Clinical application research on the quantitative measurement of supraspinatus muscle fatty degeneration based on PACS system to improve preoperative assessment.

OBJECTIVE

To evaluate the clinical efficacy of a novel quantitative method using the Picture Archiving and Communication System (PACS) for multiplane assessment of supraspinatus muscle fatty infiltration (FI) and compare its reliability and accuracy with traditional single-plane visual evaluations (Under Direct Vision-FF) in preoperative planning for rotator cuff tear (RCT) patients.

METHODS

A retrospective analysis was conducted on patients undergoing arthroscopic rotator cuff repair (ARCR) between January and June 2023. Preoperative 3.0 T MRI scans were analyzed using PACS to measure FI in three sagittal planes (medial, Y-plane, lateral). Four orthopedic surgeons performed Goutallier classification and manual FI assessments under direct vision and via PACS. Intra- and interobserver reliability were evaluated using intraclass correlation coefficients (ICCs), while Bland-Altman analysis and paired -tests compared measurement consistency and differences.

RESULTS

PACS-based measurements (PACS-FF) demonstrated superior reliability (intraobserver ICC: 0.973-0.996; interobserver ICC: 0.940-0.978) compared to direct vision assessments (intraobserver ICC: 0.538-0.967; interobserver ICC: 0.864-0.940). Significant discrepancies were observed between methods, with direct vision underestimating FI ( < 0.05-0.0001). Multiplane analysis revealed heterogeneous FI distribution, with lateral-plane FI significantly higher than medial and Y-plane values ( < 0.001). Bland-Altman analysis showed 60%-85% of direct vision measurements exceeded clinically acceptable limits of agreement (±10%).

CONCLUSIONS

Quantitative multiplane PACS-based FI assessment improves accuracy and reliability over traditional single-plane visual evaluation, better reflecting heterogeneous fat distribution in the supraspinatus muscle. This method enhances preoperative risk stratification and surgical outcome prediction for RCT patients. Future integration of automated tools may further optimize clinical efficiency.