ATLANTA — Patent foramen ovale closure with a septal repair implant gave a modest benefit to patients with migraine and PFO, according to preliminary results of a study with 163 patients.
The enrollment phase of the study also showed that right-to-left cardiac shunts occurred in 60% of the patients with migraine who were screened, Dr. Peter Wilmshurst reported at the annual meeting of the American College of Cardiology.
This prevalence is “very, very high,” commented Dr. David O. Williams, director of interventional cardiology at Rhode Island Hospital in Providence. “The usual prevalence [in the general population] is about 15%–20%,” he said.
Insertion of a STARFlex septal repair implant provided complete migraine relief in 3 of 74 patients (4%) during 6 months of follow-up, the same rate as in the 73 patients who received a sham procedure. Thus the device, developed by NMT Medical Inc. (which also sponsored the study), failed to achieve the primary end point of complete headache relief.
“However, using more conventional migraine trial end points, significant differences were found,” noted study investigator Dr. Andrew Dowson, a headache specialist at King's College Hospital in London. Overall, 42% of patients receiving the implant had a 50% reduction in headache days, compared with 23% of sham-treated patients, a statistically significant difference.
The Migraine Intervention with STARFlex Technology (MIST) trial was a randomized, placebo-controlled, double-blind study that initially screened 432 individuals with migraine. The study was done in 2005 at 13 centers in the United Kingdom. Patients were aged 18–60 years, with a minimum 1-year history of migraine with an age at onset no later than 50 years, frequent migraines (at least 5 days per month but at least 7 headache-free days per month). The study was restricted to those with migraine with aura, because previous studies showed an association between PFO and these types of migraines, Dr. Dowson said.
Contrast transthoracic echocardiograms showed that 72 patients had small shunts (atrial and pulmonary), 22 had large pulmonary shunts, 3 had atrial septal defects, and 163 had large patent foramen ovales (PFOs), for a total of 260 shunts, said Dr. Wilmshurst, a coinvestigator and cardiologist at Royal Shrewsbury (U.K.) Hospital.
The 163 patients with large PFOs were targeted for the study, and after 16 were excluded 147 patients were randomized to receive either PFO closure or a sham operation. Patients underwent a 3-month healing phase after surgery, followed by a 3-month analysis phase in which migraine occurrences were continually monitored.
Patients continued their prophylactic migraine medication, but those who overused migraine medication were excluded from the study. Other exclusion criteria included prior stroke or transient ischemic attack, and cardiac contraindications.
In addition to the significant reduction in headache days, the implant also provided a significantly greater reduction than did sham in headache burden, a measure that incorporates both the frequency and duration of headaches.
Patients receiving the implant achieved a 37% reduction in headache burden, from 136 at baseline to 86 at last follow-up, whereas those receiving the sham operation improved from 117 at baseline to 96 at the last 6-month follow-up, a 17% reduction. The investigators noted that the difference in the magnitude of the reduction between the implant and sham groups was statistically significant.
The results send “a little bit of a mixed message” Dr. Williams commented. The low rate of complete headache resolution suggests that PFOs may not have an etiologic role in migraine. “On the other hand, there was some improvement, and we wonder whether the closure process may not have been complete,” he said.
The high prevalence of cardiac shunts in patients with migraine suggests that screening selected patients with transthoracic echocardiography and then performing repairs when indicated may be a good idea. “It's something to consider for disabled patients,” Dr. Williams said. “In skilled hands, the risks of PFO closure are very low. The procedural risk is like that for cardiac catheterization.”
In the implant group, five serious adverse events occurred in one patient each: cardiac tamponade, pericardial effusion, retroperitoneal bleeding, atrial fibrillation, and chest pain. In the sham group, one patient had incision site bleeding, two patients had effects of antiplatelet therapy (anemia, nosebleed), and one patient had a brainstem ischemic stroke 4 months into the follow-up.
The researchers plan to evaluate residual shunting in patients who had repairs. Longer follow-up may show greater benefit from implants, Dr. Dowson noted.
Philadelphia bureau chief Mitchel L. Zoler contributed to this report.
Significantly more patients with the implant had a 50% reduction in headache days, compared with controls. DR. DOWSON