Send letter to journal:
Re: PFO and migraine: pearls and pitfalls in the theorizing process

Dear Editor,

Wilmshurst et al.,[1] link inheritance of atrial shunts to inheritance of migraine with aura in some families. The link between migraine with aura and patent foramen ovale (PFO) appears to be gathering strength and the plea for randomized controlled trials of PFO-closure in migraine patients are beginning to appear justified.[2-4] The conclusion that closure of atrial shunts usually improves or cures migraine with aura[1] reflects the philosophic commitment and excitement of the therapists.

As the momentum for PFO closure for prevention of migraine builds up, it becomes increasingly difficult to maintain scientific equipoise.[5-8] One, closure of both PFO (right-to-left shunt) as well as closure of atrial septal defect (left- to-right shunt) has been proposed to prevent migraine attacks.[5,6]

Two, migraine is generally a disorder with a predictable onset and recurrences, and, is a characteristically lateralizing disorder, with the headache persistently occurring unilaterally (right or left), bilaterally, or in a side-shifting manner. The Valsalva manœuvre can aggravate existent migraine headache but is not known to precipitate migraine headache. A MEDLINE search does not reveal any report of a direct link between Valsalva manœuvre and precipitation of migraine. Cerebrovascular dysfunction secondary to a congenital defect such as PFO cannot explain the late appearance (in teens or twenties), disappearance (second and third trimesters of pregnancy and in later decades), or periodic recurrences (menstrual migraine) of headache or aura or both. Moreover, with the concept of paradoxical embolization of gas, thrombi or vasoactive neuromediators, these potential precipitants are presumed to be streamed regularly over decades to the same brain parenchymal site or circulatory segment in order to produce lateralizing headache or aura – such a pathophysiological mechanism is highly unlikely.[7-9]

Three, cluster headache (CH) is a strictly lateralized headache without implication of brain ischemia in its pathogenesis; a remarkably high incidence of right-to-left shunt (42.5%) has been seen in CH patients.[10] To conclude a striking effect on prevention of migraine aura or headache or both simply by PFO closure is premature as we need to know: (ii) why the incidence of PFO is higher in migraine or CH patients (or other hitherto unknown subsets) than in the general population; and (ii) why PFO-associated right-to-left shunt does not cause aura-like brain dysfunction in most patients.

Four, drugs that do not cross the blood- brain barrier and cannot critically influence brain neuronal function, are effective migraine prophylactic (atenolol, verapamil or nadolol) or aura abortive (nifedipine or isoproterenol) agents.[9] Migraine prevention does not appear convincingly related to crucial modulation of a particular brain neuronal function; a peripheral extra-CNS origin for migraine appears to be distinctly plausible.[9,11]

Five, headaches are less frequent, less severe and shorter in migraine with aura patients; the likelihood of a placebo effect of PFO closure is much more likely in this subset[5,8]

Six, migraine is more than twice as common in women than in men but the incidence of PFO does not show a similar pattern.[1]

Seven, it is debatable whether aspirin can be stopped altogether in migraine patients after PFO closure, as implantation of an occluder device with a left-atrial disc would augment platelet aggregation and embolism.

Eight, a large PFO or small atrial septal defect as seen in migraine patients[1,3] would tend to maintain a left-to-right shunt under most circumstances. Access to the left heart at the level of the atria in PFO by air-bubbles during Valsalva manœuvre is a laboratory artifact that does not reflect a real-life situation; in paradoxical embolism, it is a platelet-thrombin plug that embolizes – such an embolus has physical properties not at all akin to air bubbles. The larger-sized PFO may permit air bubbles to easily cross over to the left heart due to rheological factors that fundamentally differ from those that affect platelet-thrombin complexes.

Nine, reduced frequency of migraine in patients started on oral anticoagulation[2] does not directly support a pathogenetic role for PFO in migraine. Besides the placebo effect of warfarin which is quite significant in a highly variable condition such as migraine, by impairing blood coagulability warfarin may increase vasopressin bioavailability; sub-clinical rises in vasopressin might, in turn, prevent recurrences of migraine.[11,12] Prevention of migraine following PFO-closure is the pursuit of possible benefit involving a remote circulatory effect on the brain circulation. While the practice of PFO-closure for migraine itself cannot be currently recommended, serious conceptual groundwork is still missing from the research endeavour. Neither the aura nor the headache represents the true onset of the migraine attack; brain circulatory changes or cortical spreading depression must not be construed as the part of the early physiological changes in migraine.[11] The primary pathogenetic aberrations in migraine lie buried in the largely clinically inaccessible but variably extended prodromal and “pre-prodromal” phases.[11] Larger studies with optimal designs will not eliminate these basic scientific issues.[5-9] Since almost 50% of migraine with aura patients do not have a PFO,[3] pathogenetic mechanisms other than PFO-related are clearly operative.[1] The emphasis on relatively larger atrial shunts in migraine with aura patients[1] does not diminish the strength of the arguments against such a pathogenetic mechanism.

To suspend clinical disbelief at this stage of evolution of migraine mechanisms does not appear justifiable. From beta-blockers to magnesium[14] to botulinum toxin,[15] serendipity has (mis)guided migraine research. Migraine therapeutics, particularly preventive or prophylactic therapy, remains purely empirical. PFO has simply been detected in a fraction of migraine patients; any conclusion that it might be involved in the pathogenesis of migraine is premature. Whether the presence of a PFO in primary headache patients confers a teleological or physiological advantage is an intriguing question. Precipitation of migraine following closure of atrial septal defect[16] suggests that a higher stoke volume / cardiac output might worsen or unmask a migrainous diathesis.[6]

No theory should be allowed to exist in a vacuum; all theories must be generalizable and should gather strength by an increasing number of logically defensible clinical associations and links with basic science tenets that, in turn, lead towards the evolution of an intelligible synthesis. The occurrence of migraine in any one individual is probably based on a polygenic trait superimposed on a highly variable state. The link between PFO and primary headaches, both migraine and cluster headache, may lie in the expression of a connective tissue disorder that affects both inter-atrial septal closure as well as visco-elastic properties and pressure-volume relationship of the corneo-scleral envelope.[17,18]

References:

1. Wilmshurst PT, Pearson MJ, Nightingale S, Walsh KP, Morrison WL. Inheritance of persistent foramen ovale and atrial septal defects and the relation to familial migraine with aura. Heart 2004;90:1315-20.

2. Zanchetta M, Rigatelli G, Ho SY. A mystery featuring right-to-left shunting despite normal intracardiac pressure. Chest 2005;128:998-1002.

3. Schwerzmann M, Nedeltchev K, Lagger F, Mattle HP, Windecker S, Meier B, Seiler C. Prevalence and size of directly detected patent foramen ovale in migraine with aura. Neurology 2005 (Published online before print September 7, 2005).

4. Morandi E, Anzola GP, Casilli F, Onorato E. Migraine: traditional or “innovative” treatment? A preliminary case-control study. Pediatr Cardiol 2005;25:1-3.

5. Gupta VK. Percutaneous closure of patent foramen ovale reduces the frequency of migraine attacks. Neurology 2004;63:1760-1761.

6. Gupta VK. Closure of atrial septal defect and migraine. Headache 2004;44:291-292.

7. Gupta VK. ASD closure for migraine: is there a scientific basis? Eur Heart J 2005; 2005;14:1446.

8. Gupta VK. Patent foramen ovale / atrial septal defect closure and migraine: searching the rationale for the procedure. J Am Coll Cardiol 2005;46:737-8.

9. Gupta VK. Non-lateralizing brain PET changes in migraine: phenomenology versus pharmacology? Brain 2004 127:E12.

10. Finocchi C, Del Sette M, Angeli S, Rizzi D, Gandolfo C. Cluster headache and right-to-left shunt on contrast transcranial Doppler. A case- control study. Neurology 2004;63:1309-1310.

11. Gupta V. Migraine, cortical excitability and evoked potentials: a clinico-pharmacological perspective. Brain 2005;128:E36.

12. Gupta VK. Does vasopressin mediate the migraine-remitting influence of warfarin? Headache 1999;39:140-141.

13. Gupta VK. A clinical review of the adaptive role of vasopressin in migraine. Cephalalgia 1997;17:561-569.

14. Gupta VK. Magnesium therapy for migraine: do we need more trials or more reflection? Headache 2004; 44: 445-446.

15. Gupta VK. Botulinum toxin type A therapy for chronic tension-type headache: fact versus fiction. Pain 2005; 116: 166-167.

16. Yankovsky AE, Kuritzky A. Transformation into daily migraine with aura following transcutaneous atrial septal defect closure. Headache 2003;43:496-498.

17. Gupta V. Does the eye modulate the clinical expression of cluster headache? BMC Neurology 2005;5:6 (Published online 17 May 2005). Available at: http://www.biomedcentral.com/1471-2377/5/6/comments#201461