Predictive value of computed tomography for eosinophilic chronic rhinosinusitis with nasal polyps in different histopathologic criteria

计算机断层扫描对不同组织病理学标准下嗜酸性慢性鼻窦炎伴鼻息肉的预测价值

阅读:1

Abstract

BACKGROUND: Computed tomography (CT) scan is a good and noninvasive prediction tool for eosinophilic chronic rhinosinusitis with nasal polyps (eCRSwNP), and how to choose the appropriate CT parameter is crucial-especially because there have not been unanimous histopathologic criteria to diagnose eCRSwNP. OBJECTIVE: This study sought to select the suitable CT parameter to predict eCRSwNP in different criteria. METHOD: This retrospective study included 147 CRSwNP patients who underwent a sinus CT scan and histopathological examination. Nine common CT parameters and 5 representative criteria of eCRSwNP (>5/HPF, >10/HPF, >70/HPF, >10%/HPF and >20%/HPF) were adopted. Logistic regression analysis and ROC analysis were performed to evaluate the predictive value of CT parameters. RESULT: Among the 9 CT parameters, olfactory cleft (OC) score, ethmoid sinus and maxillary sinus ratio (E/M ratio), and posterior ethmoid (PE) score were significantly associated with eCRSwNP. In the 5 representative criteria of eCRSwNP, all the results showed that the OC score was not only the significant predictor in the univariate analysis, but also the most significant one in the multivariate analysis. Meanwhile, both OC score and E/M ratio were included in the multivariable logistic regression model. The clinically convenient cut-off points of model were OC score >2 and E/M ratio >2.5. CONCLUSION: The OC score, E/M ratio, and PE score were significantly associated with eosinophilia of nasal polyp tissue. Therein, OC score was the best marker to predict eCRSwNP among the 5 representative criteria. The combination of OC score and E/M ratio can obtain a better predictive value of eCRSwNP.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。