Intra- and Interreader Variability of Orbital Volume Quantification Using 3D Computed Tomography for Reconstructed Orbital Fractures

• Autor(a) principal: Gomes de Oliveira, P
• Data de Publicação: 2019
• Outros Autores: Perry da Câmara, C, Valejo Coelho, P
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.17/3617
• topic: Humans; Imaging, Three-Dimensional; Orbit; Reconstructive Surgical Procedures; Reproducibility of Results; Retrospective Studies; Tomography, X-Ray Computed; Enophthalmos; Orbital Fractures; HSJ CIR MAXFAC; HSJ NRAD

Purpose: Management of orbital fractures continues to present some difficulties, particularly regarding the prediction of late complications. Radiographic assessment provides a detailed evaluation, but the results lack consistency to be considered a standard factor in the decision-making process. Studies focusing on reliability of post-operative imaging are lacking. Materials and methods: We performed a retrospective study using patients from a major trauma center with unilateral orbital floor fracture who underwent surgery. Using three-dimensional volume assessment software, we performed a volume calculation and determined the intra- and interreader variation by intraclass correlation coefficient analysis. Results: Twenty-four orbits were assessed. Mean orbital volume (SD) was 24.02 (2,43) cm3 for reader 1 and 24.08 (2,51) cm3 for reader 2. The intraclass correlation coefficient (95% CI) was 0.95 (0.91-0.98) between readers and 0.96 (0.91-0.98) for intra-reader variability. Normal and reconstructed orbits assessed separately also showed very high correlation coefficient for both intra- and inter-subject variability. Conclusion: Results show an almost perfect agreement of volume assessment between readers. The presence of reconstruction material does not seem to add variability. Although reproducible and reliable, radiological volume assessments have not yet shown a clear correlation with clinical outcomes and post-operative management decisions should be based mainly on clinical findings.