Role of Gamma Knife radiosurgery in trigeminal neuralgia - Its long term outcome and prediction using Artificial...

We aimed to study the long-term efficacy, prognostic factors and complications associated with the GKRS for trigeminal neuralgia.

Prospectively created database was analyzed for these patients. We created an Artificial neural Network - using Multilayer perceptron model in SPSS 23 by including all variables whose p value were<0.5 in univariate analysis.

A total of 36 patients were included in the study. Patients pain free at 6 months were 25(69.44%), which reduced to 6(40 %) at 6 years. Median time to pain relief was 18.5 days. Only 5(13.88%) of them developed new onset or worsened numbness post Gamma Knife radio surgery. Median radiation dose was 80 Gy (prescribed at 100 percent isodose line). 1st GKT (P value < 0.05) and post GKT numbness (P value < 0.05) were the only factors favouring good outcome. Prior history of MVD was associated with poor pain relief post GKRS although p value was not significant (p = 0.136). ANN model could predict with 90.0 percent accuracy the favourable or unfavourable response on 11 Tested cases. In ANN model, a greater number of Pre GKT medications, previous MVD history, V2 dermatome involvement and negative history of post GKT numbness were negative prognostic factors.

Lesser number of pre GKRS drugs used, involvement of V1 dermatome, post GKT numbness are favourable prognostic factors. Also, history of failed MVD for trigeminal neuralgia is associated with poor outcome. Repeat GKRS failed to show improvement in BNI grades.