1. 2022;25;E1249-E1255Analysis of Risk Factors and Prediction Model for Recurrence of Hemifacial Spasm After Radiofrequency Ablation
   Retrospective Study
   Ming Yao, PhD, Jianmei Xia, MD, Bing Huang, MD, Yong Fei, MD, Yi Wu, MD, and Enming Zhang, MD.

BACKGROUND: Computed tomography (CT)-guided percutaneous stylomastoid foramen puncture radiofrequency ablation for the treatment of hemifacial spasm has a significant clinical effect; however, related risk factors for recurrence have not been studied.

OBJECTIVE: To investigate the risk factors for the recurrence of hemifacial spasm after radiofrequency ablation and construct a model for predicting recurrence.

STUDY DESIGN: This is a single-center retrospective observational study.

SETTING: The study was conducted at the Pain Department of the Affiliated Hospital of Jiaxing College in Jiaxing, China.

METHODS: A retrospective analysis was performed on 99 patients diagnosed with primary hemifacial spasm (HFS) admitted to the Affiliated Hospital of Jiaxing University between August 2018 and December 2021. All patients underwent CT-guided percutaneous stylomastoid foramen radiofrequency ablation. Kaplan-Meier survival analysis, log-rank test, and Cox proportional risk regression model were used to analyze the clinical factors that affect the recurrence of patients with HFS after radiofrequency ablation, and a recurrence prediction model was established.

RESULTS: Follow-up was 3-12 months; recurrence rates were 20.2%, 36.4%, and 71.9% at 3, 6, and 12 months postoperatively, respectively. Univariate analysis showed that puncture approach, operation time, and facial paralysis level were factors influencing recurrence in patients with HFS after radiofrequency ablation (P < 0.05). The multivariate Cox proportional risk regression model showed that the operative time and facial paralysis grade were independent factors for recurrence after radiofrequency ablation in patients with facial spasms. The recurrence risk function model of patients with facial spasms after radiofrequency ablation was expressed as h(t) = h0exp(-0.619X1-2.589X2), where X1 and X2 represent the operation time and facial paralysis grade, respectively. The likelihood ratio of the model was statistically significant (chi squared = 55.769, P < 0.001).

LIMITATIONS: We look forward to increasing the sample size in follow-up studies and exploring relevant conclusions in randomized controlled trials.

CONCLUSION: Long operation times and high-grade facial paralysis can reduce the risk of recurrence in patients with facial spasms. The constructed recurrence prediction model could serve as a reference for clinical diagnosis and treatment.

KEY WORDS: Hemifacial spasm, radiofrequency ablation, recurrence, risk-factors, prediction model