Fornix Infarction due to Involvement of Posterior Circulation

Letter to the Editor Fornix Infarction due to Involvement of Posterior Circulation Dear Editor, We read with great interest the article entitled ‘‘Vertebral artery dissection leading to fornix infarction: a case report,’’ by Kurokawa et al.1 The authors reported the case of a 26-year-old woman with bilateral fornix infarctions caused by an artery-to-artery embolism that occurred after vertebral artery dissection. These infarctions occurred in the left thalamus (anterior pole and posterolateral part, including the pulvinar thalamus), body of the left caudate nucleus, and left occipital lobe, as well as the bilateral fornices. The authors concluded that these bilateral fornix infarctions were caused by a vertebral artery embolus that migrated to the left posterior choroidal artery (LPChA) via the left posterior cerebral artery (PCA). The authors interpreted the involved culprit vessels for infarctions in the fornix and the body of the caudate nucleus as the LPChA; in the thalamus as the thalamotuberal artery (TTA), the thalamogeniculate artery, and the LPChA; and in left occipital lobe as the cortical branches of the PCA, based on magnetic resonance angiographic observations of the obstruction and subsequent recanalization of the left PCA and bilateral posterior communicating arteries. However, we have several concerns regarding the authors’ discussion. Our first concern is related to the authors’ conclusion regarding the vascular supply to the fornix. Although the authors cited an article describing LPChA perfusion of the crus, body, and part of the anterior columns of the fornix,2 we can state that the detailed anatomy of the vascular supply to the entire fornix is unclear except for the subcallosal artery, the unpaired and largest perforator from the anterior communicating artery (ACoA). The subcallosal artery likely supplies the pars libera and pars tecta of the fornical column on both sides and is most likely responsible for the postoperative amnesia of the ACoA aneurysm (Fig 1).3 Meanwhile, in their case, Kurokawa et al suggested that a fornix infarction with apparent bilateral involvement of the pars libera of the fornical column and unilateral or bilateral involvement of the fornical body was caused by an LPChA obstruction. Certainly, we also believe that the left LPChA was involved in this case, given the involvement

The authors declare that they have no disclosures. 1052-3057/$ - see front matter Ó 2015 by National Stroke Association

of the posterolateral part of the thalamus, including the pulvinar and lateral wall of the lateral ventricle (including the body of the caudate nucleus), both of which were reported to be perfused by the LPChA.4,5 However, we would like to suggest an additional culprit vessel for this fornix infarction, in addition to the LPChA: the medial posterior choroidal artery (MPChA). The LPChA usually arises from the P2 segment of the PCA and courses along the choroid plexus of the lateral ventricle, and presumably supplies the crus and (in part) body of the fornix, which are located along the choroid fissure of the lateral ventricle. On the other hand, the MPChA also usually originates from the P2 segment of the PCA, initially courses backward, encircles the midbrain medial to the PCA, and finally courses forward along the choroid plexus of the third ventricle, in the roof of the third ventricle (Fig 1).5,6 The MPChA often anastmoses with its counterpart, the opposite MPChA, in the roof of the third ventricle7 and with the LPChA, particularly at or around the foramen of Monro, and has been reported to supply the fornix.7-9 Meanwhile, the body of the fornix is located near the roof of the third ventricle toward the foramen of Monro, where it divides into the bilateral columns of the fornix. Near the foramen of Monro, the MPChA from the PCA and the subcallosal artery from the ACoA are thought to form a boundary zone at the pars libera of the fornical column, according to descriptions in the literature.8,10,11 Therefore, the MPChA could partially supply the fornix on its way along the roof of the third ventricle to reach the foramen of Monro, especially the anterior portion of fornical body and the pars libera of the fornical column, the latter of which is also known to be within the supply area of the unpaired subcallosal artery. For this reason, we suspect that the MPChA might also have been involved in this fornix infarction in the case of Kurokawa et al. Our second concern is the possibility that involvement of the left TTA might have been responsible for the amnesia in this case. We agree with the authors that thalamic infarction, including the anterior pole, and initial obstruction and subsequent recanalization of the posterior communicating arteries suggested TTA involvement. Meanwhile, infarctions in the TTA region of the dominant brain hemisphere are often associated with intellectual, memory, and emotional disturbances, because of disruption of the Papez neural circuit without fornical involvement.12 Therefore, to us, the infarcted thalamic foci in this case suggest that the amnesia might have been caused by not only fornical involvement but also

Journal of Stroke and Cerebrovascular Diseases, Vol. 24, No. 12 (December), 2015: pp 2883-2885

2883

LETTER TO THE EDITOR

2884

Figure 1. Mid-sagittal view of the minimum intensity projection (slice thickness 1 mm) of the three-dimensional constructive interference in steady state shows the anatomical relationship between the fornix (Fx) and the subcallosal artery (SbA) or the medial posterior choroidal artery (MPChA).3 The SbA originates from the posterosuperior aspect of the ACoA and ascends dorsally into the lamina terminalis cistern and curves upward and forward, thus exhibiting a characteristic S-shaped curve. The MPChA originates from the posterior cerebral artery, enters the roof of the third ventricle, runs forward along the internal cerebral vein, and distributes to the choroid plexus of the third ventricle (3CP). Note that the SbA comes close to the pars libera of the FxCo from the front and the MPChA does from behind. Abbreviations: ACoA, anterior communicating artery; AC, anterior commissure; CCr and CCg, the rostrum and genu of the corpus callosum, respectively; FxB, body of fornix; FxCo, column of fornix; ICV, internal cerebral vein; MB, mammillary body.

callosum. Based on these findings, we believe that a lack of involvement of the genu or rostrum of the corpus callosum does not always indicate a lack of involvement of the subcallosal artery. The final concern is that in the case presented by Kurokawa et al, a signal change in the fornical body on fluid-attenuated inversion recovery images at 4 weeks postonset might not represent an infarction, but rather Wallerian degeneration that is possibly associated with changes in intensity in the mammillary body. Indeed, a recent diffusion tensor imaging study also demonstrated reduced fractional anisotropy and mean diffusivity with a decreased fornical body volume after treatment for an ACoA aneurysm,15 which we believe was due to a secondary injury or Wallerian degeneration. Therefore, we are unsure whether the intensity change in the fornical body on fluid-attenuated inversion recovery images at 4 weeks postonset (but not on acute-phase diffusionweighted images) represents an actual infarction or subsequent Wallerian degeneration resulting from an infarction in the pars libera of the fornical column, as demonstrated by acute-phase diffusion-weighted images.

Shunji Mugikura, MD, PhD Shoki Takahashi, MD, PhD Department of Diagnostic Radiology Tohoku University, Graduate School of Medicine Sendai, Japan http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2015.07. 009

References involvement of the anterior thalamic nuclei and/or mammillothalamic tract within the TTA region. A detailed neuropsychological assessment and a detailed 3D magnetic resonance imaging would have been useful for clarifying the type of amnesia in this case. Next, we do not completely agree with the authors’ emphasis that the lack of involvement of the genu and/or rostrum of the corpus callosum suggests a lack of involvement of the subcallosal artery in their case, although we agree that the subcallosal artery was not involved. Indeed, subcallosal artery occlusion often does not involve such regions, likely because of anastomoses between the distal branches of the subcallosal artery that perfuse the genu or rostrum of the corpus callosum and the distal branches of the anterior cerebral artery.13 Actually, 3 of our 10 patients with subcallosal artery infarctions after surgical ACoA aneurysm clipping3 and 6 of 17 cases (including cases unrelated to aneurysmal surgery)14 had infarctions limited to the column of the fornix, without involvement of the rostrum or genu of the corpus

1. Kurokawa T, Baba Y, Fujino K, et al. Vertebral artery dissection leading to fornix infarction: a case report. J Stroke Cerebrovasc Dis 2015;24:e169-e172. 2. Stephens RB, Stilwell DL. Vertebral-basilar system. Arteries and veins of the human brain. Springfield, IL: Charles C Thomas 1969:96-99. 3. Mugikura S, Kikuchi H, Fujii T, et al. MR imaging of subcallosal artery infarct causing amnesia after surgery for anterior communicating artery aneurysm. AJNR Am J Neuroradiol 2014;35:2293-2301. 4. Saito R, Kumabe T, Sonoda Y, et al. Infarction of the lateral posterior choroidal artery territory after manipulation of the choroid plexus at the atrium: causal association with subependymal artery injury. J Neurosurg 2013; 119:158-163. 5. Takahashi S, Goto K, Fukasawa H, et al. Computed tomography of cerebral infarction along the distribution of the basal perforating arteries. Part II: thalamic arterial group. Radiology 1985;155:119-130. 6. Saito R, Kumabe T, Kanamori M, et al. Medial posterior choroidal artery territory infarction associated with tumor removal in the pineal/tectum/thalamus region through the occipital transtentorial approach. Clin Neurol Neurosurg 2013;115:1257-1263.

LETTER TO THE EDITOR 7. Wolfram-Gabel R, Maillot C, Koritke JG. The vascular pattern in the tela choroidea of the prosencephalon in man. J Neuroradiol 1987;14:10-26. 8. Lasjaunias P, Berenstein A, ter Brugge Karel. Intradural arteries. In: 1. Clinical Vascular Anatomy and Variations, Surgical Neuroangiography. 2nd ed. Berlin: Springer 2001:479-630. 9. Morris PP. The posterior cerebral artery. In: Practical Neuroangiography. 2nd ed. Philadelphia: Lippincott Williams & Wilkins 2006:226-239. 10. Pullicino PM. Diagrams of perforating artery territories in axial, coronal and sagittal planes. Adv Neurol 1993;62:41-72. 11. Takahashi S. Intracranial arterial system: basal perforating arteries. In: Takahashi S, ed. Neurovascular Imaging: MRI & Microangiography. London: Springer-Verlag 2010:53-130.

2885 12. Nishio Y, Hashimoto M, Ishii K, et al. Multiple thalamocortical disconnections in anterior thalamic infarction: implications for thalamic mechanisms of memory and language. Neuropsychologia 2014;53:264-273. 13. Marinkovic S, Milisavljevic M, Marinkovic Z. Branches of the anterior communicating artery. Microsurgical anatomy. Acta Neurochir 1990;106:78-85. 14. Meila D, Saliou G, Krings T. Subcallosal artery stroke: infarction of the fornix and the genu of the corpus callosum. The importance of the anterior communicating artery complex. Case series and review of the literature. Neuroradiology 2015;57:41-47. 15. Hong JH, Choi BY, Chang CH, et al. Injuries of the cingulum and fornix after rupture of an anterior communicating artery aneurysm: a diffusion tensor tractography study. Neurosurgery 2012;70:819-823.