Objective  To investigate the correlation between digital breast tomosynthesis (DBT) image features, pathological features, and the results of breast-conserving surgery (BCS) in early stage breast cancer.

Methods We retrospectively analyzed 422 cases of BCS resulting in successful breast preservation and 211 BCS cases that were followed by mastectomy. All of the patients underwent surgery at Tianjin Medical University Cancer Institute & Hospital between January 2019 and December 2023. Preoperative DBT images and clinicopathological features were assessed. A univariate Logistic regression analysis was performed to screen out the characteristics associated with BCS surgical results, after which multivariate Logistic regression analysis was performed to screen out the characteristics.

Results Univariate Logistic regression analysis showed that age (P=0.020), architectural distortion (P<0.001), breast composition (extremely dense, P=0.001), and molecular subtype (Her2, P=0.001) were statistically different between the two groups. Multivariate Logistic regression analysis showed that architectural distortion (P<0.001), breast composition (extremely dense breast, P=0.003), and molecular subtype (Her2, P<0.001) differed significantly between the two groups.

Conclusions Breast composition, architectural distortion, and molecular subtype correlated with BCS results. Extremely dense breast composition, architectural distortion, and Her2 subtype are associated with a higher possibility of conversion to mastectomy. These factors serve as effective predictive indicators of BCS results and thus aid clinicians in deciding the appropriate surgical strategies in the treatment of breast cancer.