Subcortical SISCOM hyperperfusion: Should we pay more attention to it?
Seizure. 2018-11-01; 62: 43-48
DOI: 10.1016/j.seizure.2018.09.017
Read on PubMed
Aupy J(1), Wongwiangjunt S(2), Wang ZI(3), Wu G(4), Alexopoulos A(3).
Author information:
(1)Cleveland Clinic, Neurological Institute, Epilepsy Center, Cleveland, OH, 44195, USA; CHU of Bordeaux, Pole de Neurosciences Cliniques, Bordeaux, France; University of Bordeaux, IMN, UMR-CNRS 5293, Bordeaux, France. Electronic address: .
(2)Cleveland Clinic, Neurological Institute, Epilepsy Center, Cleveland, OH, 44195, USA; Faculty of Medicine, Siriraj Hospital, Mahidol University, Department of Internal Medicine, Bangkok, Thailand.
(3)Cleveland Clinic, Neurological Institute, Epilepsy Center, Cleveland, OH, 44195, USA.
(4)Cleveland Clinic, Department of Nuclear Medicine, Cleveland, OH, 44195, USA.
PURPOSE: Demonstrating cerebral blood flow changes during seizures, ictal-interictal single photon emission computed tomography (SPECT) with co-registration to MRI (SISCOM) reflects brain activation and its pathways of spread. To investigate subcortical ictal hyperperfusion patterns during focal seizures, we retrospectively reviewed SISCOM analysis of patients who became seizure-free after cortical resection. Our aim was to evaluate the relationship between epileptogenic zones and subcortical hyperperfusion.
METHOD: 67 patients were identified as having SISCOM evaluation and having remained seizure-free for at least one year after surgical resection. SISCOM analysis was blindly reviewed for localization of basal ganglia (BG), thalamic (TN) and cerebellar (CH) hyperperfusion based on three different thresholds. Subcortical activation and epilepsy characteristics were then compared between patients. For a given region of interest and threshold, the sensitivity, specificity and positive and negative predictive value for correct lateralization of the epilepsy side was calculated.
RESULTS: Depending on the threshold used, BG hyperperfusion was found in 37.3-73.9% of patients, TN hyperperfusion in 31.3-68.1% and CH hyperperfusion in 13.5-29%. For a threshold of 1.5, the best predictive positive value for correct lateralization of the epilepsy side was obtained with BG/CH coactivation (89%). For a threshold of 2.0 and 2.5, it was obtained with BG/TN coactivation (88%) and BG activation (82%), respectively.
CONCLUSION: Subcortical SISCOM hyperperfusion could offer additional clues in terms of lateralization.