Diffusion-weighted imaging and status epilepticus during vagus nerve stimulation

Epilepsy & Behavior Epilepsy & Behavior 5 (2004) 411–415 www.elsevier.com/locate/yebeh

Case Report

Diffusion-weighted imaging and status epilepticus during vagus nerve stimulation W.O. Tatum IV,a,* A. Malek,a M. Recio,a J. Orlowski,b and R. Murtaghc a

c

Department of Neurology, University of South Florida, Tampa, FL, USA b University Community Hospital, Tampa, FL, USA Department of Neurosurgery and Radiology, University of South Florida, Tampa, FL, USA Received 3 December 2003; revised 18 February 2004; accepted 20 February 2004 Available online 19 March 2004

Abstract Purpose. Transient abnormalities have been reported on diffusion-weighted imaging (DWI) during status epilepticus. Vagus nerve stimulation (VNS) is a therapy for epilepsy that has previously demonstrated alteration in regional cerebral blood flow on functional neuroimaging. We describe the peri-ictal DWI abnormalities in a patient with status epilepticus. Methods. A 21-year-old woman with pharmacoresistant localization-related epilepsy was treated with VNS and underwent brain magnetic resonance imaging (MRI) with DWI for clinical purposes. Results. Transient and reversible hyperintense signal abnormalities were noted on DWI at the site of seizure onset, in addition to the thalamus and midbrain bilaterally. A concomitant decrease in the apparent diffusion coefficient mimicked ischemia, yet complete clinical, and electrographic resolution occurred following successful termination of status. Conclusions. High-energy brain MRI sequences using DWI were safely performed in our epilepsy patient with a vagus nerve stimulator who experienced status epilepticus. This case highlights the bilateral and robust involvement of subcortical structures present immediately following status epilepticus. Additionally, bilateral abnormalities in the thalamus and midbrain in addition to the region of seizure origin, were observed in our patient implanted with a vagus nerve stimulator. Modulation of regional cerebral blood flow is one potential mechanism of action for VNS in humans; therefore, these regions of involvement could reflect the effects of status epilepticus, activation or facilitation by VNS, or both. Ó 2004 Elsevier Inc. All rights reserved. Keywords: Status epilepticus; Seizures; Magnetic resonance imaging; Diffusion-weighted imaging; Vagus nerve stimulation

1. Introduction Magnetic resonance imaging (MRI) is an essential neuroimaging technique in assessing the underlying brain anatomy in patients with epilepsy. Transient focal abnormalities on MRI have been described in patients experiencing status epilepticus [1,2]. Diffusion-weighted imaging (DWI) is a form of echo planar imaging that is sensitive to detecting the molecular motion of water and highlighting differ-

*

Corresponding author. Fax: 1-813-971-6951. E-mail address: [email protected] (W.O. Tatum IV). URL: http://www.WOTatumIV.com. 1525-5050/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.yebeh.2004.02.009

ences in brain tissues [2]. The most common site for reversible signal abnormalities on DWI in patients with partial status epilepticus is the site of seizure origin, though other locations have been identified [2– 4]. Vagus nerve stimulation (VNS) is an effective adjunctive therapy for patients with pharmacoresistant localization-related epilepsy, with more than 22,000 patients implanted worldwide [5]. The precise mechanism of action for VNS is unknown though multiple mechanisms probably are operational. Anatomic–physiologic abnormalities, as well as alterations in regional cerebral blood flow occur [6]. We report a patient treated with VNS for refractory epilepsy who safely underwent peri-ictal DWI after status epilepticus.

412

Case Report / Epilepsy & Behavior 5 (2004) 411–415

2. Case report A 24-year-old Phillipino-American woman was in good health until 11 September 2001, when she developed refractory generalized convulsive status epilepticus due to encephalitis. Following a resistant and protracted 1-month course necessitating drug-induced coma with pentobarbital, she survived, though with severe memory deficits and pharmacoresistant epilepsy. Drug resistance or allergy to eight different antiepileptic drugs (AEDs) occurred with ongoing, frequent near-daily, complex partial and secondarily generalized seizures singly and in clusters despite phenytoin and phenobarbital. She was evaluated for epilepsy surgery at a tertiary care epilepsy center, with bilateral seizure onset and multifocal interictal epileptiform discharges captured on scalp EEG. MRI demonstrated severe bilateral hippocampal formation atrophy, and neuropsychological testing revealed severe memory deficits. She was felt to be a limited candidate for resective surgery and was implanted with the vagus nerve stimulator.

Parameters were titrated to 1.25 mA, 25 Hz, 250 ls at 30 s on and 5 min off. Almost 1 year after initial onset, she developed serial complex partial seizures that prompted hospitalization. Video-EEG demonstrated electroclinical complex partial status epilepticus of right temporal lobe origin (see Fig. 1), which was terminated with intravenous lorazepam and valproate in combination with high-dose phenytoin and phenobarbital. CT and lumbar puncture were unremarkable. She became comatose with hyperammonemia following valproate administration, with diffuse slowing of the background and typical triphasic waves on EEG. Brain MRI was performed within 24 h of electroclinical seizure termination. The vagus nerve stimulator was deactivated immediately prior to all imaging procedures. MRI using DWI and apparent diffusion coefficient maps suggested ischemia, and evaluation with brain MRA, carotid dopplers, echocardiogram, cardiac telemetry, and laboratory studies was normal. Hyperammonemia resolved after discontinuation of valproate, and she recovered to baseline cognitive and physical function with improved

Fig. 1. Ictal electroencephalogram demonstrating right temporal seizure onset.

Case Report / Epilepsy & Behavior 5 (2004) 411–415

413

Fig. 2. (A–C) Transverse axial DWI immediately following status epilepticus in a patient implanted with a vagus nerve stimulator, demonstrating (A) ipsilateral right temporal, (B) bilateral posterior-lateral thalamic, and (C) bilateral diffuse midbrain hyperintensity. (D) Representative T2 coronal MRI scan 6 weeks after status epilepticus with resolution of the previous abnormalities. Note bilateral hippocampal formation atrophy, which is unchanged from baseline.

stabilization of baseline seizure frequency. Her acute and follow-up brain MRI scans are depicted in Fig. 2.

3. Neuroimaging A baseline brain MRI scan was obtained 18 months prior to admission, in the immediate peri-ictal period following status, and 6 weeks following recovery. Im-

ages were obtained on a GE Signa (Madison, WI, USA). An established epilepsy protocol was used, and sagittal, axial, and coronal images were obtained. The examination was performed obtaining sagittal T2, axial T2, FLAIR and diffusion, apparent diffusion coefficient maps, and coronal T2 and FLAIR images through the brain. Two signals with repetition times of 10,000 and 8000 ms and echo times of 107 and 83 ms were used for axial and coronal images. A 40  20-cm field of view and

414

Case Report / Epilepsy & Behavior 5 (2004) 411–415

256  256 matrix were used at 5 mm skip 0. Flair images were used repetition times of 8000 and 8802 ms and echo times of 157 and 127 ms at 4 mm skip 0 for coronal images.

4. Discussion Status epilepticus is a common life-threatening neurologic disorder [7], that presented as serial complex partial seizures in our patient implanted with a vagus nerve stimulator. During status, increased cerebral blood flow is known to occur [3] and involves a complex interaction of excitatory and inhibitory mechanisms between cortical and subcortical generators [8]. Early reports of postictal T2 changes emphasized subcortical white matter changes due to vasogenic edema [1], whereas later reports described cytotoxic edema affecting the cortical gray matter [8]. Magnetic resonance imaging may provide information central to the etiology for patients who experience complex partial seizures [8,9], and DWI abnormalities reflecting the distribution of water diffusibility may help elucidate the pathophysiology. Vagus nerve stimulation is used for patients with refractory partial-onset seizures and is composed of a subcutaneously implanted infraclavicular metallic generator and lead projected to helical electrodes placed on the left vagus nerve in the neck [10]. Clinical experience has supported the safety of performing MRI in patients with VNS [11]. Use of a send-and-receive head coil is recommended by the manufacturer [5] for patients with VNS who undergo brain MRI; though high-energy sequences including functional MRI and DWI are not approved, they have been safely performed [12]. DWI abnormalities on MRI initially felt to represent irreversibly lost brain tissue [1,2] may be transient and reverse after seizures (3). Transient physiologic changes on MRI associated with recurrent or prolonged seizures have previously been identified at the site of seizure origin [1–4]. Subcortical involvement has been reported in the thalamus [13], cerebellum [14], splenium [15], and white matter [16] and is felt to reflect transient cytotoxic edema induced by the seizures [3]. Abnormalities have been noted in the hippocampus [17] that may be bilateral [18]. Single case reports have demonstrated bilateral thalamic involvement [19], as in our patient, as well as basal ganglia involvement [20], and claustrum abnormalities have been noted on autopsy [21]. Postictal DWI for the localization of focal epileptic areas in localization-related epilepsy has a low yield when a single seizure occurs [22]. We found bilateral thalamic and ipsilateral temporal hyperintense abnormalities in addition to diffuse midbrain involvement on DWI in our patient with status epilepticus of right temporal origin. Peri-ictal DWI ab-

normalities have been shown to persist following the termination of SE [16]. Polysynaptic mechanisms are likely operational in the spread of epileptiform activity to adjacent and subcortical regions in the brain [8,23,24]. Bilateral involvement in two or more vascular territories in a patient with prolonged seizures usually permits differentiation of the transient physiologic effects of SE from ischemia [3]. Concomitant perfusion changes were addressed for clinical purposes only. Abnormalities in DWI in patients with status epilepticus may mimic acute cerebral ischemia [1,3], and probably reflect the transient cytotoxic and vasogenic edema induced by seizures [16]. We suspect that the hyperammonemia probably had a limited role in the abnormalities seen on DWI. The comatose state that was noted clinically was probably multifactorial and was caused by status, the effects of high doses of antiepileptic drugs, as well as the hyperammonemia. Subcortical hypoperfusion is more likely with metabolic aberrations [25], and valproate-induced hyperammonemic encephalopathy has been associated with bilateral lesions in the globus pallidus and cerebellar white matter [26], which were not observed in our patient. Excitatory neurotransmitters probably mediate the postsynaptic neuronal necrosis in SE [8,16,23]. In our patient, severe bilateral hippocampal formation atrophy and memory loss followed refractory generalized convulsive status epilepticus but did not worsen after partial status recurred. Both VNS and SE can alter cerebral blood flow [3,6,23]. Pressure-dependent cerebral blood flow may have accounted for the changes that occurred at the site of seizure origin and bilaterally in the thalamus and midbrain level due to status, activation or facilitation by VNS, or both. Multiple mechanisms probably underlie the antiseizure effects of VNS [5,6,24]. Brainstem polysynaptic projections from the vagus nerve to the nucleus tractus solitarius in the medulla project from a wide variety of sources including the thalamus and dorsal tegmental nucleus of the midbrain [6]. Vagus nerve afferents have robust projections to the thalamus, and H2 15 O PET studies with both high- and low-level VNS demonstrate an increase in thalamic blood flow [6]. Additionally, VNS activation of the thalami has been noted by functional MRI in seizure patients [12,27]. The amygdala and hippocampus are also common sources of partial-onset seizures [9], as well as a target for VNS [6,10], though right temporal DWI abnormalities probably resulted from status itself. Clinically, a sustained effect from VNS may outlast the stimulus following deactivation [28], and it is possible that the VNS may have augmented or influenced the subcortical distribution of MRI signal changes observed during status epilepticus in our patient given the similar regions of involvement.

Case Report / Epilepsy & Behavior 5 (2004) 411–415

In conclusion, peri-ictal DWI was feasible and safely performed in our patient with status epilepticus and refractory localization-related epilepsy treated with VNS. Transient signal abnormalities occurred bilaterally in subcortical structures of the thalamus as well as the midbrain, in addition to the site of seizure onset. These regions are important areas for initiation and termination of seizures and the robust reversible abnormalities on DWI. We feel the changes on DWI are important as they mark sites of potentially irreversible effects of status epilepticus. Further peri-ictal DWI of other patients implanted with vagus nerve stimulators who experience status epilepticus would be helpful in clarifying whether VNS contributed to the robust reversible changes that occurred.

Acknowledgments The authors acknowledge and thank Dr. Ron Lesser for his help with the presurgical evaluation and implantation of the vagus nerve stimulator, as well as for his time reviewing and editing this manuscript. We also thank Dr. Carlos Martinez for his expertise and help with additional neuroimaging.

References [1] Kramer RE, Luders H, Lesser RP, et al. Transient focal abnormalities of neuroimaging studies during focal status epilepticus. Epilepsia 1987;28:528–32. [2] Kim JA, Chung JI, Yoon PH, et al. Transient MR signal changes in patients with generalized tonic–clonic seizures or status epilepticus: periictal diffusion-weighted imaging. AJNR 2001;22:1149–60. [3] Landsberg MG, OÕBrien MW, Norbash AM, Moseley ME, Morrell M, Albers GW. MRI abnormalities associated with partial status epilepticus. Neurology 1999;52:1021–7. [4] Hattori H, Matsuoka O, Ishida H, Hisatsune S, Yamano T. Magnetic resonance imaging in occipital lobe epilepsy with frequent seizures. Pediatr Neurol 2003;28:216–8. [5] Cyberonics, Data on file. [6] Henry TR. Therapeutic mechanisms of vagus nerve stimulation. Neurology 2002;59(Suppl. 4):S3–14. [7] Tatum IV WO, French JA, Benbadis SR, Kaplan PW. The diagnosis and treatment of status epilepticus. Epilepsy Behav 2001;2:311–7. [8] Chu K, Kang D-W, Kim J-Y, Chang K-H, Lee SK. Diffusionweighted magnetic resonance imaging in nonconvulsive status epilepticus. Arch Neurol 2001;58:993–8.

415

[9] Engel Jr J, Wiebe S, French J, et al. Practice parameter: temporal lobe and localized neocortical resections for epilepsy. Epilepsia 2003;44:741–51. [10] Schachter SC, Wheless JW. The evolving place of vagus nerve stimulation therapy. Neurology 2002;59(Suppl. 4):S1–2. [11] Benbadis SR, Nyhenhuis J, Tatum W.O. IV, Murtagh FR, Gieron M, Vale FL. MRI of the brain is safe in patients implanted with vagus nerve stimulation. Seizure 2001;10:512–5. [12] Narayanan JT, Watts R, Haddad N, Labar D, Li M, Fillippi CG. Cerebral activation during vagus nerve stimulation: a functional MR study. Epilepsia 2002;43:1505–14. [13] Nagasaka T, Shindo K, Hiraide M, Sugimoto T, Shiozawa Z. Ipsilateral thalamic MRI abnormality in an epilepsy patient. Neurology 2002;58:641–4. [14] Stubgen J-P. Crossed cerebellar diaschisis related to recurrent focal seizures. Epilepsia 1995;36:316–8. [15] Oster J, Doherty C, Grant PE, Cole AJ. Diffusion-weighted imaging abnormalities in the splenium after seizures. Epilepsia 2003;44:852–4. [16] Kim JA, Chung JI, Yoon PH, et al. Transient MR changes in patients with generalized tonic–clonic seizures and status epilepticus: periictal diffusion-weighted imaging. Am J Neuroradiol 2001;22:1149–60. [17] Chan S, Chin SS, Kartha K, et al. Reversible signal abnormalities in the hippocampus and neocortex after prolonged seizures. AJNR 1996;17:1725–31. [18] Sirven JI, Zimmerman RS, Carter JL, Drazkowki JF, Larson JS. MRI changes in status epilepticus. Neurology 2003;60:1866. [19] Singhal AB, Newstein MC, Budzik R, et al. Diffusion-weighted magnetic resonance imaging abnormalities in Bartonella encephalopathy. J Neuroimaging 2003;13:79–82. [20] Dematteis M, Kahane P, Vercueil, Depaulis A. MRI evidence for the involvement of basal ganglia in epileptic seizures: an hypothesis. Epileptic Disord 2003;5:161–4. [21] Nixon J, Bateman, Moss T. An MRI and neuropathological study of a case of fatal status epilepticus. Seizure 2001;10:588–91. [22] Diehl B, Najm I, Ruggieri P, et al. Post diffusion-weighted imaging for the localization of focal epileptic areas in temporal lobe epilepsy. Epilepsia 2001;42:21–8. [23] Wasterlain CG, Fujikawa DG, Penix L, Sankar R. Pathophysiological mechanisms of brain damage from status epilepticus. Epilepsia 1993;34(Suppl. 1):S37–53. [24] Ko D, Heck C, Grafton S, et al. Vagus nerve stimulation activates central nervous system structures in epileptic patients during PET H2 /15 O blood flow imaging. Neurosurgery 1996;39:426–31. [25] Seo DW, Na DG, Na DL, Moon SY, Hong SB. Subcortical hypointensity in partial status epilepticus associated with nonketotic hyperglycemia. J Neuroimaging 2003;13:259–63. [26] Ziyeh S, Thiel T, Spreer J, Klisch J, Schumacher M. Valproateinduced encephalopathy: assessment with MRI imaging and 1H MR spectroscopy. Epilepsia 2002;43:1101–5. [27] Liu W-C, Mosier K, Kalnin AJ, Marks D. BOLD fMRI activation induced by vagus nerve stimulation in seizure patients. J Neurol Neurosurg Psychiatry 2003;74:811–3. [28] Takaya M, Terry WJ, Naritoku DK. Vagus nerve stimulation induces a sustained anticonvulsant effect. Epilepsia 1996;37: 1111–6.