Tuesday, May 22, 2007

Zapping Seizures Away

Continued from page 1

By Emily Singer

"I think this concept has a stronger scientific basis than constant medication for a condition that comes and goes," says Ivan Osorio, a neurologist at the University of Kansas Medical Center, who has developed a similar system. "It may also increase [effectiveness] because it addresses changes at the time they occur."

It's not yet clear if the therapy will be successful: an ongoing clinical trial of 180 patients is only about 25 percent complete. But early results are promising. The device appears to be well tolerated and does suppress seizures in some patients. "There's not a lot of efficacy data out there yet," says Brian Litt, a neurologist and bioengineer at the University of Pennsylvania, who developed some of the algorithms used in the device. "But I've seen some beautiful anecdotal recordings where you can see the seizure starting and the device deploying and stopping the seizure without the patient ever knowing."

The same factors that make the Neuropace device unique also make it more challenging to administer. The location of a seizure and its electrical characteristics can vary from person to person. And neurologists must find the optimal detection and stimulation settings for every individual, a task that involves a certain amount of trial and error.

Each day, the patient uses a wand to wirelessly download electrical data recorded by the device to a laptop, where it is then uploaded for the patient's doctor to peruse. The doctor can monitor the effectiveness of the therapy and change parameters as needed. "We don't yet know what the ideal stimulation parameters are," says Gregory Bergey, director of the Johns Hopkins Epilepsy Center, who is leading one arm of the ongoing trial.

But the data generated during these studies should help. Scientists can analyze the reams of data to create better prediction and detection algorithms, a process that is already under way. "This device is making big contributions to the field, though that doesn't mean it will make big clinical contributions in its first run," says Litt.