Athletes demand a lot from their knees. Sudden stops and starts, twists and turns; the need to jump, land, and pass a basketball down the court. Such movements, though, can go wrong and harm the knee. For instance, coming to a sudden stop and then pivoting can rip the anterior cruciate ligament, or ACL, from the bone.

The ACL is part of a patchwork of tendons and ligaments that give knees flexibility and stability.

Orthopedic surgeons repair more than 200,000 torn ACLs each year in the United States. The majority of these injuries occur in young athletes from 15 to 25 years old. About 70 percent of ACL tears happen without contact from another player.

These athletes want to return to the same sport at the same competitive level, said Dr. Rick Wright, an orthopedic surgeon specializing in sports medicine at the Washington University School of Medicine. ACL surgery indeed allows athletes to return to the playing field, but those same athletes face a higher risk — up to 8 percent — of reinjuring the repaired ACL.

Here's the new problem those athletes face: The second round of ACL repair doesn't bring results equal to the first surgery. The failure rate on the second ACL repair can be as high as 14 percent.

Wright says that sometimes patients experience pain and swelling after the second reconstruction, ending up with a knee that doesn't function at the same level as the first surgery allowed.

"It's probably somewhat amazing that the ACL graft tear rate isn't any higher than it is," Wright said. "If athletes tear the one God gives them; they'll tear the one a surgeon gives them."

To figure out why subsequent ACL repairs fail, Wright is heading a large, nationwide study called the Multi-center ACL Revision Study, or MARS. The MARS study has been going on for about four years. Recently, the team at Washington University received a $2.6 million grant from the National Institute of Musculoskeletal and Skin Diseases to figure out why second ACL reconstructions fail at a higher rate than the first ACL repair. The effort includes 87 surgeons from 57 institutions.

The team will study many factors including surgical techniques. When surgeons reconstruct the ACL, they can use a ligament from a cadaver or a tendon from the patient. There are also several techniques they can use to attach the replacement graft to the knee.

An injury severe enough to rip the ACL from the bone may also damage the meniscus cartilage, the spongy pad that cushions joints, which could in turn, set the stage for problems, Wright said. Other factors are being compared as well, such as, rehabilitation techniques.

Patients will be followed for two years to try to identify which factors predict problems after a second ACL surgery. Pinpointing why the first graft failed will guide surgeons' efforts to prevent subsequent failures.

In the first four years of the study, the team analyzed data from about 400 patients and found that the strongest predictor for a bad outcome after ACL surgery is whether that surgery was the initial reconstruction or a later procedure.

For the second phase of the study, Wright's team will enroll 1,000 patients nationwide to identify which are the strongest predictors for problems after ACL surgery.

"The reason we're doing this study is because none of us know. We have our biases but we need data," Wright said.