Investigating risk factors for ACL injuries

A new study on risk factors for anterior cruciate ligament (ACL) injuries may have revealed a previously unidentified risk factor. This factor, which the researchers call “early peaks,” is more common in girls than in boys, and it has also been found that specific movements in the hips, knees, and trunk that often occur alongside ACL injuries also occur alongside early peaks.

“ACL injuries are an important research topic because the consequences of these injuries can last a lifetime. The causes of ACL injuries and the movements that increase their risk have long been sought, but with limited success. This study is therefore about identifying risk factors for ACL injuries so that they can be prevented in the future,” says Haraldur Björn Sigurðsson, associate professor of physiotherapy at the University of Iceland.

Haraldur explains that such research is challenging because many studies have been conducted in this area and many factors have been examined, but so far, the risk factors found in one study have not been confirmed in the next. “It’s like looking for a needle in a haystack. There are countless ways to analyse the data, making it statistically likely that something will come out as a risk factor purely by chance.” One possible explanation for the differences between studies could be that the data are not analysed systematically based on how ACL injuries occur. “By finding such a connection, it would be possible to significantly reduce the number of variables examined and thus reduce the likelihood of identifying a risk factor by chance,” Haraldur explains.

ACLs tear in one-twentieth of a second

“The idea for this study arose during my PhD, and it all stems from the question of whether movement analysis data could be examined entirely in light of what is known about the forces acting on the knee of a person who tears an ACL,” says Haraldur. Around 2016, there was a surge in research abroad on forces, including the development in the U.S. of hardware that could create ACL injuries in knees, mimicking those that occur in people. Around the same time, a group of researchers in Norway was analysing videos of ACL injuries with great precision using motion analysis software.

“These foreign studies showed that knee valgus, or the inward force on the knee, plays an important role in the injuries. They also showed that it happens extremely fast; an ACL tears in one-twentieth of a second from the moment the athlete’s foot touches the ground,” says Haraldur, pointing out that this speed makes studying ACL injuries so demanding. The key question was thus how to identify, during that extremely short period, a part of the movement that could lead to an ACL injury in a healthy individual.

Data processing methods at the heart of the study

The research team uses precise motion capture equipment — infrared cameras that track reflective markers attached to athletes to analyse potential risk factors for ACL injuries. “The athletes then perform rapid and demanding changes of direction, which are the movements that most often cause ACL injuries,” Haraldur explains.

He says the data processing methods are at the heart of the study. “We calculate the forces acting on the knee, but although the equipment is good, it’s not perfect. In most studies, the peak values of the forces are examined, but this approach has major flaws because the peaks usually don’t occur as early in the movement as the ACL injuries do.” Even a slight shift in the reflective markers can significantly change the calculations, especially early in the movement, particularly with faster athletes. “To solve this problem, we developed a computational method that analyses the shape of the force curve at the beginning of the stance phase. The idea is that a certain shape, called early peaks, could be linked to ACL injury risk. It’s a shape where the peak knee valgus occurs early enough to align with when an ACL injury could happen,” Haraldur explains.
 

An international research team

The study originally began as Haraldur’s doctoral research, supervised by Kristín Briem, professor at the University of Iceland’s Faculty of Medicine. On his doctoral committee were Jón Karlsson, professor of orthopaedics at the University of Gothenburg; Þórarinn Sveinsson, professor emeritus at the University of Iceland’s Faculty of Medicine; and Lynn Snyder-Mackler, professor of physiotherapy at the University of Delaware.

An international team is now conducting the project. “As the results began to take shape, interest spread, and part of the work was subsequently carried out in the U.S. at Snyder-Mackler’s lab,” says Haraldur. Tron Krosshaug, professor at the Norwegian School of Sport Sciences, also joined the team. “Through our collaboration with him, we gained access to the largest database of ACL injuries with motion analysis data that exists — data Krosshaug has collected over many years. And of course, we mustn’t forget the many master’s students who have contributed both work and great ideas,” Haraldur says.

Bridging the Gap Between Research Questions and Complex Datasets

Asked where his main interests lie, Haraldur answers without hesitation: “In short, the causes and consequences of ACL injuries.” However, since the project began, he has also become very interested in bridging the gap between measurements from complex datasets and research questions. “When I started digging into the motion analysis data from the study, I quickly realised these were extremely complex datasets.”

The data collection is extensive. Everyone is asked to move in a specific way while being measured with dozens of reflective markers and a force plate on the floor. About 400 frames per second are collected, and joint angles, forces, accelerations, and more are calculated. “In the end, we have hundreds of data points for each measured movement, far more data than can be fully anticipated in advance. That’s why the data must be transformed in some way to extract useful truths. As I delved into these complex datasets, my interest quickly shifted to this transformation — how to bridge the gap between measurements and research questions and how best to extract meaning from complex datasets.”

Early Peaks More Common in Girls

The study's final results are not yet available, but much exciting information has emerged since it began as Haraldur’s PhD research. “We first chose to study this phenomenon, early peaks, to assess whether it was theoretically possible that it could be a risk factor. We found that early peaks are more common in girls than in boys, and girls are at higher risk of ACL tears than boys. We also found that certain movements in the hips, knees, and trunk that often occur alongside ACL injuries also occur alongside early peaks,” he explains.

With help from Snyder-Mackler’s lab, the research team also discovered that preventive training that reduces ACL injuries also reduces early peaks. “We were, however, moderately optimistic in the first prospective analysis because we only had a small group of injured individuals — out of a sample of 177 — but early peaks were still a strong risk factor for injuries in this group,” says Haraldur.

Currently, with Krosshaug’s help, the team is working to confirm the risk factor in a database with numerous ACL injuries. “Those results have not yet been published, but the analysis already conducted certainly gives reason for optimism,” Haraldur says, eager for what lies ahead.

Results Could Inspire Innovation in Biomechanics

The novelty of Haraldur’s research lies in examining the effect of specific movements on the risk factor, namely early peaks, rather than on the injuries themselves. The main obstacle to preventive studies is how time-consuming they are. “Hundreds of individuals are needed in an intervention to see a difference in risk because, although ACL injuries are serious, they thankfully occur relatively rarely,” he explains.

Intervention studies can be accelerated by using a strong risk factor and examining its effects instead of focusing directly on injuries. This opens pathways for preventive training by identifying a strong, confirmed risk factor. “For every ACL injury prevented, there’s one less person who needs surgery, who won’t develop degenerative changes in the knee due to the injury, and who can maintain high health-related quality of life into older age.

The scientific value of the study is no less important than its potential impact on athletes’ health. “I dare to dream that by demonstrating the usefulness of our analysis methods, which are very unconventional, and by moving toward examining movement patterns in a broader context, we could open the door to all sorts of innovation in biomechanics,” says Haraldur. “I may be overly optimistic, but just as we’re now looking at 12-year-old data with these innovative methods, researchers will be able to take these new tools into all kinds of older databases and, of course, new studies as well. Time will tell how many seize this opportunity.”