The interobserver reliability of the diagnosis and classification of scaphoid fractures using high-resolution peripheral quantitative CT. (April 2020)

• Record Type: Journal Article
• Authors: Daniels, Anne M.; Wyers, Caroline E.; Janzing, Heinrich M. J.; Sassen, Sander; Loeffen, Daan; Kaarsemaker, Sjoerd; van Rietbergen, Bert; Hannemann, Pascal F. W.; Poeze, Martijn; van den Bergh, Joop P.
• Is Part Of: Bone & joint journal. Volume 102B:Number 4(2020)
• Page Start: 478
• Page End: 484
• Publication Date: 2020-04
• Subjects: Interobserver agreement -- Interobserver variability -- Reliability -- Scaphoid fracture -- Diagnosis -- Imaging -- High-resolution peripheral quantitative CT -- HR-pQCT; Bones -- Surgery -- Periodicals; Joints -- Surgery -- Periodicals; Orthopedic surgery -- Periodicals; 617.47005
• DOI: 10.1302/0301-620X.102B4.BJJ-2019-0632.R3
• Languages: English
• ISSNs: 2049-4394

Abstract

Abstract : Aims: Besides conventional radiographs, the use of MRI, CT, and bone scintigraphy is frequent in the diagnosis of a fracture of the scaphoid. However, which techniques give the best results remain unknown. The investigation of a new imaging technique initially requires an analysis of its precision. The primary aim of this study was to investigate the interobserver agreement of high-resolution peripheral quantitative CT (HR-pQCT) in the diagnosis of a scaphoid fracture. A secondary aim was to investigate the interobserver agreement for the presence of other fractures and for the classification of scaphoid fracture. Methods: Two radiologists and two orthopaedic trauma surgeons evaluated HR-pQCT scans of 31 patients with a clinically-suspected scaphoid fracture. The observers were asked to determine the presence of a scaphoid or other fracture and to classify the scaphoid fracture based on the Herbert classification system. Fleiss kappa statistics were used to calculate the interobserver agreement for the diagnosis of a fracture. Intraclass correlation coefficients (ICCs) were used to assess the agreement for the classification of scaphoid fracture. Results: A total of nine (29%) scaphoid fractures and 12 (39%) other fractures were diagnosed in 20 patients (65%) using HR-pQCT across the four observers. The interobserver agreement was 91% for the identification of a scaphoid fracture (95% confidence interval (CI) 0.76 to 1.00) and 80% for other fractures (95% CI 0.72 to 0.87). The mean ICC for the classification of a scaphoid fracture in the seven patients diagnosed with scaphoid fracture by all four observers was 73% (95% CI 0.42 to 0.94). Conclusion: We conclude that the diagnosis of scaphoid and other fractures is reliable when using HR-pQCT in patients with a clinically-suspected fracture. Cite this article: Bone Joint J 2020;102-B(4):478–484.