Determinants of Breast Nodules in Perimenopausal Women: A Prospective Cohort Study

Background: Perimenopausal women often require hormone replacement therapy (HRT), which is associated with an increased risk of developing breast nodules compared to women in other age groups. Consequently, this study aimed to identify risk factors for breast nodule development in perimenopausal women and to develop a predictive model to mitigate these risks. Methods: This prospective cohort study included 436 perimenopausal women who underwent breast ultrasound examinations at the Affiliated Hospital of Guizhou Medical University, China. Clinical data were collected, with 304 cases (70%) assigned to the modeling group, while the remaining 132 cases (30%) were allocated to the validation group using a computerized randomization method. Subsequently, participants in each group were categorized into either the control group or the disease group based on the presence or absence of breast nodules. Risk factors associated with the occurrence of breast nodules in perimenopausal women from the modeling group were analyzed using univariate analysis and multivariate logistic regression. A nomogram predictive model was subsequently constructed using R software. The predictive accuracy and discriminative ability of the model for perimenopausal breast nodules were evaluated in both the modeling and validation groups using goodness-of-fit curves and receiver operating characteristic (ROC) analysis. Results: Factors exhibiting significant differences in the univariate analysis were included in the multivariate logistic regression model. The results revealed that family relationships, the modified Kupperman score, depression, dietary status, estradiol (E2), triglycerides (TG), and total cholesterol (TC) were independent risk factors for the development of breast nodules during perimenopause. In contrast, elevated high follicle-stimulating hormone (FSH) levels were identified as a protective factor against perimenopausal breast nodules. A nomogram predictive model was developed to assess the predictive validity of breast nodules occurrence during perimenopause, using goodness-of-fit curves. The results showed χ2 = 4.936, p = 0.764 for the training group, and χ2 = 8.642, p = 0.071 for the testing group. The model’s discrimination was evaluated by the ROC curve, with the results showing an area under the curve (AUC) of 0.941 for the training model, along with a specificity of 91.7%, and sensitivity of 96.4%. In the testing model, the AUC was also 0.941, with a sensitivity of 90.2% and specificity of 98.4%. Conclusions: Poor family relationships, unhealthy dietary habits, severe menopausal symptoms, severe depression, elevated estrogen levels, and elevated blood lipid levels were identified as independent risk factors for the development of breast nodules during perimenopause. In contrast, high FSH levels serve as a protective factor against perimenopausal breast nodules. The predictive model developed using this approach demonstrates strong predictive accuracy and discriminative power.