Validation of a System to Predict Recanalization After Endovascular Treatment of Intracranial Aneurysms

BACKGROUND: Anatomic and functional hemispherectomies are relatively infrequent and technically challenging. The literature is limited by small samples and single institution data. OBJECTIVE: We used the Nationwide Inpatient Sample (NIS) database to report on a large population of hemispherectomy patients and their in-hospital complication rates over a 23-year period. METHODS: Between 1988 and 2010, we identified 304 pediatric hospitalizations in the NIS database where hemispherectomy was performed. Using the NIS weighting scheme, this inferred an estimated 1611 hospitalizations nationwide during this time period. Descriptive statistics were calculated on this inferred sample for patient and hospital characteristics and stratified by the presence of in-hospital complications. The adjusted odds of in-hospital complications and nonroutine discharge were estimated using multivariable models. RESULTS: The mean age of the patients was 5.9 years; 46% were female, and 54% were white. In the inferred series, 909 hospitalizations (56%) encountered at least 1 in-hospital complication; 42% were surgery related, and 25% were related to the hospitalization itself. For every 1-year increase in age, there was a corresponding 8% increase in the odds of a nonroutine discharge, adjusting for other potential confounders (95% confidence interval: 1.01-1.16). The most common in-hospital complication was the need for a blood transfusion (30%), followed by meningitis (10%), hydrocephalus (8%), postoperative hematoma/stroke (8%), and adverse pulmonary event (8%). Thirty-three mortalities (2%) were inferred from this series. CONCLUSION: This is the largest study to date examining hemispherectomy and associated in-hospital complication rates. This study supports early surgery in patients with medically intractable epilepsy and severe hemispheric disease. ABBREVIATIONS: ICD-9-CM, International Classification of Diseases, 9th Revision-Clinical Modification NIS, Nationwide Inpatient Sample