Establishment and validation of a screening and risk predication model for rotator cuff tear and shoulder stiffness.

BACKGROUND

Rotator cuff tears (RCTs) with shoulder stiffness have a high association rate; however, the mechanism and possible risk factors are unclear. This study aims to collect the factors that may affect RCT and shoulder stiffness, screen out the relevant risk factors through statistical analysis, and establish a simple model to predict the risk of RCT combined with shoulder stiffness.

METHODS

A retrospective analysis was conducted on 406 patients diagnosed with RCT through arthroscopic surgery at the Department of Joint and Sports Medicine, the First Affiliated Hospital of Dalian Medical University, from December 2019 to June 2023. The analysis comprised two groups: 213 patients with both RCT and shoulder stiffness, and 193 patients without shoulder stiffness. A total of 21 potential risk factors associated with RCT and shoulder stiffness were considered, and a prediction model was developed using single-focus logistic regression analysis and multifocal logistic regression analysis in the training set (N=284), which was presented as nomograms. The validation set (N=122) was used to assess the model's discrimination, calibration and clinical practicability. The proportion of patients with RCT combined with shoulder stiffness in both the training set and the validation set was 52.5%.

RESULTS

The study identified eight pertinent risk factors: gender, dominant side, smoking, hypothyroidism, depression, hyperlipidemia, type III acromion, and partial tear. Based on these factors, a clinical prediction model was developed. The model demonstrated excellent predictive performance with an area under the receiver operating characteristic curve (AUROC) of 0.856 [95% confidence interval (CI): 0.812-0.900] for the training set and 0.867 (95% CI: 0.807-0.928) for the validation set. Calibration curves exhibited strong agreement between the actual disease probabilities and predicted probabilities using the model in both datasets. Decision curve analysis (DCA) further confirmed the clinical utility of the model.

CONCLUSIONS

Based on routine data, the prediction model offers clinicians a simple and reliable tool for predicting the combination of RCT and shoulder stiffness.