Côrte Filipa Camacho, Orfao Tiago, Dias Cláudia Camila, Moura Carla Pinto, Santos Margarida
Department of Otorhinolaryngology, Hospital de São João EPE, Porto, Portugal; University of Porto Medical School, Porto, Portugal.
Department of Otorhinolaryngology, Hospital de São João EPE, Porto, Portugal; University of Porto Medical School, Porto, Portugal.
Auris Nasus Larynx. 2018 Jun;45(3):471-475. doi: 10.1016/j.anl.2017.07.021. Epub 2017 Aug 24.
Analyse and compare the characteristics of patients with epistaxis admitted to the otolaryngology emergency department with those provided by a control group. Establish a model to identify epistaxis predictive factors.
Prospective analysis of 283 consecutive adults, admitted to the otolaryngology emergency department of a tertiary referral centre between 25 January and 25 February 2014. Comparison of gender, age, co-morbidities, usual medication, history of epistaxis or nasal trauma, presence of septal deviation and blood pressure value on admission, between the elements that were admitted to the emergency due to epistaxis (group 1) and a group composed of patients with other symptoms (group 2). Intergroup variations were analyzed using t student and chi-square tests. Multivariate logistic regression and a receiver operating characteristic curve were used to establish a predictive model and test its suitability.
Male gender (OR=2.57, 95% CI 1.1-6.0, p=0.029), older age (OR=1.03, 95% CI 1.0-1.1, p=0.002), existence of peripheral vascular disease (OR=13.47, 95% CI 1.9-95.3, p=0.009), cardiovascular disease (OR=3.91, 95% CI 1.6-9.7, p=0.003) and previous history of epistaxis (OR=5.53, 95% CI 2.5-12.1, p<0.001) were predictors of epistaxis when adjusted for the presence of elevated blood pressure, history of hypertension, cerebrovascular disease and chronic use of anticoagulants or antiplatelet drugs. The model revealed a good applicability (area under the curve of 0.852).
The only predictive factors of admission to the emergency department due to epistaxis were male gender, older age, peripheral vascular disease, cardiovascular disease and history of epistaxis.
分析并比较耳鼻喉科急诊科收治的鼻出血患者与对照组患者的特征。建立一个识别鼻出血预测因素的模型。
对2014年1月25日至2月25日期间在一家三级转诊中心的耳鼻喉科急诊科连续收治的283名成年人进行前瞻性分析。比较因鼻出血入院的患者（第1组）和由有其他症状的患者组成的一组（第2组）在性别、年龄、合并症、常用药物、鼻出血或鼻外伤史、鼻中隔偏曲情况及入院时血压值等方面的差异。采用t检验和卡方检验分析组间差异。使用多因素逻辑回归和受试者工作特征曲线建立预测模型并检验其适用性。
在调整了血压升高、高血压病史、脑血管疾病以及长期使用抗凝药或抗血小板药物的因素后，男性（比值比=2.57，95%可信区间1.1 - 6.0，p = 0.029）、年龄较大（比值比=1.03，95%可信区间1.0 - 1.1，p = 0.002）、存在外周血管疾病（比值比=13.47，95%可信区间1.9 - 95.3，p = 0.009）、心血管疾病（比值比=3.91，95%可信区间1.6 - 9.7，p = 0.003）以及既往鼻出血史（比值比=5.53，95%可信区间2.5 - 12.1，p < 0.001）是鼻出血的预测因素。该模型显示出良好的适用性（曲线下面积为0.852）。
因鼻出血而入住急诊科的唯一预测因素是男性、年龄较大、外周血管疾病、心血管疾病和鼻出血史。
