Metabolic Correlates of the Ictal-Interictal Continuum: FDG-PET During Continuous EEG

Abstract

Background: Ictal-interictal continuum (IIC) continuous EEG (cEEG) patterns including periodic discharges and rhythmic delta activity are associated with poor outcome and in the appropriate clinical context, IIC patterns may represent "electroclinical" status epilepticus (SE). To clarify the significance of IIC patterns and their relationship to "electrographic" SE, we investigated FDG-PET imaging as a complementary metabolic biomarker of SE among patients with IIC patterns.

Methods: A single-center prospective clinical database was ascertained for patients undergoing FDG-PET during cEEG. Following MRI-PET co-registration, the maximum standardized uptake value in cortical and subcortical regions was compared to contralateral homologous and cerebellar regions. Consensus cEEG review and clinical rating of etiology and treatment response were performed retrospectively with blinding. Electrographic SE was classified as discrete seizures without interictal recovery or >3-Hz rhythmic IIC patterns. Electroclinical SE was classified as IIC patterns with electrographic and clinical response to anticonvulsants; clonic activity; or persistent post-ictal encephalopathy.

Results: Eighteen hospitalized subjects underwent FDG-PET during contemporaneous IIC patterns attributed to structural lesions (44 %), neuroinflammatory/neuroinfectious disease (39 %), or epilepsy (11 %). FDG-PET hypermetabolism was common (61 %) and predicted electrographic or electroclinical SE (sensitivity 79 % [95 % CI 53-93 %] and specificity 100 % [95 % CI 51-100 %]; p = 0.01). Excluding electrographic SE, hypermetabolism also predicted electroclinical SE (sensitivity 80 % [95 % CI 44-94 %] and specificity 100 % [95 % CI 51-100 %]; p = 0.01).

Conclusions: In hospitalized patients with IIC EEG patterns, FDG-PET hypermetabolism is common and is a candidate metabolic biomarker of electrographic SE or electroclinical SE.

Keywords: EEG; FDG-PET; Ictal-interictal continuum; Status epilepticus.

Fig. 1

Representative cases of patterns of PET hypermetabolism with corresponding electrographic and MR data. 13-s cEEG samples and FDG-PET imaging [rainbow color scheme; 10 % gradations from hypometabolic (blue) through green and yellow to hypermetabolic (red), cerebellum set to yellow, L demarcates left orientation] and FLAIR MR images for six subjects. a, b are representative cases of focal hypermetabolism. a (SUBJ 3) is a 32-year-old woman with a history of peri-natal stroke that presented with epilepsia partialis continua, MRI had T2 hyperintensity near the remote infarct with corresponding PET hypermetabolism in the same region. EEG had LPDs and intermittent discrete seizures. She required surgery to remove the ictal focus, which resulted in cessation of seizure activity, but a worsened hemiparesis. b (SUBJ 1) Is a 25-year-old woman that presented with convulsive status epilepticus transitioning to persistent encephalopathy with initial AED treatment. cEEG had persistent 0.5 Hz right posterior quadrant LPDs, MRI revealed a posterior temporal T2 hyperintensity with an adjacent developmental venous anomaly. PET revealed focal hypermetabolism in the region of the MRI abnormality. Treatment was escalated with IV sedation and patient made a full recovery. Follow-up MRI revealed a resolution of T2 hyperintensity. c, d are representative cases of regional hypermetabolism. c (SUBJ 15) is a 77-year-old man presenting with hypertension and T2 hyperintensities of the left occipital/parietal cortex and scattered white matter T2 hyperintensities, DWI changes, and microhemorrhages. Initial cEEG revealed discrete electrographic seizures. After anti-seizure medication (top panel), cEEG displayed an IIC pattern (left LPDs) associated with focal FDG-PET hypermetabolism in left parietal, occipital, posterior-frontal right thalamic regions. (Bottom panel) Anesthetic burst-suppression was employed and repeat PET yielded resolution of hypermetabolism 12 days later concurrent with burst-suppression on cEEG. LPDs subsequently resumed with weaning sedation and his persistent encephalopathy prompted family to elect palliative care, autopsy revealed inflammatory cerebral amyloid angiopathy. d (SUBJ 13) is an 84-year-old woman with dementia s/p VP shunt placement for NPH that developed discrete focal seizures, which resolved with AED treatment. Patient had persistent encephalopathy and was found to have LPDs on cEEG and regional hypermetabolism on PET. MRI revealed pachymeningeal thickening and enhancement as well as a right frontal focal area of hemorrhage and surrounding T2 hypermetabolism along the shunt tract. Patient did not improve and was transitioned to palliative care. e, f are representative cases of diffuse hypermetabolism. e (SUBJ 11) is 44-year-old man with subacute cognitive decline presenting with clinical seizures and worsening encephalopathy. cEEG had 4 Hz GRDA and diffuse (L > R) cortical and subcortical hypermetabolism. Found to have VGKC antibody, return to baseline with AEDs and IVIG. f (SUJB 5) is a 23-year-old man presenting with NORSE, his clinical seizures improved with treatment, but clinical deficits persisted. cEEG revealed LPDs. PET has bilateral hypermetabolism primarily in the basal nuclei. MRI was normal other than mild generalized volume loss. Patient recovered with a residual modest spastic tetraparesis and cognitive recovery. Etiology of illness remains cryptogenic (Color figure online)

Fig. 2

The ictal-interictal-injury-metabolism continuum. This diagram is a graphic conceptualization adapted from Chong and Hirsh [24] using the current ACNS terminology and with the addition of PET metabolism ranging from hypo/normo metabolic in blue to hypermetabolic in red. This conceptualization is meant only to reflect scalp EEG patterns. X-axis represents the spectrum of interictal to ictal, i.e., the spectrum of cerebral dysfunction, which is potentially reversible with anti-seizure treatment. Y-axis is secondary neuronal damage attributable to the EEG pattern. SB suppression burst, RDA rhythmic delta activity, LPD lateralized periodic discharges, SW spike wave, GPD generalized periodic discharges, SIRPIDs stimulus-induced rhythmic, periodic, or ictal discharges, NCS non-convulsive seizures, GCSE generalized convulsive status epilepticus, NCSE non-convulsive electrographic status epilepticus, EPC epilepsia partialis continua, HYPO hypometabolism, HYPER hypermetabolism (Color figure online)