Editor's note (12/14/16): After weighing the latest science on general anesthetic and sedation exposure in young children, the U.S. FDA is requiring that warnings be added to the labels of these drugs about potential risks for children’s developing brains related to repeated or lengthy exposures.

The game is a contemporary of the original Nintendo but it still appeals to today’s teens and lab monkeys alike—which is a boon for neuroscientists. It offers no lifelike graphics. Nor does it boast a screen. Primate players—whether human or not—are simply required to pull levers and replicate patterns of flashing lights. Monkeys get a banana-flavored treat as a reward for good performance whereas kids get nickels.

But the game's creators are not really in it for fun. It was created by toxicologists at the U.S. Food and Drug Administration in the 1980s to study how chronic exposure to marijuana smoke affects the brain. Players with trouble responding quickly and correctly to the game’s commands may have problems with short-term memory, attention or other cognitive issues. The game has since been adapted to address a different question: whether anesthetics used to knock pediatric patients unconscious during surgery and diagnostic tests could affect a youngster's long-term neural development and cognition.

Despite 20 years’ worth of experiments in young rodents and monkeys, there have been few definitive answers. To date, numerous studies suggest that being put under with anesthesia early in life seems somehow related to future cognitive problems. But whether this association is causal or merely coincidence is unclear.

Researchers do know that the young human brain is exceptionally sensitive. When kids are exposed to certain harmful chemicals in their formative years, that experience can fundamentally alter the brain’s architecture by misdirecting the physical connections between neurons or causing cell deaths. But unraveling whether anesthetics may fuel such long-term damage in humans remains a challenge.

The connection does seem plausible. Anesthetics are powerful modulators of neurotransmission, or communication between neural cells, so the idea that early exposure to these chemicals may alter brain development does not seem far-fetched. Moreover, anesthesia exposure in animals has been linked to long-term learning and memory problems for almost all commonly used anesthetics.

Merle Paule, director of the division of neurotoxicology at the FDA’s National Center for Toxicological Research, has spent decades studying how a variety of chemicals affect animals. Four years ago he and his colleagues reported that when rhesus monkeys are put under with ketamine—an anesthetic sometimes used for kids during short, painful procedures—it is associated with lasting damage to the brain as compared with control group monkeys that were not exposed. When the monkeys were five or six days old, they were put under using ketamine for a 24-hour period. The anesthetized monkeys, as a group, subsequently performed worse than control monkeys in tests on learning and discriminating by color and position. Some three years after that exposure the ketamine monkeys were often unable to select as many levers per second compared with the control animals. The differences, however, were relatively subtle and perhaps would not make much of a difference in the lives of individual monkeys. Yet on a larger level, because the subjects performed slightly worse than the controls, it gives researchers and clinicians pause. And still, seven years after their ketamine treatment, those monkeys continue to show below-normal brain function, Paule says.

What’s more, his team also showed in separate work that similarly exposed monkeys suffer more brain-neuron deaths. Most recently Paule’s team has found in preliminary work that when monkeys were put under with a mix of isoflurane and nitrous oxide—similar to what is often used in young humans—just one eight-hour period of anesthesia was linked with long-term development and learning issues in those nonhuman primates. But translating that finding to humans is not perfect: Pediatric surgery in humans rarely takes that length of time.

The new monkey work has not yet been submitted to a peer-reviewed publication but Paule presented preliminary results during a recent FDA Science Forum open to the public in May 2015 at the agency’s offices in Silver Spring, Md. Based on these kind of findings, Paule says, researchers need to explore if there is a harm threshold for each anesthetic regimen used in humans and determine if there is anything that can be done to ameliorate or prevent the adverse effects already seen in nonhuman primates.

In humans, a growing body of work is already suggesting there may be cause for concern. One retrospective study published in Pediatrics in 2011 found that children who had multiple anesthesia exposures before two years of age were twice as likely (compared with those who were not exposed) to be diagnosed with a learning disability—even when overall health was taken into account. Children who had only a single course of anesthesia, however, did not exhibit elevated levels of such disorders.

There are dueling research findings, however. Another study, published in Pediatrics in 2012, found that when children under three years old had even one surgical procedure that required general anesthesia, those children appeared to be more likely to have difficulties with abstract reasoning and language by age 10.

Despite conflicting results like these, groups including the FDA and American Academy of Pediatrics decided the same year there was enough evidence to endorse a consensus statement which stated that “increasing evidence…[now] suggests the benefits of these agents should be considered in the context of their potential to cause harmful effects.” Still, their statement stopped short of recommending avoiding anesthesia all together. Instead, it said that in the absence of conclusive evidence it would be unethical to withhold sedation and anesthesia when necessary.

Since then additional study findings have heightened concerns. Another report, published in Pediatrics this month, found that children under four years of age who had been under general anesthesia for an average of 37 minutes tended to score lower as a group on listening comprehension and performance IQ tests than those who were not exposed.

Such cognitive deficits in the anesthetized kids were also associated with brain changes in the occipital cortex and cerebellum. Yet, like the ketamine monkey studies, these types of deficits may not be significant in the daily lives of children. “Maybe scoring three, four or five points worse on IQ tests may not mean much for an individual, but if you lower the IQ in all the kids that have anesthesia exposure early in life, that could put a big burden on society in general,” says lead author Andreas Loepke, a professor of clinical anesthesia and pediatrics at the University of Cincinnati College of Medicine.

But is it the anesthesia that fueled future issues or might that exposure simply be a stand-in for some other larger problem affecting these children—say sicker kids needed surgery and went on to have future cognitive issues stemming from those health problems? Perhaps. “Even if you see an association, you don’t know if it’s anesthesia," says David Warner, a professor of anesthesiology at Mayo Clinic who is overseeing new long-term analysis of children. "With surgery other things like inflammatory response could perhaps cause issues and anesthesia may just be a marker, so that’s why a study like this can never be definitive,” he says. “We can never overcome that limitation but we are trying to account for variations.”

That’s where the game comes into the picture. For the past three years his team has worked on an observational study that aims to explore how such anesthesia use relates to future learning disabilities. The same mechanized light game played by monkeys has been getting almost daily play from kids at Mayo who are participating in the research. If the research team can find anything in kids consistently affected from these exposures, say specific aspects of learning and memory and specific brain changes, that would be an important win for the field.

The kids are doing more than gaming. They each come in for one four-hour testing session at Mayo. One hour is spent on the video game. (Kids typically win around $5 in nickels alongside the $100 they are paid for participating in the experiment.) The other three hours are spent completing a battery of various tests involving memory recall, card sorting and other widely accepted psychological tests. The lab, on average, sees one study child each workday and Warner expects to complete the study in spring 2017. Some of the kids had multiple anesthesia exposures or had experience with various anesthesia chemicals so the study may help shed light on differences there, too, or if there may be differences by sex. (Young children are usually completely anesthetized and not just sedated with lower doses of drugs for most medical procedures, which is another reason why their exposure levels may be high.)

The study, like others that came before it, is observational rather than the ideal gold standard study where patients can be randomized to specific treatments. But the Mayo work can still help answer some as-yet unsettled questions. Warner believes his study is attractive because it will use methods to evaluate kids similar to those the FDA already used in monkeys, which will allow for direct comparison of the primate findings with human data. They both use the same test game with similar rewards—although children learn to play after watching a short instructional video whereas monkeys need to be extensively trained.

The Mayo group has been following a group of middle and high school–aged teens who had general anesthesia before three years of age and comparing them with children who did not undergo anesthesia at that age. The control group is matched by birth weight, gestational age (for example, if they were born prematurely), parental education levels and if they, too, would have been likely to receive anesthesia but never did—say they were ill but their parents elected to postpone surgery because the condition was not life-threatening.

Yet even with these results it will still be murky what to do next. Only a tiny fraction of kids—in the single digits—are put under at a young age. But those numbers as a whole shake out to mean that at least half a million children under three years old are exposed to anesthetic agents each year.

Many of these surgeries are unavoidable. They treat life-threatening illnesses, avert serious health complications or substantially improve quality of life. The most common type of birth defect—congenital heart defects affecting the structure of an infant’s heart and its function—is one such example. About one in four babies born with heart defects need surgery or other procedures during the first year of life.

And the same chemicals used for surgical anesthesia are also used to anesthetize kids during nonsurgical procedures such as MRI scans and CT scans to ensure patients do not move. “I think if the kids need these tests and they need to hold still for these tests, then we have to use the drugs,” Loepke says. “We are between a rock and a hard place there because if the kid doesn’t have these diagnostic tests and we don’t know what’s wrong with the kid, then the kid may suffer more because we didn’t figure out what was needed.” For his part, Warner says he hopes his team’s findings may prompt future research into alternative anesthesia formulas or the development of drugs to boost brain health in the aftermath of surgery.

One area researchers may explore, for example, is if behavioral therapy to give kids more stimulation following surgery may offset anesthesia’s effects—something that has shown some promise in rodents.

To truly confirm the link between anesthesia and deficits, however, a randomized study would need to be done. One such work is already underway, headed up by an Australian researcher. It compares infants undergoing hernia repair under general anesthesia versus those getting the surgery while they are anesthesized only in a specific region. Then the kids undergo neurocognitive testing at age five. Results from that study are expected in the next couple years. But for now doctors and researchers are carefully watching for the results from the Mayo study. As that research team doles out bags of nickels, parents and physicians are banking on a big return.