| Abstract |
Background
According to the new definition, status epilepticus (SE) is defined as a minimum of 5 or more episodes of continuous clinical and/or electrographic seizure activity or recurrent seizure activity without recovery (returning to baseline) between seizures. Refractory status epilepticus (RSE), is defined as persistent epileptic activity despite first- and second- line antiepileptic treatment and usually require suppression by continuous infusion of anesthetic drugs. A subcategory of RSE is Super Refractory Status Epilepticus (SRSE), which includes seizures that present as SE and persist for 24 hours or more after the introduction of general anesthesia, or recur when anesthesia is reduced or stopped. Magnetic Resonance Imaging (MRI aims to find epileptogenic lesions. The role of MRI in investigating brain pathology associated with RSE and SRSE, in both children and adults, is of particular scientific interest, especially for patients who present to the emergency department (ED) or are admitted to the ICU with SE.
Objective
The aim of this study was to record the MRI findings in children hospitalized in PICU with status epilepticus with emphasis in topography and qualitative characteristics, and to investigate possible correlations between these qualitative and quantitative characteristics with clinical-laboratory data, the etiology, SE types, prognostic indicators and outcome.
Methods
In this retrospective study, children and adolescents aged <18 years, hospitalized in the pediatric intensive care unit (PICU) of PAGNI due to status epilepticus between the years 2009-2020, were included. The demographic, clinical, and laboratory data were collected from available electronic and documentary patient records. Data recorded included: seizure type (SE, RSE, SRSE), the time elapsed between PICU admission and the MRI, the number of MRIs during PICU hospitalization, clinical findings at admission and during hospitalization, the etiology of SE, the prognostic indicators (PRISM), the length of PICU stay in days, duration of ventilation in days, final diagnosis and outcome. In addition, all imaging studies, performed from the child's admission to the hospital discharge, were retrieved and re-evaluated. Τhe type and location of abnormal MRI features were recorded, as well as qualitative and quantitative characteristics in different areas of the cerebral parenchyma. MRIs were categorized into normal, abnormal with periictal abnormalities and abnormal without periictal abnormalities. Quantitative parameters like diffusion values in ADC maps and the signal intensity in regions of interest (ROI) in FLAIR sequences, were measured, utilizing the software of the EVORAD system of our hospital. For each test in FLAIR sequences, the absolute (δROI) and percentage (κ) difference in signal intensity between the qualitatively abnormal and the contralateral normal cortex were calculated.
Results
The study included a total of 40 patients with status epilepticus who underwent MRI during their hospitalization at the ICU, from which 15 (37.5%) were boys and 23 (57.5%) girls with a mean age of 3.5 (2-9) years. Five children (12.5%) were diagnosed with SE, nineteen with RSE (47.5%) and fifteen with SRSE (37.5%). Out of 40 patients, 18 had normal MRI (45%), 14 (35%) abnormal with periictal abnormalities and 8 (20%) abnormal without periictal abnormalities. The age (mean, IQR) differed significantly between seizure type (SE 0.83 years (0.25-3.5) vs RSE 3 years (2-5) vs SRSE 8 years (3-10.75, P=0.023). There was a positive correlation (p = 0.041) between final diagnosis and MRI category, with FIRES being the most common final diagnosis (28.6%) in children with abnormal MRI with periictal abnormalities followed by NORSE (21.4%) and autoimmune disease (21.4%). Genetic disease (75%) was the most common diagnosis in children with abnormal MRI without periictal abnormalities. The etiology, the type of seizures, and the clinical features studied, were not associated with the category of MRI. Regarding MRI sequences, agreement was found on the detection of hyperintensity between FLAIR and T2-w sequences with statistical significance (p <0.001). Lack of agreement, with detection of hyperintensity in FLAIR, and no hyperintensity in T2-w sequences was observed at the perisylvian cortex (7.1%), the hippocampus (10%), the parietal cortex (23.1%), the temporal cortex (9.1%), the occipital cortex (16.7%) and thalami (16.7%). There was no MRI in which hyperintensity was detected in T2-w sequence and not in FLAIR sequences. The detection of hyperintensity in FLAIR was more common in the group of patients with abnormal MRI with periictal abnormalities at the perisylvian cortex (p <0.001), the hippocampus (p <0.001), the parietal cortex (p <0.001), the temporal cortex (p = 0.001), the white matter (p = 0.003) and occipital cortex (p = 0.004). The PICU outcome was associated with the hyperintensity in FLAIR sequence in the 1st MRI, at the basal ganglia (p = 0.012), the parietal cortex (p = 0.024), the hippocampus (p = 0.025), the brainstem (p = 0.026), the corpus callosum (p = 0.026), the frontal cortex (p = 0.003) and at the occipital cortex (p = 0.003). Regarding the quantitative measurements, there was a correlation (p = 0.031) between signal intensity values in FLAIR at the right hippocampus (transverse plane) and the outcome. The highest intensity values were associated with death (361 ± 9.9), while the intensity values in the group of patients with better outcome had a mean value 327 ± 23.3. The percentage difference κ in FLAIR sequence was associated with the category of MRI. The mean value of κ in normal MRI was 5%, while in abnormal-periictal MRI the mean value of κ was 16%, meaning 16% higher than the optically normal cortex.
Conclusions
Periictal abnormalities were seen in MRI of children with SE by order of occurrence at the perisylvian cortex, the hippocampus, the parietal cortex, the temporal cortex, the white matter, the frontal and the occipital cortex. Periictal edema was detected more frequently in FLAIR, than in T2-w sequences. The absolute and percentage differences of signal intensities between qualitatively abnormal and contralateral more normal cortex was higher in the group of patients with abnormal-periictal MRI, compared to the group of patients with normal MRI. Hyperintensity in FLAIR located at the basal ganglia, hippocampus, brainstem, corpus callosum, and at the parietal, frontal and occipital cortex was associated with poor outcome.
|
Η μαγνητική τομογραφία σε παιδιά με status epilepticus
