A systematic literature review of tibial plateau fractures: What classifications are used and how reliable and useful are they?

• Published in: Injury 2018-03, Vol.49 (3), p.473-490
• Main Authors: Millar, Stuart C., Arnold, John B., Thewlis, Dominic, Fraysse, François, Solomon, Lucian B.
• Format: Article
• Language: English
• Subjects: Classification; Imaging; Morphology; Reliability; Tibial plateau fracture
• ISSN: 0020-1383; 1879-0267
• DOI: 10.1016/j.injury.2018.01.025

Classification systems such as the Schatzker and AO/OTA have been proposed for standardised assessment of tibial plateau fractures and to guide clinical decision making. However, there has been no comprehensive literature review of all classification systems for tibial plateau fractures, including assessment of their reliability. The aim of this systematic review was to identify and appraise previously established classification systems for tibial plateau fractures and determine their reliability for fracture classification. Six databases were searched from inception until October 2016. Classification systems for tibial plateau fractures were identified. No restriction was placed on imaging modality (plain film X-ray, CT, MRI). Data synthesis was performed to identify common features of the systems, their prevalence within the literature and studies of intra and inter-rater reliability of fracture classification using Kappa coefficient (κ). Thirty-eight classification systems were identified, five of which were a sub-classification of a single fracture type from a previous tool. The Schatzker and AO/OTA classification systems were the most commonly reported. Of the tools identified only five have been tested for inter and intra-observer reliability (Schatzker, AO/OTA, Duparc, Hohl and Luo). Reliability of more simplistic classification systems, such as that by Luo et al. (three-column) was typically high (intra-κ = 0.67–0.81, inter-κ = 0.71–0.87), but with the disadvantage of providing less information on fracture patterns and morphology. Intra and inter-observer reliability using plain film X-ray was frequently moderate (κ = 0.40–0.60), with 2D and 3D CT typically improving reliability of classification. Only 11 of the 32 complete classification systems identified association of fracture classification with clinical outcome. Frequently used systems for classification of tibial plateau fractures display moderate intra and inter-observer reliability. More sophisticated imaging modalities such as 2D and 3D CT typically improve reliability estimates. Using fracture classification based on imaging findings to predict clinical outcome was not a commonly reported goal of newly developed systems. More detailed assessment of fracture patterns and morphology, in conjunction with information on surgical fixation, may be desirable for predicting outcomes and to guide clinical decision making.