Reproducibility of digital measurements of lower-limb deformity on plain radiographs and agreement with CT measurements

Abstract

IntroductionFive angles (HKA, HKS, alpha, beta, tibial slope) are used for goniometry in total knee arthroplasty. The reproducibility of the measurement of these angles has been assessed on plain and digitized x-rays, but to our knowledge, this has not been confirmed on x-rays taken on the PACS system and they have not been compared to computed tomography (CT) measurements, the reference for angle measurement. This prospective study aimed to: (1) evaluate the inter- and intrarater reliability of the measurement of these angles on digital x-rays taken on a PACS; (2) determine the agreement of these measurements with those obtained using a CT protocol. HypothesisThe measurements of these five angles on digitized radiographs are reproducible and in agreement with CT values. Material and methodsForty-two patients suffering from knee osteoarthritis and scheduled for total knee arthroplasty were included in the study. Each patient had a PACS digitized x-ray and a CT intended to produce patient-specific instrumentation (Symbios, Yverdon, Switzerland) including measurements of the angles evaluated. Four senior orthopaedic surgeon-raters measured all the angles twice. Inter- and intrarater reliability was then calculated as well as the agreement between the second measurement of each rater and the CT measurement using interclass correlation and kappa coefficients (data provided as means and 95% confidence intervals). ResultsThe inter- and intrarater reliability values were excellent for the HKA, alpha, and beta angles (with, respectively, a coefficient of 0.99 [0.97–0.99], 0.84 [0.76–0.9], and 0.94 [0.86–0.96] interrater reliability and 0.98 [0.96–0.99], 0.86 [0.75–0.92], and 0.65 [0.44–0.8] intrarater reliability). Interrater reliability was low for HKS and tibial slope angles (coefficients all<0.4 for interrater reliability and <0.7 for intrarater reliability). The x-ray/CT agreement was very good for the HKA, alpha, and beta angles (0.81 [0.67–0.99], 0.74 [0.56–0.91], and 0.74 [0.45–0.92], respectively) and low for the HKS and tibial slope angles (all<0.45). Discussion/conclusionThe HKA, alpha, and beta angles were reproducible for digital radiographs and showed good agreement with CT measurements. HKS and tibial slope angles should be used with greater caution, and other navigation methods or patient-specific instrumentation should be explored. Level of evidenceLevel III, prospective, comparative diagnostic case-control study.