Correlation of Non-destructive Electromechanical Probe (Arthro-BST) Assessment with Histological Scores, Biochemical Composition and Mechanical Properties in Human Knee Joints

Methods: Electromechanical properties were first mapped manually (4 measurements per cm2, 1 measurement per site, 30 minutes per joint) using the Arthro-BST on the entire articular surfaces. The device calculates a quantitative parameter (QP) of cartilage electromechanical activity corresponding to the number of microelectrodes in contact with the cartilage when the sum of their streaming potential reaches 100 mV. A high QP indicates weak electromechanical properties and vice-versa. The articular surfaces were graded according to the ICRS lesion grading system and a total of 156 osteochondral cores were harvested from normal and lesional regions. Safranin O-Fast Green-stained paraffin sections3 were scored with the Mankin histological-histochemical grading system4. From those cores, 50 were tested in unconfined compression to obtain the fibril modulus (Ef), equilibrium modulus (Em), and permeability (k) prior to histoprocessing. Furthermore, 64 of the cores were analyzed with the polarized light microscopy (PLM) qualitative scoring system that scores stratified collagen structure of articular cartilage. Biochemical analyses were done on 48 cores to obtain water content, glycosaminoglycan content, number of cells and collagen content. Core locations were precisely identified for each articular surface using the camera-registration system and the Arthro-BST’s electromechanical QP corresponding to the cored site was calculated as the average of all QPs measured within 6 mm from the core center location.

Results: To assess correlation of electromechanical QP with the total Mankin score (scale 0 (normal) to 14 (totally degraded)), cores were first separated into three Mankin score groups (0-2, 3-5 and 6-14). For the first group (0-2), the histological slides were normal (Fig. 1a). For the intermediate group (3-5), decreased Safranin O staining and structural alterations were apparent in the superficial zone (Fig. 1b). Clefts and reduced Safranin O staining were observed for the degenerated group (6-14) (Fig. 1c). Similarly, cores assessed by the PLM collagen score (scale 0 (entirely degraded) to 5 (normal)) were divided into two groups: PLM score 3-5 samples (Fig. 1d) are stratified into superficial, transition, radial zones as in normal cartilage and PLM score 0-2 samples (Fig. 1e) do not have the 3 zones present. Using linear regression and a one-way analysis of variance (ANOVA) a positive correlation was found between electromechanical QP and Mankin score (R2=0.55, p < 0.001; Fig. 2a), while a negative correlation was observed between QP and PLM score (R2=0.50, p < 0.001; Fig. 2b). Negative correlations were also found between QP and either Ef (R2 = 0.62, p < 0.0001; Fig. 3a) or Em (R2 = 0.52, p < 0.0001; Fig. 3b) while a strong positive correlation was found between QP and permeability (log (k)) (R2 = 0.53, p < 0.0001; Fig. 3c). Results from biochemical analyses revealed a positive correlation between QP and water content (R2=0.29, p < 0.0001; Fig. 4a) and a negative correlation between QP and GAG content (chondroitin sulfate per wet weight) (R2=0.25, p < 0.0001; Fig. 4b), while there was no significant correlation between QP and the number of cells per wet weight (Fig. 4c) nor the collagen content (Fig. 4d).

Discussion: The measured electromechanical quantitative parameter (QP) correlated strongly with Mankin score, PLM score, water content and unconfined compression mechanical parameters; while weaker correlations were observed with the biochemical composition (GAG and collagen content (both relative to dry or wet weight)). These results suggest that the evaluation of the cartilage’s electromechanical properties by the Arthro-BST is more sensitive to the structure of the extracellular matrix than to its biochemical composition. In the interpretation of these results, it is also important to keep in mind that about 90% of the area of the articular surfaces was visually normal (ICRS score = 0). The inclusion of more degenerated articular surfaces in the study could have strengthened the observed correlations. Considering the fact that the non-destructive mapping of an entire distal femur with a high spatial resolution takes about 30 minutes, we believe that the Arthro-BST can provide a rapid and reliable tool for articular cartilage assessment. Significance: The electromechanical properties of articular cartilage are representative of its biochemical composition, structure and stiffness. Since histological, biochemical and biomechanical properties requires destructive processing of the samples, non-destructive electromechanical properties of complete and intact articular surfaces could be used as a biomarker for the diagnosis and the evaluation of treatments of articular cartilage degeneration. Acknowledgements: Funding provided by the National Sciences and Engineering Research Council (NSERC), the Fonds québécois de la recherche sur la nature et les technologies (FQRNT) and Biomomentum Inc. We acknowledge the technical contributions of Geneviève Picard, Marie-Helene Boulanger and Sylvain Gaufrès.