An MRI can diagnose a tear in the meniscal cartilage of the knee, but it does not reliably predict which tears can be repaired with surgery, a new study finds.

It's estimated that more than 850,000 Americans undergo surgery each year for meniscal injuries in the knee joint. That surgery can involve either suturing the tear back together or removing the damaged portion of the meniscal tissue altogether.

At present, surgeons cannot tell whether the tear will be reparable until the surgery is underway.

Knowing ahead of time would be helpful for both surgeons and patients, according to Dr. Nicholas M. Bernthal, the lead researcher on the new study.

For patients, it matters because repaired meniscus tears have a more involved recovery compared with surgical removal of the tissue.

When the injured tissue is simply trimmed off, "people can basically get up and walk home," said Dr. Bernthal, an orthopedic surgeon at the University of California, Los Angeles.

They'll have some pain, he noted in an interview, but there are few restrictions on what they can do, and most can fully return to their usual physical activities within about two weeks.

But such partial meniscectomies only work well if the tear is small. For large tears, removing a sufficient portion of the meniscus often leads to knee joint degeneration.

Patients who have the meniscal tear repaired, on the other hand, need about four to six weeks of recovery, often with physical therapy.

So it would be nice, Dr. Bernthal said, to be able to tell patients ahead of time which surgery they are going to get.

"This is one of the few instances in surgery where we can't tell people what to expect," Dr. Bernthal said.

MRI scans are widely used to diagnose meniscus tears, and they are accurate up to 99% of the time, Dr. Bernthal noted. But MRI is not good with the details, where the tear begins and ends, or how complex it is.

Recent debate about whether today's more advanced MRI technology might be better at predicting the complexity of meniscus tears prompted Dr. Bernthal and his colleagues to test the idea.

They asked two senior radiologists to examine the MRI scans of 119 patients who had already undergone surgery for a meniscal tear, either a surgical repair or meniscectomy.

The radiologists tried to predict the reparability of each tear.

They often got it wrong. One doctor was correct 63% of the time, and the other only 58% of the time.

And when the researchers looked at the specific reparability "scores" each doctor gave the patients, they found that the radiologists were in agreement only 38% of the time.

Dr. Bernthal and his colleagues reported the findings online December 20th in the American Journal of Sports Medicine.

According to Dr. Bernthal, the two radiologists in the study are very experienced and "renowned" in their field. So it's likely that radiologists in community hospitals would also fare poorly if asked to predict the reparability of a meniscal tear.

For now, he said, patients having surgery for a meniscal tear will continue to have to wait until the procedure itself to know just what to expect afterward.

"My assumption is, somewhere down the line, we'll get to the point of being able to predict (reparability)," Dr. Bernthal said. "But we're just not there yet."

Not all meniscal tears require surgery. Some patients recover with simple measures like rest, ice and over-the-counter pain medication.

Am J Sports Med. Posted online December 20, 2010. Abstract

Amy Norton • Reuters Health Information © 2010