by Deborah Blum
An anesthesiologist’s job is surprisingly subjective. The same patient could be put under general anesthesia in a number of different ways, all to accomplish the same fundamental goal: to render him unconscious and immune to pain. Many methods also induce paralysis and prevent the formation of memory. Getting the patient under, and quickly, is almost always accomplished with propofol, a drug now famous for killing Michael Jackson. It is milky and viscous, almost like yogurt, in a fat syringe. When injected, it has a nearly instant hypnotic effect: blood pressure falls, heart rate increases, and breathing stops. (Anesthesiologists use additional drugs, as well as ventilation, to immediately correct for these effects.)
Other drugs in the anesthetic arsenal include fentanyl, which kills pain, and midazolam, which does little for pain but induces sleepiness, relieves anxiety, and interrupts memory formation. Rocuronium disconnects the brain from the muscles, creating a neuromuscular blockade, also known as paralysis. Sevoflurane is a multipurpose gaseous wonder, making it one of the most commonly used general anesthetics in the United States today—even though anesthesiologists are still relatively clueless as to how it produces unconsciousness. It crosses from the lungs into the blood, and from the blood to the brain, but . . . then what?
Other mysteries have been untangled. Redheads are known to feel pain especially acutely. This confused researchers until someone realized that the same genetic mutation that causes red hair also increases sensitivity to pain. One study found that redheaded patients require about 20 percent more general anesthesia than brunettes. Like redheads, children also require stronger anesthesia; their youthful livers clear drugs from the system much more quickly than adults’ livers do. Patients with drug or alcohol problems, on the other hand, may be desensitized to anesthesia and require more—unless the patient is intoxicated at that moment, in which case less drug is needed.
After delivering the appropriate cocktail, anesthesiologists carefully monitor a patient’s reactions. One way they do this is by tracking vital signs: blood pressure, heart rate, and temperature; fluid intake and urine output; oxygen saturation in arteries. They also observe muscles, pupils, breathing, and pallor, among many other indicators.
One organ, however, has remained stubbornly beyond their watch. Even though anesthesiologists are not entirely sure how their drugs work, they do know where they go: the brain. All changes in your vital signs are only the peripheral reverberations of anesthetic drugs’ hammering on the soft mass inside your skull. Determining consciousness by measuring anything besides brain activity is like trying to decide whether a friend is angry by studying his or her facial expressions instead of asking directly, “Are you mad?”
In lamenting how little we know about the anesthetized brain, Gregory Crosby, a professor of anesthesiology at Harvard, wrote in the New England Journal of Medicine in 2011, “The astonishing thing is not that awareness occurs, but that it occurs so infrequently.”
This ignorance gap seems almost absurd in the context of today’s dazzling array of medical technologies. Doctors can parse your brain with innumerable X-ray slices and then collate them into a three-dimensional grayscale image in a process called computed tomography, or CT. They can send you into a tube where powerful magnets flip the spin of protons on water molecules in your brain; when the protons flip back into position, they emit radio waves, and from that information a computer can generate a comprehensive image known as an MRI (for magnetic resonance imaging). Positron emission tomography, or PET, scans provide detailed maps of metabolic activity. Yet we are in the Dark Ages when it comes to determining whether the brain is conscious or not? We can’t figure out whether patients are awake, or what being awake even means?
Due to their hulking size, CT scanners and MRI machines are rarely, if ever, brought into an operating room. But other technologies are more mobile. For example, doctors can measure electrical activity in the brain using a machine that, with the help of a few electrodes attached to your scalp, generates what’s known as an electroencephalogram (EEG)—essentially a snapshot of your brain waves. An EEG is printed in undulating longitudinal lines, like the scribbled outline of a mountain range: sometimes smooth and regular like the Appalachians, at other times rough and craggy like the Rockies, and in death or deep coma more like the Great Plains.
This technology is regularly used in sleep studies and to diagnose epilepsy and encephalitis, as well as to monitor the brain during certain specialized surgeries. But the problem with an EEG is interpreting it. The data come at a constant, unforgiving pace, with lines stacked one on top of another like a page of sheet music (high-density versions can have up to 256 lines). Before digitization, EEG printers disgorged paper at 30 millimeters per second, resulting in 324 meters of print for just three hours of surgery. And even today’s machines provide data that are next to impossible to analyze on the fly, at least with any sort of detail or depth.
In 1985, when a twenty-three-year-old doctoral student named Nassib Chamoun first looked at the sheet music of an EEG, he saw a symphony—albeit one he could not yet read. Chamoun was then an electrical engineer doing a research fellowship at the Harvard School of Public Health, working on decoding the circuitry of the human heart. When an anesthesiologist he worked with argued that the brain was much more electrically interesting than the heart, Chamoun agreed to attend a demonstration of EEG, which was then a relatively new technology in the operating room, during a surgery at a Harvard hospital. The EEG printer was an old model, spilling reams of paper that piled near the head of the gurney. As the anesthesiologist injected the patient with drugs, the machine’s pen danced wildly, ink splattering off the page. Chamoun was entranced by the complexity of the patterns he saw that day. He couldn’t stop thinking about how to engineer these data into something that would be more useful for surgeons and anesthesiologists. He left his doctorate program and embarked on a twenty-five-year quest to decode the brain—and, ultimately, to quantify and measure consciousness.
As a child in Lebanon, Chamoun had been fascinated by taking things apart and putting them back together. During the 1970s, as ethnic tensions there boiled into civil war, Chamoun spent a lot of time cooped up at home when school was canceled or when it was unsafe for him to venture outside. The soldering iron and circuit board became his playground. Family members asked him to fix televisions, tape recorders, and radios. His parents gave him a microscope as a birthday present. He made his way to the United States for college, eager to expand his study beyond home electronics.
As it happened, the mid-1980s were an auspicious time for a young technologist with a promising idea. When Chamoun began working with EEG, he had early access to mainframe computers at Harvard and Boston University. More important, he had access to surgeries. He wheeled his digital EEG machine into Harvard operating rooms, fixed electrodes to patients, and recorded millions of data points. Then, using computers, he began to sift through the oceans of information, searching for a unifying pattern. Meanwhile, he was enlisting his old mentor, a Nobel Prize–winning Harvard professor, and courting venture-capital firms for seed money. The result was Aspect Medical Systems, a biotech firm he founded in 1987 with a singular goal: to build a monitor that anesthesiologists could use to discern their patients’ level of awareness.
Chamoun turned out to have a pivotal ally in a family friend, Charlie Zraket, the CEO of a big defense contractor called Mitre. In the 1960s, mathematicians had developed a statistical method called bispectral analysis, which breaks down waveforms to find underlying patterns. This method was originally used for studying waves in the ocean, but Mitre applied it to voice-recognition software and later to sonar on war submarines and radar on airplanes. If bispectral analysis could be used to interpret patterns in ocean, radio, and sound waves, Zraket and Chamoun reasoned, why couldn’t it be applied to brain waves?
Chamoun ended up banking everything on the belief that if he collected enough EEG data, he could hack the patterns using bispectral analysis. But by 1995, eight year
s in, the entire project was collapsing. Chamoun had gathered more than $18 million from investors, credit lines, and friends, and had spent it all, but still his algorithms could not reliably predict a patient’s level of awareness. Just as he was confronting bankruptcy, he secured a $4 million investment from a well-known venture capitalist. This bought the time Chamoun needed. From that point, he and his team achieved a series of breakthroughs that caused them to fundamentally reframe the way they thought about consciousness. Chamoun had never believed that the brain was something with a simple on/off switch, but he had been looking for one master equation—a sort of electrical fingerprint of consciousness—that would connect all the dots. Only when he let go of the idea of a single equation did a new, more viable model come into view: consciousness as a spectrum of discrete phases that flowed one into the next, each marked by a different electrical fingerprint. Fully conscious to lightly sedated was one phase; lightly to moderately sedated was another; and so on. Once he realized this, Chamoun was able to identify at least five separate equations and arrange them in order, like snapping a series of photos and compiling them into a broad panorama shot.
The end product was a shoebox-size blue machine that used EEG data to rank a patient’s level of awareness on an index of zero to 100, from coma to fully awake. Chamoun called it the Bispectral Index, or BIS, monitor. To use the BIS, all that anesthesiologists had to do was connect a pair of disposable electrode sensors to the machine, apply them firmly to a patient’s forehead, and wait for a number to appear on the box’s green-and-black digital display. They would then administer anesthesia and watch the number drop from a waking average of 97 to somewhere in the ideal “depth of anesthesia” range—between 40 and 60—at which point they could declare the patient ready for surgery.
The FDA cleared the BIS monitor in 1997. When Time interviewed Chamoun about the revolutionary device, he called it anesthesia’s “Holy Grail.” Two years later, Aspect Medical’s quarterly revenue surpassed $8 million; the company soon went public. In 2000 Ernst & Young named Chamoun the Healthcare and Life Sciences Entrepreneur of the Year for the New England region.
Enthusiasm for the BIS monitor grew in 2004, when The Lancet published a groundbreaking study reporting that the device could reduce the incidence of anesthesia awareness by more than 80 percent. This nearly pushed the BIS into the realm of medical best practices. By July 2007 half of all American operating rooms had a BIS monitor. By 2010 the device had been used almost 40 million times worldwide. At his home in the suburbs of Boston, Chamoun has the 10-millionth sensor memorialized in a sealed plastic case.
The BIS monitor fundamentally changed the way scientists thought about consciousness. It compressed an enigmatic idea that had long mystified researchers into a medical indicator that could be quantified and measured, like blood pressure or body temperature. One effect of the accessibility of Chamoun’s invention was that it was occasionally used outside the operating room, for purposes he had not foreseen. In a 2006 injunction involving a North Carolina death-row inmate named Willie Brown, a federal judge ruled that performing a lethal injection on a conscious prisoner could cause excessive pain. North Carolina requires prisons to anesthetize inmates before killing them, but the judge worried about the possibility of anesthesia awareness. Only when prison officials purchased a BIS monitor did he allow them to proceed with Brown’s execution. So on April 21, 2006, attendants hooked Brown up to the monitor, injected him with a sedative, and watched his BIS value drop. At approximately two o’clock in the morning, once the number had fallen below 60, an attendant administered a lethal dose of pancuronium bromide and potassium chloride.
In the centuries before EEG and computers, the most active contemplators of consciousness were not doctors but philosophers. The seventeenth-century French thinker René Descartes proposed an influential theory that leaned on neuroanatomy as well as philosophical inference. He declared that the pineal gland, a pea-size glob just behind the thalamus, was the seat of consciousness, “the place in which all our thoughts are formed.” But Descartes was a dualist: he believed that body and mind are separate and distinct. Within the physical matter of the pineal gland, he reasoned, something inexplicable must lie, something intangible—something that he identified as the soul.
This idea has been rejected by reductionist thinkers, who believe that consciousness is a scientific phenomenon that can be explained by the physiology of the brain. In an attempt to understand various sensory functions, a nineteenth-century cohort of reductionist biologists burned, cut, and excised lumps of the brain in rabbits, dogs, and monkeys, eventually pinpointing centers for hearing, vision, smell, touch, and memory. But even the most extreme experiments of the period failed to identify a center for consciousness. In 1892 a German scientist named Friedrich Goltz, who rejected this notion of cerebral localization and hypothesized instead that the brain operated as a cohesive unit, cut out the majority of a dog’s cerebral cortex over the course of three operations. The animal managed to survive for eighteen months; it even remained active, walking its cage and curling up to sleep, and reacted to noises and light by flipping its ears and shutting its eyes. Yet other things had changed. The dog required assistance with eating, and its memory seemed to have been destroyed. “The condition was that of idiocy but not of unconsciousness,” wrote one scientist.
Today’s neuroscientists, most of whom are reductionists, have offered multiple hypotheses about where consciousness resides, from the anterior cingulate cortex, a region also associated with motivation, to some parts of the visual cortex, to the cytoskeleton structure of neurons. Some theories peg consciousness not to a particular part of the brain but to a particular process, such as the rhythmic activation of neurons between the thalamus and the cortex.
David Chalmers, an Australian philosopher who has written extensively about consciousness, would refer to this neurobiological hunt as the “easy problem.” With enough time and money, scientists could ostensibly succeed in locating a consciousness center of the brain. But at that point, Chalmers argues, an even bigger mystery would still remain, one that he calls the “hard problem.” Say you and a friend are looking at a sunset. Your body is processing a huge variety of sensory inputs: a spectrum of electromagnetic waves—red, orange, and yellow light—which focus on your retina; the vibrations of your friend’s voice, which bounce along the bones of your inner ear and transform into a series of electrical signals that travel from neuron to neuron; memories of past sunsets, which spark a surge of dopamine in your mesolimbic pathway. These effects coalesce into one cohesive, indivisible experience of the sunset, one that differs from your friend’s. The hard problem, or what the philosopher Joseph Levine called the “explanatory gap,” is determining how physical and biological processes—all of them understood easily enough on their own—translate into the singular mystery of subjective experience. If this gap cannot be bridged, then consciousness must be informed by some sort of inexplicable, intangible element. And all of a sudden we are back to Descartes.
In 2004 a sixty-year-old man checked in for open gastric-bypass surgery and a gall-bladder removal at Virginia Mason Medical Center in Seattle. Simon, as I’ll call him, stood five feet nine inches tall and weighed approximately three hundred pounds. In an open gastric bypass, the surgeon penetrates mounds of flesh and fat before finding the peritoneum, the glossy membrane that holds the abdominal cavity intact. Many surgeons use a space-age device called a Harmonic Scalpel, which cuts tissue while simultaneously blasting it with ultrasound waves to stop the bleeding. Once the surgeon uncovers the stomach and yards of folded, tubular intestines, she uses metal retractors to pull the skin apart and clear away slippery membranes, juicy organs, and fatty layers of tissue. Then to business: cut, suture, cut, suture, cauterize, cut.
No surgeon could have imagined a procedure of this magnitude 167 years ago, in the days before anesthesia. It would have been impossible to endure, both for the patient and the surgeon. Simon’s anesthesiologist, Micha
el Mulroy, was particularly worried about him because of his hypotension and reliance on painkillers, both of which increased his risk of awareness. To make sure that Simon didn’t drift into consciousness, Mulroy decided to use a BIS monitor.
Surgery records show that throughout the three-hour procedure, Simon’s BIS value hovered between 37 and 51, well below the threshold for sedation. Mulroy had given Simon a relatively light dosage, reluctant to risk further deflating his patient’s dangerously low blood pressure, but he took comfort in the fact that the BIS told him that Simon was unconscious and unaware.
After the surgery, in the postoperative recovery room, nurses asked Simon whether he was in pain. “Not now,” he said, “but I was during surgery.” Simon reported memories that began after intubation, including “unimaginable pain” and “the sensation that people were tearing at me.” According to a clinical report, he heard voices around him and “wished he were dead,” but when he tried to alert the surgical team, his body did not respond to his brain’s commands.
The news of Simon’s experience devastated Mulroy. He explained to his patient that he had used the BIS monitor and that it had confirmed Simon’s unconscious state throughout the procedure. In the end, Mulroy says, all he could do was apologize and arrange for a psychiatrist. He hasn’t seen Simon since, but he published the case in a 2005 issue of the journal Anesthesia & Analgesia. Mulroy felt that the BIS monitor had betrayed him; he might have done more to deepen Simon’s sedation if the BIS had not reassured him that everything was fine.
Mulroy was one of the first to question the BIS, but his concern was soon echoed in other corners of the medical community. In 2008 the New England Journal of Medicine published a study comparing nearly 2,000 surgery patients at high risk of awareness: 967 patients were monitored by the BIS, and 974 via attention to changes in the amount of anesthetic gas they exhaled throughout a procedure. The author, a researcher at the Washington University School of Medicine in St. Louis named Michael Avidan, found that both groups of patients experienced awareness at similar rates. In other words, the BIS was no more effective than a much cheaper and more standard method. After questions were raised about his methodology, Avidan repeated the experiment with a broader sample and found the same thing. Chamoun’s window to the brain, it turned out, was not especially enlightening.