Third Nerve Palsy Due to Posterior Communicating Artery Aneurysm Rupture Into the Subdural Space

We describe a unique case of intracranial aneurysm causing third nerve palsy (3NP) rupturing nearly entirely into the subdural space (SDS). The presence of large amount of blood in SDS mimicked dural thickening on computed tomography (CT), misdirecting neuro- radiological interpretation of the initial imaging.

Clinical Correspondence Section Editors: Robert Avery, DO Karl C. Golnik, MD Caroline Froment, MD, PhD An-Guor Wang, MD Third Nerve Palsy Due to Posterior Communicating Artery Aneurysm Rupture Into the Subdural Space Trishal Jeeva-Patel, MD, Daniel M. Mandell, MD, PhD, Edward Margolin, MD W e describe a unique case of intracranial aneurysm causing third nerve palsy (3NP) rupturing nearly entirely into the subdural space (SDS). The presence of large amount of blood in SDS mimicked dural thickening on computed tomography (CT), misdirecting neuroradiological interpretation of the initial imaging. A 65-year-old woman noticed progressively worsening headache for 2 months. Her past medical history was remarkable only for well-controlled hypertension. Two weeks before presentation she developed left upper lid ptosis, binocular oblique diplopia, and pain behind the left eye. CT head without contrast was normal. The headaches gradually worsened and she was eventually referred for neuro-ophthalmic evaluation. On examination, the left pupil was 2 mm larger than right and unreactive to light; there was almost complete left ptosis with only 30% adduction, 20% supraduction, and 20% infraduction of the left eye. Diagnosis of left pupil-involving partial 3NP was made. CT angiography (CTA) was arranged for the same day; however, while waiting for the scan, patient suddenly collapsed in the hospital cafeteria. She spontaneously regained consciousness after 30 seconds and underwent immediate CT head without contrast and CTA. CT demonstrated smooth hyperdense thickening along the falx, tentorium, posterior aspect of the clivus, and along the floor of posterior fossa (Fig. 1). No intracranial aneurysm was reported on CTA. Radiological differential diagnosis for the diffuse dural thickening included inflammatory and infiltrative etiologies (IgG4 disease and neoplasms), and thus, CT body was performed and was interpreted as normal. Patient’s headache was worsening, and it was felt that she must have a compressive etiology responsible for 3NP. MRI and MR venography (MRV) of the brain was obtained and Departments of Ophthalmology and Vision Sciences (TJ-P, EM), and Medical Imaging (DMM), Faculty of Medicine, University of Toronto, Toronto, Canada; and Division of Neurology, Department of Medicine (EM), Faculty of Medicine, University of Toronto, Toronto, Canada. The authors report no conflicts of interest. Address correspondence to Edward Margolin, MD, Division of Neurology, Departments of Medicine, Ophthalmology and Visual Sciences, University of Toronto, 801 Eglinton Avenue West, Suite 301, Toronto, ON M5N 1E3, Canada; E-mail: edward.margolin@uhn. ca e738 showed that hyperdense thickening along the dura seen on CT was due to acute blood in SDS rather than thickening of the dura itself: it bloomed on susceptibility weighted imaging, was isointense on T1, and did not enhance (Fig. 2). Because of the clinical suspicion for compressive lesion and presence of subdural blood, MRI/MRV and previous CTA were re-examined and a small left posterior communicating artery (PCom) aneurysm was identified (Fig. 3A). Digital subtraction angiography (DSA) demonstrated 3.5-mm aneurysm at the origin of left Pcom from internal carotid artery (ICA), which was treated by endovascular coiling (Fig. 3B). There was no blood in subarachnoid space (SAS) on CT and only a very small amount of blood identified there on MRI, and thus it was concluded that aneurysm has ruptured nearly entirely into SDS. One month later, patient was asymptomatic with almost normal ocular motility. Rupture of intracranial aneurysm usually results in subarachnoid hemorrhage (SAH). The prognosis of ruptured aneurysms is generally poor with associated mortality rate of up to 50% and a high rate of residual neurological impairment in survivors (1,2). Aneurysmal rupture into SDS rather than SAS is rare with few cases reported in the literature (1–3). It tends to occur more commonly at Pcom origin as in our case (1–3). The large intracranial arteries travel through SAS and thus hemorrhage into this space can result in cerebral vasospasm, cerebral ischemia, and impair cerebrospinal fluid resorption resulting in hydrocephalus and elevated intracranial pressure. Thus prognosis of aneurysms rupturing into SDS is much better than those producing SAH (1). Although prognosis for aneurysmal rupture into the SDS is much better, it is more difficult to diagnose; it is rare and may be unfamiliar to interpreting radiologists and hyperdense thickening along the dura has a broader differential diagnosis (2,4). In our case, unusually diffuse distribution of subdural hemorrhage was interpreted as dural thickening and thus distracting from more careful evaluation of ICA–PCom junction for the presence of aneurysm. Small size of the aneurysm in our case was also unusual, and one recent article concluded that it is the distance between the ICA and anterior–posterior clinoid process that is most relevant to whether the aneurysm will produce 3NP (5). This study showed that Jeeva-Patel et al: J Neuro-Ophthalmol 2021; 41: e738-e740 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical Correspondence FIG. 1. Nonenhanced CT brain axial images. Images show smooth, homogeneously hyperdense abnormal thickening along the dura of the falx (left image), tentorium cerebellum (middle image), clivus, and floor of the posterior fossa (right image). 13 Pcom aneurysms causing 3NP were sized 6.89 ± 2.22 mm (range 4.3–12.3 mm) vs 35 PCom aneurysms not causing 3NP sized 6.75 ± 2.51 mm (range 2.9– 12 mm). Hence, in that study, all the PCOM aneurysms 4 mm or less did not produce a 3NP. Two mechanisms explaining aneurysmal rupture into SDS and not SAS were proposed. There could be small successive bleeds from the stretched aneurysmal site resulting in tight adhesions to the arachnoid membrane creating a channel for final rupture through the arachnoid membrane and into the SDS. It is also possible that in cases involving high pressure aneurysmal rupture there is rapid accumulation of blood from leaking aneurysm with resultant velocitydirected rupture through the arachnoid membrane directly into SDS (1–4). One study that reviewed the most common reasons for false-negative neuro-imaging reporting of Pcom aneurysms found that it was not the choice of imaging or its FIG. 2. Nonenhanced CT image (A) sagittal plane shows hyperdense thickening along the floor of the posterior fossa (arrows) extending into the upper aspect of the spinal canal. Susceptibility weighted MRI (B) axial image shows corresponding hypointensity consistent with blood. Nonenhanced T1 sagittal MRI (C) images shows corresponding isointensity consistent with acute blood. Contrast-enhanced T1 sagittal MRI (D) image shows that the abnormal thickening (small arrows) is nonenhancing consistent with blood and separate from the dura (long arrow). Jeeva-Patel et al: J Neuro-Ophthalmol 2021; 41: e738-e740 e739 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical Correspondence FIG. 3. Contrast-enhanced MR venogram (A) sagittal image shows a 3–4-mm rounded structure (arrow) in the left cavernous sinus region suspicious for an aneurysm. Catheter angiography (B) left internal carotid artery injection shows a corresponding 3–4-mm posterior communicating artery aneurysm (arrows). limitations but rather incomplete or erroneous communication to the radiologist and the training of the radiologist that was the most common culprit (6). In our case, although the study was interpreted by an experienced neuro-radiologist in an academic center, the diagnostic challenge was due to the very uncommon bleeding pattern. CTA is considered the ideal initial examination to assess for aneurysms, and a recent meta-analysis reported its sensitivity of 95% (95% confidence interval = 93%–96%) and specificity of 96% (95% confidence interval = 93%–98%) for diagnosis of intracranial aneurysms with DSA as the reference standard. Sensitivity was slightly lower for detection of small aneurysms, with mean sensitivity of 94% for aneurysms #4 mm using 16- or 64-detector CT scanners (7). It is important to recognize that in some patients the lumen of a ruptured or leaking aneurysm can contain thrombosed blood preventing its visualization using techniques such as CTA, contrast-enhanced MRA, and DSA that rely on filling the aneurysm with intravascular contrast for visualization. Thus, if clinical suspicion for ruptured aneurysm is high, DSA is warranted, and if it is negative and suspicion for aneurysm is very high, CTA should be repeated after 1–2 weeks. Our case highlights the importance of ruling out aneurysmal rupture in every patient with pupillary involving 3NP and the need for close collaboration between clinician and radiologist to ensure that high clinical suspicion of PCom aneurysm is communicated to ensure that PCom aneurysm is excluded before any other entities are considered in differential diagnosis. Astute clinicians should also be aware of the possibility of aneurysmal rupture into SDS and its associated unusual clinical and neuro-radiological presentation. e740 STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: E. Margolin, T. Jeeva-Patel, and D. Mandell; b. Acquisition of data: E. Margolin, T. Jeeva-Patel, and D. Mandell; c. Analysis and interpretation of data: E. Margolin, T. Jeeva-Patel, and D. Mandell. Category 2: a. Drafting the manuscript: E. Margolin, T. Jeeva-Patel, and D. Mandell; b. Revising it for intellectual content: E. Margolin, T. Jeeva-Patel, and D. Mandell. Category 3: a. Final approval of the completed manuscript: E. Margolin, T. Jeeva-Patel, and D. Mandell. REFERENCES 1. Marbacher S, Tomasi O, Fandino J. Management of patients presenting with acute subdural hematoma due to ruptured intracranial aneurysm. Int J Vasc Med. 2012;2012:753596. 2. Feng Z, Tan Q, Li L, Chen Z. Subdural hematoma caused by rupture of a posterior cerebral artery aneurysm. Neurosciences. 2016;21:161–163. 3. Gong J, Sun H, Shi XY, Liu WX, Shen Z. Pure subdural haematoma caused by rupture of middle cerebral artery aneurysm: case report and literature review. J Int Med Res. 2014;42:870–878. 4. Inamasu J, Saito R, Nakamura Y, Ichikizaki K, Suga S, Kawase T, Hori S, Aikawa N. Acute subdural hematoma caused by ruptured cerebral aneurysms: diagnostic and therapeutic pitfalls. Resuscitation. 2002;52:71–76. 5. Anan M, Nagai Y, Fudaba H, Kubo T, Ishii K, Murata K, Hisamitsu Y, Kawano Y, Hori Y, Nagatomi H, Abe T, Fujiki M. Third nerve palsy caused by compression of the posterior communicating artery aneurysm does not depend on the size of the aneurysm, but on the distance between the ICA and the anterior-posterior clinoid process. Clin Neurol Neurosurg. 2014;123:169–173. 6. Elmalem VI, Hudgins PA, Bruce BB, Newman NJ, Biousse V. Underdiagnosis of posterior communicating artery aneurysm in noninvasive brain vascular studies. J Neuroophthalmol. 2011;31:103–109. 7. Yoon DY, Lim KJ, Choi CS, Cho BM, Oh SM, Chang SK. Detection and characterization of intracranial aneurysms with 16-channel multidetector row CT angiography: a prospective comparison of volume-rendered images and digital subtraction angiography. AJNR Am J Neuroradiol. 2007;28:60–67. Jeeva-Patel et al: J Neuro-Ophthalmol 2021; 41: e738-e740 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.