Sinonasal Outcome Scores And Imaging: A Concurrent Assessment Of Factors Influencing Their Association

Objective The overall discriminatory ability of validated instrument scores for computed tomography (CT) findings of chronic rhinosinusitis has limitations and may be modified by multiple factors. To support optimal methods for assessment, we studied which factors could influence this relationship, including the concurrent impact of multiple discrete CT scoring mechanisms, colocalized imaging findings, and nasal comorbid conditions. Study Design Observational outcomes study. Setting Academic medical center. Methods Patients with sinonasal complaints who completed the 22-item Sinonasal Outcome Test (SNOT-22) and underwent CT were included. Multivariate ordinal regression was utilized to assess associations. CT data were quantified with the Lund-Mackay system, Zinreich system, and a direct measure of maximal mucosal thickness. The impact of incidental findings (mucous retention cysts, periapical dental disease) and nasal comorbid conditions was also assessed. Results A total of 233 patients were included. SNOT-22 nasal scores were significantly associated with CT results when those with incidental findings were excluded, regardless of the radiologic scoring mechanism utilized: Lund-Mackay regression coefficient, 0.321 (P = .046); Zinreich, 0.340 (P = .033); and maximum mucosal thickness, 0.316 (P = .040). This relationship subsided when incidental findings were present. SNOT-22 overall scores, sleep scores, and psychological domain scores had no significant association with imaging results, regardless of radiologic scoring system utilized. Nasal comorbid conditions had inconsistent associations. Conclusions SNOT-22 nasal domain scores were associated with all 3 radiologic scoring systems when incidental findings were absent but not when they were present. Delineating the presence or absence of these colocalized findings affected the relationship between SNOT-22 scores and radiological results, beyond other concurrent factors.