A reliable grading system allows the clinician to classify disease severity, monitor progress and evaluate treatment efficacy. There is no currently accepted grading system for vocal process granuloma of the larynx.
To evaluate the reliability of a new grading system for vocal process granuloma.
All vocal process granuloma images from a digital laryngeal image library were abstracted. Granulomas were graded on a one to four system, as follows: grade one, sessile, non-ulcerative granuloma limited to vocal process; grade two, pedunculated or ulcerated granuloma limited to vocal process; grade three, granuloma extending past vocal process but not crossing midline of airway in fully abducted position; and grade four, granuloma extending past vocal process and past the midline of the airway in the fully abducted position. The granulomas were additionally graded A if unilateral and B if bilateral. Two laryngologists and two otolaryngology residents rated the granulomas on two separate occasions. Intra- and inter-observer reliability was evaluated with the kappa (κ) test statistic.
Thirty-five vocal process granulomas were identified. The percentage intra-observer agreement for the two laryngologists was 97 and 100 per cent (κ = 0.94 and 1.00, respectively). The percentage inter-observer agreement between the two laryngologists was 91 per cent (κ = 0.83). The percentage intra-observer agreement for the two residents was 89 and 91 per cent (κ = 0.83 and 0.77, respectively). The percentage inter-observer agreement between the two residents was 83 per cent (κ = 0.67).
The proposed grading system for vocal process granuloma displayed excellent intra- and inter-observer reliability among residents and experienced laryngologists.