A PFO is present in about 25 % of the general population. Many epidemiological studies showed that there is an association between migraine and particularly large PFOs. Especially in patients with migraine with aura, a PFO was present in up to 60 % of patients in some studies. In this context an autosomal-dominant inheritance of migraine and PFO was suggested. Some observational studies suggested that a closure of the PFO may treat migraine. About 80 % of patients who underwent PFO closure for nonmigraine indications reported cessation or improvement of migraine attacks after PFO closure. However, these data were limited because of retrospective nonrandomized study designs and highly selected patients. Therefore, there was a need for a randomized controlled study to assess the effect of the PFO closure compared with a sham procedure in migraine treatment. A blinded study design is mandatory as a high placebo effect can be expected in migraine trials in general and in an interventional trial in particular. The MIST study (Migraine Intervention with STARFlex® Technology) addressed this important question for the first time in a prospective randomized double-blind sham-controlled design. Patients with frequent migraine attacks and aura were included that had failed at least two classes of prophylactic medication. In total, 423 patients were screened for a PFO. In 38 % (163 patients) a moderate or large PFO could be diagnosed, confirming the high prevalence of right-to-left shunts in migraine patients with aura. The primary endpoint of complete cessation of migraine attacks was reached in 4 % of the patients in both intervention and sham intervention group which has no significant difference. The second endpoint was also not met as there was no significant difference regarding migraine frequency (Intervention: 3.23 ± 1.80 versus Sham procedure: 3.53 ± 2.13). In the context of the intervention, serious side effects were observed. One patient of the placebo group had an ischemic stroke. In conclusion, the MIST trial was not able to show the efficacy of PFO closure at least in the investigated patient population. Therefore, based on the currently available study data, PFO closure should not be performed for migraine treatment. A screening of migraine patients for a PFO is not necessary and useful as results do not change treatment recommendation but might worry some patients.