Quantification of the cell infiltrate in synovial tissue by digital image analysis

Objective. The experience with digital image analysis (DIA) in the assessment of synovial inflammation is limited. In this study we compared DIA with two currently applied methods for the evaluation of the synovium. Methods. Synovial tissue (ST) specimens were obtained by arthroscopy from knee joints of rheumatoid arthritis (RA) patients with variable disease expression and in control subjects. CD68 + macrophages and CD3 + T-cells were detected by immunohistochemical staining using two different labelling techniques. The labelled ST sections were quantified using three different analysis techniques: manual cell counting (MC), semi-quantitative analysis (SQA) and DIA. Results. We observed strongly positive correlations between the three methods when examining T-cell infiltration: between MC and SQA: sigma = 0.91 (P < 0.0001), between MC and DIA: sigma = 0.95 (P < 0.0001), and between SQA and DIA: sigma = 0.83 (P < 0.0001). Similarly, for the analysis of synovial intimal macrophages, positive correlations were noted between MC and SQA (sigma = 0.62; P = 0.002), MC and DIA (sigma = 0.56; P < 0.01), and SQA and DIA (sigma = 0.79; P < 0.0001). Finally, the analysis of synovial sublining macrophages revealed positive correlations between MC and SQA (sigma = 0.95; P < 0.0001), MC and DIA (sigma = 0.64; P = 0.001) and SQA and DIA (sigma = 0.69; P < 0.0001). Conclusion. These three different methods generated similar results. DIA offers the opportunity of a reliable and time-efficient analysis of the synovial infiltrate.