Atlanta, GA - Hopes that closing a patent foramen ovale (PFO) could vanquish migraine with aura sank here today, with the final results from the Migraine Intervention with STARflex Technology (MIST) I trial showing no difference in headache cure between closure- and sham-treated migraineurs. Presenting the headache data here at the i2 Summit at the American College of Cardiology 2006 Scientific Sessions, Dr Andrew Dowson (King's College Hospital, London, UK) insisted that the strategy still shows promise. In fact, while PFO closure was no better than a sham procedure in preventing all headaches, it was associated with significant reductions in other end points more typical of headache trials, he said.

"Further analysis of MIST data may indicate which patients are most likely to show a significant treatment response," Dowson said. In the meantime, close review of MIST will help refine design of the MIST II study, he added.

Emerging from the MIST

The 147 migraine patients who took part in the MIST trial ranged in age from 18 to 60 and, in preliminary results previously reported by heartwire, were found to have PFOs that were significantly larger than PFOs found in the general population. To be included in the trial, migraine patients had to be refractory to at least two classes of migraine medication and have a one-year history of migraine.

In the trial, all participants underwent contrast transthoracic echocardiography to establish shunt size, then were randomized to either PFO-closure-device implantation with the STARflex septal-repair implant (NMT Medical, Boston, MA) or to a sham procedure consisting of general anesthesia and groin incision. All patients received identical ward management, were prescribed aspirin and clopidogrel for three months, and were followed in an identical fashion for an additional three months.

As Dowson told the late-breaking clinical-trials session audience today, three patients in each arm of the trial achieved the primary end point of complete cessation of headache. Other end points, however, favored PFO closure: more patients who had their PFOs closed had a 50% or greater reduction in headache days, and more PFO-closure patients had a reduction in headache burden, calculated as headache frequency times duration.

MIST I results

End point	PFO closure	Sham	p
Headache cessation (n)	3	3	NS
50% reduction in headache days (%)	42	23	0.038
Reduced headache burden (%)	37	17	0.033

Primary end point MIST

Dowson commented that other factors deserve closer attention in the MIST II trial, now enrolling in the US. For one, longer follow-up may be warranted, particularly since MIST I, as well as observational studies, have indicated that migraine improvement tends to increase over time. As well, it may be important to better identify the type of headache that is most affected by PFO, if any. Type of aura, frequency of aura, and migraine triggers may warrant closer examination, he said.

Presenting the cardiac results from MIST, co-primary investigator Dr Peter Wilmshurst (Royal Shrewsbury Hospital, UK) emphasized that the trial investigators have not yet analyzed the data according to whether or not patients had residual shunts. If blood-borne substances are responsible for migraine, significant residual shunting in patients treated with PFO closure would influence the results, he said.

Discussing the MIST findings in a press conference, Dr David O Williams (Rhode Island Cardiology Center, East Greenwich), who was not involved in the trial, called the results "a little bit of a mixed message."

"I was a little bit disappointed," Williams admitted. While device closure did not appear to cure migraine altogether, "on the other hand, there was some improvement in headache episodes."

Commenting on the study for heartwire, Dr Gilles Montalescot (Hôpital Pitie Salpetriere, Paris, France) was adamant that the strategy should not be abandoned as a potential treatment for migraine with aura.

"I think it works. It certainly seems to work better than drugs, and all the details that they showed on the frequency of migraines point to benefit. We'll need the second study just to make up our minds, but I think it probably works."

Montalescot believes the choice of headache cessation was probably too ambitious for such a small trial, with limited follow-up. "It's intriguing that there is no difference in the primary end point, but I think it was a difficult end point to reach—complete cure of migraine is very difficult to achieve. However, there is a strong trend, so we will have to wait for a reevaluation of these secondary end points."