The moment she saw him, Judy Cox knew her son was dead. It was an October morning in 2008, and she had just stepped out the door to run an errand when she found him lying faceup in the driveway, ghost white, covered in purple splotches. He wasn’t breathing, and when she couldn’t revive him, she ran screaming into the house where her husband, Wayne, was still asleep. “Chris is dead,” she cried. “Call 911!”
Wayne jumped out of bed and raced down to the driveway, where he knelt over his son’s limp frame and tried frantically to elicit a breath or a heartbeat. As he pumped Chris’s chest and scooped out the vomit that had collected in his mouth, Judy ran to the kitchen and steadied herself long enough to call for an ambulance.
Chris was 26. He had not been well. An A.T.V. accident the previous August left him with debilitating back pain that physical therapy did nothing to alleviate. His doctor had recently prescribed Oxycontin. His parents learned later that he had taken too much.
By the time the ambulance arrived, Chris’s heart had been still for at least 15 minutes. It took the paramedics another 15 to get it pumping again; even then, doctors had little hope he would survive. Brain cells begin dying off just five minutes after blood stops delivering oxygen. After 30 minutes, there is likely to be more dead tissue than living.
Nonetheless, the emergency-room staff members at the local hospital did their best. They hooked Chris up to a tangle of tubes and machines and injected him with drugs to stabilize his heart rate. Wayne and Judy watched helplessly from the hallway. After four hours, a doctor finally summoned them to a secluded corridor.
Chris was in a coma, the doctor said, and in all likelihood had suffered severe, irreversible brain damage. He was breathing only with the help of a ventilator and would probably have a series of heart attacks in the night.
“First they asked us to let them pull the plug,” Judy recalled one recent afternoon, as we sat in the living room of the Coxes’ house in a Memphis suburb. “Then they tried getting us to sign a do-not-resuscitate order.” Without one, the doctor explained, hospital staff would be forced to revive Chris each time he started slipping away, which could mean cracking his ribs and shocking him with electricity. Even if they managed to keep his body alive, what was left of his brain would surely die in the days ahead.
Wayne and Judy refused to sign. “This is not some dog we’re talking about putting down,” Wayne shouted. “This is our son.” Chris still lived with his parents. He was a good kid, a joker, but bashful, especially around girls. He liked playing basketball and fishing in the pond near his house. He was planning to take over the family repo business when Wayne retired in a few years. Before the A.T.V. accident, he’d never given them much trouble at all. He deserved every chance the hospital could give him.
The heart attacks never came. Four days later, Chris woke up.
It was not the awakening of Hollywood movies in which the patient comes to, just as he was, speaking full sentences and completely mobile. Three years later, Chris still cannot talk. Although he breathes on his own, his lungs battle a steady barrage of infections; a feeding tube provides all his sustenance, and his muscles have contracted into short, twisted knots. He can move only the slightest bit — his fingers and eyelids twitch, but his arms and legs remain mostly immobile — and his neck is not quite strong enough to hold up his head, which leans against a crescent-shaped support around his wheelchair headrest.
Still, Wayne and Judy say that his cognition is improving. On good days, they say, he can respond to basic commands — blink his eyes for yes, wiggle his finger for no, give a thumbs up when asked. Doctors agree that Chris has progressed beyond a vegetative state, to a hazy realm known as minimal consciousness. What that means — what it says about his experience of the world around him or his prospects for further recovery — is something they are still trying to figure out.
Convinced that the son they know and love is still “in there,” Chris’s parents have spent the past three years searching for a way to bring him back out. So far, their best hope has come from an unlikely source: Ambien. A growing body of case reports suggests that the popular sleep aid can have a profound — and paradoxical — effect on patients like Chris. Rather than put them to sleep, both Ambien and its generic twin, zolpidem, appear to awaken at least some of them. The early reports were so pronounced that until recently, doctors had a hard time believing them. Only now, more than a decade after the initial discovery, are they taking a closer look.
The first report of a zolpidem awakening came from South Africa, in 1999. A patient named Louis Viljoen, who, three years before, was declared vegetative after he was hit by a truck, had taken to clawing at his mattress during the night. Thinking he was suffering from insomnia, his family doctor suggested zolpidem to help him sleep. But 20 minutes after his mother ground the tablet up and fed it to him through a straw, Viljoen began to stir. His eyes, which normally wandered the room, vacant and unfocused, flickered with the light of consciousness. And then he began to talk (his first words were “Hello, Mummy”), and move (he could control his limbs and facial muscles). A few hours later he became unresponsive. But the next day, and for many days after that, zolpidem revived him, a few hours at a time.
Here was a case worthy of Hollywood: three years was well past the point at which doctors would expect any sort of spontaneous recovery. Viljoen awoke with the ability to speak in complete sentences. Not only did he recognize his mother, but he also recognized the voices of people who had spoken to him only when he was apparently vegetative. He remembered nothing of the mysterious realm he kept receding back into. When doctors asked him what it was like to slip away, he said he felt no changes at all. But he could recall conversations from the previous day’s awakening, along with bits and pieces of his former life: his favorite rugby team, specific matches he attended, players that he rooted for and against. As time passed, his cognition improved. He could laugh at jokes, and his awakenings stretched from a few hours to entire days. Eventually, he no longer needed zolpidem.
In the years that followed, a steady trickle of similar reports emerged — some from doctors who tried zolpidem after hearing about the Viljoen case, others from those who discovered its benefits accidentally, as Viljoen’s doctor had. The drug did not work for everyone, and even when it did, its effects typically wore off after an hour or two. But for a lucky few, those effects were profound. People who seemed vegetative for years were waking up.
There are roughly 200,000 patients in the United States trapped in the borderlands between consciousness and oblivion. Until recently, most doctors believed that recovering from this condition was not possible. Vegetative states were considered permanent after three months if the injury was caused by oxygen deprivation, or one year if it was caused by blunt trauma. And since minimally conscious patients did not fare much better than those who were vegetative, most doctors did not bother to draw the distinction.
But in the past decade, a series of developments have coalesced into a far more complicated picture than previously imagined. In 2003, an Arkansas man named Terry Wallis emerged, after 19 years, from a minimally conscious state. Neuroimaging suggested that his brain had essentially reconfigured itself — surviving neurons bypassed dead ones and forged new connections to one another. In a 2007 Nature paper, Nicholas Schiff, a neurologist from Weill Cornell Medical College, and his colleagues showed that deep brain stimulation — surgically implanting a “brain pacemaker” that sends electrical impulses to specific regions of the brain — can help some severely injured patients recover the ability to speak and eat, years after the injury. And just this month, Adrian Owen, a British neuroscientist, reported in the journal The Lancet that the brains of some patients who seemed vegetative responded to basic commands: their bodies didn’t move, but distinct patterns of neuronal firing were detected on EEG scans when these patients were told to make a fist (which triggered one region of the premotor cortex) or wiggle their toes (which triggered another).
This year, scientists at Moss Rehabilitation Research Institute and at the University of Pennsylvania, both in the Philadelphia area, began the first large-scale clinical study of zolpidem as a treatment for disorders of consciousness. (Amantadine, a drug used to treat Parkinson’s disease, and the anti-anxiety medication Ativan also show promise in increasing awareness in minimally conscious patients.) So far, the evidence suggests that less than 10 percent of brain-injured patients will experience the drug’s paradoxical effects, and that among those, only a few will respond as profoundly as Viljoen did. For families like the Coxes, such odds provide a tortured kind of hope. For doctors, they bring questions. Why does a sleeping pill induce awareness in some patients but not others? And what can these bizarre awakenings tell us about the brain’s ability to heal?
Two weeks after Chris first emerged from the coma, he began tracking objects with his eyes. At one month, he could follow simple commands. “His friends would come in the room, and there’d be two or three on each side of the bed,” Judy recalled. “And eventually, when they’d say, ‘Look at Jim,’ or ‘Look at Bob,’ he’d fix his eyes on the right guy.” Wayne and Judy asked for a follow-up M.R.I., but their neurologist said it would be pointless. Chris’s behaviors were entirely reflexive, he said; they were produced by his brainstem, which regulates basic functions like breathing and body temperature, not by his cortex, the region responsible for higher-order thinking. That Chris’s friends and family saw him following commands was proof of their denial, not of Chris’s recovery.
“Every couple days, the doc would stop in the doorway and shout Chris’s name to see if Chris responded,” Judy said. “But he wouldn’t come in the room and look at Chris up close. So one day, I practically grabbed his arm and dragged him into the room, over to Chris’s bed.” She told Chris to blink his eyes. He did. Then she made the doctor walk across the room and told Chris to keep his eyes on the doctor. He did. Finally, with the doctor standing across the room, eyes fixed on Chris, she asked Chris to give her a thumbs up. When he wiggled his thumb, just the tiniest bit, the doctor’s jaw dropped. Chris was not in a vegetative state after all. He was minimally conscious.
Still, there was little that the community hospital could do for him. It had neither the resources nor the expertise to tease out a prognosis or chart a course of therapy. The same was true of local nursing homes, which is where many patients like Chris end up.
So Wayne and Judy took over their son’s care, bringing him first to a premier brain-injury center in Atlanta (where Chris had a device implanted in his spine, which releases drugs to help with spasticity) and then to a clinic in Destin, Fla. (where he tried an experimental treatment known as hyperbaric oxygen therapy). They had just made their way back home to Tennessee when a friend told them about the Ambien paradox and the clinical trial in Philadelphia.
One hallmark of the minimally conscious state is a rapid fluctuation between levels of awareness. Spend 10 or 20 minutes with Chris Cox, and you might conclude that there is nothing going on upstairs. But spend a full hour, and at some point you’ll see his puppy-dog eyes come into focus. They will appear to search for one of his parents, or to settle quizzically on the new person in the room. Ask him to say something, and he’ll smack his lips frantically before leaning forward and tapping his feet in apparent frustration. You’ll swear that he is there with you and that only his physical infirmities (he cannot quite swallow or control his jaw) prevent him from describing the netherworld from which he has just emerged.
And then, a few minutes later, he’ll slip away again.
This fluidity makes diagnosis a challenge. “If a patient follows every command you give them, you know that,” says Dr. John Whyte, director of the Moss Institute and lead investigator on the zolpidem trial. “If a patient has never, ever followed a command, you know that too. But if you tell a patient to wiggle their finger, and they do it occasionally — which is the case for most of these folks — how do you figure out if that ‘occasionally’ means something or not?”
Whyte has spent his entire career trying to answer this question. His first job after his residency was at a facility with a large number of vegetative patients. While working there, he was struck by the amount of contention over diagnoses. For all their experience with this population, clinicians could not seem to agree on whether any given patient was actually conscious. Family members also argued, with one another and with staff, over the meaning of every wince, twitch and eye flutter.
It turned out that a lot of people — staff members included — were drawing their conclusions from pure coincidence. Whyte told me about one mother who insisted that her son would point down toward his feeding tube to indicate that fluid was leaking onto his stomach, causing irritation. “He did it while I was there,” Whyte says. “And she lifted his shirt and said: ‘See, doctor, there’s the liquid. He’s communicating with us.’ And I said: ‘How often do you look under there when he isn’t pointing like this? Never? Not even once?’ ” It was possible that the pointing corresponded to the leak, Whyte explained. But it was also possible that the leaking was constant and the pointing was random. There were countless other examples. “Behaviors would be exceptions if they happened at the wrong time, and evidence if they happened at the right time,” Whyte says.
To help eliminate this bias, Whyte developed what he calls the single-subject assessment, in which doctors design a set of tests specific to each patient’s idiosyncrasies to determine whether the patient is vegetative or minimally conscious. It is painstaking work, but the information it yields is significant. “Patients who achieve minimal consciousness early tend to have a better prognosis,” Whyte says. “And you can at least try to build a communication system with them, because you have a foundation to work from.”
With a reliable assessment method in place, he began searching for ways to build on that foundation. Then the curious Ambien awakenings caught his attention.
It’s not entirely surprising that Ambien would arouse instead of sedate. The pill has long been linked to reports of bizarre sleepwalking behavior (not to mention sleepeating, sleeptalking, even sleepdriving). Some scientists call this phenomenon “paradoxical excitation.” So far, none of the accepted determinants of prognosis — age, overall health, the nature of the initial injury or the extent of brain damage as determined by an M.R.I. — have proved useful in predicting which brain-injured patients will experience it and which won’t. To begin answering that question, Whyte says, you need to study both responders and nonresponders in an unmedicated state.
One morning this past March, I met Chris, Wayne and Judy at the University of Pennsylvania’s main hospital, where they had been flown in from Tennessee, at the study’s expense, so that Chris could be tested in an unmedicated state. From the corner of a small hospital room, we watched as Whyte’s research assistant, Andras Szeles, attached dozens of tiny electrodes to Chris’s face and scalp, then fitted him with a large headset. The electrodes would measure Chris’s brain activity as Szeles administered a series of cognitive tests.
For one test, Szeles placed a rubber glove on Chris’s right hand. A voice coming through the headset told Chris to either “squeeze glove” or “squeeze bare,” several times over. Chris did not seem to be responding at all, but Szeles explained that the electrodes would measure what the naked eye could not. “We’re not so interested in whether or not Chris can squeeze,” he said. “We just want to know if he’s trying to squeeze.” Different neurons fire when you move your left hand versus your right hand. They also fire if you imagine moving it, prepare to move it or start to move it but stop, all of which the electrodes would detect.
“The term ‘consciousness’ can be a real can of worms,” Szeles said. “There are degrees of awareness, and it’s not always clear what the threshold should be. What we’re really looking for here is evidence of comprehension and will.” If Chris understood the words “squeeze” and “glove,” knew that this very specific thing was being asked of him, and possessed, at the most basic level, the will to respond, a distinct pattern of brain waves would show up in the results.
After Whyte and his team have tested 80 patients, they will compare the results of zolpidem responders to those of nonresponders and look for clues that might help explain the difference — maybe a specific brain region that lights up unexpectedly, or a pattern of neuronal firing common to one group but not the other. Any such discovery could light a path not only through the labyrinths of Chris’s fractured mind but to a better understanding of consciousness itself.
Wayne and Judy have a more immediate question: they want to know their son’s long-term prognosis. Has he reached the pinnacle of his cognitive recovery? Or is it a launching pad from which greater heights might be reached?
“Once a patient progresses to minimal consciousness, we can’t predict what’s going to happen,” says Dr. Joseph J. Fins, chief of medical ethics at Weill Cornell Medical College and author of a coming book, “Rights Come to Mind: Brain Injury, Ethics and the Struggle for Consciousness.” Some patients have recovered full consciousness, but many more remain stuck in limbo. The only way to know the outcome is to give the patient time.
But offering time is a complex proposition. “Early on, when families have the option to pull the plug, it’s almost impossible to tell what the long-term prognosis will be,” says Dr. Soojin Park, a neurointensivist at the University of Pennsylvania Hospital, and an investigator on the zolpidem trial. “And then later, when we have the certainty — that this is as good as it’s going to get — that option is gone. Because by then, the patient is breathing on their own. There’s no more plug to pull.” At that point, families who want to end a loved one’s suffering must either have the feeding tube removed, or agree to let the next bacterial infection win out, unhindered by antibiotics. Many families find choosing these deaths much more difficult than turning off a ventilator. It’s an instinct reinforced by religious edicts that forbid the withholding of basic sustenance but allow, for example, unplugging artificial respirators.
It is not uncommon for doctors to assume the worst and advise family members to withdraw care early. They do so in part because they see their duty as helping loved ones face reality. But Fins argues that this is a cop-out. “It’s glossing over all the unknowns for the sake of a quicker, cleaner solution,” he says. “It’s wrong to be so uniformly fatalistic so early on, especially with all the data emerging about the prospects for later-stage recovery.”
According to several studies, about 40 percent of patients who have been declared vegetative are actually minimally conscious. Other studies have shown that a surprising number of vegetative and minimally conscious patients made huge strides toward recovery much later than conventional wisdom would predict.
Park says that more doctors are trying drug therapy on vegetative and minimally conscious patients, but for the most part, they are groping in the dark. “We still don’t understand which drugs should work on which patients, or at what dosage, or at what point in their recovery,” she says. “And that makes it tough for families to know when they should fight and when they should give up.”
There is also the matter of cost. Treating and monitoring patients like Chris — designing and performing single-subject assessments that can discern random twitches from deliberate behavior, managing the host of medical complications that can stymie brain recovery and continually evaluating progress — is significantly more expensive than placing them in nursing homes, where they receive basic care but have no access to brain-injury specialists. Proponents argue that the measures will save money in the long run — if the patient is able to go home, for instance. Still, it’s unclear whether even the most aggressive care will make much difference for many patients. “The payers need a better sense of what the likely outcome is for any given patient,” says Tom Smith, program director at Moss, “so that they can say with confidence which patients are likely to benefit from treatment, and how significant that benefit is likely to be. And I hate to sound this way, but then it’s basically: ‘Am I going to invest this amount of money to get this outcome? Is that worth it?’ And that is a tough, tough question to answer.”
When Chris first returned from the hospital, the Coxes’ house was flooded with well-wishers. A former teacher brought a quilt that Chris and his classmates made in grade school. Old girlfriends, acquaintances and childhood pals visited regularly. His high-school friends even held a fund-raiser to help pay for some of his therapy. But as time wore on, and it became clear that he would neither die nor fully recover, the guests dwindled. Besides Chris’s sister, Amber, who visits often, Wayne and Judy are now mostly on their own.
The couple surrendered their bedroom, which is on the ground floor of their split-level house, to Chris. They take turns sleeping there, on a small cot in the corner of the room, surrounded by medical equipment: a hospital bed with a queen-size air mattress that helps prevent bed sores, a special chair called a “sit-stand,” which can be cranked into an upright position and is supposed to help ease muscle contractions. A mechanical lift. A breathing machine.
Judy is especially keen to show me Chris’s latest gadget, a device called an Eyegaze, that looks a bit like an iPad, and which Chris is learning to use to communicate. She mounts it to a handlebar on Chris’s wheelchair. At first, Chris’s eyes move rapidly across several rows of icons — words with pictures, partial sentences like “I want” or “I like,” and cartoon images of friends and family members with their names in bold. Tracking his eyes, the computer reads: “Green. Brit. I want.” And then, “Brit–Brit–Brit–Brit–Brit.” Brit is his cousin. It’s unclear if he’s asking for her, or if his eyes are just not cooperating.
Judy puts the computer away. He still needs practice, she says. When they first tried it, in February, during outpatient rehab, the therapist asked Chris what color went with Valentine’s Day, and he was able to look at the “red” icon. But so far, “I want a drink,” is the only full sentence he has managed.
Each day begins with the same routine: Judy helps Chris up, forcing him to sit on the edge of the bed, without neck or back support, for several minutes. The goal is to strengthen his neck muscles so that one day he can hold his own head up. A nurse helps her bathe him and then lift him into his chair, so that he can be wheeled into the living room or taken outside. Three times a week, a van delivers mother and son to a rehab center, where a therapist works to stretch and stimulate Chris’s contracted muscles. And almost as often, either Wayne or Judy spends several hours on the phone, battling Chris’s insurance company, which they say has covered private nursing and weekly therapy but denied an extended stay at a brain-injury rehab center. The couple arrange their schedules so that their son is never without at least one parent.
The Coxes have surrounded themselves with a new group of friends — other parents with children like Chris. The families in this informal group share strategies and trade information on emerging research and experimental treatments. They also talk of weariness and isolation. “At the end of the day, you feel like you’re a thousand years old,” Judy says. “And you have no idea how you’re going to get up and do it all over again tomorrow.” Both Wayne and Judy admit to being frustrated by the slow pace of Chris’s progress. But he’s come this far; they can’t help hoping that he might come further still. “I know that some people, even people who loved Chris, think maybe he should have just died,” Wayne says. “But he didn’t die. He lived. And as long as he’s still breathing, we have to do the absolute best we can for him.”
The reports on zolpidem are still mixed. Viljoen and a few others have improved steadily over time; some of them are now fully conscious on their own, without medication. (Viljoen is confined to a wheelchair and has cognitive disabilities but has improved over the years.) But such improvement is rare. According to Whyte, most responders fall into one of two categories: those who can take zolpidem daily, with no appreciable loss of efficacy, and those for whom the “awakenings” wane with continued use. The latter type, he says, may be the most common.
With no way of knowing which type of responder Chris might be, the Coxes play it safe. They give Chris Ativan every day; it has a similar though less profound effect on his behavior. But zolpidem they hoard like pixie dust, giving it only on special occasions, when friends and family can be there.
They gave him some when I visited, so I could see how it works.
A few minutes after receiving the zolpidem, Chris opened his eyes and smiled. Judy sat on the bed, facing him. “Hey, buddy!” she said. “Can I have a kiss?” She leaned in, cheekward, but he turned his head up in what looked like a deliberate slight. “Chris!” she admonished, playfully. He smiled and grunted. He was teasing her. Just as she pulled away, he thrust forward and grazed her cheek with puckered lips. Confident that he was there with us, Judy took out a marker and some paper and wrote one command after another, each of which Chris followed: Stick out your tongue, give me five, give me a thumbs up. And then, “Show us your Elvis grin.” Chris curled his upper lip into a sneer. When Judy ran out of commands, Chris began smacking his lips and moving his tongue. “Talk to us, buddy,” Wayne said. “Say, ‘Mom,’ ” Judy said. After several moments, Chris managed a loud, slow “Maaaa.”
