Stuck with a drowsy patient, evoke the Percheron
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.162045
Source of Support: None, Conflict of Interest: None
Background: Strokes caused by normal variants of the cerebral circulation can be difficult to diagnose, hence a high index of suspicion is needed. This case series discusses the clinical and radiological aspects of one such stroke caused by occlusion of the artery of Percheron (AOP).
Keywords: Bilateral thalamus; Percheron; thalamic infarction
Normal variants of the cerebral circulation can be seen commonly; however, strokes due to occlusion of these variant arteries are uncommon and can cause difficult-to-diagnose strokes. The artery of Percheron (AOP) is one such variant artery. In a patient presenting with altered sensorium with no other obvious cause, AOP infarction should be considered in the differential diagnosis.
The thalamus has a blood supply with contributions from both the anterior and posterior circulations [Figure 1]a. Based on the blood supply, the thalamus can be divided into 4 vascular regions.  (1) Anterior region: Supplied by the polar (or thalamotuberal) arteries arising from the posterior communicating artery (PcomA); (2) Paramedian region: Supplied by the paramedian (or thalamoperforating) arteries which are branches from the P1 segment of the posterior cerebral artery (PCA); (3) Inferolateral region: Supplied by thalamogeniculate arteries arising from the P2 segment of the PCA; and, (4) Posterior region: Supplied by the posterior choroidal arteries from the P2 segment of the PCA.
There can be several normal variants making its complicated blood supply even more complex.
In 1973, Percheron described a variant supply to the paramedian thalamus.  Normally the paramedian region is supplied by the thalamoperforate arteries (paramedian arteries) which originate from the P1 segments of the posterior cerebral artery (PCA). Rarely, both the paramedian regions can get supplied by a single arterial trunk, the artery of Percheron (AOP). This artery arises from the P1 segment of one PCA and divides to supply bilateral paramedian thalami [Figure 1]b. The paramedian mesencephalic arteries, which usually arise from the P1 segment to supply the superior medial parts of the midbrain, can originate from the AOP. In such patient, an occlusion of the AOP can also involve the rostral midbrain medially.
The anterior thalamus is usually supplied by the polar artery, which arises from the PcomA. The polar artery can be absent in 30-60% of the population. , In such cases, the paramedian arteries can be seen supplying both the paramedian area and the anterior thalamus.
Hence, an occlusion of the AOP can cause bilateral paramedian thalamic infarctions with or without midbrain infarction, and in the absence of the polar artery, the anterior thalamus also will be affected [Table 1]. Thus, the artery of Percehron occlusions can result in a variety of clinical presentations, which can mislead a clinician.
In this case series, we have discussed the clinical and radiological aspects of this rare but interesting stroke.
This study was conducted in the Department of Neurological Sciences of a quaternary level teaching hospital. Patients with AOP infarction over a period of 12-years (2002-2014) were identified from the computerized discharge summaries. The patient data was collected from the inpatient records and the stroke clinic where the patients were being followed up. Radiological images were reviewed from picture archiving and communication system (PACS, GE). The diagnosis of AOP infarction was based on magnetic resonance imaging (MRI) and magnetic resonance angiography. Any other etiologies which can mimic AOP infarction like a deep venous system infarction, tumor etc., were excluded after carrying out other appropriate imaging/tests. Furthermore, patients with top of the basilar artery occlusions with bilateral thalamic infarcts were excluded.
The modified Rankin Scale (mRS)  was used to assess the outcome. Two-tailed Fisher's exact test was used to look for significant associations. Statistical analysis was done using a statistical package (STATA, StataCorp).
Over a 12-year period, out of a total of 3589 patients with strokes (both ischemic and hemorrhagic), 17 (0.47%) patients had an AOP infarction. The male: female ratio was 0.8 (8 males and 9 females). The mean age was 50 years (range: 31-72 years).
The most common presenting symptom was its association with disorders of consciousness (16/17; 94%) [Table 2]. Sudden loss of consciousness of varying intervals ranging from 3 h to 24 h was seen in 12/17 (70.5%) patients, and 4/12 (33.3%) had fluctuations in their consciousness as their presenting symptom. Disorders of gaze were seen in 9/17 (53%), upgaze restriction 5/17 (29%), both upgaze and downgaze restriction in 2/17 (12%), and skew deviation in 3/17 (18%) patients.
Apathy, muteness, and hypersomnolence were seen in 5/17 (29%) patients. Memory (anterograde amnesia) was impaired in 4/17 (24%) patients. All 4 patients, who had a memory impairment, had involvement of the anterior thalamus on imaging. Headache was a prominent presenting complaint in 3/17 (18%) patients. At follow-up, 2/17 (12%) had patients developed extrapyramidal features as a late complication.
All patients were evaluated by MRI performed on Philips 3T, 1.5T or Siemens 1.5T scanners. In all of them, bilateral paramedian thalamic infarcts were seen. The anterior thalamus was also involved in 5/17 (30%) patients (bilateral involvement in 2 patients and a right sided involvement in 3 patients) [Figure 2].
Midbrain involvement was seen in 10/17 (59%) cases. Four had symmetrical involvement; in 4, the midbrain involvement was more on the left side; and in 2, the involvement was predominantly on the right. The circle of Willis was incomplete in 7/17 (41%) patients out of which 4 had absent unilateral or bilateral PComs, and 3 had absent unilateral or bilateral P1 segments.
A computed tomography (CT) scan was done in 11/17 patients prior to the MRI scan and in 6/11 (55%) patients, the AOP infarct was visible on the CT scan.
Magnetic resonance venogram was done in 5 patients to rule out suspected deep venous system thrombosis. In 7/17 (41%) patients, the imaging was repeated. There was no extension of the infarct in any of them. Two had a hemorrhagic transformation of their infarcts.
Using the Trial of Org 10172 in Acute Stroke Treatment criteria,  the most common cause of the infarct was cardioembolic in 8/17 (47%) patients, rheumatic heart disease in 3 patients, right to left shunts and atrial fibrillation in 2 patients each, and dilated cardiomyopathy in one patient. Small infarcts due to simultaneous embolic occlusion in the cerebellum and in the anterior circulation were seen in 3 patients.
Small vessel occlusion was seen in 7/17 (41%) patients. These patient had the presence of multiple risk factors for atherosclerosis. One patient had the presence of systemic vasculitis with central nervous system involvement and in another, the evaluation was incomplete.
Two out of the 17 (12%) patients died in the acute phase. A patient, who had a mitral valve replacement for rheumatic mitral involvement, had developed infective endocarditis and sepsis with shock. Another patient, who was treated with intravenous tissue plasminogen activator, had an intracranial bleed with intraventricular extension and succumbed to his illness.
A 6 months follow-up was available for 12/17 (71%) patients. An MRS of ≤2 was seen in 8/17 (47%) at 6 months. In the remaining 4 patients, with an MRS of >3, the follow-up was available for 2 years; however, there was no significant improvement noted in these patients. At follow-up. these patients were noted to have developed severe extrapyramidal features and a debilitating cognitive impairment.
In 2 large series studying the distribution of strokes, the characteristic AOP infarct pattern was estimated to have occured in 0.1% and 0.3% of all ischemic strokes. , In our series, this figure was 0.47%; this slightly higher percentage could have been due to a referral bias.
The mean age of patients in our series was 50 years; this was similar to that seen in the study reported by Lazzaro et al.,  where the mean age was 59 years, and the meta-analysis by Yarmohammadi et al., where the mean age of 50 years was exactly similar to ours. 
Disorders of consciousness and gaze disorders are the most common presenting features in AOP infarcts. Bilateral paramedian thalamic infarction can result in a decrease in consciousness ranging from somnolence to coma. This is usually transient. The paramedian region of the thalamus takes part in reward learning  and a lesion involving these areas can cause apathy and lack of motivation.  A presleep behavior and a tendency to assume a sleep inducing body posturing have also been reported. 
If the paramedian upper midbrain is also involved, there can be impairment of oculomotor functions. The vertical eye movement restrictions, especially the down gaze, can be affected. Pupillary abnormalities, ptosis, and adduction deficits are common with rostral midbrain lesions, producing a "mesencephalothalamic" or "thalamopeduncular" syndrome. Other associated findings can be hemiplegia , cerebellar ataxia and movement disorders. 
If the anterior thalamus is involved, memory impairment can occur due to the involvement of the mammillothalamic tract, and the inferior thalamic pedicle as well as anterior and dorsomedial nuclear , confabulations can be a frequent association. 
In this series, the thalamic infarcts due to the AOP infarcts were picked up only in 6/11 (55%) of the patients who had a CT scan done prior to the MRI. Furthermore, it is worth noting that the AOP is usually not seen on the conventional angiography.  Hence, MRI with diffusion-weighted imaging (DWI) sequence is best to pick up early ischemic changes caused by AOP infarction. Very rarely, visualization of the acute infarction may be absent on the initial DWI, and if the index of suspicion is high, a repeat imaging should be done.  Lazzaro et al.  demonstrated 4 distinct patterns of AOP infarction. These include bilateral paramedian thalamic with rostral midbrain (43%); bilateral paramedian thalamic without midbrain (38%); bilateral paramedian and anterior thalamic with midbrain (14%); and, bilateral paramedian and anterior thalamic without midbrain (5%) infarctions. Comparing our series, we found the similar frequency in the patterns detected, the rarest being bilateral paramedian and anterior thalamic without midbrain infarction [Table 3]. Furthermore, with AOP infarcts, the midbrain "V" sign can be seen.  A distinct pattern of V-shaped hyperintensity is seen in the axial fluid-attenuated inversion recovery and/or diffusion weighted image along the pial surface of the midbrain adjacent to the interpeduncular fossa. This sign signifies that the midbrain perforators are also taking off from the AOP. This V sign signifying medial midbrain involvement was seen in 5/10 (30%) patients.
In the acute setting, the imaging differentials will include a deep venous system thrombosis. , Susceptibility weighted imaging sequences on the MRI can be helpful in demonstrating the thrombosed veins.  If the distal basilar artery gets occluded, both the thalamoperforate arteries (paramedian arteries) and the paramedian mesencephalic arteries on both sides may also get occluded mimicking an AOP occlusion. Clinically, these patients will usually have other features of 'top-of-the-basilar' syndrome. 
Wernicke encephalopathy may present with the classical triad of ataxia, altered consciousness, and abnormal eye movements. T2-weighted MR changes will be seen in the mammillary bodies, tectal plate, periaqueductal gray matter, dorsal medulla, and the medial aspects of thalami. There may be restriction of diffusion in these areas mimicking an arterial infarction. ,,
The most common etiology of the infarction in our series was cardioembolic (8/17; 47%); in other series ,,,,, also, cardioembolism was a major cause for AOP infarcts [Table 4].
A 6 month follow-up was available for 12/17 (71%) patients. An MRS of ≤2 was seen in 8/17 (47%) patients at 6 months. In the remaining 4 patients with an MRS of ≥3, a follow-up was available for 2 years. It is noteworthy that not only was there lack of further improvement seen in these patients but they also developed severe extrapyramidal features and debilitating cognitive impairment.
The mortality in our series was 12% which was similar to that seen in other studies [Table 4]. It was also noted that if cognitive recovery does not occur in the acute phase, it is unlikely to improve at follow-up.
AOP infarcts can present with sudden alterations in sensorium and disorders of gaze may give a clue towards the diagnosis. Since the CT scan may not pick up the AOP infarcts, MRI with DWI/ADC should be done in suspected cases. An embolic source should be actively looked for in these patients. If the patient does not improve in the initial phase, the chance for further recovery appears to be low; hence, in the appropriate patient, thrombolytic therapy should be considered. 
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]