An interesting question! While researching, I found it is a common (and perhaps controversial) belief amongst hypnotherapy discussion forums that it is inappropriate to hypnotise epileptics, but no evidence to support that view.
Many of the issues were related to professional hypnotists, and the risks involved if the patient was to (perhaps coincidentally) have a fit while under their care. Another concern was the use of flashing lights to induce hypnosis might also trigger an episode for photosensitive epilepsy sufferers.
For every warning against hypnotising epileptics, there was another case study of someone doing so with no problems. I found no case studies where it actually caused a problem.
Ref: An unedifying example of such a discussion
A Hypnosis Training Site tried to get to the bottom of this by interviewing a number of experts, including William Broom from the General Hypnotherapy Register (GHR) who explained:
Although there are those both within orthodox medicine and outside of it who suggest that hypnotherapy (or rather any relaxation techniques and procedures that hypnotherapy may include) could well trigger a seizure, the GHR has not seen any compelling evidence to support this view. Whiles we continue to advise that some manifestations of epilepsy could contra-indicate hypnotherapy in certain circumstances [...], as a general rule we confirm that it is perfectly acceptable to proceed with therapy provided that the therapist concerned is entirely comfortable in doing so.
While I am reluctant to use the Membership Secretary of a Industry Body as an expert to evaluate scientific evidence (especially in an area some consider to be pseudo-scientific), I am nonetheless citing him to support my claim of the absense of evidence - it isn't just my poor searching skills that failed to find anything.
Back in 1950, psychiatrists were happy to hypnotise epileptics to help with diagnosis: Role of hypnosis in differentiation of epileptic from convulsive-like seizures. It isn't clear to me whether they were trying to trigger seizures, or just relax the patients into remembering more about them.
Off on an aside: In 2008, researchers went further. They hypnotised 9 children who had had seizures, and then via hypnotic suggestion tried to trigger the seizures. (Note: It wasn't the hypnosis itself that was triggering the seizures, but a deliberate suggestion.) They succeeded with 8 of the 9, and used measurements taken during the seizure to confirm that they did not have epilepsy. The fact that they eventually diagnosed the children as NOT having epilepsy makes this less relevant. Obviously there weren't overly concerned about an epileptic being triggered by hypnotherapy as that was the intention of the session. (I would like to see a control here, where 9 healthy children also underwent hypnotic suggestion, to ensure that the seizures were truly representative of the patient's normal condition and not an artifact of suggestability. Is that unethical?)
Ref: Hypnosis-provoked nonepileptic events in children, Donald M. Olson, Neva Howard, Richard J. Shaw, Epilepsy & Behavior, Volume 12, Issue 3, April 2008, Pages 456-459
doi:10.1016/j.yebeh.2007.12.003
You may notice there isn't a lot in the way of peer-reviewed research. Searching Cochrane Reviews and Google Scholar failed to come up with any relevant papers. This is either (limited) evidence of absence or a sign that hypnotherapy and its risks are not being researched/accepted scientifically.
Conclusion: Unsurprisingly for an area with a poor history of quality research, I found no evidence of novel dangers to epilepsy sufferers from hypnotherapy.