Fusion also eliminates motion and function, something that discs in the spine, which have no blood supply, need. Motion signals the body to provide nutrients to the discs.

Besides that, when part of the spine can't move, the function is transferred to another area, an adjacent segment of spine, and that degenerates.

Not treating a severe curvature is no option, either. "A big thoracic curve can be painful. It can create trunk imbalance," a sort of listing to one side, Braun said. And it crowds the organs, reducing heart and lung function." The effect is to crowd one side of the chest with the bony curve of spine while organs on the other side gets squished.

Braun tells Clark and Parsley to stick with the brace for now. The near future, he says, holds promise that simply isn't there today.

He believes surgery within a year or two will allow a physician to guide growth of the spine straight without the trauma and invasiveness, using an arthroscope and devices even now in development that are fusionless, meaning the spine does not have to be made rigid. He's spent eight years testing and refining such devices, and others are also working on it.

He often collaborates with Ogilvie, his long-time mentor. Ogilvie designed a staple for use in the spine. The latest version is made of shape-memory alloy, which returns to its original shape at a certain temperature. In its first major study in 2003 at a Shriners Hospital in Philadelphia it "showed nice results for children."

That's still not ideal, Ogilvie and Braun agree, because a spine moves and that asks for a staple to fail, kind of like twisting a staple on a stack of papers and expecting it to hold.

Now Braun's working on a bone anchor with hollow threads that screws into the bone, which would then grow into it and around it, making it part of the body for a very strong joining. That bone anchor has been tested in pre-human trials with a ligament loop that allows the surgeon to tether the spine to correct the curvature without taking away the spine's mobility.

Even that's "clumsy" compared to what children need, Braun said.

He believes the best solution will be found in genetic study. Genes make proteins that come out of cells and "if we could figure that out, we could attack this much more elegantly."

Five years ago, Braun began a genetic study of scoliosis, and now researchers know the region of the human genome where the deformity starts.