开发和验证一种非侵入性预测模型，用于识别嗜酸性粒细胞性哮喘。

Development and validation of a noninvasive prediction model for identifying eosinophilic asthma.

机构信息

Department of Respiratory and Critical Care Medicine, Clinical Research Center for Respiratory Disease, West China Hospital, Sichuan University, Chengdu, PR China; Department of Respiratory and Critical Care Medicine, First Affiliated Hospital of Kunming Medical University, Kunming, PR China; Laboratory of Pulmonary Immunology and Inflammation, Frontiers Science Center for Disease-related Molecular Network, Sichuan University, Chengdu, PR China.

Pneumology Group, Department of Integrated Traditional Chinese and Western Medicine, West China Hospital, Sichuan University, Chengdu, PR China.

出版信息

Respir Med. 2022 Sep;201:106935. doi: 10.1016/j.rmed.2022.106935. Epub 2022 Jul 19.

DOI:10.1016/j.rmed.2022.106935

PMID:35926430

Abstract

BACKGROUND

Identification of eosinophilic asthma (EA) using sputum analysis is important for disease monitoring and individualized treatment. But it is laborious and technically demanding. We aimed to develop and validate an effective model to predict EA with multidimensional assessment (MDA).

METHODS

The asthma patients who underwent a successful sputum induction cytological analysis were consecutively recruited from March 2014 to January 2021. The variables assessed by MDA were screened by least absolute shrinkage and selection operator (LASSO) and logistic regression to develop a nomogram and an online web calculator. Validation was performed internally by a bootstrap sampling method and externally in the validation cohort. Diagnostic accuracy of the model in different asthma subgroups were also investigated.

RESULTS

In total of 304 patients in the training cohort and 95 patients in the validation cohort were enrolled. Five variables were identified in the EA prediction model: gender, nasal polyp, blood eosinophils, blood basophils and FeNO. The C-index of the model was 0.86 (95% CI: 0.81-0.90) in the training cohort and 0.84 (95% CI: 0.72-0.89) in the validation cohort. The calibration curve showed good agreement between the prediction and actual observation. The decision curve analysis (DCA) also demonstrated that the EA prediction model was clinically beneficial. An online publicly available web calculator was constructed (https://asthmaresearcherlimin.shinyapps.io/DynNomapp/).

CONCLUSION

We developed and validated a multivariable model based on MDA to help the diagnosis of EA, which has good diagnostic performance and clinical practicability. This practical tool may be a useful alternative for predicting EA in the clinic.

摘要

背景

通过痰分析鉴定嗜酸性粒细胞性哮喘（EA）对于疾病监测和个体化治疗很重要。但是，它既费力又技术要求高。我们旨在开发和验证一种使用多维评估（MDA）有效预测 EA 的模型。

方法

从 2014 年 3 月至 2021 年 1 月，连续招募了接受成功痰诱导细胞学分析的哮喘患者。通过最小绝对收缩和选择算子（LASSO）和逻辑回归筛选 MDA 评估的变量，以开发列线图和在线网络计算器。通过 bootstrap 抽样方法进行内部验证，并在验证队列中进行外部验证。还研究了该模型在不同哮喘亚组中的诊断准确性。

结果

共纳入了 304 例训练队列和 95 例验证队列患者。EA 预测模型中确定了五个变量：性别、鼻息肉、血嗜酸性粒细胞、血嗜碱性粒细胞和 FeNO。模型在训练队列中的 C 指数为 0.86（95%CI：0.81-0.90），在验证队列中的 C 指数为 0.84（95%CI：0.72-0.89）。校准曲线显示预测与实际观察之间具有良好的一致性。决策曲线分析（DCA）也表明 EA 预测模型具有临床获益。构建了一个在线公共可用的网络计算器（https://asthmaresearcherlimin.shinyapps.io/DynNomapp/）。