Epilepsy patients can sometimes predict seizures

By Anthony J. Brown, MD

NEW YORK (Reuters Health) - While they can't always tell, epilepsy patients quite often know when a seizure will occur, a new study indicates. Younger patients and those with frequent seizures are particularly good at predicting an attack.

According to the report in the medical journal Neurology, when an epilepsy patient predicts a seizure in the next 24 hours, the likelihood that it actually will occur more than doubles. Still, the results also indicate that epilepsy patients fail to predict most seizures.

"We were surprised by the findings," lead author Dr. Sheryl R. Haut, from Montefiore Medical Center in New York, told Reuters Health. "We have seen patients in the clinic who state that they can predict when a seizure will occur," she explained, but her group didn't think that the predictive ability would be as strong as it was or that it was present to some extent in most patients.

Haut's team studied 71 adult patients with epilepsy who had at least one seizure in the last 12 months. The subjects kept daily seizure diaries and were asked to rate the likelihood that a seizure would occur in the next 24 hours: extremely likely, somewhat likely, somewhat unlikely, and extremely unlikely.

In the overall analysis, when subjects predicted a seizure they were 2.25-times more likely to experience one than when they did not, the report indicates. Yet, they only predicted 32 percent of the seizures that occurred.

Twelve patients were better than the others at predicting seizures. When these patients, who were younger and had frequent attacks, predicted a seizure, the odds of one actually occurring increased more than three-fold. Even this group, however, predicted just 37 percent of seizures.

Being able to predict when a seizure will occur is important for three reasons, according to Haut. "First, it gives them more peace of mind and a sense of control. Second, it allows them to better plan their daily activities. And lastly, it opens the way for preemptive therapy," which is currently not used for epilepsy.

Haut believes that with appropriate training in recognizing early warning signs, the accuracy of predicting seizures could improve. Along these lines, she said that her group is now involved in research to identify exactly what symptoms the patients are experiencing that make them believe a seizure is likely to occur in the coming hours.

SOURCE: Neurology, January 23, 2007.