Intraobserver and Interobserver Reliability of Cone Beam Weightbearing Semiautomatic Three- Dimensional Measurements in Symptomatic Pes Cavovarus

CATEGORY: Basic Sciences/Biologics, Hindfoot INTRODUCTION/PURPOSE: Pes cavovarus is a three-dimensional complex foot deformity variably involving a varus hindfoot, high longitudinal arch and forefoot adduction. Two-dimensional radiographs may be flawed by rotational bias and operator-related bias, while standard computed tomography (CT) cannot be perfomed in loading conditions. Three-dimensional (3D) cone beam weightbearing computed tomography (WBCT) may overcome these drawbacks, obtaining physiological weightbearing images with low-dose radiation. New 3D semiautomatic tools have been proposed to measure hindfoot alignment, but reliability in pes cavovarus has never been tested. The aim of this study was to assess intra and interobserver reliability of 3D biometrics on WBCT imaging in pes cavovarus. Our hypothesis was that foot and ankle offset, calcaneal offset and hindfoot angle were reliable measures regardless of type and severity of deformity. METHODS: Cone beam WBCT anonymised datasets from 34 pes cavovarus (PC) (17 neurological, 17 non-neurological) and 17 normal feet were retrospectively reviewed. All WBCTs were performed during routine investigation. Inclusion criteria consisted of no previous ipsilateral foot/ankle surgery and ability to heel weightbear. Three foot and ankle surgeons (blinded to patient diagnosis) independently measured the following variables on 51 feet: foot and ankle offset (FAO), calcaneal offset (CO) and hindfoot angle (HA) using dedicated software. Each observer repeated all measurements two-weeks apart. The Shapiro–Wilk test was used to assess normality of data distribution. The Pearson or Spearman correlation test and Intraclass Correlation Coefficients (ICCs) and were used to assess intra and interobserver reliability, respectively. Subgroup analysis was performed to assess whether clinical diagnosis (neurological PC, non-neurological PC, normal controls) or severity of varus deformity (4 groups based on FAOs 25th, 50th and 75th centiles) could affect reliability of FAO measurements. RESULTS: Intra and interobserver reliability for FAO (r= 0.98; ICC: 0.98), CO (r=0.94; ICC 0.89) and HA measurements (r=0.93; ICC:0.89) were excellent. Subgroup analysis showed that FAOs intra and interobserver reliability remained excellent in neurological PC (r= 0.96; ICC: 0.97), non-neurological PC (r=0.95; ICC: 0.96) and normal feet (r=-0.97; ICC: 0.94). Also, intraobserver (r=0.95, r=0.76, r=0.82 and r=0.92 for FAO >0%, from -7% to 0%, from -13% to -7%, and <-13%, respectively) and interobserver reliability for FAO measurements (ICC: 0.78, ICC: 0.76, ICC: 0.85 and ICC: 0.92) scored excellent regardless of the deformity. CONCLUSION: This study confirms that three-dimensional semiautomatic tools used on weightbearing CT datasets have excellent intra and interobserver reliability in the assessment of hindfoot alignment in pes cavovarus. The nature and severity of the deformity does not affect the reliability of the measurements. This infers that these 3D biometric tools on WBCT images can be reliably used in severe neurological cavovarus feet to evaluate and monitor cavovarus deformity and assess response to intervention.