Abstracts

Rationale: Stereotactic radiofrequency thermocoagulation (SRT) has been used for treatment of epileptic hypothalamic hamartomas (HHs) since 1997 in our hospital. Our strategy with combination of multiple trajectories and coagulations enables complete disconnection for every size and types of HH. Although we had already reported favorable seizure outcomes (Kameyama, 2016; Shirozu, 2020), long-term outcomes have not been disclosed.

Methods: 131 patients with follow-up period of at least 3 years after the last SRT were retrospectively reviewed. Regular follow-up was performed at 3 or 6 months, 1 year, 2 years, and 3 years after SRT. Additional follow-up was also performed according to their seizures or requests. Seizure outcomes were evaluated by ILAE classification for gelastic seizure (GS) and non-gelastic seizure (nGS) separately. ILAE class 1 and 2 were considered seizure free in this study. Seizure-free rates after both the first and last SRT were analyzed by Kaplan-Meier survival curves. Long-rank test and Cox-hazard model were used to evaluate the risk factors for seizure outcomes.

Results: All patients had GS and 107 patients (81.7%) had nGS as well. Age at surgery, GS onset, and nGS onset ranged 1.7-51 (median, 8), 0-11 (1), 0-20 (3) years, respectively. 10 patients (7.5%) were related to genetic syndrome. 34 patients (30.0%) had undergone previous surgical treatments before our SRT. Median follow-up period after the last SRT was 60 months (36-202). 34 patients (26.2%) performed reoperation. Finally, 116 patients (88.6%) were free from GS and 85 of 108 patients (79.4%) were free from nGS. Kaplan-Meier analysis revealed that mean GS-free survival after the first SRT was 64.1 months (95%CI 57.3-70.9) and improvement into 80.2 months (75.7-84.8) after the last SRT. In contrast, mean seizure-free survivals of nGS did not change from 84.4 months (73.0-90.7) after the first SRT to 83.1 months (74.1-92.0) after the last SRT. Log-rank test revealed genetic syndrome (P=0.01) as a risk factor for GS-free survival, but no factors on Cox-hazard model. On the other hand, longer durations of GS (P=0.03) and nGS (P=0.02) and multiple previous treatments (P=0.006) were related to nGS-free survivals on Log-rank test. Cox-hazard model showed multiple previous treatments such as open surgery, endoscopic surgery, and radiosurgery (HR 15.65, 95%CI 1.79-137-16, P=0.01) was the risk factors for nGS outcomes.

Conclusions: The last SRT was equivalent to achievement of complete disconnection of HH. Because improvement and sustained GS-free survivals were obtained after the last SRT, complete disconnection is essential to eliminate GS. In contrast, nGS was considered to be attributed to secondary epileptogenesis, therefore, complete disconnection did not affect nGS outcomes. Early surgical indication should be considered and ineffective other treatment should be avoided.

Funding: Please list any funding that was received in support of this abstract.: This study was supported by the Ministry of Health, Labour Welfare, Japan Research program on rare and intractable diseases (JPMH20FC1029).