The Clinical Utility of Surgical Histopathology in Predicting Seizure Outcomes in Patients with Rasmussen Encephalitis Undergoing Hemispherectomy.

OBJECTIVE

The objectives of this study were to determine the relationship between the severity of pathology and seizure outcomes in patients who underwent hemispherectomy for Rasmussen encephalitis (RE) and to investigate which clinical factors correlated with severity of pathology.

METHODS

In this retrospective cohort study, we collected and reviewed pathology and clinical variables. We ascertained seizure outcomes using Engel's classification, and Pardo stages were used to grade pathology.

RESULTS

We included 29 unique patients who underwent 34 hemispherectomy procedures for analysis. There was no statistically significant correlation between Pardo stage and seizure outcome (P = 1). Increasing duration of epilepsy (β = 0.011, P = 0.02) and duration of hemiparesis (β = 0.024, P = 0.01) were significantly associated with a more severe Pardo stage. In contrast, the presence of epilepsia partialis continua had a negative relationship with Pardo stage (β = -0.49, P = 0.04). Twenty-six (89.75%) patients were Engel class I at the last follow-up, including all 5 patients who underwent redo hemispherectomy in our cohort.

CONCLUSIONS

Consistent with the progressive nature of RE, more severe pathology was associated with a longer duration of epilepsy and longer duration of hemiparesis, while the presence of epilepsia partialis continua was associated with less severe pathology. Results from this series suggest the degree of cortical involvement with RE as assessed on surgical histopathology does not correlate with seizure outcome after hemispherectomy, which appears to be more dependent on surgical technique/complete disconnection.