Researchers have known for some time that female athletes experience higher rates of concussion than their male counterparts, and also often suffer harsher symptoms and take longer to recover. But why women seem more vulnerable to such injuries has long remained a puzzle.

Concussion symptoms range from headache, dizziness and confusion to memory loss, noise or light sensitivity, and irritability. Most people recover quickly but some develop problems lasting a year or more. A 2010 study led by neurologist Jeffrey Bazarian of the University of Rochester found that women—especially those of child-bearing age—had worse symptoms measured three months after injury.

Several explanations have been proposed including sex hormones, neck structure and cerebral blood flow, but no one really knows what is to blame. Now, however, a study led by Douglas Smith, director of the Center for Brain Injury and Repair at the University of Pennsylvania, adds a new candidate: differences in axons—the output “wires” of neurons.

Smith and his colleagues discovered differences in the size and structure of male and female axons and found the female structure was more susceptible to damage. “The findings are intriguing,” says neuropsychologist Donna Broshek of the University of Virginia, who was not involved in the study. “Many theories have been put forth, including that—because of differences in cultural socialization—women are more likely to endorse symptoms.” But the new results, published recently, “suggest that women report more symptoms because they are ... experiencing more symptoms,” Broshek says.

Each neuron has many inputs, called dendrites, but generally only one output: its axon. Along with transmitting the electrical signals underlying cognition, axons transport cargo of various kinds—including proteins that are used mainly for repair and maintenance. Other specialized proteins haul this cargo via the axons along train track–like structures called microtubules. A major sign of concussion is damaged axons. “There's an emerging consensus that one underlying issue with concussion is diffuse axonal injury,” Smith says. This led him to wonder if the sex differences in concussions might have something to do with axons. “Is there something different about brain architecture,” he says, “that given the same mechanical forces, a female brain may be more injured than a male brain?”

To test this idea, the team first used a cell-culture model in which two populations of neurons are separated by tiny channels, along which axons grow. The researchers then used pulses of air to rapidly stretch the axons, mimicking sudden head trauma. “We found a dramatic difference,” Smith says. “The female axons had many more undulations, which were bigger, suggesting more structural damage.” Undulations occur when microtubules break. Their cargo then spills and builds up in the axon, causing swelling. Sodium, which is needed for normal transmission of electrical signals, then rushes in—but too much of it actually disrupts this signaling. “In concussion you have an immediate change in mental status, because the electric grid has been taken out,” Smith says. “It's like a blackout or browning out of the city.” The excess sodium also causes calcium to flood in, activating enzymes that destroy structures inside axons. Twenty-four hours after being hit with air pulses, the female axons had more swellings and signaling loss. This was true for both neurons taken from rats and human neurons derived from stem cells that were genetically male or female.

The researchers next examined the axons with electron microscopy. They found the female axons were thinner, with fewer microtubules—a difference that has been hinted at previously but never directly demonstrated. They then used a computational model to compare how the different male and female structures responded to the same mechanical force. The female structure suffered more breaks. Microtubules are connected by proteins that become stiffer the faster they are stretched. And if axons are stretched too rapidly, these proteins pull on microtubules and break them, Smith says. Having more microtubules gives the whole structure more stability.

The relatively rapid recovery most people experience may reflect the time it takes to repair axons—but if the damage is too severe, axons may degenerate. “We think that's the route that has persisting symptoms,” Smith says. “Up to 20 percent of individuals have persisting dysfunction; we're really interested in those people, because we think they have permanent damage.” If he is right, recovery from long-term symptoms may reflect the time it takes brain plasticity to compensate for lost cells.

How much of the concussion sex differences might be explained by this discovery is not yet clear. “I'm almost certain this cannot be the only thing that contributes,” Smith says. “It's probably one of several factors; time will tell how large a role it plays.” Differences in rates may be “primarily about axonal damage,” but recovery time is a different question, Bazarian says. “There's a host of processes that take place, to clean up, that may also have sex differences.” Most important is the brain's inflammatory response, which must be precisely controlled to clean out waste without doing damage. Hormones regulate this process, which could explain findings like worse outcomes in child-bearing years. How effective these processes are—plus injury severity—are likely what determine recovery times. “It's a combination of the two,” Bazarian says.

A caveat is that cell cultures and models are a far cry from the immensely complex environment of a real brain. “This sets the stage for the next step, which is finding some way to verify these findings in living human brains,” Bazarian says. This will be challenging, as researchers cannot easily examine live human axons directly. There are ways around this, however. For example, researchers can look at less-direct evidence of the damage, such as changes in white matter. (Nerve fiber tracts appear white due to the coloring of the myelin sheaths that cover and insulate axons.) “This will open the door to looking at white matter with advanced neuroimaging,” Smith says. “Given the same head impacts, do females have more changes in white matter?”

Another promising approach is measuring levels of proteins in blood. “We're realizing we see these axon proteins in the blood after injury, which are only there if the axon degenerates,” Smith says. “Hopefully in the next few years we’ll have a test to identify individuals who are going to have long-term problems.” This would be a boon not just for diagnosis but for testing treatments. One reason clinical trials fail is if the participants are too diverse, so the treatment is not appropriate for all. If a blood test could identify patients likely to develop persistent problems, they could be selected and enrolled in trials targeting the axon damage flagged by the test.