Patellar height after barrel-vault high tibial osteotomy.

The purpose of the study was to assess the influence of barrel-vault high tibial osteotomy on patellar position and to evaluate the reliability and interobserver variability of three patellar height ratios. The radiographs of 24 knees that had undergone barrel-vault height tibial osteotomy with available preoperative as well as postoperative radiographs of the last follow-up were enrolled in the study. The patellar position was evaluated with the use of the Insall-Salvati, Blackbourne-Peel and Caton-Linclau methods. The indices were calculated with the tibial slope angle measured by the Moore-Harvey method. The average patellar index in the Blackburne-Peel and Caton-Linclau measurements had a statistically significant tendency to increase after barrel-vault tibial osteotomy. The results of the Insall-Salvati assessment method showed no statistically significant differences. Intrasession ICCs (intraclass correlation coefficient) varied between 0.8 and 0.99. The kappa value for the inter-observer agreement of all three ratios determined varied between 0.48 for the Blackburne-Peel index and 0.5 for the Insall-Salvati and Caton-Linclau indices. The average tibial slope before the surgery was 11.8 degrees (+/- 3.6) and decreased to 6.6 degrees (+/- 4.3). No significant associations between the index change and tibial slope change were detected. The data suggest that since some of patellar height indices measure different anatomical relationships, comparable types of indices should be used in order to estimate the patellar height or patellar relation to the joint line.