Modified 20% anatomical axis method demonstrates the highest reliability for posterior tibial slope measurement on radiographs.

Abstract

Several methods are described for measurement of the posterior tibial slope (PTS), with no clear consensus on which method is most reproducible and precise for measurement of the PTS. This study aimed to identify the most reliable method for measuring PTS which is easily reproducible and precise. Patients who underwent unilateral Medial Opening Wedge High Tibial Osteotomy (MOWHTO) for varus deformity were retrospectively analysed using their pre-operative and 6 months post-operative Electronic Optical Scan (EOS) scans for measurement of PTS. Four commonly used methods to measure PTS were used for comparison of intra and interobserver reproducibility: Modified 20% anatomical axis (AA) method, mechanical axis (MA) estimation method, 2-circle AA method and 5-15cm AA method on short lateral knee radiographs. Three methods used AA estimation whereas one method used MA estimation for measuring PTS. Two observers performed two blinded measurement sessions >1 week apart for each patient. Intraclass correlation coefficient (ICC) and 95% Confidence intervals (95% CI) was used to measure the intra, interobserver reproducibility and precision for each method. Cohen's D measure was used to ascertain the effect size. Amongst the 30 patients analysed for this study, the median age of patients included was 46 years with 21 males and 9 females. The modified 20% method was consistently found to be most reproducible and precise for measurement of PTS amongst all methods on comparing ICC and 95 % CI. The other methods had comparable values of ICC and 95% CI. The mean difference of measured PTS when using short or long knee radiographs with AA estimation methods was negligible. The measured PTS was found to be, on an average, 2° higher when using AA estimation methods compared to MA estimation methods. The modified 20% AA method was found to be most reproducible and reliable in measuring the PTS. There was negligible difference in the measured PTS on long and short radiographs using AA estimation methods. On an average, the PTS was found to be 2° higher when measured using the AA estimation methods compared to the MA estimation methods. Retrospective observational study (Level IV).