The correlation and reproducibility of histological scoring systems in cartilage repair

In recent years, the field of cartilage tissue engineering has seen a sharp increase in publications using many tissue engineering techniques and various analysis methods. Comparison between studies remains difficult, due to a lack of uniformity in methods used. A broad range of histological scoring systems is used to examine cartilage quality. Unfortunately, so far little is known on the reliability and correlation of these scoring systems. The objective of this study was to compare two frequently used cartilage repair grading scales, namely, the comprehensive O'Driscoll and the simple Pineda scale. We determined the intra- and interobserver variability of each score as well as the correlation between them. Thirty-eight joint section samples with variable cartilage quality were examined. Three observers documented their findings with both systems at two points in time. Statistical analysis showed very good intra- and interobserver reliability as well as a good correlation between the two scores. For the intraobserver variability of the O'Driscoll scale, we found an average difference of 0.05 with a SD of 0.93 in a 24-point score and a kappavalue of 0.87. For the interobserver reliability, the average difference was 0.001, SD 2.25, and a kappavalue of 0.92. The Pineda scale showed an average difference of 0.86 with a SD of 1.38 in a 14-point score and a kappavalue of 0.86 for the intraobserver reliability, whereas values for the interobserver reliability were average difference 0.82, SD 0.96, and a kappavalue of 0.89. The comparison between the two scales showed a high, inversely related correlation with a correlation coefficient of 0.71. From these results, we concluded that both the O'Driscoll and the Pineda scales are reliable semiquantitative cartilage scoring systems and that acceptance for general use of these two scores will benefit the reliability of literature on tissue engineering for cartilage repair. Thus, the added strength of comparison between published study results allows better understanding of cartilage repair publications and increases the impact of their results.