Evaluation of IVIM and FACT Imaging for Early Detection of Rotator Cuff Muscle Degeneration and Physiotherapy Exercise Effects

Ling Teng¹Mingru Huang²Hongyue Tao¹Xiao Zhang²Sijia Feng²Xiner Du²Zhiwei Qin³Xiangwen Li¹Yiwen Hu¹Changyan Liu¹Yuxue Xie¹Shuang Chen^1*

• ¹Huashan Hospital Fudan University Department of Radiology, Shanghai, China
• ²Huashan Hospital Fudan University Department of Sports Medicine, Shanghai, China
• ³Shanghai United Imaging Healthcare Co Ltd, Shanghai, China

Abstract Background: This study investigated novel magnetic resonance imaging (MRI) techniques, including intravoxel incoherent motion (IVIM) and fat analysis & calculation technique (FACT) to detect early pathologic changes in rotator cuff (RC) muscle in patients with shoulder pain and evaluate rehabilitation effects. Methods: MRI examinations, including IVIM and FACT, and clinical scores were collected in patients with unilateral shoulder pain at baseline (n=48) and 3-month follow-up (n=42) after physiotherapy exercise. The contralateral shoulder served as control. IVIM-derived parameters (apparent diffusion coefficient, ADC; diffusion coefficient, D; pseudo-diffusion coefficient, D*; perfusion fraction, f; and their product, fD*) and FACT-derived parameters (fat fraction, FF) were measured in four RC muscles. Clinical assessment included Constant Score (CS), the American Shoulder and Elbow Surgeons (ASES), and the Visual Analog Scale (VAS). Differences in imaging parameters were compared between the affected and the contralateral shoulder. Changes in imaging parameters and clinical scores after exercise rehabilitation were analyzed between baseline and 3-month follow-up. Correlation analysis was conducted between the change in imaging values and the change in clinical scores. Results: At baseline, D* values of the affected sides in four RC muscles were lower compared to the contralateral sides (p<0.05). f and fD* values of the affected sides were higher and lower respectively, compared to the contralateral sides and showed a difference in infraspinatus and subscapularis (p = 0.03 and p = 0.04). No significant differences for FF and other IVIM-derived parameters in either muscle were observed. After the physiotherapy exercise, all of the clinical scores showed significant improvements compared to the baseline (all p<0.05). Regarding imaging parameters, only D* values showed a significant increase in the supraspinatus muscle (p < 0.05). Additionally, the change in D* and f correlated with CS, ASES scores and VAS scores respectively. FF showed a decreasing trend but lacking significance (p>0.05). Conclusion: Capillary flow impairment (reflected by D*) appears more prominent than fat infiltration (reflected by FF) in the early RC muscle degeneration. IVIM-derived parameters, particularly D* may assist clinicians in detecting early-stage muscle degeneration and provide a non-invasive method for monitoring rehabilitation outcomes.