Going Under

Emery Brown knows how to take the sting out of surgery. As an anesthesiologist, he has steered hundreds of patients to pain-free oblivion, allowing doctors to go about their business resetting bones, repairing heart valves or removing tumors. During surgery he continually monitors his patients, keeping tabs on their heart rate, blood pressure and breathing. Recently, he has also been eyeing what happens in their brains.

Rather than going under the knife, some of the people in Brown’s care are going into scanners to reveal how the brain responds when people are knocked out. These deep glimpses could answer vexing questions about how people enter the state of unconsciousness known as general anesthesia and what happens in the brain while they are there.

Although it is widely used and remarkably effective, anesthesia’s neural mechanisms have long remained mostly mysterious. While every anesthetic drug has its own effect, scientists know little about how the various versions work on the brain to transport patients from normal waking awareness to dreamless nothingness.

Understanding how such compounds meddle with the nervous system might lead to anesthetics capable of tweaking neural circuits more precisely, delivering only what is needed where it’s needed, says Brown, of Massachusetts General Hospital in Boston. Fine-tuning the drugs’ effects could also help doctors bring patients into and out of consciousness more quickly and safely, avoiding the side effects that can occur when medications act at brain sites other than those intended or act at targeted sites for too long a time.

That may be reassuring to the more than 20 million patients who are put under general anesthesia each year in the United States. Though new drugs and procedures have made anesthesia safer and more comfortable, patients still may experience nausea, abnormal heart rhythms or fogginess after surgery. In extremely rare cases more serious effects such as brain injury or death can result.

“We’ve made a lot of progress in anesthesiology as far as taking care of patients,” Brown says. “But it’ll be much more reassuring when we can say to them, ‘We have a good idea what’s happening inside your brain.’ ”

Beyond all that, mapping the pathways of neural activity taking the brain to this particular type of unconsciousness may help scientists understand how it compares with sleep and coma, which may ultimately lead to new sleep medicines or new ways to help patients recover after a severe brain injury.

Painless experiments

Before general anesthesia’s discovery in the 1840s, patients simply had to endure the trauma of surgery, although alcohol or opiates sometimes numbed the pain. After observing in 1844 that nitrous oxide — or laughing gas — could stifle pain, dentist Horace Wells had a tooth pulled while on the gas. Taking several deep inhalations, he nodded off, giving his colleague ample time to yank the tooth. Feeling no pain during the procedure, Wells took his discovery to the medical community: At Massachusetts General Hospital, he gave a patient a whiff of nitrous oxide before extracting a tooth.

The demonstration didn’t go as planned, as the patient moaned during the procedure. But another dentist, William Morton, began experimenting with ether. In 1846 Morton used ether to knock out a patient while a surgeon removed a neck tumor. By 1847 either ether or chloroform was routinely administered during surgery to put patients into a dreamless, pain-free state. In the early 20th century, as surgical procedures advanced, these gases were replaced by mixtures of nitrous oxide and oxygen or intravenous narcotics, ultimately giving doctors more control over zonked-out patients.

Today, anesthesiologists administer about a dozen drugs to produce the desired effects. Sedatives help with relaxation, opiates take away the pain and a muscle relaxant paralyzes the body. Other drugs added to the mix render patients unconscious and make sure they don’t remember the experience.

During surgery, anesthesiologists know when to intervene. They use a mechanical ventilator to control breathing, ensuring that a patient is inhaling and exhaling slowly, deeply and rhythmically, and they continually check for signs of perspiration. Beeping monitors signal trouble if it should arise.

Though not standard practice, brain activity is recorded during some surgeries. Electroencephalograph, or EEG, signals track brain cell firings by measuring surface electrical activity through sensors usually placed on a patient’s forehead. Some EEG monitors do additional number crunching — translating readings into ballpark assessments of a patient’s level of consciousness and spitting out a number ranging from 0 to 100. A value of 100 means that the patient is fully awake, while a score of 0 — a flatline — indicates no brain activity at all.

Anesthesiologists generally want to keep a patient’s score within a range of 40 to 60.
While such measurements give anesthesiologists a rough way to gauge how much anesthetic is needed — preventing too much or too little — the devices can’t always tell whether or not a patient retains any sensory awareness. Anesthesiologist George Mashour of the University of Michigan in Ann Arbor says that doctors still don’t have a monitor that can reliably detect consciousness in a paralyzed and otherwise unresponsive patient.

“The reality is, there is no standard device or monitor for the brain during surgery,” Mashour says. “Which is pretty interesting if you consider the fact that the brain is one of the main target organs of general anesthesia.”

Lights out

One of the difficulties in learning how the brain suspends consciousness is the way in which general anesthesia is induced, Brown says. Drug mixtures are administered first intravenously and then by inhalation, sending multiple drugs throughout every part of the brain and nervous system within seconds. Because the drugs go everywhere, it has been difficult to discern which circuits need to be hit for a person to reach a surgery-ready state.

But a picture is beginning to emerge. Using PET (positron emission tomography) and functional MRI scanners, scientists can actually image brain activity when people go under. Though such techniques can’t be used during surgery, they help reveal which brain areas are affected by anesthetic agents and when.

Images studied so far suggest that anesthetic drugs wield their effects by altering connections between nerve cells, making it difficult for different brain areas to talk. Areas highly affected by the communication breakdown include the cerebral cortex, the wrinkled layer of “gray matter” at the surface of the brain; the thalamus, a ball of tissue at the center of the brain; and the brain’s arousal centers, located at the top of the brain stem, the intersection between the brain and the spinal cord.

Studies looking at how various knockout drugs break up the organized patterns of activity among these brain regions reveal disruptions in several circuits. A key circuit involves the cortex and thalamus. The cortex plays a role in attention, language and information processing, and the thalamus acts as a relay station for sensory information flowing into the brain. By passing signals to each other via nerve cells, the two areas help people make sense of what they see, hear and feel.

Studies in animals and humans on anesthetics show that blood flow to the thalamus is reduced, disrupting this crucial connection. But last summer, scientists directed by Irene Tracey of Oxford University in England reported that another brain structure, the putamen, is actually the first to unhook from the rest of the brain under anesthesia. Located within the basal ganglia, a bundle of nuclei deep in the brain, the putamen plays a role in controlling movement. Though this disconnection had been noted in animal studies, Tracey’s team was the first to document it in humans.

Eight healthy volunteers were placed in an fMRI machine after receiving a dose of the widely used anesthetic propofol. As the volunteers became unconscious, they were intermittently asked to reply to verbal cues or were zapped with a pain-causing device.

Tracey’s team noticed that as people became unresponsive, connectivity between the putamen and other brain regions decreased steadily while the connection between the thalamus and cortex remained intact. The findings, reported in the Journal of Neuroscience, fit with observations from the operating room, where patients often exhibit brief, jerky movements as they slide into a surgery-ready state, Tracey says. Such movements may represent the uncoupling of normal controls from the main motor system in the brain. Tracey’s group is now combining fMRI with measurements of electrical activity in the brain to further explore how this uncoupling occurs.

“The EEG is brilliant at telling you something happened over on the right-hand side and in a millisecond moved on to the left, but you don’t know where on the right or quite where on the left,” Tracey says. By pairing that info with fMRI, the scientists hope to pick up on subtle connectivity changes as they occur.

Similar studies are under way at Massachusetts General, where Brown and researcher Patrick Purdon are combining fMRI and EEG to document how different knockout drugs act in various brain regions. These studies and others may help answer another key question about the ability of anesthetics to induce unconsciousness: Is the brain taken down in a single step, like turning off a light in a room, or is there a hierarchy of switches to be flipped?

Tracey says that, from a biological or evolutionary perspective, one might think of anesthesia “as a house with lots of rooms, where you take the lights out in a different order.” Such a step-by-step disconnection might allow the brain to better preserve some functions until unconsciousness is reached.

Wake-up call

Scientists see a similar series of disconnections in brain areas when people catch their daily dose of z’s. Sights, sounds and other distractions often melt away as the chatter between the thalamus and cortex quiets and sleep ensues. But even at the deepest stage of sleep, a barking dog or a good shake can rouse a person. Most certainly, the slice of a knife will do the job.

In contrast, people under general anesthesia can’t be stirred. As the drugs take effect, patients are rendered senseless, numb to the world around them and within. Though the anesthetized brain uses some of the same neural machinery as sleep, studies of brain activity show that the two states are in fact entirely different. The steady, slow-rolling EEG patterns induced under general anesthesia more closely resemble those seen in coma. Brown, Nicholas Schiff of Weill Cornell Medical College in New York City and Ralph Lydic of the University of Michigan reviewed the similarities between general anesthesia and coma in the Dec. 30 New England Journal of Medicine.

Schiff, a neurologist who helps brain-injured patients recover consciousness, is poring over the EEG patterns seen in anesthesia cases looking for common links to those found in patients recovering from coma. Such commonalities may point to a “signature” for awareness that could be detected with EEG.

While he has yet to find such a marker, Schiff has identified brain behaviors that are similar in the two groups. As drugs administered to patients undergoing general anesthesia begin to take effect, activity between neurons may actually increase for a short period. (The same effect can explain the excitable feeling you get after a single glass of wine.) As more medication is given, patients give in to sedation.

Schiff sees similar patterns of temporary excitability in some brain-injured patients. Those who are “minimally conscious” — meaning they show only occasional awareness of themselves and their surroundings — can respond to low doses of sedative drugs by waking up. Here, the anesthetic drugs “kick-start” the brain, helping it to turn itself back on.

“Patients might start to speak, or suddenly gain the ability to stand or walk or chew or swallow,” Schiff says.

Over time, with repeated doses, the mere act of being alert can help the underactive brain make gains toward recovery, Schiff says. In the January 2010 Trends in Neurosciences, he suggests that a switching mechanism explaining this excitatory effect may rely on links between the thalamus, the cortex and two brain regions — the globus pallidus and the striatum — that help regulate activity between the cortex and thalamus.

Back in the operating room

While Schiff looks for a signature of awareness for patients in coma, other labs are looking for markers that might help anesthesiologists track surgery patients’ awareness. 
By correlating the squiggly EEG signals of anesthetized brains to images showing changes in blood flow in specific brain regions, Purdon and Brown hope to identify patterns that could be used in the operating room.

“Once you know that connection, you can then identify specific EEG patterns and make inferences about what different brain systems are doing when you see that pattern,” Purdon says.

Mashour and University of Michigan physicist UnCheol Lee are taking advantage of graph theory, a mathematical approach used to study networks, to link the EEG patterns seen in anesthetized patients to knowledge about the underlying brain networks. Ideally, such a system could be set up to track the moment-to-moment changes seen in patients during surgery.

The team has already found that clinicians may get more information on patient awareness by gluing the EEG monitors farther back on the head, rather than on the forehead. In the April Anesthesiology, Mashour’s team showed that networks in the parietal region display greater levels of disruption during anesthesia than those in the front of the brain. “The parietal lobe may be an important hub or point of convergence for information processing in the brain,” he says.

Such studies may allow anesthesiologists to fine-tune their procedures during surgery, knowing exactly where to steer the brain to balance drugs’ main effects against a host of potential side effects.

“Certain brain areas, if activated, can cause nausea and vomiting. Others can adversely affect your respiratory system,” Brown says. “If I can have a drug avoid going there, I’d love to do that.”

Michael Alkire, who studies the neural biology of anesthesia at the University of California, Irvine, says that by knowing how anesthetics affect different brain areas, researchers may be able to develop new therapies or find ways to customize treatments. He foresees a day when a patient’s genes are analyzed before surgery to determine sensitivity to drugs or potential to suffer certain side effects. “If you know that a patient is very anxious, and maybe that anxiety is related to the amygdala function in the brain, you might want to use a different agent or anesthetic for that case,” he says.

Understanding how anesthetics work in the brain could also lead to entirely new ways of creating anesthetic states. Future anesthesia may rely on Star Trek–like devices with energy fields capable of disrupting key circuits, for example.

In such a scenario, patients wouldn’t need to take any type of medication at all, meaning side effects would be minimal, Alkire says. “But first,” he says, “we have to figure out the anesthetics.”

---

Bridges in the brain

Mathematician Leonhard Euler proved in the 18th century that there was no journey through Königsberg, Prussia, that would take a walker across each of the city’s bridges once and only once. By simplifying the city’s layout — four land areas connected via seven bridges (as shown) — into a series of nodes connected via paths, Euler established the foundations for modern network theory. Today researchers study the brain by similarly reducing it to an architecture of nodes and paths. Neurons or groups of neurons are Königsberg’s landmasses, and connections between those neurons are the city’s bridges. Recent work suggests anesthesia can disrupt the brain by interfering with the system in two ways — changing the layout of the nodes and altering the efficiency of the pathways.

Credit: Janel Kiley