Features of Importance in Nasal Endoscopy: Deriving a Meaningful Framework

Introduction: Uncertainty exists about the findings of importance in nasal endoscopy, including the specific abnormalities of interest in distinct clinical scenarios. We sought to identify concordance among clinicians with regards to features encountered during nasal endoscopy.

Methods: An online survey was sent in July 2023 to a diverse sample of 20 expert rhinologists in the United States, asking them to rank the importance of anatomical structures and pathological findings on nasal endoscopic examination with respect to five different cases (post-nasal drip, nosebleed, nasal polyps, nasal congestion, facial pressure). The survey was constructed using JotForm® software with a set of 4 identical questions for each case. Three additional questions were posed to assess the perceived anatomic boundaries of structures in the nasal cavity and the preferred behavior of a clinical decision support algorithm. Normalized attention scores (NAS) and weighted normalized attention scores (W-NAS) were computed to aggregate clinicians’ ranking preferences into a single value, ranging from 0 to 1, representing the perceived importance of examining the structure or finding.

Results: General concordance was found in the endoscopic structures and abnormal findings of importance within each case. The perceived features of importance differed between cases based on clinical presentation. In case one, the middle meatus (NAS, 0.73) was selected as the most important structure to examine when assessing for post-nasal drip, with mucus (W-NAS, 0.66) selected as the most important abnormal finding. The primary feature of interest for mucus was whether it was purulent or not (W-NAS, 0.67). Similar analyses were performed for features in each case.

Conclusions: Similarities in approach to anatomy and features of interest exist amongst clinicians who perform nasal endoscopy. These findings may help establish standardized examination and diagnostic criteria and improve diagnostic accuracy, augment instruction of novices, and inform the development of artificially intelligent endoscopic clinical decision support software.