Dr. H. had a struggling heart and a consequential decision to make. He needed an operation to fix a faulty valve but postponed it, afraid that the surgery and anesthesia would irrevocably damage his brain. The man made this decision more than a decade ago based on a phenomenon that was reported in an alarming article all the way back in 1955—one that many doctors still see first hand—of older patients who deteriorate cognitively after surgery.
“When patients are asked, ‘what is the thing that you fear the most?’ it’s being disabled,” says Mervyn Maze, an anesthesiologist and researcher at the University of San Francisco. “It’s not dying. It’s living with a cognitive disability. And that becomes even more scary the older that you get.”
The author of the 1955 The Lancet article was a surgeon and consulting physician in England who reviewed thousands of charts and zeroed in on 18 “extreme” cases of people who appeared to be cognitively normal before the operation but became severely demented afterwards. He concluded that “operations on elderly people should be confined to unequivocally necessary cases.”
It was a bold statement, and one that likely overstated the risk. Today, you would be unlikely to find a surgeon who would advise against surgery that could significantly improve your quality of life. But the potential for damage remains. Dallas Seitz is a young geriatric psychiatrist at Queen’s University in Ontario, Canada, who says his colleagues are familiar with memory or cognitive problems after surgery, what’s been known as postoperative cognitive dysfunction, or POCD. “I would be hard pressed to find a colleague who said I’ve never heard of it or I don’t think that that exists,” he says, though he’s quick to acknowledge that such anecdotes are not scientific evidence. “It’s not an experimental study. But I think we can’t discount what our patients tell us and what our experience tells us as well.”
Because some 60,000 people undergo surgery daily in the U.S., and the numbers continue to climb, quantifying the risk is essential. It hasn’t helped that the term “POCD” is vague. “We’re talking about things that happen immediately after operations, like delirium, to things that happen perhaps a decade after an operation like dementia or Alzheimer’s,” notes Roderic Eckenhoff, an anesthesiologist and researcher at the University of Pennsylvania. New, more precise terms will soon be announced, and even though the picture is muddled, there is hope that the tools of modern science—from animal models to well-designed clinical trials—will bring more clarity.
Delirium is the most likely complication of surgery in older adults, with a risk as high as 60%. Patients with delirium are in a temporary and sometimes fluctuating type of acute brain failure: they can be confused, sometimes hallucinatory, aggressive, agitated, or sleepy and passive. Some small subset have mild cognitive impairment that can last for months before they recover. Their memory is not as clear and they describe themselves as being in a fog. They are generally less competent at planning or making decisions. And, worryingly, some remain permanently impaired or even descend into dementia, with studies suggesting that this outcome is more likely in patients who experience delirium compared to those who don’t.
But it turns out that many older people were already on the road to dementia before their surgery. “It’s a standard sort of thing that we hear, that we are contacted by people who say, ‘you know, Mom was perfectly fine or Dad was perfectly fine until they had their knee replacement,’ says Lis Evered, an anesthesiologist and researcher at St. Vincent’s Hospital Melbourne in Australia. “But when you drill down on the ‘perfectly fine,’ they weren’t perfectly fine. It’s just that they were coping. And people develop strategies that can cover a decline quite markedly.”
Between 10–34% percent of people over the age of 65 are already in a state of mental decline ranging from mild cognitive impairment—problems that go beyond what you would expect in normal aging—to dementia, including the most common form, Alzheimer’s disease. Evaluating someone’s mental status can be tricky, and not just because they’ve developed coping strategies. A visit to the surgeon before the operation occurs at a particular moment in time, but patients have good and bad days, and cognitive decline doesn’t always follow a smooth trajectory. It’s only after surgery that an underlying vulnerability may be unmasked. A recent study published in The Neurobiology of Aging showed that in apparently normal adults undergoing elective surgery with general anesthesia, a particular brain signature associated with the development of Alzheimer’s disease was tied to how severe the postoperative delirium was.
“Patients are saying, ‘look this is my husband, he has cognitive impairment, and I’m really worried about conversion to Alzheimer’s,’” says Dr. Charles Brown, a cardiac anesthesiologist and researcher at Johns Hopkins University, “‘How does this particular procedure affect that? Should I get it? Should I not? If I do get it, how are you going to manage me differently?’ “
Quantifying the Risk
“It’s very important to really be careful about what we know, about what we don’t know, and what we anticipate or hypothesize,” cautions Lars Eriksson, an anesthesiologist and clinical researcher at Karolinska University Hospital in Sweden. Many studies have significant weaknesses, adds Michael Avidan, an anesthesiologist and researcher at Washington University in St. Louis, and that includes overlooking shared risk factors.
“If you have a cohort of people who have a disease that is known to be associated with cognitive decline,” Avidan says, “like hypertension, people who are heavy smokers, who are obese, who have diabetes, who have peripheral vascular disease, who have arthritis—whether or not they undergo surgery I can predict that those people are going to decline cognitively more than people who are marathon runners who are fit and all the rest of it.” Moreover, hospitalization itself, with or without surgery, has been shown to be a risk factor for cognitive decline, for reasons that are not entirely clear.
Avidan believes the evidence for the idea that anesthesia and surgery cause long term cognitive decline is “incredibly flimsy,” and points to a 2013 randomized controlled clinical trial published in The Annals of Thoracic Surgery as strong support for the idea that the risk is overblown, or perhaps even non-existent. The study found that patients who had open heart coronary artery bypass surgery with general anesthesia were better off cognitively 7.5 years after surgery than those who were treated with a less invasive non-surgical procedure that used only regional anesthesia. (Because regional or local anesthesia does not enter the brain, it’s assumed that it has no impact on brain function.)
In contrast, a 2017 study in Alzheimer Disease and Associated Disorders reached the opposite conclusion, finding an association between coronary artery bypass graft surgery and long-term dementia risk. The researchers controlled for many typical risk factors but the study was not randomized.
These are just two studies out of hundreds of others published over the past decade, each with different groups of patients or types of operations. That makes it difficult to reach a definitive conclusion on whether surgery and anesthesia cause permanent cognitive decline.
Managing the Immune Response
Perhaps it shouldn’t be so surprising that some elderly patients experience profound changes to their mental capacities after an operation. Surgery is a violent assault on the whole body, of which the brain is a part. Anesthesiologists administer drugs that push the brain into a reversible state of amnesia, immobility, loss of consciousness, and freedom from pain. Surgeons cut open bodies, stop hearts, replace hips or knees with titanium, and stitch stomachs smaller and tighter. In response, the traumatized tissues release substances that engage the immune system, activating a host of key cells and proteins that encourage healing. A mere 12 hours after surgery, signs of inflammation can also be detected in the cerebrospinal fluid. This is actually a good thing.
“There is a signal within the body after peripheral injury that’s purposeful, that reaches the brain,” Eriksson says. “And that contributes to a condition that you could call sickness behavior, or the injured phenotype. It’s time to rest—to not be out hunting for food—and to heal. But, of course, when the injury becomes so large and big and it goes over the top, the situation can be different.”
An immune response that doesn’t know when to quit is likely to be dangerous. In the elderly, the blood brain barrier is weaker, giving immune cells easier access to the brain. In aged rodents, scientists have found inflammatory molecules crossing the blood brain barrier and setting up shop in areas that are essential for learning and memory, such as the hippocampus. In human patients, researchers also found amyloid beta and tau—two proteins associated with Alzheimer’s disease—went from normal levels before cardiac surgery to levels associated with mild cognitive impairment six months after the operation. Other studies have shown similar changes as soon as 48 hours after surgery.
This year, an international study by Eriksson and his colleagues pulled off a pioneering feat—using PET scans to image the immune response in patients before surgery, three to four days afterwards, and again at three months. In four of the eight patients, they saw profound and global changes in the brain’s immune system three months after surgery, long after the patients were home and going about their daily lives.“There was an association between the amount of immune activity and the decline in cognitive capacity,” Ericsson says. “The more unregulated your immune system was at three months, the worse was the cognitive function.”
But finding a way to safely and effectively intervene in the immune response is notoriously difficult. In the case of surgical patients, extinguishing the initial inflammation could disrupt important processes like wound healing. “I suppose until we actually understand the mechanisms that are occurring in the perioperative period, trying to address what interventions might have an impact is very difficult,” says Evered, the Melbourne-based anesthesiologist. Many experts underscore the importance of using animal models to suggest possible mechanisms for the disease process so as to develop treatments. But a key interim step is to make sure that that same mechanism translates in humans. This is what makes Eriksson’s recent PET imaging study so important—it makes the connection between what we see in rodent models and what we see in us.
Regardless of the actual mechanism, doctors and researchers are hard at work to find anesthetics and postoperative practices that will minimize the risk of any sort of cognitive decline. “If you are going to have a procedure, what are the ways that we can protect the brain the most?” asks Dr. Brown, who is investigating whether optimizing blood flow to the brain during surgery can make a difference. “Maybe you’re someone with mild cognitive impairment—and that is such a strong predictor of further cognitive decline—maybe you need to take that into account a lot more.”
Optimizing anesthesia could be one way to mitigate risk. Anesthesia was the original leading contender for cognitive decline, but while studies in rodents have shown toxic effects, the data for human trials has not. In a 2014 study of more than 12,000 patients, Dallas Seitz from Ontario found that the risk of dementia in patients who had an operation under general anesthesia versus regional, or local pain control, was “almost identical in the two groups. There was no statistical difference.” This is but one of many studies that lead anesthesiologists like Eckenhoff to believe that “the anesthetic is a minor player.”
While we don’t know if anesthesia can hurt the human brain, what if certain types of anesthetic drugs could help protect it? Until recently, investigating this idea has been hamstrung by the simple fact that it would be unethical to perform surgery—whether on people or lab animals—without any anesthesia. But a discovery by Maze and colleagues provided a neat workaround. They showed that a protein called HMGB1, which launches the immune response to trauma, can simulate the inflammatory effect of surgery. “We now don’t have to do surgery to cause the surgical effect,” Maze says. “We can just provide an injection of HMGB1 and start the process that way.” Maze is a co-founder of a start-up company called NeuroproteXeon that is in a phase 3 clinical trial of xenon. Xenon is not the only drug or class of drugs being investigated for neuroprotective qualities. For example, in small clinical trials, the anesthetic agent ketamine has been shown to reduce postoperative cognitive impairments, perhaps by modulating the immune response.
Another way to optimize anesthesia could be to fine-tune the amount a patient receives. Typically anesthesiologists track heart rate and blood pressure as their guide for deciding whether the patient has the right amount of drugs on board. But Avidan and others are investigating whether monitoring brain waves during surgery using an EEG can identify a “goldilocks” amount of general anesthesia that could decrease the risk of delirium. It’s easy to give more drugs than necessary, particularly in the elderly. “We’re trying to prevent or to avoid a pattern called EEG suppression during general anesthesia, where the EEG looks like it’s flatlined, and to see if that decreases postoperative delirium,” says Avidan.
Optimizing Patient Care
Other measures could include making sure sleep and nutrition are optimal and that patients are given the right antibiotics at the right time to decrease the likelihood of surgical wound infections.
And then there’s a decidedly low-tech approach: reorienting patients quickly. This includes having friends or family nearby as soon as possible after the operation, giving back their glasses, dentures, and hearing aids, or putting clocks and calendars on the wall. While it’s unproven whether this reduces delirium, “there’s no downside to doing these things,” Avidan says. “And it’s kind. If nothing else it’s respecting people’s humanity and dignity, and making their surgery experience more pleasant. So you have enough reason to do that regardless of whether or not you think it’ll improve short or long term outcomes.“
None of these approaches are likely to tackle the problem on their own, though. Robert Whittington, an anesthesiologist and researcher at Columbia University Medical Center, addressed this question at the 2017 meeting of the American Society of Anesthesiologists, pointing out that given the complexity of the brain, rather than trying to find the magic bullet therapy, there should be a multi-pronged approach similar to treatment strategies for HIV.
As with many medical procedures, and as with Dr. H., who chose not to get heart surgery, making tough decisions goes back to looking at risk and benefit. Surgery is a powerful stress test that can unmask underlying cognitive problems. But by the same token, surgeries that repair your heart or your hip, that allow you to exercise, or that reduce pain can improve your cognition.
Older people should be aware that their brains may still be healing long after they go home. But there is still an unanswered question: can surgery and anesthesia accelerate or even cause a permanent cognitive decline, even if the surgery and aftercare are optimal? On top of all the vulnerabilities that elderly patients may enter the operating room with—from previous risk factors for cognitive decline, to leaky blood brain barriers, to the mere fact of being in a hospital—is there a specific surgical vulnerability for permanent cognitive decline that has yet to be nailed down?
“There can be a little subgroup, perhaps not very big, that has a very unfavorable outcome,” says Eriksson. “And if you don’t do standard studies, using standard designs, those people may go undetected. The biological signal might be very small. There might be vulnerable individuals out there where a particular trigger—in this case surgery—where the pathways are very important. We might not be able to detect them. There could be vulnerability out there that we need to define.”