Abstract
Introduction
It has been suggested that corrosion and fretting at the taper junctions of stemmed metal-on-metal hip replacements may contribute to their high failure rates. A peer-reviewed semi-quantitative scoring system [Goldberg et al., 2002] has been used to visually assess the severity of corrosion and fretting of the taper junction but has not been validated using multiple examiners. The aim of this study was to assess the inter-observer variability of this method.
Method
Macroscopic and stereomicroscopic examinations of the femoral head and stem tapers of 100 retrieved large diameter metal on metal (MOM) hip components were performed by two independent observers using the methods defined by Goldberg et al. [2002] to quantify corrosion and fretting. Scores ranging from 1 (none) to 4 (severe) were assigned to the medial, lateral, posterior and anterior quadrants of the neck taper and the distal and proximal regions of the head taper. An overall score was then assigned to each surface as a whole.
Cohen's weighted Kappa statistic (κ) was used to measure the inter-observer agreement. A quadratic weighting scheme, that allocated weights to the importance of disagreements that are proportional to the square of the number of categories apart, was used to take account of scaled disagreement.
Kappa values were assessed using previously established criteria where κ ≤ 0 = poor, 0.01 to 0.20 = slight, 0.21 to 0.40 = fair, 0.41 to 0.60 = moderate, 0.61 to 0.80 = substantial, 0.81 to 1 = almost perfect.
A sample size of 100 was used in order to detect a coefficient of 0.60 to within 0.25 with 95% confidence with two experienced observers. Statistical analysis was performed using Stata/IC version 12.1 (StataCorp, College Station, TC, USA) and a p value < 0.05 was considered statistically significant.
Results
Figure 1 presents the observed and expected percentage agreement and kappa values. The observed agreement for the overall corrosion and fretting scores for the head taper were 95% and 82% respectively. The reliability of the proximal and distal head taper corrosion scores was moderate (κ = 0.54 to 0.57), whilst reliability of the overall head taper corrosion score was substantial (κ = 0.63). The reliability of the proximal, distal and overall head taper fretting scores was fair (κ = 0.21, 0.31 and 0.31 respectively).
The observed agreement for the overall corrosion and fretting scores for the stem taper were 90% and 85%. The reliability of the overall stem taper corrosion (κ = 0.58) and the individual stem taper corrosion (κ = 0.54 to 0.56) scores was moderate.
The reliability of the individual stem taper fretting scores ranged from slight to fair (κ = 0.14 to 0.24), whilst the reliability of the overall stem taper fretting score was fair (κ = 0.24).
Discussion
The results of this study suggest that the Goldberg scoring system is a reliable method for visually quantifying stem and head taper corrosion. There is however a greater variability in fretting assessments between different examiners, which may be due to difficulties in distinguishing between insertion and/or retrieval damage and actual fretting.
