Clinicians order skin biopsy and genotyping for EPM2A and EPM2B variants when cognitive decline or ataxia start in childhood or adolescent epilepsies. Consequently, diagnosis of LD is made late when the disease is already midcourse. To solve this problem, we studied prospectively 15 living LD patients and compared results with 15 deceased LD patients retrospectively analyzed.

CWS, AVD-E and RMD independently extracted clinical outcome measures, namely, seizures, cognition, speech, gait ataxia, motor system and quality of life. Using scales including MOCA and SARA, each constructed an LD Progression Scale for 30 patients from seizure onset to cognitive decline to beginning status. JB, VH-N and AA biostatistically measured these clinical elements during early stages of progression with multiple time points, analyzed by Kaplan-Meir (K-M) survival analysis.

All prospectively and retrospectively studied had PAS+ inclusion bodies on skin biopsies. Amongst prospective patients, nine had EPM2A variants (eight homozygous and one compound heterozygous); four had EPM2B compound heterozygosities. First symptoms were visual seizures only with EEG occipital spikes in 6/15, visual plus myoclonic-tonic-Clonic (mTC) and absence seizures in 4/15, mTC only in 2/15 and myoclonic, absence and mTC seizures in 3/15. Mean age onset was 11.54 years with dementia established (MOCA <22/30) at 15 years and 5 months, 2.8 years (S.D. 1.3) from disease onset. First bouts of status epilepticus appeared 3.9 years from disease onset (S.D.1.7 years). 

In our cohort, LD starts at 11.5 year, often with visual seizures (10/15), EEG occipital spikes and regional 2FDG PET hypometabolism in temporal-occipital area. Myoclonic, mTC and absence seizures eventually appear in all. LD has a tight window of 2.8 years before dementia and 3.9 years before status epilepticus. Results of K-M curves will be discussed.