Evaluation of patello-femoral alignment by CT scans: interobserver reliability of several parameters

Patello-femoral malalignment (PFM) is a common cause of disability often related to patello-femoral syndrome (PFS). Several causes have been taken into account; a proper diagnosis requires instrumental imaging and a methodical evaluation of different parameters. The aim of the present study was to identify the most reliable parameters for measuring patello-femoral and inferior limb alignment by CT. Twenty randomly selected patients suffering from PFS for a total of 40 knees were studied by static CT scans in order to assess patellar tilt, patellar displacement, patellar and trochlear morphology and inferior limb alignment. All known parameters were measured; the variability of the measurements between observers was evaluated by boxplots, Pearson's correlation coefficients, and infraclass correlation coefficient [ICC(2,1)] based on a two-way random effect model. Bland-Altman mean differences and 95 % limits of agreement were computed for each pair of measurements. Patellar tilt parameters appeared equally reliable; patellar displacement is best measured with BoTot that showed an ICC of 0.889; morphology is best measured with WibergTot, with an ICC of 0.862; lastly, for the inferior limb alignment parameters' analysis, FTV outperformed the others in terms of reliability. The present study allowed us to select a limited number of reliable parameters in the evaluation of patello-femoral and inferior limb alignment. The use of these parameters may also result in a more reliable comparison of studies on PFM and in a better evaluation of the treatment outcomes.