0148-396X/82/1102-0247$02.00/0 NEUROSURGERY Copyright © 1982 by the Congress of Neurological Surgeons Vol. 11. No. 2, 1982 Printed in U.S.A Hemianopia: A Presenting Feature of Acute Epidural Hematomas Cornelius Matwijecky, M.D. and Paul Steinbok, M.B. Division of Neurosurgery, Department of Surgery, University of British Columbia, Vancouver, British Columbia Visual field abnormalities were studied in a retrospective analysis of 199 patients with epidural hematomas. One hundred three patients had been alert enough at the time of hospital admission to have had their visual fields examined, but in only 27 patients were visual fields documented in the charts. Six of the 27 patients (22%) had a correctly localizing homonymous hemianopia. The case histories of these 6 patients are detailed and the pathogenetic mechanisms of the visual field abnormalities are discussed. The importance of recognizing that homonymous hemianopia can be an early sign of an epidural hematoma is stressed. (Neurosurgery 11:247-249, 1982) Key words: Epidural hematoma, Homonymous hemianopia, Pathogenesis, Visual field abnormality INTRODUCTION Homonymous hemianopia is generally considered to be a sign of a deep-seated cerebral lesion. However, we are reporting several cases in which a homonymous hemianopia was a presenting sign of an acute epidural hematoma. This occur- rence has not received much attention in the literature. METHODS The hospital records of 199 patients who were admitted to Vancouver General Hospital between 1953 and 1980 and whose final diagnosis was an acute epidural hematoma were reviewed. Details about visual field testing were sought to ascertain the nature, frequency, and clinical importance of visual field abnormalities in patients with acute epidural he- matomas. Other information that might be of relevance, such as the level of consciousness and the site of the hematoma, was analyzed also. RESULTS In our series of 199 cases, 103 patients initially were alert enough to have their visual fields tested, but only 27 patients actually had their visual fields documented. Ninety-six patients were either unconscious or too uncooperative for such an examination. Six patients presented with a homonymous hem- ianopia. Their cases are described in detail. The series includes 154 cases of epidural hematoma that occurred before the availability of the computed tomographic (CT) scanner and 45 cases that presented after its installation in 1977. Twenty-four of the 45 patients treated during the CT scan era were conscious when first examined. Visual field testing was documented in 15 of these 24 patients, and ho- monymous hemianopia was detected in 4 patients (Patients 2, 4, 5, and 6). In the six patients with hemianopia, the epidural hematomas were located in the posterior fossa-occipital region in three patients, in the posterior temporoparietal region in two patients, and in the posterior parietal region in one patient. CASE REPORTS Case 1 This 50-year-old alcoholic woman was found unconscious and was taken to the hospital. Two years previously, she had 247 undergone the evacuation of a left subdural hematoma and after that procedure had been neurologically normal. On ar- rival at the hospital, her blood pressure was 170/100 and her pulse was 96 per minute. Her level of consciousness had improved. She was drowsy but roused to verbal stimuli and was thought to be globally aphasic. Her right pupil was slightly larger than the left, both pupillary responses were brisk. The fundi were normal. A right facial droop and dense right hemiparesis and right hemianesthesia were present. It was thought that she had a postictal Todd’s palsy. However, her deficit persisted, and 2 days later she was examined by a consulting neurologist, who found a dense right homonymous hemianopia in addition to the previously noted neurological deficits. Bilateral carotid angiograms showed a left cerebral surface space-occupying lesion. A large parietal extradural hematoma was removed at craniotomy. Postoperatively the patient made a complete recovery. Case 2 This 32-year-old alcoholic man was admitted with ataxia and bifrontal headache radiating into the neck. Examination revealed that he was drowsy and disoriented. A right homony- mous hemianopia and horizontal nystagmus on left lateral gaze were present. The left pupil was 3 mm and sluggish. The right pupil was 4 mm and brisk. There was no limb weakness. but the right lower limb was spastic and the right plantar response was extensor. A skull x-ray film showed a linear left occipital fracture. A radionuclide brain scan showed uptake in the left occipital region, which was thought to be due to the skull fracture. Arteriography was considered but was not done as the pyramidal signs and nystagmus disappeared over the next 2 days. The patient’s hemianopia persisted, and this caused his physicians to diagnose an intrinsic cerebral lesion. A lumbar puncture was performed and the nystagmus returned. Six days after admission, the patient suffered a respiratory arrest and was resuscitated. On examination, he was apneic. with flaccid, areflexic limbs, pinpoint nonreactive pupils, ab- sent corneal reflexes, and a gaze palsy to the right shown by oculocephalic testing. He did not respond to pain. Some 2 hours later, he was fully alert and carrying out commands. His eyes were conjugately deviated to the right and ocular bobbing was noted. An asymmetrical quadriparesis was present. Cere- bral angiograms showed a left-to-right shift with straightening of the left posterior cerebral artery and depression of the left posterior inferior cerebellar artery. The diagnosis was unclear wa 248 MATWIJECKY AND STEINBOK and pneumoencephalography was done. During ventriculog- raphy. the patient had a respiratory arrest and subsequently died. Autopsy revealed a large occipito-posterior fossa epidural hematoma weighing 170 g. Case 3 This 44-year-old right-handed man sustained a head injury in a motor vehicle accident and did not lose consciousness. He was admitted with a severe right temporal headache and pro- gressive drowsiness. On examination, he was found to have a dense left homonymous hemianopia, mild left upper limb weakness. and an extensor left plantar response. A CT scan showed a right posterior temporoparietal hema- toma. The clot was evacuated through a right parietotemporal craniotomy. When the patient was examined a few hours later, his hemianopia had resolved. Case 4 This 42-year-old woman sustained a brief loss of conscious- ness after she fell down some stairs and struck her head. On ad ion, she was alert but confused and disoriented. There was a subgaleal hematoma in the left occipitoparietal region. She had a dense right homonymous hemianopia and a mild right upper limb weakness. She gradually became drowsier. A CT scan showed a left posterior fossa-occipital extradural hematoma. A craniotomy was performed and the clot was removed. The visual field defect was completely resolved on the 2nd postoperative day. Case 5 This 33-year-old woman had a syncopal episode while jog- ging and struck her occiput on the sidewalk, She complained of severe, excruciating headache and was vomiting repeatedly. On examination, she had a right hemotympanum and an occipital subgaleal hematoma, She was awake, alert, and ori- ented. Her pupils were equal and reacted to light. A dense left homonymous hemianopia and right-sided hearing loss were found. Her left plantar response was extensor. The remainder of the neurological examination was normal. A CT scan showed a right occipito-posterior fossa epidural hematoma. En route to the operating room, her right pupil became dilated, but it returned to normal after the intravenous administration of mannitol. The hematoma was removed through a craniotomy. The patient was able to count fingers in her left visual field on the Ist postoperative day, but the fields were still constricted to a 5-mm white object. Her visual fields were normal by the 2nd postoperative day. Case 6 This 23-year-old man fell and struck his head on the pave- ment. sustaining a brief loss of consciousness. He was admitted to a peripheral hospital with severe headache, vomiting, and a right temporal bone skull fracture. He developed a left hemi- paresis and was transferred to Vancouver General Hospital. On admission. his blood pressure was 110/70 and his pulse was 76 per minute and regular. There was a subgaleal hema- toma in the right frontotemporal region. The patient was drowsy but easily alerted and was oriented. There were no pupillary signs, but a dense left homonymous hemianopia was found, along with a mild left hemiparesis. A flexor response was elicited by plantar stimulation of each foot. Proprioception was impaired on the left and there was sensory extinction on the left side to double simultaneous stimulation. A CT scan showed a right temporoparietal epidural hematoma, which was removed at craniotomy. Neurosurgery, Vol. 11, No. 2 The patient's visual field deficit was improved on the Ist postoperative day and was resolved on the 2nd postoperative day. His left hemiparesis resolved 3 days later. DISCUSSION Hemianopia is generally not recognized as a sign of a surface intracranial lesion, However. several authors have described homonymous hemianopia secondary to subdural hematoma (5, 6. 9). In several large series of patients with epidural hematomas. there is no mention of visual field defects (1, 2, 4). McKissock et al. noted a contralateral hemianopia in 2 of 125 patients with an epidural hematoma (7). Guthkelch reported | case of cor- tical blindness from a bilateral occipital epidural hematoma (3). In our series of 199 cases of traumatic epidural hematoma, 6 patients (3%) presented with homonymous hemianopia. How- ever, the visual fields were documented in only 27 patients. Therefore, the incidence of hemianopia in patients whose visual fields were documented was 6 of 27, or 22%. In most of the patients presented above (Cases 1, 3, 4, 5, and 6), the hemiano- pia was an important sign of the epidural hematoma and helped to direct further radiological investigations. However, in | patient (Case 2), the presence of a homonymous hemiano- pia caused the attending doctors to suspect an intracerebral lesion and delayed the recognition of the epidural hematoma. It is interesting that visual fields were not documented in 76 of 103 patients who were well enough to have had this test performed. In view of the significant incidence of homonymous hemianopia in this series (22%), more attention should be given to the examination of visual fields in patients with head injuries. Furthermore, the presence of hemianopia should not rule out an intracranial surface lesion. Several mechanisms have been proposed to account for the visual field defects in patients with intracranial surface hema- tomas. Pevehouse et al. (9) and Sunderland (10) have listed the possible mechanisms as (a) compression of the optic tract: (b) compression of the anterior choroidal artery affecting the blood supply to the optic tract, the lateral geniculate body, or the optic radiation in the internal capsule: (c) compression of the posterior cerebral artery or its calcarine branch; and (d ) direct compression of the surface of the cerebral hemisphere. Although Meyer (8). Maltby (5), and Sunderland (10) em- phasize compression of the posterior cerebral artery and its calcarine branches by transtentorial herniation in the patho- genesis of visual field defects, Pevehouse et al. (9) and Mc- Kissock et al. (7) have suggested that direct brain compression is more important in hemianopias related to surface lesions. Guthkelch’s one case of cortical blindness that resolved after the evacuation of a bilateral occipital extradural hematoma (3) also lends support to direct brain compression as an important pathogenic factor. In five of our six cases, pupillary signs of transtentorial herniation were absent at the time of the initial findings of the hemianopia. Furthermore. all six cases had posteriorly located epidural hematomas. These findings suggest that direct compression is the most common mechanism producing hem- ianopia in patients with surface blood clots. The prognosis for recovery of vision is excellent if the hematoma is removed. as occurred in five of five patients. In conclusion. we think that visual field testing should be performed routinely in patients with head injuries. and the finding of a homonymous hemianopia should not be taken as evidence of an intracerebral lesion because this physical sign is not uncommon in patients with an epidural hematoma. The finding of hemianopia in a patient suspected of having an August 1982 epidural hematoma suggests a more posterior location for the hematoma and may be of value in planning surgical treatment when CT scanning is not done preoperatively. ADDENDUM Since the preparation of this report. we have enountered another case of a posterior temporoparietal epidural hematoma that presented with a homonymous hemianopia. Received for publication, March 12, 1982: accepted. May 1, 1982. Reprint requests: Dr. C. Matwijecky, c/o Dr. P. Steinbok, Depart- ment of Surgery, C Floor, 700 West 10th Avenue, Vancouver, British Columbia V5Z 1LS, Canada. REFERENCES 1, Gallagher JP, Browder EJ: Extradural hematomas: Experience with 167 patients. J Neurosurg 29:1-12, 1968. Gurdjian ES. Webster JE: Extradural hemorrhage: A collective review of the literature and report of 30 cases of middle meningeal hemorrhage and 4 cases of dural sinus hemorrhage treated surgi- cally. Int Abstr Surg Suppl, Surg Gynecol Obstet 75:206-220, 1942. 3. Guthkelch AN: Extradural haemorrhage as a cause of cortical blindness. J Neurosurg 6:180-182, 1949. 4. Jamieson KG, Yelland JDM: Extradural hematoma: Report 167 cases. J Neurosurg 29:13-23, 1968. 5. Maltby GL: Visual field changes and subdural hematomas. Surg Gynecol Obstet 74:496-498, 1942 6. McKissock W, Richardson A, Bloom WH: Subdural hematoma review of 389 cases. Lancet 1(7139):1365-1369, 1960. 7. McKissock W, Taylor JC, Bloom WH, Till K: Extradural hema- toma. Lancet 2(7143):167-172, 1960. 8. Meyer A: Herniation of the brain. Arch Neurol Psychiatry 4:387- 400, 1920. 9, Pevehouse BC, Bloom HW, McKissock W: Ophthalmologic as- pects of diagnosis and localization of subdural hematoma: An analysis of 389 cases and review of the literature. Neurology (NY) 10:1037-1041, 1960. 10. Sunderland S: A tentorial notch and complications produced by herniations of the brain through that aperture. Br J Surg 45:422-— 438, 1958. nN COMMENTS The authors correctly assert that homonymous hemianopia is not generally well recognized as a complication of epidural, or subdural, hematoma. This is probably because of the com- mon failure to assess the visual fields in patients with head injury who are being rushed to the computed tomography suite to get “the facts.” Two authoritative references on the neuro-ophthalmological HEMIANOPIA 249 complications of epidural and subdural hematoma reveal gross pathology specimens that support the contention that the hem- lanopic visual field defect is attributable to the local compres- sive effects of the clot upon the brain cortex or the posterior cerebral artery blood supply to the posterior hemisphere (1. 2). I agree with the suggestion that routine assessment of visual fields should be integral to the examination of all patients with head injury. If a visual field anomaly is found, it need not be the result of an expanding mass within the brain parenchyma, and the failure to find such a mass should prompt a careful reappraisal for a missed surface lesion over the posterior hem- isphere. John A. McCrary III, M.D. Houston, Texas |. Lindenberg R, Walsh FB, Sacks JG: Neuropathology of Vision: An Atlas. Philadelphia, Lea and Febiger, 1973, p 472. 2. Walsh FB, Hoyt WF: Clinical Neuro-Ophthalmology. Baltimore. Williams and Wilkins, 1969, ed 3, vol 3, pp 2415, 2426. Although I agree that neurologists and neurosurgeons should be aware that a visual field defect does not rule out an extracerebral mass lesion (the delay in diagnosis in the two patients in their series before CT scanning is impressive), I am doubtful that this awareness will assist planning for surgical treatment because it is unlikely that craniotomy is ever per- formed today without a CT scan. Taking note of their statement, “In view of the significant incidence of homonymous hemianopia in this series (22%) ....” T understand the rationale with which this figure was produced. but I do not think it justifies even an implication that the incidence of visual field defects in epidural hematomas is likely to be 22% overall. As the authors themselves note, these hematomas were mostly posterior and in half of the cases they involved the tentorium itself. In all patients who were not comatose, the incidence of visual field defects was about 6%. more than that of McKissock et al. but within the 10% generally accepted for extracerebral hematomas. Infarction of the occipital lobe due to posterior cerebral artery occlusion during herniation is well documented as a cause of visual field defects. However, I agree with the authors that compression of the cortex is the most likely cause of the defects seen in the cases they describe and probably in a substantial number of patients with extracerebral hematomas and visual field defects. John Carter, M.D. Columbia, Missouri