IMPROVING ACCURACY IN WEAR MEASUREMENT ON CLINICAL RADIOGRAPHS-INTRODUCTION TO POWERPOINT METHOD

Abstract

Wear of the polyethylene liner in Total hip arthroplasty (THA) is associated to aseptic component loosening. With low wear bearing surfaces and metal backing in acetabular components the manual methods of measurements have not fared well. Computerized methods increased the ease and accuracy of wear measurement. The average clinician has no access to these methods. In this study we proposed to develop a method of manual wear measurement (PowerPoint – PP method) using a simple office PC and

1. quantify its accuracy and reproducibility

2. compare the accuracy with Livermore and Dorr method and

3. determine the accuracy in different degrees of wear.

The study population was divided into class 1 (C1), Class 2 (C2) and Class 3 (C3) group. C1 group had 20 patients who had undergone liner change for high wear. This class simulated a high wear situation. C2 group had 24 patients who were implanted with HXLP. This class simulated very low wear situation. 10 patients were included in C3 group. The same 6week postoperative radiograph was paired as a set of x rays for analysis. This mimics a zero wear situation.

PP method had more consistency with Livermore method for C1 group. For C2 and C3 groups all the three methods did not provide consistent results. The correlation coefficient values for wear measurement by PP method showed good correlation between observers in C1 and C2 wear (P values < 0.05). For C3 with true zero wear there was poor correlation between the observers (r −0.659, 0.028, 0.638). The paired T test P values for all classes and both observers were > 0.05. There was no statistically significant difference in the reading of the two observers. Pearson correlation coefficient for all methods showed good correlation for C1and C2 groups. All the methods had errors while measuring true zero in C3. The one way ANOVA analysis was done to identify the ability of the three methods differentiate between C2 and C3. The PP method had the ability (P value < 0.05) to differentiate between C1, C2 and C3. The Dorr’s and Livermore’s methods could only differentiate the C1 from C2andC3.

Computerized methods have certain limitations. Matthew Collier et al reported a mean linear wear rate of 0.4(0.04–0.86) and 0.27 (0.01–0.56) by computerized methods in radiographs with true zero wear. In C3 group the average wear rate by PP method was 0.22 ± 0.206 mm. In PP method ability to work at 400% magnifications, ability to correct for rotation on X axis, grouping function of PowerPoint program leads to less chances of errors. PP method has a good reproducibility for clinical use (r> 0.930). The ability of the PP method to differentiate between C2 and C3 should make it a preferred manual method of wear assessment.

The PP method has limitations. The least measurement is limited by diameter of the femoral head. It cannot be utilized for wear analysis in cup loosening or migration. It can be regarded as a supplement to the existing methods of manual wear measurements.