Every Patient Tells a Story

by Lisa Sanders

  Anyone who’s been through training won’t doubt the accuracy of this young man’s story. In residency, it often seems that no one cares if the patient is examined or not. Small wonder that many of the finer points of the exam simply slip away. And once they’re gone, it practically takes a miracle to get them back. And yet with a patient like Patty Donnally, these skills can unravel a mystery.

  A Kink in the System

  Patty Donnally is a youthful-appearing fifty-eight-year-old woman who has had high blood pressure since she was a teen. And no matter how many medications she’s taken—and she has taken many—it’s never been well controlled. Her internist tried for years to tame it. He put her on every combination of medications he could think of. Her blood pressure came down—but was never normal. Not even close. Occasionally he wondered if she was even taking her medicines. But she came to all her appointments, was aggressive in following up, even read up on her problem. That wasn’t the behavior of someone who didn’t take her meds. And when asked, she could recite her most current medication regimen no matter how many times it had changed. No. It was clear—this lady took her medicines. But her blood pressure remained high. After almost a decade her internist gave up and referred her to a specialist in hypertension. The specialist was baffled too and eventually he referred her to the hypertension clinic at Yale.

  At Yale she was seen by Dr. Bill Asch, a young enthusiastic hypertension fellow whose cheerful disposition often made her forget the frustrations of her apparently untreatable disease. His wit made the schlep to New Haven almost worth it. So she was disappointed and a little annoyed when a new doctor walked through the door.

  “Where’s my regular doctor?” she asked the young woman who entered the tidy exam room. A trace of annoyance colored her voice, and the lines between her brows deepened in the top half of a frown. Dr. Shin Ru Lin sighed inwardly. She had finished her residency training a few weeks before, and had just started at Yale’s hypertension subspecialty training program. She was getting to know the patients she inherited from Asch, who was doing research this year and not seeing patients. Serious and shy, she’d been a little hurt by the disappointment expressed by more than a few of his patients when they found that she was now going to be their doctor.

  And she was more than a little intimidated by this case in particular. Ms. Donnally was on six potent hypertension drugs, and yet, according to the nursing note on the front of the chart, her blood pressure was still too high. The patient had seen many doctors, had had scores of tests. The chart was inches thick, and still no one understood what was going on. Lin had only just begun her graduate fellowship in hypertension—how was she supposed to figure this out? What could she possibly have to offer?

  “When were you first diagnosed with hypertension?” the doctor asked tentatively.

  “I’ve had it forever—you know, it’s all in my records.” Patty waved toward the thick chart. “My blood pressure is too high, I’m always tired, and my legs hurt when I walk. Nothing changes—except my doctors.”

  In a specialty clinic like this one for hypertension at Yale–New Haven Hospital, patients have already been to several doctors, and often they are as frustrated as the physician who referred them. Each specialist, each series of tests, eliminates more of the likely causes of the problem, and the diagnostic question seems increasingly difficult to answer. And in an academic medical center, patients are often seen by trainees, like Lin, who change every year.

  Lin was overwhelmed. Waiting outside the exam room as the patient undressed for the physical exam, she opened the thick chart. She knew it would take her hours to go through it properly and she still had several more patients to see. Lin scolded herself for not reviewing it more thoroughly before meeting her for this first visit. She quickly paged through it. High blood pressure—okay. Also high cholesterol. She took a medicine for that. She didn’t smoke or drink. She carefully kept track of her blood pressure at home. Before the doctor could get much further, it was time to go back in.

  On exam, the patient’s blood pressure was—as expected—very high. But there were unexpected findings as well. As Lin listened to the patient’s neck over the carotid arteries, she heard a soft rhythmic whooshing noise over the normally silent vessels. This sound, known as a bruit, is caused by an unnatural turbulence in the flow of blood. It often indicates a narrowing of the arteries caused by atherosclerosis, commonly referred to as a hardening of the arteries.

  She moved her stethoscope down to the chest. She heard more unexpected noises. In between the lub and dup of the normal heartbeat there was a brief, harsh murmur—like the snarl of an angry animal. Was this a new symptom? She would have to check the chart. It was audible everywhere she placed her stethoscope on the left side of the chest, though it seemed loudest at the top. Atherosclerosis could affect the valves of the heart as well as the arteries. This raspy murmur suggested that the disease may have narrowed the patient’s aortic valve, one of the four valves of the heart. Could that be driving her blood pressure up? It seemed unlikely.

  Then, in the abdomen, she found yet another noise: a soft shush-shush over the renal arteries. As she completed the exam, Lin remembered the patient’s other complaint and examined her legs and feet. They looked fine—no lesions, redness, or rashes—but she couldn’t find a pulse at either ankle. Was this more evidence of hardened arteries diminishing blood flow to her feet? That could explain the pain in her legs.

  Finally, she asked herself the question all doctors must ask at the end of a visit: what could she do for this patient today? She added yet another medicine for the high blood pressure. And she would need to check the patient’s cholesterol. Even though she was on one cholesterol medication, if all this noise and the leg pain were from narrowing of the arteries, it would be essential to bring her cholesterol down as low as possible.

  What about the heart murmur? Although Lin couldn’t imagine how a narrowed valve could drive the patient’s blood pressure up, she thought it made sense to be thorough in a case this elusive. An echocardiogram would show whether the noise was coming from an abnormal cardiac valve.

  That evening Lin sat down with the patient’s chart. Before figuring out what she could do to solve this puzzle, she needed to know what had already been done. The most striking feature in this patient’s case was a remarkably high level of renin, a chemical made by the kidney to increase blood pressure. When the kidneys receive too little blood, they release this enzyme, which increases blood flow to the kidneys by increasing the pressure in the arterial system—the way you might get water to a distant flower bed by increasing the pressure in a garden hose. This woman produced one hundred times the normal amount of renin. No wonder her blood pressure was abnormal.

  So what in the world could cause the kidney to produce so much renin? Most commonly that occurs when atherosclerotic disease, the thickening and hardening of the vessels of the body, blocks the arteries supplying the kidney with blood. Perhaps that was the problem, she thought triumphantly. No, she realized moments later. An earlier angiogram had showed there was nothing blocking the arteries that carried the blood from the aorta to the kidneys.

  Could she have a renin-producing tumor? There have been cases of these types of tumors in the kidney. No, an MRI of the kidney hadn’t shown any tumors. Adrenaline makes your renin go up. Could she have an adrenaline-producing tumor? That had already been ruled out too. As Lin closed the chart and packed up to leave, she worried that she would have nothing new to offer the patient when she returned.

  The following week, Lin ran into the attending doctor with whom she had seen the patient. “Hey, Shin, did you see the results of the echo?” he asked, referring to the echocardiogram and brimming with excitement. “Do you know what it showed?” He paused dramatically. “Aortic coarctation.” Lin felt her eyes widen. She had found the cause of the hypertension—but that disease hadn’t even crossed her mind. It was a diagnosis made by accident.

  The aorta is the large, muscular vessel tha
t takes blood from the heart and delivers it to all the parts of the body. A normal aorta is about three centimeters wide, about the size of a half dollar. In coarctation, the aorta develops abnormally, and instead of being a wide-open tube, it has a kink, narrowing the tube and limiting the flow of blood. The kidneys weren’t getting enough blood, just as Lin and the other doctors had suspected. They had looked for such a blockage, but in the wrong places. Instead of being next to the kidneys, it turned out it was just inches from the heart.

  Once Lin confirmed the diagnosis with an MRI, the patient was referred to Dr. John Fahey, a cardiologist with experience in the delicate process of repairing the aorta. The day after her surgery, Ms. Donnally told me, she needed only one medication to control her blood pressure. It was, she said, a miracle. And the pain in her legs diminished. Like her kidneys, the muscles in her legs must have been starved for blood.

  The Old/New Science of the Physical Exam

  Why hadn’t Lin, or any of the patient’s previous doctors, considered coarctation of the aorta? If you look at a list of causes of hard-to-treat hypertension, it’s always on that list. And yet it had been missed. Certainly it’s an unusual cause of hypertension in an adult—mostly because it’s ordinarily picked up in childhood. It’s the number one cause of high blood pressure in children but far down the list of causes of adult hypertension. And yet doctors often think of diseases that are just as unusual. High on Dr. Lin’s differential diagnosis was a renin-producing tumor. An exceptionally rare disease. This patient had already been tested for this and other diseases even less common than coarctation.

  Moreover, Donnally had all the classic signs and symptoms. She had the murmur that was heard throughout her chest, neck, and abdomen. She had no pulses at all in her lower extremities and pain in her legs when walking. And of course she had high blood pressure. Yet it was still missed, not by one doctor, but by many. I spoke at length with Dr. Lin and Dr. Asch about why this diagnosis was missed. Both confessed that they hadn’t done the one physical exam test that would have most strongly suggested this diagnosis: comparing the blood pressure in the arms to the blood pressure in the legs. Normally the blood pressure in the legs is the same or higher than that in the arms. But because of the narrowing of the aorta, patients with coarctation provide less blood to the lower half of the body than normal. And because there’s less blood, the blood pressure taken in the legs would be lower rather than higher.

  When they finally checked, indeed the blood pressure in this patient’s legs was much lower than that found in her arms. Both Asch and Lin say they now do that exam routinely on patients with resistant hypertension. But they didn’t do it then. Of course since both doctors were in training, they were supervised in their care of this patient. Dr. John Hayslett, a well-known researcher and hypertension expert, carefully reviewed the care given to each patient at Yale’s hypertension clinic. His work has appeared in the most prestigious journals in medicine and his clinic at Yale is considered one of the best in the country. He never asked about this particular physical exam test. Says Asch, he probably assumed it had been done in the course of doing a thorough physical exam—if not by these fellows then by any of the dozen or so doctors who had already seen this patient.

  Hayslett couldn’t know whether this particular exam had been done because he hadn’t seen these postgraduate fellows do the exam. The assumption is that by the time you get to this level of training, assessment of the basics—like the physical exam—is simply not necessary.

  This is a common assumption, says Dr. Eric Holmboe. “We send a resident or medical student into a room with a patient, telling them to take a history and perform a physical exam. They come out and we ask them what they found. That’s like sending a music student into a soundproof room with a piano and a piece of sheet music and asking them when they came out, So, how’d you do? It’s crazy. How would they even know? You’d fire the music teacher who taught that way.” Maybe at some point in the past there was no need to evaluate the basic data gathering—though Holmboe is not sure there has ever been a time when teachers could assume these basic skills were done well. “There’s a tendency to think that back in some previous golden era things were better. I call that Nostalgialitis imperfecta,” he continued with a smile. “But there’s plenty of evidence that there were significant inadequacies in the way doctors took a history and performed a physical exam starting as early as the 1970s.”

  Eric wants to change all that. An energetic man in his forties with a rangy build, broad smile, and loping gait, he greeted me enthusiastically when I appeared at one of his workshops taking place in Boston. Eric is in charge of developing programs to shore up the physical exam training in medical residency programs for the American Board of Internal Medicine. One of the principal ways he does that is by teaching teachers how to teach. His focus is to convince teachers to actually watch residents as they examine their patients and then teach them how to fix what they find. “The way I was taught the physical exam was just crazy,” he told me. “No one ever watched me. How could they help me get better? I could count on one hand the number of times I was observed performing the most basic parts of my job.”

  When Eric finished his training in internal medicine at Yale, he returned to Bethesda Naval Hospital to complete his military service. His job was to teach residents who were training at the hospital. Fresh out of his own residency, Eric recalled his frustrations with the system and began observing residents on the job, as they evaluated the patients who’d come to the hospital or the clinic. At first the residents were anxious about his presence. No one had ever done this before. Some were worried that they were being singled out. Had Eric heard something that made him question their abilities? Over time, Eric was able to convince his residents that this was an important and useful practice for everyone in training—not just those with problems.

  “It didn’t take long before the trainees in our program began to welcome these observed encounters. I wouldn’t say they begged for them, but they were happy to have me there and I think they found the feedback extremely useful.” And, he continued, they needed it.

  “I couldn’t believe what these residents were doing. Examining people fully clothed. Listening to the heart and lungs through layers of clothing, placing the stethoscope in the wrong places. Poking, prodding, and thumping in places where it just won’t tell them anything.” And he found residents almost universally grateful when he showed them a better way of doing it. “The physical exam just becomes a much more useful tool when you use it correctly.”

  In a paper first promulgating the use of direct observation as a tool in evaluating residents, Eric wrote: “Direct observation of trainees is necessary to evaluate the process of data acquisition and care. A trainee’s ability to take a complete history; perform an accurate, thorough physical examination; communicate effectively; and demonstrate appropriate interpersonal and professional behavior can best be measured through the direct sampling of these clinical skills.” It seems obvious and yet it’s been a remarkably hard sell—not just to residents but to training programs as well. It’s time-consuming and many physicians are not comfortable enough with their own physical exam skills to feel competent to supervise the skills in someone else. And it simply wasn’t the way things were done—traditionally.

  That tradition is summed up in one phrase that I heard frequently in my own training: see one, do one, teach one. It’s how residents have been taught to do procedures for decades. It also describes how many were taught the physical exam. A study published recently shows how inadequate this style of teaching is. A group of residents in nine teaching hospitals in England were asked to describe how they were taught to perform seven relatively simple procedures—from giving a shot to taking an EKG. They were also asked about their confidence in their own ability to perform this procedure the first time they did it. The same questionnaire was given to a group of nurses who traditionally get highly structured training in the performance of procedures. O
ver a third of doctors said they received no training at all before performing the procedure and nearly half said they felt unqualified when they first performed them. Nearly half performed these procedures unsupervised when they did it the first time. Doctors are often sent out into the wards to perform on their patients—with inadequate training, and sometimes no training at all—procedures that carry some, usually small, risk to the patient if done incorrectly. And yet we continue to allow medical students and residents to perform these procedures without adequate training. The same is true of the noninvasive clinical stuff—the taking of a history or the performing of a physical exam that doctors do where there is no risk of directly doing harm, only of missing something important.

  So Eric has spent the past several years as a one-man sales force, traveling from training program to training program selling the idea that direct observation of residents in training is the right thing to do. He has developed a four-day course to teach the teachers how to observe. One of the problems, says Holmboe, is that since many doctors themselves were not given any formal training in these skills, most doctors haven’t developed formal criteria of how to talk to a patient and how to examine a patient. If doctors aren’t certain that they are doing it right, how can they know if a student is doing it right? As a teaching tool, Eric scripted and videotaped three clinical encounters where a resident was shown taking a history, performing the physical exam, and counseling the patient. He taped three versions of each of three scenarios: one of poor quality, one of moderate quality, one of high quality. Then he asked teachers to grade each encounter. The grades were all over the map. The encounters that were poor were given grades as high as the high-quality encounters. No one had a clue. This class helps teachers develop criteria for each component of the clinical exam and teaches them to apply them when they watch a trainee. Teachers are also coached in how to provide feedback in a constructive and useful way.