Prognostic factors in epileptic encephalopathies at onset in the first 2 years of life: The experience of a tertiary healthcare center in Italy

Introduction: The aim of this retrospective cohort study was to identify some prognostic factors in anamnestic/clinical/instrumental data at the onset of epileptic encephalopathy (EE), for multiple outcome measures. Methods: We recruited patients diagnosed as affected by EE at Sant'Anna University Hospital, with onset in the first 24 months of life, with follow-up lasting longer than 3 years. Results: At the end of the follow-up, 6 patients (14%) died within 2 years of age; 20 patient (49%) had a drug-resistant epilepsy (DRE); 9 patients (22%) had a language development delay (LDD); 12 patients (30%) had an autism spectrum disorder (ASD); 20 patients (49%) had a global psychomotor impairment (GPI); 9 patients (22%) needed palliative care; and nobody had a normal psychomotor development. Preexisting developmental delay predicts death (p = 0.009), and in survivors, it is associated with a GPI (p < 0.001); patients with normal neurological examination at the onset of EE only develop a LDD (p = 0.020). Neuroimaging structural alterations are associated with DRE (p = 0.012) and with a GPI (p = 0.013). The history of perinatal risk factors predicts the worst prognosis (death: p = 0.035, GPI: p = 0.015, and access to palliative care: p = 0.007). The absence of early response to treatment is correlated to a poor long-term prognosis (GPI, p = 0.019; DRE, p = 0.001). The multivariate analysis confirms that a normal development at onset predicts the most favorable prognosis, both in terms of survival and cognitive outcome (OR [odds ratio] = 0.1). An early response to treatment is a protective factor for DRE (OR = 0.1). A perinatal pathology is confirmed as an independent prognostic factor of severe comorbidities (access to palliative care: OR = 10.4). Significance: This study was conducted to recognize possible prognostic factors among onset data of patients with EE, considering multiple outcome measures. This study design represents an innovative element compared to available papers, which were centered on isolated endpoints of prognosis, such as the prediction of neurocognitive development impairment or drug resistance. The data obtained from the study confirm that EEs prognosis is generally, but not universally, poor. Structural etiology and/or lack of response to antiepileptic drug (AED) within three months are main risk factors for DRE. Normal development at the onset of EEs and early response to treatment are the main positive prognostic factors.