Premium
Reproducibility of volume measurements of vestibular schwannomas – a preliminary study
Author(s) -
Cross J.J.,
Baguley D.M.,
Antoun N.M.,
Moffat D.A.,
Prevost A.T.
Publication year - 2006
Publication title -
clinical otolaryngology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.914
H-Index - 68
eISSN - 1749-4486
pISSN - 1749-4478
DOI - 10.1111/j.1749-4486.2006.01161.x
Subject(s) - medicine , reproducibility , schwannoma , vestibular system , reliability (semiconductor) , nuclear medicine , audiology , radiology , statistics , mathematics , power (physics) , physics , quantum mechanics
Objective: A preliminary study to investigate the intra‐observer and inter‐observer variability of measurements of vestibular schwannoma volume using an area‐tracing and linear dimension measurement method. Design: Prospective blinded measurements by two observers (one a consultant and one a sub‐specialty trainee in neuroradiology). Setting: Ambulatory patients with known vestibular schwannoma attending a tertiary referral centre. Participants: Twenty‐three patients with 26 vestibular schwannomas aged 29–80 years old. Main outcome measure: Quantification of the variability in measured volume due to intra‐ and inter‐observer measurement differences, expressed as reliability coefficients within which 95% of repeated measurements are calculated to lie from each other in relative percentage terms. Results: For the linear measurement method, intraobserver variability was calculated to have a reliability coefficient of 65% and for interobserver variability this was 155%. For the area tracing method, the corresponding coefficients were 26% for intraobserver variability and 44% for interobserver variability. Conclusions: Volume measurements in vestibular schwannoma are variable even when the measurements are made by the same observer on identical images. The area tracing method, commonly regarded as the gold standard, is less variable than a linear method but still introduces more variability than commonly realized (for 95% of patients, repeated measurements by the same observer lie within around 25% of each other). Manual area tracing is not widely used in practice because it is time consuming and reliable automated methods are eagerly awaited.