Reliability of a new semi-automatic image analysis method for evaluating the Doppler signal and intratendinous vascular resistance in patellar tendinopathy

Abstract

The aim of this study was to determine the intra- and inter-rater reliability of a new semi-automatic image analysis method for quantification of the shape of the Doppler signal and the intratendinous vascular resistance in patellar tendinopathy. Thirty athletes (27.4 y, standard deviation = 8.57 y) with patellar intratendinous vascularity were included in a cross-sectional study (42 tendons analyzed). The intratendinous blood flow was assessed with power Doppler and ImageJ (Version 1.50b, National Institutes of Health, Bethesda, MD, USA) quantification software over a manually selected region of interest. Two blinded observers performed the analysis of the Doppler signal (vascular resistance) and shape descriptors (number of signals, pixel intensity, area, perimeter, major diameter, minor diameter, circularity and solidity). The intraclass correlation coefficient (ICC) was calculated, and the Bland-Altman mean of differences (MoD) and 95% limits of agreement (LoA) were determined. Also, small real differences (SRDs) and the standard error of measurement (SEM) were calculated. Intra-rater reliability was at a maximum for area (ICC = 0.999, 95% confidence interval [CI] = 0.998-0.999) and at a minimum for solidity (ICC = 0.782, 95% CI: 0.682-0.853). The MoD and 95% LoA were very low, and the relative SRD and SEM were below 5.3% and 2%, respectively. The inter-rater reliability was the maximum for area (ICC = 0.993, 95% CI = 0.989-0.996) and the minimum for circularity (ICC = 0.73; 95% CI=0.611-0.817). The MoD and 95% LoA were low, with the SRD and SEM below 6% and 2.2%. The proposed quantitative method for studying the intratendinous Doppler signal in the patellar tendon is reliable and reproducible.