Abstracts

Rationale: The ictal-interictal continuum (IIC) is a spectrum of EEG patterns found in critically ill patients. Understanding the metabolic and neurophysiologic effects of IIC patterns is critical to understanding their impact on neurologic injury and defining interventions. Hypermetabolism on FDG-PET is associated with ictal activity and is a marker of cerebral function distinct from EEG. We performed an observational study assessing cerebral metabolism via FDG-PET in patients undergoing continuous EEG (cEEG).Methods: We identified subjects from the MGH cEEG database who underwent FDG-PET. Subjects included had IIC EEG patterns at the time of injection the FDG-PET. Structural imaging, FDG-PET findings, medical records, and EEG data were collected. FDG-PET analysis was performed blinded to EEG, using MRI coregistration for identification of frontal, temporal, parietal, occipital, sub-cortical and cerebellar regions of interest. The maximum standardized uptake value (SUVmax) was measured for each region and compared as an SUV ratio to SUVmax of the cerebellum; an SUV ratio >1.20 was considered hypermetabolic. Two clinical neurophysiologists blinded to clinical/FDG-PET data assessed cEEG data using ACNS Standardized Critical Care EEG Terminology. Discrepancies were resolved by consensus. A fourth investigator blinded to both FDG-PET assessed clinical records for etiology of neurologic injury as well as EEG and clinical response to AEDs. Definite electrographic status epilepticus (SE ) was defined as a rhythmic pattern >3 Hz or discrete electrographic seizures without recovery of background. Possible electroclinical SE was defined as an IIC pattern without definite electrographic seizures but either: 1) associated with clonic activity, 2) electrographic and clinical response to AED, or 3) ongoing encephalopathy following a clinical or electrographic seizure and a pattern on the IIC.Results: Of 18 subjects with IIC cEEG activity undergoing FDG-PET between 2008 and 2014 (mean age 56.6), FDG-PET was hypermetabolic in 11 (61%), with diffuse or focal hypometabolic in 5 (28%) and normal in 2 (11%). The final diagnosis was an acute structural lesion in 8 subjects (44%), neuroinflammatory or neuroinfectious process in 7 (39%), preexisting epilepsy in 2, and Creutzfeltd-Jakob Disease (CJD) in 1 subject (Table 1). FDG-PET metabolism was assessed for its dependence on pre-specified candidate variables (Table 2). Several parameters such as >1-Hz or high-amplitude PDs were highly specific (1.0) but with broad confidence intervals because of limited number of subjects. FDG-PET predicted either definite possible SE with 79% (95% CI 0.53-0.93) sensitivity and 100% (0.51-1.0) specificity, with a Fisher exact test result of p=0.01.Conclusions: This observational study demonstrates that FDG-PET hypermetabolism may be observed in patients with IIC activity, including those without definite electrographic SE, particularly in those with LPDs, GPDs, or possible electroclinical SE. These data suggest that IIC patterns with appropriate clinical accompaniments have cerebral metabolic effects akin to definite electrographic SE.