Watanabe Tetsuo, Suzuki Masashi
Department of Otolaryngology, Head and Neck Surgery, Oita University Faculty of Medicine, 1-1, Idaigoka, Hasama-machi, Yufu-city, Oita 879-5593, Japan.
Ear Nose Throat J. 2018 Jul;97(7):E36-E40. doi: 10.1177/014556131809700706.
We performed a cluster analysis to classify the audiogram shape in patients with idiopathic sudden sensorineural hearing loss (ISSNHL). We also investigated whether the audiogram shape is a prognostic indicator in the management of ISSNHL. A total of 115 inpatients with ISSNHL treated between 2001 and 2010 were analyzed. The data collected included age, sex, duration of hearing loss at the time of treatment, and the presence or absence of tinnitus, vertigo, diabetes, nystagmus, and canal paresis. A hierarchical cluster analysis was performed using the hearing threshold for each frequency on audiograms as variables. A logistic regression model was used for the prognostic analysis. The audiogram shape was classified into four clusters: (1) crossing horizontally pattern of all tones; (2) up-sloping pattern of low-tone loss; (3) deaf pattern; and (4) down-sloping pattern of high-tone loss. The age of the patient, presence of canal paresis, and audiogram shape showed statistically significant relationships with hearing improvement. The audiogram shape based on the cluster analysis demonstrated a significant relationship with hearing improvement in patients with ISSNHL. Further studies are needed to elucidate the underlying etiology of each audiogram shape.
我们进行了聚类分析,以对特发性突发性感音神经性听力损失(ISSNHL)患者的听力图形状进行分类。我们还研究了听力图形状是否是ISSNHL治疗中的一个预后指标。对2001年至2010年间接受治疗的115例ISSNHL住院患者进行了分析。收集的数据包括年龄、性别、治疗时听力损失的持续时间,以及耳鸣、眩晕、糖尿病、眼球震颤和管麻痹的有无。使用听力图上每个频率的听力阈值作为变量进行层次聚类分析。采用逻辑回归模型进行预后分析。听力图形状分为四类:(1)所有音调水平交叉型;(2)低音调损失上升型;(3)全聋型;(4)高音调损失下降型。患者年龄、管麻痹的存在以及听力图形状与听力改善呈统计学显著关系。基于聚类分析的听力图形状与ISSNHL患者的听力改善呈显著关系。需要进一步研究以阐明每种听力图形状的潜在病因。