Abstracts

Rationale: Infantile spasms (IS) are characterized by a triad of clinical spasms, hypsarrhythmia on electroencephalogram (EEG), and developmental delay. Some children with IS go on to develop Lennox Gastaut syndrome (LGS), a severe age-related epilepsy syndrome characterized by intellectual disability and multiple seizure types generally refractory to medical treatment. The diagnostic criteria for this syndrome are vague, with no specific set of criteria defined by the International Leage Against Epilepsy (ILAE). Many experts utilize varying combinations of the following criteria for diagnosis: paroxysmal fast activity, slow spike and wave on EEG, current developmental delay, multiple seizure types, or current tonic seizures. Our objective was to assess how many children in a high-risk cohort for LGS actually received this diagnosis. Methods: We analyzed the medical charts of children with IS who were diagnosed and treated at Children’s Hospital Colorado between the years of 2012 and 2018. Fifteen patients were excluded as they had either moved away or did not have follow-up for 3 years prior to 11/2018. LGS was defined as the presence of 3 out of 5 of the following characteristics: paroxysmal fast activity, slow spike and wave on EEG, current developmental delay, multiple seizure types, or current tonic seizures. Descriptive statistics were performed with mean and median. The relationship between variables was assessed using either Chi-square or Fisher’s exact test. Results: There were 98 children with IS, 51% were male. The median age of IS diagnosis was 7 months (IQR 5,9). Thirty-six patients met our criteria for LGS diagnosis (37% of the IS cohort), while only 10 patients (10%) actually received a formal diagnosis in the medical chart. Patients were more likely to be diagnosed with LGS if they met our criteria (8 met criteria and were diagnosed, 28 met criteria but not diagnosed, p=0.0045). Children with a known etiology were more likely to meet criteria for LGS (33 with known etiology, 3 without; p=0.016). Children who received a formal diagnosis of LGS were also more likely to have an identified etiology, although this was not significant (9 with known etiology, 1 with unknown etiology, p=0.68). There was no significant difference in number of medications received between children who met criteria for LGS and those who did not (p=0.52). Children receiving the diagnosis were not more likely to be on more medications (one on 3 medications, one on 4 medications; p= 0.48). Participants with developmental delay prior to onset of IS were more likely to meet criteria for LGS (p=0.002), however they were not more likely to receive a formal diagnosis (p=0.71).  Conclusions: Less than 1/3 of patients who meet the diagnostic criteria for LGS are being given the formal diagnosis at our institution. This may be due to the fact that patients who meet criteria for the diagnosis of LGS already have an underlying etiology that is more specific than LGS. The underdiagnosis of LGS may also be attributed to poorly defined criteria. This underdiagnosis may have significant implications for outcomes research that depends on use of diagnostic codes for LGS, as those identified with a diagnosis of LGS may not represent the larger population that may meet criteria for this disease. However, given these findings, more information regarding the relevance of diagnosing this syndrome is necessary. Funding: No funding