Chul-Hyun Cho1, Joo Han Oh2, Gu-Hee Jung3, Gi-Hyuk Moon4, In Hyeok Rhyou5, Jong Pil Yoon6, Ho Min Lee7. 1. Pain Research Center, Department of Orthopedic Surgery, Dongsan Medical Center, Keimyung University School of Medicine, Daegu, South Korea oscho5362@dsmc.or.kr. 2. Department of Orthopedic Surgery, Bundang Hospital, Seoul National University College of Medicine, Seongnam, South Korea. 3. Department of Orthopedic Surgery, Gospel Hospital, Kosin University School of Medicine, Busan, South Korea. 4. Department of Orthopedic Surgery, Pohang St Mary's Hospital, Pohang, South Korea. 5. Upper Extremity and Microsurgery Center, Department of Orthopedic Surgery, Pohang Semyeng Christianity Hospital, Pohang, South Korea. 6. Department of Orthopedic Surgery, Kyungpook National University Hospital, Daegu, South Korea. 7. Department of Orthopedic Surgery, Gyeongju Hospital, Dongguk University College of Medicine, Gyeongju, South Korea.
Abstract
BACKGROUND: As there is substantial variation in the classification and diagnosis of lateral clavicle fractures, proper management can be challenging. Although the Neer classification system modified by Craig has been widely used, no study has assessed its validity through inter- and intrarater agreement. PURPOSE: To determine the inter- and intrarater agreement of the modified Neer classification system and associated treatment choice for lateral clavicle fractures and to assess whether 3-dimensional computed tomography (3D CT) improves the level of agreement. STUDY DESIGN: Cohort study (diagnosis); Level of evidence, 3. METHODS: Nine experienced shoulder specialists and 9 orthopaedic fellows evaluated 52 patients with lateral clavicle fractures, completing fracture typing according to the modified Neer classification system and selecting a treatment choice for each case. Web-based assessment was performed using plain radiographs only, followed by the addition of 3D CT images 2 weeks later. This procedure was repeated 4 weeks later. Fleiss κ values were calculated to estimate the inter- and intrarater agreement. RESULTS: Based on plain radiographs only, the inter- and intrarater agreement of the modified Neer classification system was regarded as fair (κ = 0.344) and moderate (κ = 0.496), respectively; the inter- and intrarater agreement of treatment choice was both regarded as moderate (κ = 0.465 and 0.555, respectively). Based on the plain radiographs and 3D CT images, the inter- and intrarater agreement of the classification system was regarded as fair (κ = 0.317) and moderate (κ = 0.508), respectively; the inter- and intrarater agreement of treatment choice was regarded as moderate (κ = 0.463) and substantial (κ = 0.623), respectively. There were no significant differences in the level of agreement between the plain radiographs only and plain radiographs plus 3D CT images for any κ values (all P > .05). CONCLUSION: The level of interrater agreement of the modified Neer classification system for lateral clavicle fractures was fair. Additional 3D CT did not improve the overall level of interrater or intrarater agreement of the modified Neer classification system or associated treatment choice. To eliminate a common source of disagreement among surgeons, a new classification system to focus on unclassifiable fracture types is needed.
BACKGROUND: As there is substantial variation in the classification and diagnosis of lateral clavicle fractures, proper management can be challenging. Although the Neer classification system modified by Craig has been widely used, no study has assessed its validity through inter- and intrarater agreement. PURPOSE: To determine the inter- and intrarater agreement of the modified Neer classification system and associated treatment choice for lateral clavicle fractures and to assess whether 3-dimensional computed tomography (3D CT) improves the level of agreement. STUDY DESIGN: Cohort study (diagnosis); Level of evidence, 3. METHODS: Nine experienced shoulder specialists and 9 orthopaedic fellows evaluated 52 patients with lateral clavicle fractures, completing fracture typing according to the modified Neer classification system and selecting a treatment choice for each case. Web-based assessment was performed using plain radiographs only, followed by the addition of 3D CT images 2 weeks later. This procedure was repeated 4 weeks later. Fleiss κ values were calculated to estimate the inter- and intrarater agreement. RESULTS: Based on plain radiographs only, the inter- and intrarater agreement of the modified Neer classification system was regarded as fair (κ = 0.344) and moderate (κ = 0.496), respectively; the inter- and intrarater agreement of treatment choice was both regarded as moderate (κ = 0.465 and 0.555, respectively). Based on the plain radiographs and 3D CT images, the inter- and intrarater agreement of the classification system was regarded as fair (κ = 0.317) and moderate (κ = 0.508), respectively; the inter- and intrarater agreement of treatment choice was regarded as moderate (κ = 0.463) and substantial (κ = 0.623), respectively. There were no significant differences in the level of agreement between the plain radiographs only and plain radiographs plus 3D CT images for any κ values (all P > .05). CONCLUSION: The level of interrater agreement of the modified Neer classification system for lateral clavicle fractures was fair. Additional 3D CT did not improve the overall level of interrater or intrarater agreement of the modified Neer classification system or associated treatment choice. To eliminate a common source of disagreement among surgeons, a new classification system to focus on unclassifiable fracture types is needed.
Authors: Evi Fleischhacker; Georg Siebenbürger; Johannes Gleich; Wolfgang Böcker; Fabian Gilbert; Tobias Helfen Journal: J Clin Med Date: 2022-09-24 Impact factor: 4.964