Modified Clavien-Dindo Classification for Adverse Events in Otolaryngology-Head and Neck Surgery.

Importance: Adverse events in otolaryngology-head and neck surgery (OHNS) are inconsistently reported. The Clavien-Dindo Classification has been modified and validated in several surgical subspecialties but has yet to be validated in OHNS. Objective: To create and validate a modified Clavien-Dindo Classification for OHNS that reflects the severity and duration of harm for complications experienced by patients undergoing OHNS. Design, Setting, and Participants: This is a multicenter survey study of subject matter experts (SMEs) from OHNS subspecialties from academic institutions within the US. A literature review identified publications reporting surgical complications in OHNS in the English language between January 2019 and April 2020. Complications were summarized and used to inform a modified Clavien-Dindo Classification. Feedback from internal SMEs was incorporated. Thirty-five clinical vignettes were created, and an electronic survey was sent to SMEs at outside institutions. Each SME was randomly assigned to 1 of 4 groups to evaluate 8 to 9 vignettes. The survey was conducted in September 2023, and data were analyzed from February to May 2024. Exposure: Participation in the electronic survey. Main Outcomes and Measures: Vignette scores were assessed and the Comprehensive Complication Index (CCI) calculator was used to assess cumulative morbidity. The scale is based on the grading by the Clavien-Dindo Classification and reflects overall morbidity on a scale from 0 (no complication) to 100 (death). Variability of the CCIs between reviewers, the primary outcome, was summarized with the SD. The percentage of reviewers with a high (≥26.2) or low (<26.2) CCI within a vignette was reported. Reliability was summarized using Krippendorff α. Results: Of 106 SMEs, 43 (response rate, 40.6%) completed at least 1 vignette. Eighteen of 35 vignettes (51.4%; 95% CI, 34.9%-68.0%) had unanimous high CCI, and 8 of 35 vignettes (22.9%; 95% CI, 8.9%-36.8%) had unanimous low CCI (overall categorical agreement, 74.3%; 95% CI, 59.8%-88.8%). On average, the Krippendorff α was 0.74, indicating moderate reliability. A total of 21 850 OHNS complications were reviewed and categorized as surgical or medical with 12 including subcategorization for specificity. Each category was then assigned a modified OHNS Clavien Dindo classification grade. Conclusions and Relevance: In this survey study of OHNS experts, a modified OHNS Clavien-Dindo Classification was developed and validated using case vignettes. The modified Clavien-Dindo Classification provided a granular subcategorization for each complication category and grade, duration of harm, and an overall summary complication score (CCI). Results suggest the modified OHNS Clavien-Dindo Classification reliably differentiated between high-grade and low-grade OHNS complications and could be utilized for assessing surgical outcomes.