Background: To assess the diagnostic value of a nomogram that integrates ultrasound and clinical features to differentiate uterine sarcoma from uterine fibroids. Methods: In this retrospective analysis, data from 60 uterine sarcoma patients and 60 uterine fibroid patients confirmed by surgical pathology at the Affiliated Hospital of Putian University (August 2024–June 2025) were examined. Clinical variables (age, disease duration, menopausal status, postmenopausal bleeding), laboratory markers (carbohydrate antigen 125 CA125, lactate dehydrogenase LDH), and ultrasound characteristics (maximal diameter, margin, echogenicity, cystic change, calcification, and Adler blood flow grading) were collected. Independent predictors were determined through both univariate and multivariate logistic regression analyses. Predictive models were constructed and evaluated via receiver operating characteristic (ROC) curves, with model robustness further assessed through tenfold cross-validation and bootstrap validation. Results: The groups were successfully matched for key baseline characteristics (age, disease duration, menopausal status; all p > 0.05). Multivariate analysis revealed that age, postmenopausal bleeding, LDH, CA125, echogenicity, and tumor margin were independent predictors. The combined model demonstrated enhanced diagnostic ability with an area under the curve (AUC) of 0.902 (95% CI: 0.847–0.935), outperforming individual models based on clinical (AUC: 0.764), laboratory (AUC: 0.651), and ultrasound (AUC: 0.804) data. The model’s generalizability was confirmed by internal validation, showing strong maintained performance (tenfold cross-validated AUC: 0.885; bootstrap-corrected AUC: 0.887). Conclusion: A nomogram based on combined clinical, laboratory, and ultrasound features provides high accuracy for differentiating uterine sarcoma from fibroids, supporting clinical decision-making.
Huang et al. (2026) studied this question.