Abstract
Introduction
Tadalafil (Cialis; Eli Lilly, Indianapolis, IN, USA) and sildenafil (Viagra; Pfizer Inc, New York, NY, USA) are both phosphodiesterase type 5 (PDE5) inhibitors approved for treatment of erectile dysfunction (1, 2). PDE5 inhibitors slow metabolism of 3′,5′-cyclic guanosine monophosphate (cGMP), a messenger used in the pathway that causes smooth muscle relaxation by nitric oxide (3, 4). Although this class of medications is known to cause headaches, we describe a patient with positive visual phenomena without headache associated the use of tadalafil.
Report of a case
A 64-year-old white man with a history of arterial hypertension and paroxysmal atrial fibrillation was sent to the neuro-ophthalmology service for evaluation. He stated that approximately 1½ years previously, he developed new-onset stereotypical positive visual phenomenon that lasted 20 min. There were no other associated symptoms, including headache, parasthesia or dysphasia.
Each episode began and ended in the same manner. The episodes were ushered in by the presence of a bright ill-defined triangular shape starting in the right visual field of what appeared to be both eyes. The apex of the triangle always pointed toward ocular fixation. The ‘object’ would gradually increase in size and slowly (over minutes) move towards the centre of his vision, eventually causing some obstruction of his ability to read. After the ‘object’ was in the centre of his visual field, it would disappear.
This gentleman had been on sildenafil for approximately 4 years without having such an episode; however, he began noticing these episodes a few months after switching to tadalafil. Quite characteristically, the episodes would occur on Mondays after taking tadalafil on Fridays. He self-discontinued tadalafil and had a recurrence of this visual phenomenon around 1 week after discontinuation. He had not taken tadalafil for 4 months prior to our examination and only had that one single occurrence when off tadalafil. In the 7 months following our examination, he had approximately five more episodes. Some of these episodes occurred after he restarted the sildenafil. Unlike with tadalafil, the patient could not identify a temporal relationship between the ingestion of sildenafil and the onset of the aura. Most notably, the frequency was markedly decreased than when he was on the tadalafil. No other changes in his medicines or his health occurred around the time these episodes began.
His past ocular history was significant only for rigid gas permeable lens wear and remote history of recurrent subconjunctival haemorrhages. Other medications included diltiazem, bisoprolol, hydrochlorothiazide, fexofenadine, aspirin and saw palmetto. His only hospitalization was within the last year for the onset of atrial fibrillation. The onset of his visual phenomenon preceded this hospitalization, and he had been on diltiazem for 5 years. His family history was negative for migraine.
Work-up initiated prior to our visit included carotid ultrasound (right common carotid artery with a small plaque with bilateral stenosis of the internal carotid artery of 0–39% stenosis), transthoracic echocardiogram (no source of emboli), MRI (normal), neurological examination (normal), ESR (normal), and CRP (normal).
Our examination, including Goldmann perimetry, was normal.
Comment
Known side-effects of tadalafil include mild systemic vasodilation (exacerbated by alcohol and nitrates) and headache (1, 5). Both tadalafil and sildenafil are very selective for PDE5; however, unlike sildenafil, tadalafil does have effects on PDE11 (1, 2, 6). The clinical significance of this is unknown. Sildenafil also affects PDE6; however, tadalafil is very non-selective for this enzyme (1, 2, 6). PDE6 is found in the retina and is involved with phototransduction (6), which may explain why visual changes are relatively rare with use of tadalafil compared with sildenafil. Further differences exist between the two in terms of half-life. Sildenafil and its metabolite have terminal half-lives of about 4 hours (2). Tadalafil has a terminal half-life of 17½ hours (1).
As PDE5 inhibitors affect smooth muscle and cause vasodilation, their association with headaches makes PDE5 inhibitors a target for understanding the mechanism of migraine. Sildenafil has been found to induce migraine attacks in patients with a history of migraine without aura; however, administration of sildenafil does not appear to affect middle cerebral, temporal or radial artery diameters (7). A similar study with sildenafil also demonstrated a significant induction of headache in healthy subjects associated with a lack of cerebral and extracerebral arterial dilation (8).
An alternative theory of migraine induction involves a spreading disruption of ionic homeostasis across the cerebral cortex known as cortical spreading depression (9). In a study of rats, cortical spreading depression did not appear to be altered by the administration of sildenafil (10).
Wolff (11) found that inhalation of a small amount of the vasodilator amyl nitrite could temporarily reverse the pre-headache scotoma in some subjects. Interestingly, an enlargement of the scotoma and generalized vasodilation was found when a larger amount of amyl nitrite was used. Wolff had concluded that the migraine aura was caused by vasoconstriction that could be overridden by the vasodilator carbon dioxide (11); however, hypercapnia inhibits the propagation of cortical spreading depression, which could explain this effect (9).
cGMP is known to be important in regulating ion channels and has been found to play an important role in neurotransmitter release and uptake, neuronal gene expression, and seizure activity (12, 13). It is plausible that a typical aura without headache could be induced by a PDE inhibitor via its action on targets different than smooth muscle.
Our patient never experienced this visual phenomenon prior to using tadalafil. The reason why there were no associated headaches and why the aura initially did not occur on sildenafil is not entirely clear. It is plausible that the aura of migraine and the pain from a headache are caused by slightly different but related mechanisms. It is also possible that this patient’s episodes occurred on tadalafil because of its longer half-life, allowing more intermediate messengers to accumulate. Drug–drug interactions could have played a role in this. In specific, sildenafil and tadalafil are both metabolized by cytochrome P450 3A4 (14, 15) and diltiazem inhibits this pathway (16, 17). This certainly could explain why the patient’s symptoms began despite the passage of several half-lives of tadalafil. It also is reasonable to suspect that the use of tadalafil may have changed the patient’s threshold for migraine induction.
It is unclear whether drug–drug interactions between the herbal saw palmetto and the phosphodiesterase inhibitors were related to the patient’s symptoms. Saw palmetto is known to have antiandrogenic and oestrogenic activity (18), which may be significant in this gentleman taking an erectile dysfunction medication.
Another possible explanation for this is the difference in selectivity of tadalafil and sildenafil on the different PDE families. Most notably, sildenafil lacks selectivity for PDE11 but tadalafil does not (1, 2, 6). At this point, there is not enough known about PDE11 to speculate about its role in migraine induction; however, the difference in half-lives seems more likely.
Conclusions
The exact aetiology of migraine is unknown. PDE inhibitors slow metabolism of cGMP, and cGMP is known to play a role in vasodilation, ion channel regulation and neurotransmitter release (3, 4, 12, 13). PDE inhibitors are known to induce migraine and may do so through any of these mechanisms. Tadalafil and sildenafil are both PDE5 inhibitors, but their selectivity on the different members of the PDE family and their durations of actions are different (1, 2, 6). Our patient experienced typical aura without headache much more frequently while using tadalafil than sildenafil. This may be due to the difference in selectivity on the different members of the PDE family or perhaps due to the longer half-life of tadalafil, which was exacerbated by the coadministration of diltiazem. It is important to note that a lack of migraine induction using one PDE5 inhibitor does not exclude the possibility of migraine induction for the whole class of medications. Furthermore, it is possible to have induction of migraine while using one medication in the class without getting migraine with another member of the class.