Brainstem Stroke Research Articles

Nomogram for predicting swallowing recovery in patients after dysphagic stroke.

To date, variables predicting the recovery of dysphagia in patients after dysphagic stroke have not been well defined. However, despite the difficulties in predicting and understanding the dysphagia recovery trajectory, its significance for stroke care cannot be understated. This study aims to identify the factors for functional swallowing recovery and develop nomograms that predict dysphagia recovery after stroke. The demographic, neurological, and swallowing characteristics were compared between patients who recovered from dysphagia and those who did not. Then, the factors with P <.1 through comparison were enrolled in the multivariable logistic regression analysis to build a prediction model. A nomogram was also built to provide a quantitative tool. Discrimination, calibration, and clinical usefulness of the prediction model were assessed by using the C index, calibration plot, and decision curve analysis. Predictors in the early-phase (T7) prediction nomogram included age, Functional Oral Intake Scale (FOIS), National Institutes of Health Stroke Scale (NHISS), hemispheric stroke, and brainstem stroke on admission. In the middle phase (T14), predictors included age, FOIS, and NHISS on admission. In the late phase (T30), predictors included age, FOIS, NHISS, bilateral stroke, and body mass index on admission. The C index for the day 7, day 14, and day 30 prediction nomograms were 0.847 (95% CI, 0.804-0.884), 0.817 (95% CI, 0.772-0.857), and 0.786 (95% CI, 0.739-0.829). These novel nomograms predicting dysphagia recovery after ischemic stroke are discriminative and well calibrated and could be used to guide enteralnutrition decision making, rehabilitation plans, and individualized care.

A rare case of lateral pontine syndrome-Marie Foix syndrome

Lateral pontine syndrome or Marie Foix Alajouanine syndrome refers to the brainstem stroke syndrome involving lateral pons due to the infarction in the distribution of the anterior inferior cerebellar artery. It involves the lateral inferior part of the pons, middle cerebellar peduncle, and floccular region. Computed Tomography (CT) is less sensitive in brainstem infarction than Magnetic Resonance Imaging (MRI). With the introduction of MRI, it is possible to precisely locate the brainstem infarcts. We present a case with a typical clinical picture of brainstem stroke with radiological findings consistent with the lateral pontine syndrome. Proper correlation with the clinical picture and radiological findings is important for the diagnosis of brainstem infarctions.

Neural correlates of cricopharyngeal dysfunction after supratentorial stroke: A voxel-based lesion-symptom mapping with propensity score matched case-control.

Poststroke cricopharyngeal dysfunction has been reported to occur in 50% of brainstem strokes; however, cricopharyngeal dysfunction also occurs commonly in patients with supratentorial stroke. The hemispheric neuroanatomical location of this dysfunction has not been clearly identified. We aimed to analyze the relationship between cricopharyngeal dysfunction and supratentorial lesion location in poststroke patients through this retrospective case-control voxel-based lesion-symptom mapping study. Cricopharyngeal dysfunction was diagnosed when the residue after swallowing (pyriform sinus) accounted for more than 25% of volume of pyriform sinus. Medical records and the video fluoroscopic swallowing studies of first-ever stroke patients who were admitted to our hospital during acute to subacute phase from 2009 to 2019 were reviewed. After propensity score matching to reduce the likelihood of selection bias, 50 patients per group were included in the cricopharyngeal dysfunction and control groups. We used a P threshold of 0.01 corrected for multiple comparisons with permutation thresholding (5000 permutations). Dichotomized diagnosis of cricopharyngeal dysfunction and the magnitude of pyriform sinus were used as dependent variables. Analysis using the Liebermeister statistics indicated that lesions of the right lentiform nucleus were associated with the development of cricopharyngeal dysfunction. After adjustment for age and total lesion volume, which are known effectors for the development of dysphagia, statistically significant correlations were found between pyriform sinus and lesions of the right lentiform nucleus and anterior corona radiata beneath the right middle frontal gyrus. Thus, our study demonstrated for the first time that damages to the right lentiform nucleus, especially globus pallidus externa, and anterior corona radiata beneath the right middle frontal gyrus are associated with the development and severity of cricopharyngeal dysfunction.

Abstract P717: Central Midbrain Strokes Are More Liable for Respiratory Failure Then Other Brain Stem Strokes

Background: Respiratory centers are known to be present in the central medulla oblongata and pons. There are multiple complex respiratory networks involving these centers. The midbrain periaqueductal grey is believed to act as a regulator of the respiratory function. The effect of brain-stem strokes on respiration remains understudied. There is a lack of clear understanding of the anatomical influence of such strokes on respiration. We attempted to identify brain-stem locations with the highest liability for respiratory failure in case of stroke. Methods: We included all ischemic and hemorrhagic brain-stem strokes from our stroke-registry between 2016 and 2018 then performed univariate/multivariate regression-analyses on variables that might predict respiratory failure and the need for intubation. The brain stem was divided into nine locations (right lateral, central, left lateral in each of the midbrain, pons, and medulla oblongata). Results: Out of 128 brain-stem strokes of different sizes and etiologies, central midbrain strokes were the only significant and independent affected location associated with respiratory failure and endotracheal intubation (coefficient= 0.1256, 95%-CI= 0.0175, 0.2338, p= 0.023). R-squared was equal to 15% when only central midbrain strokes stayed in the model. Conclusions: While one might assume that central medullary and pontine strokes have the most impact on respiration; our results show that central midbrain is the most impactful, accounting for about 15% of respiratory instability associated with brain-stem strokes. This can be explained by the adaptive nature of respiratory circuits within the medulla and pons. Central periaqueductal grey within the midbrain controls the rate and depth of respiration and might not have the same flexibility present elsewhere.

Abstract P382: Assessing the Impact of Damage to Brainstem Tracts on Motor Ability in Ischemic Stroke Patients

The extent of disruption to the corticospinal tract has been associated with motor impairment and worse recovery, but the role of other motor and sensorimotor tracts in impairment following brainstem stroke is poorly understood. Additionally, the impact of brainstem strokes on connectivity to distal cortical regions innervated by these tracts has not been well established. In 23 first-episode unilateral brainstem ischemic stroke patients, the extent of disruption to brainstem pathways was determined using the Dice score between binarized lesion masks obtained from T1 images and each of 23 brainstem tracts defined using a recently published high-resolution atlas of the brainstem based on connectome imaging data. Motor impairment was determined using the Fugl-Meyer assessment. White matter disruption due to the stroke was estimated using the Network Modification Tool 2.0 (NeMo) tool, which calculates the proportion of fiber tracts disrupted by the lesion in each voxel. Principal components analysis (PCA) was used to determine disruption patterns across patients. PCA of Dice scores identified one component which explained 65% of the variance, which corresponded to overlap with the corticospinal (CST), frontopontine (FPT), and parieto-occipito-temporo-pontine (POTPT) tracts, as well as the middle cerebellar peduncle. Component scores significantly correlated with Fugl-Meyer scores (corr. = -0.82, p &lt; 0.01). PCA of cortical disconnectivity revealed one principal component which explained 75% of the variance, which correlated most highly with cortical regions connected to the CST, FPT, and the POTPT and whose scores also significantly correlated with Fugl-Meyer scores (corr. = -0.65, p &lt; 0.01).

When assistive eye tracking fails: Communicating with a brainstem-stroke patient through the pupillary accommodative response – A case study

PurposePoor control of eye movement and coordination may impair the use of eye-trackers for communication in patients affected by severe motor diseases. Recently, the “voluntary” pupil accommodative response (PAR) was suggested as a possible alternative to traditional assistive technology. Aim of this study is to provide a proof of concept of this methodology in a clinical setting. Materials and methodsA low-cost communication system was implemented, which detects the accommodative pupillary constrictions in real time and generates trigger events to drive a commercial scanning-selection interface. As a first implementation, a simple binary yes/no selection interface was designed to be tested with a brainstem stroke patient, unable to use standard communicators based on eye tracking. The patient was instructed to operate the intended selection by switching the focus of attention from a far to a near target, and was then presented with 10 questions with obvious answer. ResultsThe patient easily understood how to perform the accommodative task. The pupillary constrictions were marked and clearly detectable in spite of the disturbing action of persistent nystagmus. On the first presentation of the device, the patient managed to correctly answer 8 out of 10 questions. ConclusionsThe present results provide a proof-of-concept for PAR-based communication in a clinical setting and support its usefulness with patients who, due to impaired control of eye movements, may be unable to use tracking-based devices.

Clinical Applications of Brain Computer Interfaces : A Boon for the Future

Brain-computer interface (BCI) is the technology act as a interface between the brain and an external device. It converts the signals emitted by the CNS of the brain to artificial output to be understood by the computer. This technology will be most useful to the severely disabled individual or people suffering from amyotrophic lateral sclerosis, brainstem stroke, or any spinal cord injury and thus are impaired of tier ability to communicate and physical functioning. With the fast-paced development and interest of various top-notch companies in this arena due to its positive future Efforts have begun recently to provide safe and secure BCI systems to severely disabled individuals to make their lives easier. In this paper, we will know about BCI , its basic functioning. We will also discuss its clinical application, p300 speller, and its potential for future. Lastly expectation from the future.

Typical somatomotor physiology of the hand is preserved in a patient with an amputated arm: An ECoG case study

Electrophysiological signals in the human motor system may change in different ways after deafferentation, with some studies emphasizing reorganization while others propose retained physiology. Understanding whether motor electrophysiology is retained over longer periods of time can be invaluable for patients with paralysis (e.g. ALS or brainstem stroke) when signals from sensorimotor areas may be used for communication or control over neural prosthetic devices. In addition, a maintained electrophysiology can potentially benefit the treatment of phantom limb pains through prolonged use of these signals in a brain-machine interface (BCI).Here, we were presented with the unique opportunity to investigate the physiology of the sensorimotor cortex in a patient with an amputated arm using electrocorticographic (ECoG) measurements. While implanted with an ECoG grid for clinical evaluation of electrical stimulation for phantom limb pain, the patient performed attempted finger movements with the contralateral (lost) hand and executed finger movements with the ipsilateral (healthy) hand.The electrophysiology of the sensorimotor cortex contralateral to the amputated hand remained very similar to that of hand movement in healthy people, with a spatially focused increase of high-frequency band (65–175 Hz; HFB) power over the hand region and a distributed decrease in low-frequency band (15–28 Hz; LFB) power. The representation of the three different fingers (thumb, index and little) remained intact and HFB patterns could be decoded using support vector learning at single-trial classification accuracies of >90%, based on the first 1–3 s of the HFB response. These results indicate that hand representations are largely retained in the motor cortex. The intact physiological response of the amputated hand, the high distinguishability of the fingers and fast temporal peak are encouraging for neural prosthetic devices that target the sensorimotor cortex.

Oculopalatal tremor and hypertrophic olivary degeneration

Background: Oculopalatal tremor is characterized by synchronous occurrence of palatal tremor and pendular nystagmus. Oculopalatal tremor (OPT) follows a destructive lesion in the Guillain- Mollaret triangle (GMT) and is associated with hypertrophic inferior olive degeneration (HOD). Case presentation: We report a case of 45 yearsold patient with palatal tremor and pendular nystagmus without ear click. He has had a brainstem stroke 11 months back. He had right side pyramidal drift, left side lower motor neuron type facial weakness and left abducens nerve palsy and mild gait ataxia in addition to oculopalatal tremor. MRI showed hypertrophic inferior olive degeneration. Conclusion: Ear click is an infrequent finding in symptomatic palatal tremor. Additional clinical features may be present in OPT depending on topography of initial structural lesion. There is a wide variety of time intervals between the occurrence of anatomical lesions and recognition of palatal tremor.

Neurogenic Dysphagia: Systematic Review and Proposal of a Classification System.

Introduction and validation of a phenotypic classification of neurogenic dysphagia based on flexible endoscopic evaluation of swallowing (FEES). A systematic literature review was conducted, searching MEDLINE from inception to May 2020 for FEES findings in neurologic diseases of interest. Based on a retrospective analysis of FEES videos in neurologic diseases and considering the results from the review, a classification of neurogenic dysphagia was developed distinguishing different phenotypes. The classification was validated using 1,012 randomly selected FEES videos of patients with various neurologic disorders. Chi-square tests were used to compare the distribution of dysphagia phenotypes between the underlying neurologic disorders. A total of 159 articles were identified, of which 59 were included in the qualitative synthesis. Seven dysphagia phenotypes were identified: (1) "premature bolus spillage" and (2) "delayed swallowing reflex" occurred mainly in stroke, (3) "predominance of residue in the valleculae" was most common in Parkinson disease, (4) "predominance of residue in the piriform sinus" occurred only in myositis, motoneuron disease, and brainstem stroke, (5) "pharyngolaryngeal movement disorder" was found in atypical Parkinsonian syndromes and stroke, (6) "fatigable swallowing weakness" was common in myasthenia gravis, and (7) "complex disorder" with a heterogeneous dysphagia pattern was the leading mechanism in amyotrophic lateral sclerosis. The interrater reliability showed a strong agreement (kappa = 0.84). Neurogenic dysphagia is not a symptom, but a multietiologic syndrome with different phenotypic patterns depending on the underlying disease. Dysphagia phenotypes can facilitate differential diagnosis in patients with dysphagia of unclear etiology.

Treatment of pseudobulbar affect (PBA) in a patient with a history of traumatic brain injury, partial brain resection, and brainstem stroke: a case report

BackgroundPseudobulbar affect is a very distressing and underdiagnosed neuropsychiatric disorder that causes contextually inappropriate episodes of laughing and crying and general emotional incontinence. Although many proposed etiologies exist, the most widely accepted theory espouses the disruption of a corticopontine–cerebellar circuit that governs the modulation of emotional response. Pseudobulbar affect is commonly diagnosed secondary to primary neurological disorders such as amyotrophic lateral sclerosis, multiple sclerosis, and traumatic brain injury. Traditional pharmacological treatment of pseudobulbar affect is largely comprised of antidepressant therapy, including tricyclic antidepressants such as amitriptyline and selective serotonin reuptake inhibitors such as fluvoxamine. However, neither of these medication classes has been studied for the treatment of pseudobulbar affect in controlled trials, and their utility remains questionable.Case presentationWe describe a case of a 62-year-old Caucasian man with history of traumatic brain injury, ischemic brainstem stroke, and depression who developed intractable pseudobulbar affect. This patient’s intensely distressing symptoms were not alleviated by amitriptyline. However, after being placed on fixed-dose 20 mg/10 mg dextromethorphan/quinidine (Nuedexta), our patient experienced complete resolution of his symptoms. He has experienced no deleterious side effects.ConclusionsThis case provides anecdotal evidence for the efficacy of dextromethorphan/quinidine in the treatment of pseudobulbar affect with remarkably swift and complete cessation of symptoms. As a secondary point, it is worth noting that our patient had experienced two devastating neurological traumas, both in anatomical areas that have been implicated in the corticopontine–cerebellar circuit thought to be responsible for pseudobulbar affect. However, only the second trauma, an acute left pontine infarction, produced symptoms of emotional disinhibition. The authors hope that reporting this case will provide both context for physicians managing this condition and hope for patients with this socially and psychiatrically damaging disease.

Intracranial Aneurysms: Does Size Really Matter?

Abstract BACKGROUND To stratify the risk of aneurysmal rupture, size remains the primary criterion as proposed by 2 ISUIA (International Study of Unruptured Intracranial Aneurysms) trials that recommend the observation of ≤7 mm aneurysms because of their low propensity to rupture. These recommendations are controversial, given the severe outcomes following hemorrhage. OBJECTIVE To retrospectively assess whether size correlates with aneurysmal rupture, and to analyze outcomes of endovascular and microsurgical procedures. METHODS All aneurysms presenting between 2016 and 2019 were reviewed. Age, gender, comorbidities, ruptured status, modes of treatment, and their outcomes were noted. Regression analysis was performed to correlate size with rupture. The type of intervention was based on clinical and aneurysmal characteristics and the likelihood of obliteration. RESULTS Of the 668 aneurysms, 116 had ruptured. In size groups 0 to 6 mm, 7 to 12 mm, 13 to 18 mm, 18 to 24 mm, and giant, the ruptured aneurysms were 60.3%, 30.2%, 1.7%, 0.9%, and 6.9%, respectively. The majority (423) were managed endovascularly, of which 84 (19.9%) were ruptured. A total of 97 (14.5%) underwent clipping, of which 31 were ruptured. The mean modified Rankin Scale for both treated groups was zero. One death in each group and a single brainstem stroke in the coiled group were noted. A total of 139 were followed, of which none had ruptured. The mean modified Raymond-Roy classification for endovascular therapy was 1.1, with 380 completely occluded; 15 had a score of 2, and 28 of score 3. CONCLUSION The ruptured aneurysms were mostly smaller than reported previously. Surgical clipping was at equipoise with endovascular treatments. Further studies to correlate aneurysmal characteristics to the risk of rupture are needed.

Mind Controlled Wheelchair

This project describes a non-invasive Brain-actuated Wheelchair which acquires the signals from various electrodes placed according to the International 10-20 electrode setup for Electroencephalograph (EEG). The brain signals are processed to determine the direction of the movement of the wheelchair. It has been shown that the patient is able to achieve EEG controlled cursor, limb movement, a prosthesis control and even has successfully communicated by means of a Brain-computer Interface (BCI). The signals are detected to observe the path to estimate the movement of the wheelchair. The work presented here mainly uses simple unipolar electrode to capture EEG from the forehead to build a control for electric wheelchairs through Bluetooth for paralyzed patients. Attention and eye-blinking signals can be collected as the control signals through a Bluetooth interface and an electrical interface is used for the electric wheelchair. Brain-computer interfaces (BCI) do not rely on muscular activity and can therefore provide communication and control for people with devastating neuromuscular disorders such as the amyotrophic lateral sclerosis, brainstem stroke, cerebral palsy and spinal cord injury.

Global multisensory reorganization after vestibular brain stem stroke.

ObjectivePatients with acute central vestibular syndrome suffer from vertigo, spontaneous nystagmus, postural instability with lateral falls, and tilts of visual vertical. Usually, these symptoms compensate within months. The mechanisms of compensation in vestibular infarcts are yet unclear. This study focused on structural changes in gray and white matter volume that accompany clinical compensation.MethodsWe studied patients with acute unilateral brain stem infarcts prospectively over 6 months. Structural changes were compared between the acute phase and follow‐up with a group of healthy controls using voxel‐based morphometry.ResultsRestitution of vestibular function following brain stem infarcts was accompanied by downstream structural changes in multisensory cortical areas. The changes depended on the location of the infarct along the vestibular pathways in patients with pathological tilts of the SVV and on the quality of the vestibular percept (rotatory vs graviceptive) in patients with pontomedullary infarcts. Patients with pontomedullary infarcts with vertigo or spontaneous nystagmus showed volumetric increases in vestibular parietal opercular multisensory and (retro‐) insular areas with right‐sided preference. Compensation of graviceptive deficits was accompanied by adaptive changes in multiple multisensory vestibular areas in both hemispheres in lower brain stem infarcts and by additional changes in the motor system in upper brain stem infarcts.InterpretationThis study demonstrates multisensory neuroplasticity in both hemispheres along with the clinical compensation of vestibular deficits following unilateral brain stem infarcts. The data further solidify the concept of a right‐hemispheric specialization for core vestibular processing. The identification of cortical structures involved in central compensation could serve as a platform to launch novel rehabilitative treatments such as transcranial stimulations.

Disrupted brain functional network topology in unilateral acute brainstem ischemic stroke.

This study aimed to investigate the topological properties of brain functional connectome in unilateral acute brainstem ischemic stroke using graph theory. Fifty-three acute brainstem ischemic stroke patients, consisted of 27 left-sided and 26 right-sided brainstem stroke patients, and 20 age, gender, and education-matched healthy controls (HCs) were recruited to undergo a resting-state functional magnetic resonance imaging (rs-fMRI) scan in this study. Graph theory analyses were then used to examine the group-specific topological properties of the functional connectomes seperately. The unilateral acute brainstem stroke patients and HCs all exhibited "small-world" brain network topology. The functional connectome of the left brainstem stroke patients showed significant differences in all topological properties while the right brainstem stroke patients showed a significant increase in clustering coefficient Cp (p < 0.001) and local efficiency Elocal (p < 0.001), and a significantly decrease in normalized clustering coefficient γ (p < 0.001) and global efficiency Eglobal (p< 0.001), suggesting both a shift toward regular networks. At the nodal level, abnormal nodal centralities were mainly observed in the defaut mode network, subcortical network, frontal and occipital lobe. The findings of disrupted topological properties of functional brain networks may help better understanding the disease characterization and innovation in management for acute brainstem ischemic stroke patients.

A Generalized Model to Estimate Reaction Time Corresponding to Visual Stimulus Using Single-Trial EEG.

The estimation of the visual stimulus-based reaction time (RT) using subtle and complex information from the brain signals is still a challenge, as the behavioral response during perceptual decision making varies inordinately across trials. Several investigations have tried to formulate the estimation based on electroencephalogram (EEG) signals. However, these studies are subject-specific and limited to regression-based analysis. In this paper, for the first time to our knowledge, a generalized model is introduced to estimate RT using single-trial EEG features for a simple visual reaction task, considering both regression and classification-based approaches. With the regression-based approach, we could predict RT with a root mean square error of 111.2 ms and a correlation coefficient of 0.74. A binary and a 3-class classifier model were trained, based on the magnitude of RT, for the classification approach. Accuracy of 79% and 72% were achieved for the binary and the 3-class classification, respectively. Limiting our study to only high and low RT groups, the model classified the two groups with an accuracy of 95%. Relevant EEG channels were evaluated to localize the part of the brain significantly responsible for RT estimation, followed by the isolation of important features.Clinical relevance- Electroencephalogram (EEG) signals can be used in Brain-computer interfaces (BCIs), enabling people with neuromuscular disorders like brainstem stroke, amyotrophic lateral sclerosis, and spinal cord injury to communicate with assistive devices. However, advancements regarding EEG signal analysis and interpretation are far from adequate, and this study is a step forward.

Predicting Surgical Intervention in Cerebellar Stroke: A Quantitative Retrospective Analysis

Debate still exists regarding whether preventive surgical decompression should be offered to high-risk patients experiencing cerebellar stroke. This study aimed to predict neurologic decline based on risk factors, volumetric analysis, and imaging characteristics. This retrospective cohort study comprised patients ≥18 years who presented with acute cerebellar ischemic stroke (CIS) between January 2011 and December 2016. Diagnostic imaging was used to calculate metrics based on individual stroke, cerebellar, and posterior fossa volumes. Head computed tomography scans on presentation and day of peak swelling were used to tabulate a CIS score. The study included 86 patients; most were male and African American. Posterior inferior communicating artery stroke was most common (50%). On initial presentation imaging, 18.6% had documented hydrocephalus, 20.9% had brainstem compression, 22.1% had brainstem stroke, and 39.5% had stroke in another vascular territory. Cardioembolic stroke was the most common etiology, followed by cryptogenic stroke. Overall, patients who underwent surgical intervention had larger stroke volumes on presentation. Patients undergoing surgical intervention also experienced faster cerebellar swelling compared with patients without intervention. Total CIS scores were statistically significant and remained significant on the peak day of swelling. CIS score was independently associated with neurosurgical intervention; patients in this group with delayed interventions (median CIS score, 6; range, 4-8) later deteriorated and required emergent surgical decompression. Eleven patients without intervention had CIS score >6; 4 patients died of stroke complications. Volumetric studies and CIS score are objective measures that may help predict decline on imaging before clinical deterioration.

Endovascular sacrifice of the proximal posterior inferior cerebellar artery for treatment of ruptured intracranial aneurysms

BackgroundRuptured aneurysms of the intracranial vertebral artery (VA) or posterior inferior cerebellar artery (PICA) are challenging to treat as they are often dissecting aneurysms necessitating direct sacrifice of the diseased...

Brainstem ischemic syndrome in juvenile NF2.

ObjectiveA new case of brainstem ischemic necrosis in a young woman with de novo neurofibromatosis type 2 (NF2) is reported, and given notable similarities to 7 prior cases of brainstem stroke in the literature, features defining a possible syndrome were sought.MethodsCase review including detailed clinical assessment, neuroimaging analysis, genetic testing, and brain biopsy, followed by a multicase analysis.ResultsBrainstem ischemia in juvenile NF2 typically occurs in teenagers without previously known NF2 as an acute, monophasic presentation with restricted diffusion in the midbrain or pons following a recent hypoperfusion event, normal vascular imaging, obvious intracranial imaging features of NF2, typical inactivating NF2 alterations, biopsy showing necrosis without small vessel pathology, and subsequent aggressive NF2 lesion progression.ConclusionsBrainstem ischemia in juvenile NF2 is a rare syndrome of unclear etiology, possibly reflecting an unknown underlying vascular abnormality; a digenic effect is not excluded.

Carotid-Cavernous Fistula Presenting With Bilateral Abducens Palsy.

Carotid-Cavernous Fistula Presenting With Bilateral Abducens Palsy.