Blink reflex: A practical test to evaluate the trigeminal nerve injury following percutaneous balloon compression for the treatment of trigeminal neuralgia

Objective To evaluate the blink reflex (BR) in estimating the potential injury of trigeminal nerve following percutaneous balloon compression (PBC) surgery, and to determine the association between BR alterations and early surgical outcomes. Methods In this single-center, prospective before-and-after study, a total of 74 patients who had primary trigeminal neuralgia and scheduled for PBC between October 2020 and June 2021 were prospectively included. BR testing and facial sensory assessment were performed pre- and post-PBC. The latency and the area under the curve (AUC) of pre- and postoperative R1 (R1(pre)/R1(post)) and R2 (R2(pre)/R2(post)) were measured. Results The BR components were noticeably delayed or diminished following PBC. R1(post) was elicited in only 26 patients, and absent in 48 patients. The residual R1(post) had markedly reduced AUC (median difference [Hodges-Lehmann]: -59.5, 95% confidence interval [CI]: -217.5 to -6.9, p = 0.023). Compared with R2(pre), the latency of R2(post) was considerably delayed (mean difference: 4.3, 95% CI: 2.9 to 5.7, p < 0.001) and the AUC was greatly suppressed (median difference [Hodges-Lehmann]: -388.4, 95% CI: -548.4 to -259.5, p < 0.001). After PBC, 58 patients had immediate total pain relief, and 16 had partial relief. The absence of R1(post) was found in 46 of 58 (79.3%) patients with complete remission, whereas in only 2 of 16 (12.5%) patients with partial relief. Association analysis showed that the absence of R1(post) was strongly associated with total pain relief (46/58 [79.3%] vs. 2/16 [12.5%], odds ratio [OR]: 26.8, 95% CI: 5.4 to 134.5, Cramer's V: 0.6, p < 0.001). The latency of R2(post) in patients with total relief was significantly delayed (mean difference: 2.5, 95% CI: 0.3 to 4.6, p = 0.028). Patients experienced graded facial numbness after PBC, of whom 31 reported mild numbness (Grades I-II) and 43 reported more severe numbness (Grades III-IV). The absence of R1(post) was significantly associated with facial numbness severity, 33/43 (76.7%) in Grades III-IV vs. 15/31 (48.4%) in Grades I-II (OR: 0.284, 95% CI: 0.105 to 0.771, Cramer's V: 0.3, p = 0.012). In patients with more severe numbness, the latency of R2(post) was significantly delayed (mean difference: 2.7, 95% CI: 0.1 to 5.3, p = 0.043), and the reduction of AUC was much greater (median difference [Hodges-Lehmann]: 17.2, 95% CI: 0.5 to 35.4, p = 0.041). Conclusion Both R1 and R2 were significantly diminished after PBC and these alterations were associated with early surgical outcomes, suggesting that the BR is useful in evaluating trigeminal injury following PBC and could provide objective information about early prognosis.