Viscoelasticity in Achilles tendonopathy: quantitative assessment by using real-time shear-wave elastography

To investigate the differences in viscoelastic properties between normal and pathologic Achilles tendons (ATs) by using real-time shear-wave elastography (SWE). The institutional review board approved this study, and written informed consent was obtained from 25 symptomatic patients and 80 volunteers. One hundred eighty ultrasonographic (US) and SWE studies of ATs without tendonopathy and 30 studies of the middle portion of the AT in patients with tendonopathy were assessed prospectively. Each study included data sets acquired at B-mode US (tendon morphology and cross-sectional area) and SWE (axial and sagittal mean velocity and relative anisotropic coefficient) for two passively mobilized ankle positions. The presence of AT tears at B-mode US and signal-void areas at SWE were noted. Significantly lower mean velocity was shown in tendons with tendonopathy than in normal tendons in the relaxed position at axial SWE (P < .001) and in the stretched position at sagittal (P < .001) and axial (P = .0026) SWE. Tendon softening was a sign of tendonopathy in relaxed ATs when the mean velocity was less than or equal to 4.06 m · sec(-1) at axial SWE (sensitivity, 54.2%; 95% confidence interval [CI]: 32.8, 74.4; specificity, 91.5%; 95% CI: 86.3, 95.1) and less than or equal to 5.70 m · sec(-1) at sagittal SWE (sensitivity, 41.7%; 95% CI: 22.1, 63.3; specificity, 81.8%; 95% CI: 75.3, 87.2) and in stretched ATs, when the mean velocity was less than or equal to 4.86 m · sec(-1) at axial SWE (sensitivity, 66.7%; 95% CI: 44.7, 84.3; specificity, 75.6%; 95% CI: 68.5, 81.7) and less than or equal to 14.58 m · sec(-1) at sagittal SWE (sensitivity, 58.3%; 95% CI: 36.7, 77.9; specificity, 83.5%; 95% CI: 77.2, 88.7). Anisotropic results were not significantly different between normal and pathologic ATs. Six of six (100%) partial-thickness tears appeared as signal-void areas at SWE. Whether the AT was relaxed or stretched, SWE helped to confirm and quantify pathologic tendon softening in patients with tendonopathy in the midportion of the AT and did not reveal modifications of viscoelastic anisotropy in the tendon. Tendon softening assessed by using SWE appeared to be highly specific, but sensitivity was relatively low.