For as long as it has been available, much of the research into managing children on the ketogenic diet has focused on two important things: 1) how to start it and 2) how to maintain it. Should children be fasted or admitted? Which ratio should be used? How frequently should patients be getting lab work, checking urine ketones, or weights? Much of this guidance was recently summarized in the 2009 Expert Consensus Statement.
However, one important aspect of clinical use of the ketogenic diet is something that is rarely discussed in reviews and books, and to date never in any research publication. Yet it is something that essentially every parent will ask their ketogenic dietitian and neurologist how to do. Most of the time even experienced ketogenic diet centers would say that we didn’t know and would make decisions based on tradition and personal experience! That important part is…how to discontinue the diet. Whether it be a situation in which the diet is not working, is completely successful (e.g. seizure-freedom), or somewhere in between…how should it be done? Traditionally, the ketogenic ratio is reduced every couple of months until after 6-9 months the diet is over. Unfortunately, there is no data to back up that approach.
This August in Epilepsy Research we finally have some scientific answers to this question. Ms. Lila Worden from our group at Johns Hopkins looked at 183 children who had the ketogenic diet stopped over a 10-year period. She only included those in whom we had good information about how it was done and what happened during the wean. Most of these children were on the diet about a year and about half had significant seizure reduction, both of which are typical for ketogenic diet patients.
What she found was surprising. First, how children were weaned off the diet was VERY variable and physician dependent: about a third each were weaned immediately (over 7 days), quickly (1-6 weeks), or slowly (>6 weeks) – the last of which is most similar to what is recommended traditionally. More of the immediate or quickly discontinued children were doing poorly and having more seizures, which makes sense: if the diet is not working, neurologists and parents usually want to stop it quickly and move on to other treatments.
What happened with the wean was just as surprising: in general, it didn’t matter how quickly it was done! Overall 14% of children had increasing seizures during the discontinuation (most of whom became better again when changes or resumption of the prior ketogenic ratio was done). If children were doing better (e.g. 50-99% seizure reduction) then they were at slightly higher risk to get worse (e.g. more to theoretically lose than a child with no seizure improvement). However, no other factors seemed to matter, including the speed of the wean, with 11% worsening in the immediate group, 11% in the quick group, and 19% in the slow group.
What does this tell us about how to wean the diet? First, we should individualize for each patient and not do things by tradition. The diet can be possibly stopped within days (or even one day) for children who aren’t doing well and within several weeks (e.g. lowering the ratio every week instead of every month) for those doing better (or seizure-free even). If seizures worsen, the ratio could either be increased back up or medications adjusted. Using KetoCalculator (www.ketocalculator.com) can make these more rapid dietary changes easier as well.
Worden LT, Turner Z, Pyzik PL, Rubenstein JE, Kossoff EH. Is there an ideal way to discontinue the ketogenic diet? Epilepsy Research 2011; 95:232-236.