This study assessed the application of SWE on the supraspinatus tendon in nonshoulder pain T2DM patients. T2DM was associated with a significantly lower tendon SWV, and the intra- and intergroup consistency was good for SWE assessment. This finding contributes to discovering the early effects of diabetes on the tendon, which can lay the foundation for clinical risk prediction and timely intervention.
One study[19] demonstrated that supraspinatus tendon tears affect tendon elasticity with a reduced SWV. To avoid the effects of tendon tears or inflammation, patients with abnormal shoulder function assessments and ultrasound indications of rotator cuff tears were excluded from the present study. Thus, SWE can be used to more objectively and accurately assess nonshoulder pain symptoms associated with purely degenerative mechanical properties of the supraspinatus tendon in T2DM patients. The results showed an increased degenerative probability and reduced SWV of supraspinatus tendons in T2DM patients compared to healthy subjects. Sneha et al.[21] also found softer Achilles tendons in diabetic patients using SWE. These results suggested that the supraspinatus tendons of T2DM patients exhibit softening earlier than those of healthy people, and the incidence of rotator cuff tears will continue to be tracked. The mechanisms might involve persistent diabetes promoting AGE and RAGE upregulation[28], causing increased NOX and ROS production[29] and contributing to the accumulation of glycosylated products in rotator cuff tissue[30], eventually leading to tendon fibrosis and local inflammation.
Herein, the supraspinatus tendon SWV was not correlated with FPG or HbA1c in T2DM patients, while patients with a long disease duration and complications had a significantly lower SWV, indicating lower tendon stiffness. FPG and HbA1c reflect short-term blood glucose levels, with immediate values and three months of control[31]. Although the included diabetic patients had a course of more than two years, most still had a relatively normal blood sugar level due to medical treatment. Therefore, the decrease in the supraspinatus tendon SWV in T2DM patients might be associated with long-term disease effects. This finding suggested that proactively controlling the disease is conducive to reducing complications and might help slow the rate of degeneration and decrease the mechanical properties of supraspinatus tendons.
The guidelines for musculoskeletal ultrasound developed by the European Society of Musculoskeletal Radiology[32] state that routine ultrasound can screen for shoulder disorders. Shingh[12] et al. suggested that the supraspinatus tendon was thicker in diabetic patients than in healthy people, but we did not find a difference between the two groups, possibly because 80% of Shingh's diabetic patients were male. However, we confirmed that male supraspinatus tendons were thicker than female supraspinatus tendons before matching. Thus, gender imbalance might interfere with thickness assessment. In the ADF mode, 96.3% of the T2DM patients had a grade 0 blood flow signal within the supraspinatus tendon, which was not significantly greater than that in healthy people. This might be because supraspinatus tendons are blood-deprived structures, and persistently high blood glucose concentrations might prevent the production of vascular endothelial growth factor[33] and thicken the capillary basement membrane[34], resulting in less intratendon blood flow.
Furthermore, upper limb exercise might affect the mechanical properties of the supraspinatus tendon. Pre-matching, post-matching, and subgroup analysis indicated that the supraspinatus tendon SWV was greater in the regular upper limb exercise group (> 5 h/week) than in the group without exercise habits, suggesting greater tendon stiffness. Exercise might promote the synthesis of growth factors and tendon collagen and decrease the glycosylation response, increasing tendon stiffness[35, 36]. These results might suggest that regular upper limb exercise can help delay the progression of supraspinatus tendon degeneration, both diabetic and healthy, to compensate for the negative effects of diabetes on tendons to a certain extent. Thus, it can be used as a guiding basis for clinical intervention and suggests that personal life habits should also be considered during individualized assessments.
Our research group[22] previously used the virtual touch tissue imaging quantification (VTIQ) technique, a shear wave ultrasound imaging technique, to quantitatively evaluate the supraspinatus tendon in healthy subjects. The deep SWV of the tendon was greater than the upper SWV, and the distal SWV was greater than the proximal SWV. However, the supraspinatus tendon SWV did not differ at different measurement points in the present study. This might be because the addition of the ultrasound gel pad in this study reduced the effect of probe pressurization on tissue stiffness, leading to more stable propagation of shear waves within the tissue and improved measurement accuracy. On the other hand, the VTIQ technique uniformly shows values exceeding 10 m/s as "out of detection range," making it difficult to display the quantitative value accurately. In contrast, the SWE technique used in this study can provide specific values above this threshold, which makes the results more objective.
Previous studies have shown[19, 37] that SWE has good feasibility for assessing the stiffness of tendons. This multicenter study has good representativeness with its large sample size covered on a geographically diverse basis, reaffirming the good reproducibility and stability of SWE in assessing the consistency of three repeated measurements across operators for the same type of subjects.
However, this study also has limitations. First, the small sample size of individuals who were left-handed and had type 1 diabetes could not reflect the effect of different dominant hands and types of diabetes. Second, since the supraspinatus tendon changes in T2DM patients were not confirmed by biopsy or surgery, the SWV cut-off value cannot be provided to indicate early degeneration, but we provide reference values for SWV in different populations.