or had a little too much wine at dinner, or had an argument with your teenager, and all of this is on your mind. You look for your wallet, and it's not on your desk where you usually leave it. You look some more and find it on the night table. You feel relieved that you found your wallet, and put it in your pocket. Now you'll make the train.

Back to a doctor trying to solve a patient's problem. The patient has a symptom that the doctor needs to explain. As he searches for the explanation, the doctor finds something wrong in the physical examination or a lab test or an x-ray. That is what happened when Dr. B jumped on the bone cysts in the MRI scan—the equivalent of locating the wallet on the night table. The problem is that there may be more than one thing to be found. Dr. Pat Croskerry put it this way: "Finding something may be satisfactory, but not finding everything is suboptimal." After putting your wallet in your pocket, walking out of your house, closing the door, and approaching your car to drive to the train, you realize that you're missing your key chain. Now you not only lack the car key, you have closed the front door and can't get back in without your house key. You were so pleased about finding your wallet that your mind shut down and you didn't consider what else was missing. Suboptimal indeed.

Dr. D was able to avoid this type of error by asking himself whether there was more to be found beyond what was seen on the plain x-rays and MRI scans. He kept searching because he was not satisfied that what he had before him was enough to account for all of my symptoms. To get me to where I needed to go, he had to find not only the wallet, but the keys.

Dr. D avoided another error in thinking, called "vertical line failure," more commonly known as thinking inside the box. Although "thinking outside the box" has become a hackneyed phrase, it still embodies the truth that sometimes "lateral thinking" that breaks out of the ordinary is vital. That "box" is the MRI scan, a revered technology that strongly constrains a doctor's thinking. Creativity and imagination, rather than adherence to the obvious, are needed in situations where the data and clinical findings do not all fit neatly together.