全聋型突发性聋伴眩晕患者虚拟现实前庭康复治疗前后前庭功能变化及其疗效的影响因素分析

Changes of vestibular function before and after virtual reality vestibular rehabilitation in patients with total deafness type sudden deafness with vertigo and factors influencing the therapeutic effect

摘要:

目的 探讨全聋型突发性聋伴眩晕患者采用虚拟现实前庭康复治疗前后前庭功能的变化及其疗效的影响因素。方法 选择2019年8月—2023年8月于北京市海淀医院初诊并采用虚拟现实前庭康复治疗的全聋型突发性聋伴眩晕患者75例为试验组，另取同期未进行前庭康复的全聋型突发性聋伴眩晕患者41例为对照组。两组患者治疗14 d后复测0.25~8 kHz平均听阈，其中39例患者受损听力提高30 dB以上纳入显效，35例患者受损听力提高15~30 dB纳入有效，42例患者受损听力提高低于15 dB纳入无效。收集患者治疗前后双温试验、视频头脉冲试验等临床资料。多因素Logistic回归分析疗效的影响因素；使用R软件评估前庭功能参数对患者疗效的预测价值。结果 治疗前，两组患者前庭功能无显著差异(P均＞0.05)。治疗后，与对照组相比，试验组前庭-眼动反射增益值较高，眩晕持续时间、扫视波异常率、不对称比异常率、单侧反应减弱指数异常率、优势偏向异常率较低(P均＜0.05)。试验组治疗显效率高于对照组(20.00% vs 2.44%，χ²=6.875，P=0.009)。除眩晕障碍量表-情感外，试验组治疗后眩晕障碍量表和医院焦虑抑郁量表评分较对照组均显著降低(P均＜0.05)。无效组患者接受虚拟现实前庭康复治疗率、水平半规管增益值和后半规管增益值较低，年龄、初诊时间、听力损失、单侧反应减弱指数异常率、优势偏向异常率、扫视波异常率较高。多因素Logistic回归结果显示，初诊时间≥14 d、听力损失≥100 dB、单侧反应减弱指数异常、优势偏向异常、扫视波异常是治疗无效的独立危险因素，患者接受虚拟现实前庭康复治疗对提高疗效具有积极作用。联合双温试验与视频头脉冲试验结果对疗效预测价值最优，联合预测参数C统计量(C-statistics)为0.861(95%CI:0.566~0.984，P＜0.001)，净重新分类指数(NRI)为0.735(95%CI:0.589~0.990，P＜0.001)，综合判别改善指数(IDI)为0.245(95%CI:0.144~0.647，P＜0.001)。结论 虚拟现实前庭康复治疗对提高全聋型突发性聋伴眩晕患者临床疗效具有积极作用，可改善前庭功能异常、减少眩晕持续时间、缓解眩晕症状和焦虑情绪，患者初诊时间≥14 d、听力损失≥100 dB、单侧反应减弱指数异常、优势偏向异常、扫视波异常是影响疗效的危险因素。

Abstract:

Objective To investigate the changes of vestibular function before and after virtual reality vestibular rehabilitation in patients with total deafness type sudden deafness with vertigo, and explore the influencing factors of therapeutic effect.Methods Seventy-five patients with total deafness type sudden deafness with vertigo who were initially diagnosed and treated with virtual reality vestibular rehabilitation from August 2019 to August 2023 in Haidian Hospital were selected as the experimental group, and 41 patients with total deafness with vertigo who did not undergo vestibular rehabilitation during the same period were taken as the control group. Average hearing thresholds of 0.25 to 8 KHz were retested after 14 days of treatment. An increase of more than 30 dB in impaired hearing was classified as the significant effect group (n=39), 15 dB to 30 dB as the effective group (n=35), and less than 15 dB as the ineffective group (n=42). Clinical data such as results of dual-temperature test and video head pulse test were collected before and after treatment. Multifactorial logistic regression was used to analyse the influencing factors of efficacy. The predictive value of vestibular function parameters for patient outcomes was assessed using R software.Results Before treatment, there was no significant difference in vestibular function between the two groups (all P>0.05). After treatment, compared with the control group, the vestibular-ocular reflex gain value of the experimental group was higher, while the vertigo duration, abnormal rate of saccade wave, abnormal rate of asymmetry ratio, abnormal rate of unilateral response attenuation index and abnormal rate of dominance bias were lower (all P<0.05). The treatment efficacy of the experimental group was higher than that of the control group (20.00% vs 2.44%, χ²=6.875, P=0.009). Except for the vertigo disorder scale-affection, all scores of the vertigo disorder scale and hospital anxiety and depression scale in the experimental group were significantly lower than those in the control group (all P<0.05). The rate of virtual reality vestibular rehabilitation, horizontal hemianopsia gain and posterior hemianopsia gain were lower in the ineffective group, whereas the age, time of initial consultation, hearing loss, the abnormal rates of unilateral response attenuation index, dominance bias, and saccade wave were higher. The multifactorial logistic regression showed that the time of initial consultation≥14 days, hearing loss≥100 dB, abnormal unilateral response attenuation index, abnormal dominance bias, and saccade wave were independent risk factors for treatment failure. Patients receiving virtual reality vestibular rehabilitation had a positive effect on improving outcomes. The combined dual-temperature test and video head pulse test had the best predictive value, with the joint predictive C-statistic being 0.861 (95%CI:0.566-0.984, P<0.001), the net reclassification index being 0.735 (95%CI:0.589-0.990,P<0.001) and the composite discrimination improvement index being 0.245 (95%CI:0.144-0.647, P<0.001).Conclusions Virtual reality vestibular rehabilitation has a positive effect on improving clinical outcomes in patients with total deafness type sudden deafness with vertigo. It can improve vestibular function, reduce vertigo duration, and relieve vertigo symptoms and anxiety. The time of initial diagnosis ≥14 days, hearing loss≥100 dB, abnormal unilateral response attenuation index, abnormal dominance bias and abnormal saccade wave are risk factors affecting the therapeutic effect.