FDG-PET assessment and metabolic patterns in Lafora disease.

PURPOSE: To describe cerebral glucose metabolism pattern as assessed by 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) in Lafora disease (LD), a rare, lethal form of progressive myoclonus epilepsy caused by biallelic mutations in EPM2A or NHLRC1. METHODS: We retrospectively included patients with genetically confirmed LD who underwent FDG-PET scan referred to three Italian epilepsy centers. FDG-PET images were evaluated both visually and using SPM12 software. Subgroup analysis was performed on the basis of genetic and clinical features employing SPM. Moreover, we performed a systematic literature review of LD cases that underwent FDG-PET assessment. RESULTS: Eight Italian patients (3M/5F, 3 EPM2A/5 NHLRC1) underwent FDG-PET examination after a mean of 6 years from disease onset (range 1-12 years). All patients showed bilateral hypometabolic areas, more diffuse and pronounced in advanced disease stages. Most frequently, the hypometabolic regions were the temporal (8/8), parietal (7/8), and frontal lobes (7/8), as well as the thalamus (6/8). In three cases, the FDG-PET repeated after a mean of 17 months (range 7-36 months) showed a metabolic worsening compared with the baseline examination. The SPM subgroup analysis found no significant differences based on genetics, whereas it showed a more significant temporoparietal hypometabolism in patients with visual symptoms compared with those without. In nine additional cases identified from eight publications, FDG-PET showed heterogeneous findings, ranging from diffusely decreased cerebral glucose metabolism to unremarkable examinations in two cases. CONCLUSIONS: FDG-PET seems highly sensitive to evaluate LD at any stage and may correlate with disease progression. Areas of decreased glucose metabolism in LD are extensive, often involving multiple cortical and subcortical regions, with thalamus, temporal, frontal, and parietal lobes being the most severely affected. Prospective longitudinal collaborative studies are needed to validate our findings.

Focal epilepsies in adult patients attending two epilepsy centers: classification of drug-resistance, assessment of risk factors, and usefulness of "new" antiepileptic drugs.

PURPOSE: To classify the grade of antiepileptic drug (AED) resistance in a cohort of patients with focal epilepsies, to recognize the risk factors for AED resistance, and to estimate the helpfulness of "new-generation" AEDs. METHODS: We included 1,155 adults with focal epilepsies who were observed consecutively after 1990 and followed regularly at two epilepsy centers. We systematically collected the clinical, diagnostic, and therapeutic data using a custom-written database. We classified the patients as seizure-free or AED resistant according to the International League Against Epilepsy (ILAE) criteria, and we evaluated the risk factors associated with AED resistance using logistic regression analysis. We further grouped AED-resistant patients in different grades (I, II, and III) according to the number of AEDs already tried as proposed by Perucca. KEY FINDINGS: AED resistance occurred in 57.8% of the 729 patients with symptomatic focal epilepsies and was positively associated with electroencephalography (EEG) abnormalities, seizure type, and the presence of mesial temporal sclerosis. Among 426 patients without detectable causes, the percentage of AED resistance was significantly lower (39.2%) and correlated with EEG abnormalities and psychiatric symptoms. Among AED-resistant patients, the majority (64.6%) had tried three or more AEDs, which fit the more severe grade III proposed by Perucca. Among seizure-free patients, more than one-half (57%) needed to try two or more AEDs before reaching seizure control (14.9% needed three or more AEDs). Furthermore, among seizure-free patients who could be previously classified as resistant to two or more AEDs, 52.2% reached seizure freedom while receiving treatment with "new generation" AEDs. SIGNIFICANCE: The ILAE classification of AED resistance, as well the graded classification proposed by Perucca, was easily exploitable in our patients, although these classifications systems appear to have a limited value in predicting seizure outcome. Actually, a small but not negligible percentage of patients reached seizure freedom after trying several AEDs (including "new" AEDs), suggesting repeated trials may be necessary for seizure control.

Myoclonus in Creutzfeldt-Jakob disease: polygraphic and video-electroencephalography assessment of 109 patients.

We used electroencephalography (EEG)-polygraphic recordings to classify myoclonus in 109 patients with Creutzfeldt-Jakob disease (CJD) on the basis of its electromyography (EMG) pattern, time course, distribution, and EEG correlates. We recorded myoclonic jerks in 55 patients (50.4%), and we classified them as periodic myoclonus in 28, rhythmic in 13, and irregular in 20 (6 patients showed two types of myoclonus). Myoclonus occurred as a prominently negative event (interrupting the EMG discharge) in 10. Periodic sharp-wave complexes (PSWCs) were present in all but one patient with myoclonic jerks but were time-locked with EMG-bursts only in case of periodic myoclonus. Jerk-locked back averaging revealed a variable EEG-EMG transfer-time commonly exceeding that characterizing cortical myoclonus. Myoclonus was frequently associated with Met/Met polymorphism at codon 129 of the prion protein gene, but it was also observed in association with Met/Val or Val/Val polymorphisms provided that the EEG showed the presence of the PSWC pattern. The presence of enlarged somatosensory evoked potentials significantly correlated with the myoclonic presentation, as did MR signal hyperintensity involving the cortical mantle. Our observations on the basis of standard polygraphic criteria suggest that CJD associates with a remarkable variety of myoclonic jerks, and therefore different brain structures are probably involved as generators. The significant association between the presence of all myoclonus types with PSWCs suggests that hyperexcitable corticosubcortical loops are always required to generate (or allow) both myoclonus and the EEG complexes, either they are time locked or not.