Development of a Computed Tomography Imaging Grading Scale for Pneumolabyrinth

Objective: To develop a reliable prognostic grading scale for pneumolabyrinth detected on CT. Design: Retrospective case series. Setting: Tertiary care institution. Participants: A total of 39 patients with pneumolabyrinth identified through CT imaging from 2010 to 2019 were included. The main outcome measures involved reviewing clinical outcomes and radiographic features. The extent of pneumolabyrinth was assessed and graded as low, intermediate, or high risk for permanent hearing loss, by expert radiologists. Main Outcomes and Measures: The primary outcome of the study was the association between the extent of pneumolabyrinth, as graded by a novel prognostic scale (low, intermediate, or high risk), and the likelihood of permanent hearing loss. Hearing outcomes and vestibular symptoms were assessed through retrospective chart review, while the grading of pneumolabyrinth extent was based on expert radiologic evaluations of CT imaging. The hypothesis regarding the predictive utility of the grading scale was formulated after data collection. Reliability of the grading scale was measured using Cohen’s kappa and Fleiss’ kappa. Results: The patient cohort had a median age of 30 years (range = 8-84 years), 33% of whom were female. Trauma, predominantly temporal bone fractures (87.5%), was the most common cause. The grading scale demonstrated excellent intra-rater reliability (Cohen’s kappa = .95) and substantial inter-rater reliability (Fleiss’ kappa = .729). High risk pneumolabyrinth cases had a higher rate of profound hearing loss (90.9%) compared low or intermediate risk cases (46.2%, P = .0136). Dizziness was not associated with pneumolabyrinth grade, but correlated with involvement of semicircular canals (72.7%) compared to patients with pneumolabyrinth limited to vestibule and/or cochlea (31.3%, P = .008). Presence of pneumolabyrinth isolated to the vestibular labyrinth was significantly associated with dizziness symptoms (P = .047). Conclusion: Our novel grading scale reliably predicted clinical outcomes, providing prognostic utility in hearing status and potential guidance in surgical approach.