Authors: Atul Gawande
But, for the Brattons, I had to wonder how useful it would be. If opinions disagreed, then what? And if they did not, wouldn’t the same fallibilities and questions remain? Furthermore, the Brattons did not know anyone to call and had to ask if
we
had any ideas.
We suggested calling David Segal, a plastic surgeon on staff who like Studdert had seen such cases before. They agreed. I called Segal and filled him in. He came down within minutes. In the end what he gave Eleanor and her father was mainly confidence, from what I could see.
Segal is a rumpled and complexly haired man, with pen stains on his white coat and glasses that seem too large for his face. He is the only plastic surgeon I know who looks like he has a Ph.D. from M.I.T. (which, as it happens, he does). But he seemed, as Bratton later put it, “not young.” And he did not disagree with what Studdert had said. He listened to Eleanor’s story and looked carefully at her leg and then said that he too would be surprised if she turned out to have the bacteria. But he agreed that it could not be ruled out. So what else was there but to biopsy?
Eleanor and her dad now agreed to go ahead. “Let’s get it over with,” she said. But then I brought her the surgical consent form to sign. On it, I had written not only that the procedure was a “biopsy of the left lower extremity” but also that the risks included a “possible need for amputation.” She cried out when she saw the words. It took
her several minutes alone with her father before she could sign. We had her in the operating room almost immediately after. A nurse brought her father to the family waiting area. He tracked her mother down by cell phone. Then he sat and bowed his head, and made some prayers for his child.
There is, in fact, another approach to decision making, one advocated by a small and struggling coterie in medicine. The strategy, long used in business and the military, is called decision analysis, and the principles are straightforward. On a piece of paper (or a computer), you lay out all your options, and all the possible outcomes of those options, in a decision tree. You make a numeric estimate of the probability of each outcome, using hard data when you have it and a rough prediction when you don’t. You weigh each outcome according to its relative desirability (or “utility”) to the patient. Then you multiply out the numbers for each option and choose the one with the highest calculated “expected utility.” The goal is to use explicit, logical, statistical thinking instead of just your gut. The decision to recommend annual mammograms for all women over age fifty was made this way and so was the U.S. decision to bail out Mexico when its economy tanked. Why not, the advocates ask, individual patient decisions?
Recently, I tried “treeing out” (as the decision buffs put it) the choice Eleanor faced. The options were simple: to biopsy or not biopsy. The outcomes quickly got complicated, however. There was: not being biopsied and doing fine; not being biopsied, getting diagnosed late, going through surgery, and surviving anyway; not being biopsied and dying; being biopsied and getting only a scar; being biopsied and getting a scar plus bleeding from it; being biopsied, having the disease and an amputation, but dying anyway; and so on. When all the possibilities and consequences were penciled out, my decision tree looked more like a bush. Assigning the probabilities for each potential twist of fate seemed iffy. I found what data I could from the medical literature and then had to extrapolate a good deal. And determining the relative desirability of the outcomes seemed
impossible, even after talking to Eleanor about them. Is dying a hundred times worse than doing fine, a thousand times worse, a million? Where does a scar with bleeding fit in? Nonetheless, these are the crucial considerations, the decision experts argue, and when we decide by instinct, they say, we are only papering this reality over.
Producing a formal analysis in any practical time frame proved to be out of the question, though. It took a couple of days—not the minutes that we had actually had—and a lot of back and forths with two decision experts. But it did provide an answer. According to the final decision tree, we should
not
have gone to the OR for a biopsy. The likelihood of my initial hunch being right was too low, and the likelihood that catching the disease early would make no difference anyway was too high. Biopsy could not be justified, the logic said.
I don’t know what we would have made of this information at the time. We didn’t have the decision tree, however. And we went to the OR.
The anesthesiologist put Eleanor to sleep. A nurse then painted her leg with antiseptic, from her toes up to her hip. With a small knife, Studdert cut out an inch-long ellipse of skin and tissue from the top of her foot, where the blister was, down to her tendon. The specimen was plopped into a jar of sterile saline and rushed to the pathologist to look at. We then took a second specimen—going deeper now, down into muscle—from the center of the redness in her calf, and this was sent on as well.
At first glance beneath her skin, there was nothing apparent to alarm us. The fat layer was yellow, as it is supposed to be, and the muscle was a healthy glistening red and bled appropriately. When we probed with the tip of a clamp inside the calf incision, however, it slid unnaturally easily along the muscle, as if bacteria had paved a path. This is not a definitive finding, but enough of one that Studdert let out a sudden, disbelieving, “Oh shit.” He pulled off his gloves and gown to go see what the pathologist had found, and I followed right
behind him, leaving Eleanor asleep in the OR to be watched over by another resident and the anesthesiologist.
An emergent pathology examination is called a frozen section, and the frozen section room was just a few doors down the hallway. The room was small, the size of a kitchen. In the middle of it stood a waist-high laboratory table with a black slate countertop and a canister of liquid nitrogen in which the pathologist had quick-frozen the tissue samples. Along a wall was the microtome that he had used to slice micron-thin sections of the tissue to put on glass slides. We walked in just as he finished preparing the slides. He took them to a microscope and began scanning each one methodically, initially under low power magnification and then under high power. We hovered, no doubt annoyingly, awaiting the diagnosis. Minutes passed in silence.
“I don’t know,” the pathologist muttered, still staring through the eyepieces. The features he saw were “consistent with necrotizing fasciitis,” he said, but he wasn’t sure he could clinch the diagnosis. He said he would have to call in a dermatopathologist, a pathologist who specializes in looking at skin and soft tissue. It took twenty minutes before the specialist arrived and another five before he could make his call, our frustration growing. “She’s got it,” he finally announced grimly. He had detected some tiny patches where the deep tissue had begun to die. No cellulitis could do that, he said.
Studdert went to see Eleanor’s father. When he walked into the crowded family waiting area, Bratton caught the expression on his face and began yelling, “Don’t look at me like that!
Don’t look at me like that!
” Studdert took him to a private side room, closed the door behind them, and told him that she appeared to have the disease. He would have to move fast, he said. He was not sure he could save her leg and he was not sure if he could save her life. He would need to open her leg up, see how bad things were, and then go from there. Bratton was overcome, crying and struggling to get out words. Studdert’s own eyes were wet. Bratton said to “do what you have to
do.” Studdert nodded and left. Bratton then called his wife. He told her the news and then gave her a moment to reply. “I will never forget what I heard on the other end of the line,” he later said. “Something, some sound, I cannot and will never be able to describe.”
Decisions compound themselves, in medicine like in anything else. No sooner have you taken one fork in the road than another and another come upon you. The critical question now was what to do. In the OR, Segal joined Studdert to offer another set of hands. Together they slit open Eleanor’s leg, from the base of her toes, across her ankle, to just below her knee, to get a full view of what was going on inside. They pulled the opening wide with retractors.
The disease was grossly visible now. In her foot and most of her calf, the outer, fascial layer of her muscles was gray and dead. A brownish dishwater fluid was seeping out with a faint smell of decay. (Tissue samples and bacterial cultures would later confirm that this was toxic group A
Streptococcus
advancing rapidly up her leg.)
“I thought about a BKA,” a below-knee amputation, Studdert says, “even an AKA,” an above-knee amputation. No one would have faulted him for doing either. But he found himself balking. “She was such a young girl,” he explains. “It may seem harsh to say, but if it was a sixty-year-old man I would’ve taken the leg without question.” This was partly, I think, a purely emotional unwillingness to cut off the limb of a pretty twenty-three-year-old—the kind of sentimentalism that can get you in trouble. But it was also partly instinct again, an instinct that her youth and fundamentally good health might allow him to get by with just removing the most infested tissue (a “debridement”) and washing out her foot and leg. Was this a good risk to take, with one of the deadliest bacteria known to man loose in her leg? Who knows? But take it he did.
For two hours, using scissors and electrocautery, he and Segal cut and stripped off the necrotic outer layers of her muscle, starting from the webbing of her toes, going up to the tendons of her calf.
They took out tissue going three-quarters of the way around. Her skin hung from her leg like open coat flaps. Higher up, inside the thigh, they reached fascia that looked pink-white and fresh, very much alive. They poured two liters of sterile saline through the leg, trying to wash out as much of the bacteria as possible.
At the end, Eleanor seemed to be holding steady. Her blood pressure remained normal. Her temperature was ninety-nine degrees. Her oxygen levels were fine. And the worst-looking tissue had been removed from her leg.
But her heart rate was running a bit too fast, one hundred and twenty beats a minute, a sign that the bacteria had provoked a systemic reaction. She was requiring large amounts of intravenous fluid. Her foot looked dead. And her skin was still burning red with infection.
Studdert stood firm with his decision not to take more, but you could see he was uneasy about it. He and Segal conferred and thought of one other thing they could try, an experimental therapy called hyperbaric oxygen. It involved putting Eleanor in one of those pressure chambers they put divers in when they get the bends—a perhaps kooky-sounding notion but not a ludicrous one. Immune cells require oxygen to kill bacteria effectively and putting a person under double or higher atmospheric pressure for a few hours a day increases the oxygen concentration in tissue tremendously. Segal had been impressed by results he had gotten using the therapy in a couple of burn patients with deep wound infections. True, studies had not proven that it would work against necrotizing fasciitis. But suppose it could? Everyone latched onto the treatment immediately. At least it made us feel as if we were doing something about all the infection we were leaving behind.
We did not have a chamber at our hospital, but a hospital across town did. Someone got on the phone and within a few minutes we had a plan for ambulancing Eleanor over with one of our nurses for two hours under 2.5 atmospheres of pressurized oxygen. We left her
wound open to drain, laid wet gauze inside it to keep the tissues from desiccating, and wrapped her leg in white bandages. Before sending her over, we wheeled her from the OR to intensive care, where we could make sure she would be stable enough for the trip.
It was eight o’clock at night now. Eleanor woke up nauseated and in pain. But she was sharp-witted enough to surmise from the crowd of nurses and doctors around her that something was wrong.
“Oh God, my leg.”
She reached down to find it, and for a few panicked moments she wasn’t sure she could. Slowly, she convinced herself that she could see it, touch it, feel it, move it. Studdert put his hand on her arm. He explained what he had found, what he had done, and what more there would be to do. She took the information with more grit and fight than I knew she had. Her whole family had now arrived to be with her, and looked as though an SUV had hit them. But Eleanor pulled the sheet back over her leg, took in the monitors flashing their green and orange lights and the IV lines running into her arms, and said, simply, “OK.”
The hyperbaric chamber that night was, as she describes it, “like a glass coffin.” She lay inside it on a narrow mattress with nowhere to put her arms except straight down or folded across her chest, a panel of thick plexiglass a foot from her face, and an overhead hatch sealed tight with turns of a heavy wheel. As the pressure increased, her ears kept popping, as if she were diving down into a deep ocean. Once the pressure reached a certain point, she would be stuck, the doctors had cautioned. Even if she should start throwing up, they could not get to her, for the pressure could only be released slowly or it would give her the bends and kill her. “One person had a seizure inside,” she remembered them telling her. “It took them twenty minutes to get to him.” Lying there enclosed, more ill than she’d ever imagined one could be, she felt far away and almost totally alone. It’s just me and the bacteria in here, she thought to herself.
The next morning, we took her back to the operating room, to
see if the bacteria had spread. They had. The skin over most of her foot and front of her calf was gangrenous and black and had to be cut off. The edges of fascia we had left behind were dead and had to be excised as well. But her muscle was still viable, including in her foot. And the bacteria had not killed anything up in her thigh. She had no further fevers. Her heart rate had normalized. We repacked her wound with wet gauze and sent her back for more hyperbaric oxygen—two hours twice a day.
We ended up operating on her leg four times in four days. At each operation, we had to take a little more tissue, but each time it was less and less. At the third operation, we found the redness of her skin had finally begun to recede. At the fourth operation, the redness was gone and we could see the pink mossy beginnings of new tissue in the maw of her wound. Only then was Studdert confident that not only had Eleanor survived, but her foot and leg had, too.