By Dara Jamieson, MD

Associate Professor of Clinical Neurology, Weill Cornell Medical College

Dr. Jamieson reports she is a consultant for Bayer and Boehringer Ingelheim.

SYNOPSIS: Right-to-left shunts, as detected by transcranial Doppler, are more common in patients with migraine with aura, but are not correlated with increased risk of silent posterior circulation infarcts or white matter lesions on MRI.

SOURCE: Koppen H, Palm-Meinders IH, Mess WH, et al. Systemic right-to-left shunts, ischemic brain lesions, and persistent migraine activity. Neurology 2016;86:1668-1675.

Studies of patients in headache clinics have determined that migraine with aura is associated with an increased risk of patent foramen ovale (PFO), with shunting of blood from the right to the left side of the heart. Koppen et al assessed whether migraine patients in the general community, as opposed to those selected from headache clinics, had an increased prevalence of systemic right-to-left shunts (RLS) as compared to a non-migraine population. They also evaluated the association of RLS with ischemic brain lesions on MRI and persistent recurrence of migraine attacks at an older age. The presence of RLS was determined by transcranial Doppler (TCD) with injection of agitated air/saline/autologous blood contrast in 166 migraineurs (mean age: 56 years; 70% women) and 69 controls (mean age: 55 years; 65% women) from the Cerebral Abnormalities in Migraine: An Epidemiological Risk Analysis Study Part 2 (CAMERA-2). In CAMERA-1 in 2000, migraine patients and controls were randomly selected from a community-based study and were assessed for ischemic brain lesions on MRI. The follow-up CAMERA-2 study assessed the prevalence, incidence, and progression of MRI lesions in migraineurs over the nine years from the original study. MRIs were evaluated for “silent infarcts,” which were defined as “non-mass parenchymal defects with a vascular distribution, isointense to CSF signal on all sequences, and when supratentorial, surrounded by a hyperintense rim on fluid attenuated inversion recovery images.” Participants in CAMERA-2 were asked if they had at least one migraine attack in the 12 months prior to the MRI scan, defining active recurrence of migraine attacks.

Valsalva-induced RLS was more frequent in participants with migraine with aura (60%) as compared to controls (42%; P = 0.02) and participants with migraine without aura (40%; P = 0.01). Participants with migraine with aura also had more frequent spontaneous RLS (35%) than participants with migraine without aura (17%; P = 0.01), but not when compared to non-migraine controls (26%; P = 0.2). Migraine with aura and spontaneous RLS predicted ongoing recurrence of migraine attacks. Silent posterior circulation infarcts were found in 9% of participants with RLS, not significantly different from 3% of participants without RLS, independent of migraine status. Higher age, but not RLS presence or migraine status, were associated with an increased infarct risk. The presence of RLS on TCD was not associated with an increased risk of deep white matter lesions on MRI. The presence of a spontaneous RLS in a migraineur was associated with ongoing recurrence of migraine attacks. However, the mean attack migraine frequency was not correlated with the presence or absence of spontaneous RLS. The authors concluded that RLS are more prevalent in migraineurs with aura. However, the presence of RLS does not explain the increased prevalence of silent posterior circulation infarcts or white matter lesions found on MRI in migraineurs.


This study reiterates the known association between migraine with aura and RLS, now shown in a general population of migraine individuals. The study also again demonstrates the lack of correlation between MRI lesions and the presence of RLS. White matter lesions on MRI are frequent incidental findings in migraineurs, both with and without aura. Although these migraine-associated lesions are assumed to be ischemic, the exact neuropathology of these lesions, as localized on MRI but not diagnosed based on tissue sampling, has not been characterized for obvious reasons. Multiple studies have determined conclusively that RLS is more common in individuals whose migraine attacks are presaged by an aura. The hypothesis that microemboli cross unfiltered through the heart to the brain, and thus induce cerebral ischemia with migraine triggering, would be buttressed by a correlation between the presence of an RLS and presumed small vessel ischemic damage as noted on MRI. However, like prior published studies, such as the shunt-associated migraine (SAM) study, this has not shown a correlation between lesions found incidentally on the MRIs of migraineurs and the presence of an RLS.

Koppen et al used TCD with bubble detection to diagnose shunting, stating that TCD cannot distinguish between cardiac and pulmonary shunts. However, TCD is more likely to pick up cardiac, as opposed to less common pulmonary shunting, and other echocardiographic studies have found a PFO and migraine with aura correlation. The reason for the association between a cardiac right-to-left shunt and migraine with aura is unclear, with genetic co-localization, rather than a causative explanation, being the most convincing hypothesis.

This study found that migraine individuals with aura, but not those without aura, more often had ongoing migraine activity over years, if they also had spontaneous RLS. The presence of a shunt may predict more persistent migraines in those with aura, identifying individuals who should be offered early and aggressive preventive therapy. However, there is no evidence that PFO closure in those with migraine with aura has any beneficial effect on migraine frequency and severity, since unconvincing results were found in numerous clinical trials. Therefore, for now, usual migraine acute and preventive therapies should be offered to all patients with migraine, regardless of the presence of systemic shunting. Likewise, the presence of characteristic MRI lesions in a migraineur does not necessitate either further investigation or different therapy beyond that offered to all migraineurs. These findings and correlations, while thought provoking, have not yet changed the way that migraine patients are evaluated or treated.