Abstracts

Rationale: Acute symptomatic SZs have been reported to occur in 73% of infants with NAHI but PTE in only 22% (Barlow 2000). Prognostic predictors for poor outcome and PTE in NAHI are lacking and needed. Aim of our study is to evaluate which clinical, EEG, seizure types, and neuroimaging findings increase the likelihood an infant with non-accidental head Injury (NAHI) will develop acute symptomatic seizures (SZs), post-traumatic epilepsy (PTE), and/or poor neurologic outcomes. Methods: We identified consecutive infants (age <12 months) hospitalized at the UNM Children s Hospital between January 2005 to April 2013 with NAHI. We retrospectively reviewed their electronic medical records, EEG and neuroimaging regarding: age; gender; clinical presentation (including early SZs within 24 h of presentation, subdural hematoma (SDH), retinal hemorrhages); EEG and neuroimaging findings; SZ types, status epilepticus (SE), nonconvulsive status epilepticus (NCSE), antiepileptic therapies (AEDs), time to control SZs, neurosurgical interventions, neurologic outcomes and PTE. We quantified outcomes using the King s Outcome Scale for Closed Head Injury (KOSCHI) for children (poor outcome < 3, favorable 4-5). Results: 34 infants (mean age 4.2 months) were diagnosed with NAHI. All survived with mean follow-up of 2.2 years. Early SZs observed in 65%. PTE developed in 47%, refractory in 11 (69%). Favorable outcomes (KOSCHI 5 in 22, 4 in 1) in 68%; poor in 32% (KOSCHI 3 in 10, KOSCHI 2 in 1). Most common SZ type was unilateral focal clonic; multifocal seizure and NCSE were associated with poor outcomes (p=.05 and p=.01 respectively). Focal SZs correlated with focal findings on EEG in 68% and with neuroimaging in 82%. Risk factors for developing poor outcomes and refractory PTE were: 1) early SZs (p=.005); 2) hypoxia or cardiac arrest at presentation needing intubation (p =.001); 3) multifocal seizures or discharges (p =.002), SE (p=.002), NCSE ( p<.001), 2 AEDs required to control SZs (p=.002) and longer time to control SZs (p=.02); 4) hypoxic-ischemic injury (p=.001), infarction (p=.04), or hydrocephalus (p=.01) and need for neurosurgical intervention. Factors associated with good outcome were unilateral focal clonic SZs easily responsive to treatment, unilateral SDH without other abnormalities on neuroimaging. Conclusions: Difficult to control multifocal acute symptomatic seizures, status epilepticus, and nonconvulsive status epilepticus in infants with NAHI increase the likelihood they will develop posttraumatic epilepsy and poor outcomes. Infants with easily controlled unilateral focal clonic seizures and a MRI showing only a unilateral SDH often had favorable outcomes. Given the high incidence of NCSE in infants expedient diagnostic testing to identify and treat SZs in these infants may improve outcomes.