Authors :
Presenting Author: Aditi Trivedi, MD – Stanford University
Erin Troiano, BS (in process) – Stanford University
Prathyusha Teeyagura, MSBI – Stanford University
Juliet Knowles, MD/PhD – Stanford University
Rationale:
Lennox-Gastaut Syndrome (LGS) is a developmental and epileptic encephalopathy that accounts for 1-10% of all childhood epilepsies. It is defined by multiple seizure types including tonic seizures and characteristic EEG patterns. Among people with LGS, the presentation can vary, and natural history studies exploring disease progression, risk factors and etiologies are lacking. Approximately 30% of patients with LGS have a history of infantile spasms (IS). This study compared clinical features in children with LGS (age at onset of LGS, severity of LGS and brain MRI findings) between those with/without prior IS. Methods:
We conducted a retrospective chart review of patients with LGS born after 2005 and managed at Stanford Children's Health. All patients had undergone MRI and EEG. LGS was defined as (1) childhood onset, medically refractory epilepsy (2) slow-spike-and-wave on EEG. Cohorts were stratified by the presence of IS (LGS+IS) or absence of IS (LGS+no IS). IS was defined as the diagnosis of epileptic spasms < 2 years old. MRI brain results were obtained by chart review, and abnormalities were categorized into six domains: disorders of cortical malformations, disorders of neuronal migration, disorders of the corpus callosum, acquired abnormality, nonspecific intracranial abnormality or normal. For some analyses, patients were sub-classified into “classic LGS”, including those with confirmed tonic seizures and slow-spike-and-wave on EEG, and “LGS-like”, defined as those without tonic seizures but meeting other criteria. We quantified differences in the age of LGS onset, LGS vs LGS-like and MRI brain findings between the two groups. Results:
The study included 78 patients, of whom 23 (29%) had a history of infantile spasms (LGS+IS) and 55 (71%) did not (LGS+no IS). Among patients with available diagnostic age data, the LGS+IS cohort had a significantly earlier age of diagnosis compared to the LGS + no IS cohort (median 3 vs. 6 years of age; n = 22 LGS+IS patients, n = 39 LGS+no IS patients; p < 0.001, t-test). Those with IS were significantly more likely to be diagnosed with classic LGS than those without a history of IS (91% vs. 53%; p = 0.001, Chi-squared test). While there was no significant difference in the distribution of MRI findings between the two groups, we observed distinct patterns. The most common finding in the LGS+IS cohort was nonspecific abnormalities (36%), while the most common finding in the LGS+no IS cohort was a normal MRI (31%).