遗传性血管水肿
医学
血管性水肿
逻辑回归
家族史
腹痛
内科学
接收机工作特性
曲线下面积
单变量分析
皮肤病科
多元分析
作者
Semra Demir,Müge Olgaç,Osman Ozan Yeğit,İlkim Deniz Toprak,Mehmet Erdem Çakmak,Merve Hörmet İğde,Nida Öztop,Pelin Korkmaz,Selda Çelik,Deniz Eyice Karabacak,Nevzat Kahveci,Işıl Göğem İmren,Faruk Erden,Raif Coşkun,Pelin Karadağ,Derya Ünal,Aslı Gelincik
摘要
Abstract Background/Aim Misdiagnosis of hereditary angioedema (HAE) leads to inappropriate management of the attacks. A scoring system that anticipates diagnosis can be beneficial for clinicians who are unfamiliar with angioedema. This study aims to develop a practical scoring system for use during acute attacks to predict HAE in patients with recurrent angioedema (RAE). Method To predict HAE, nine HAE experts unanimously identified five predictive items (PIs); absence of urticaria, presence of abdominal pain episodes, family history, early onset of attacks and previous unresponsiveness to anti‐histaminergic treatments. The researchers questioned 106 patients with HAE and 155 patients with mast cell‐mediated angioedema (MMAE) about PIs. A score was attributed to each significant PI based on OR values obtained through logistic regression analysis. The cut‐off point for the prediction of HAE and its sensitivity and specificity were determined by ROC curve analysis. Results In a univariate analysis, all items showed significant differences between HAE and MMAE patients. Regression analysis attributed scores as follows: 23 points for the absence of urticaria, 11 points for the abdominal pain episodes, 9 points for family history, and 53 points for unresponsiveness to antihistaminergic treatments. No score was attributed to early onset of age ( p > 0.05). The ROC analysis revealed an area under the curve of 0.990, with a total score of ≥38 demonstrating the best sensitivity (96.4%) and specificity (96.1%). Conclusions HAEps is a valuable tool for diagnosing HAE in patients with RAE. A score of 38 or more indicates the possible presence of HAE with substantial sensitivity and specificity.
科研通智能强力驱动
Strongly Powered by AbleSci AI