Unique Historical Records Prove Valuable Discovery to Epilepsy Foundation Researchers

The story of how Epilepsy Foundation research fellow Nicholas Poolos, M.D., Ph.D. discovered the superior effectiveness of a two anti-epilepsy drug (AED) combination reads like a page-turning historical novel with some well-placed mystery twists.

Poolos—a 2012-13 fellowship grant recipient studying synergism in antiepileptic drugs—began his work with a unique and valuable discovery. “It was hidden in plain sight,” he says. For 30 years, nurses at two Washington state assisted living facilities for the developmentally disabled had recorded seizure occurrence and AED regimens among 148 male and female adults with refractory epilepsy: at least one seizure per year despite at least two different AED treatments. The records had remained filed away, however, and for the first time ever were now presented to epilepsy researchers.

Carefully handwritten at the Rainier Residential Habilitation Center in Buckley, Wash., and on Excel spreadsheets at the Fircrest Habilitation Center in Shoreline, Wash., the monthly charts allowed Poolos and his team to study the effects of 8 commonly used AEDs alone and in 32 different combinations.

They discovered lamotrigine and valproate in combination had superior efficacy.

The study also found what scientists call a “Goldilocks Effect”—a “just right” AED regimen. Seizures respond best to two AEDs. Three-AED combinations provide no further benefit (but the potential for other side effects) and one AED is not as beneficial.

The results—which Epilepsy Foundation professional advisory board chair Joe Sirven, M.D. said “represent a paradigm shift in epilepsy”—were published in a Dec. 2011 issue of Neurology, the journal of the American Academy of Neurology.

“I honestly had no idea what we would learn, or how this would all turn out,” says Poolos, a neurologist at the University of Washington (UW) Regional Epilepsy Center in Seattle. “Our first discovery was the handwritten notes, in which I’m not sure we would have spotted any meaningful patterns. When we found the Excel spreadsheets, I knew we had a workable database.”

Like finding gold at the bottom of the sea, however, the discovery was just the first step in a long approval and recovery process. “We had to do a lot of paperwork with the state,” Poolos explains.

After state officials released the records, Poolos and UW neurobiology student Lindsay Warner set out to find out what they had, a central question or hypothesis—the first step in the scientific method—guiding the way. “Was there any combination of anti-epilepsy drugs patients did better on than any others? That was our primary question,” Poolos says.

The challenge, which took over a year to surmount, was significant: organizing and cataloguing three decades of patient records from two different hospitals in two different formats charted by dozens of nurses and physicians. To make the mountain even steeper, “I didn’t have any funding at that point,” Poolos explains. “I did computer coding on my lunch hours to crunch the data. It took hundreds of hours, especially to convert material from the handwritten charts into a form we could use.”

But the nature and rarity of the records made the work well worth it. “Keeping records like these is tough,” Poolos says. And institutional settings present the most challenging of all patient populations: people with serious epilepsy whose seizures respond to almost nothing: so-called “medically-refractory epilepsy.”

At the same time, institutions employ trained staff who record treatments and responses with a regularity, detail, and completeness seen almost nowhere else. Poolos and his team—which also included UW School of Pharmacy researcher Sophia Humphreys, Ph.D. and Rainier School clinical pharmacist Stephen Williams, Ph.D.— “weren’t getting records kept by patients themselves or family,” Poolos explained. “We were getting records from professional staff.”

The record-keeping also included a simple innovation. Medication histories—dose, blood serum levels, side effects, efficacy—are an important part of healthcare, but most medication information suffers from lack of continuity, Williams says. “The data does not continue—it simply stops after it is gathered, becoming outdated quickly,” he explains. To innovate around the problem, Rainier and Fircrest pharmacists created a “living” medication history—one that didn’t simply record the past, but was continually updated.

An example of an individual patient's information from this detailed record-keeping is shown in the accompanying graphic. It depicts monthly seizure frequency (in red), along with daily dosage amount of AEDs, in this case lamotrigine (green) and valproate (blue). For this patient, treating with lamotrigine or valproate alone had little effect in controlling seizures, even at high doses. But the combination of two was dramatically effective, in this case producing seizure freedom.

While his study of historical records was revealing, Poolos says its small sample size and retrospective design are fundamental drawbacks that a forward-looking study with hundreds if not thousands of participants could correct.

His new Foundation-funded study, “Synergism in comparative efficacy of antiepileptic drugs,” will examine why lamotrigine (LTG) and valproate (VPA) are more effective than other combinations, statistically analyzing the relationship between AED dosages, blood levels, and seizure-reducing effects, if any. The Poolos team also hopes to answer a larger question: why anti-epilepsy drugs seem to exhibit synergism. They are routinely more successful in combination than as single drug therapies.