Development and validation of a nomogram model for sleep disorders in patients with recurrent implantation failure based on physiological and lifestyle factors.

Objective: To establish and validate a nomogram model for the quality of sleep in patients with recurrent implantation failure (RIF) and to evaluate its performance.

Methods: From January 2023 to June 2023, 484 RIF patients who underwent ART fertilization treatment at the Reproductive Medicine Center of Tongji University-affiliated Obstetrics and Gynecology Hospital were selected as the modeling set and internal validation. Additionally, from July to September 2023, 223 RIF patients who underwent ART fertilization treatment at the Reproductive Medicine Center of Tongji University-affiliated Obstetrics and Gynecology Hospital were chosen as the external validation set. Their clinical data was collected. Lasso regression was used to screen potential predictive variables and multifactor logistic regression analysis was conducted to determine the final predictors. A nomogram model was established, and the model was evaluated using methods such as plotting receiver operating characteristic (ROC) curves, calibration curves, Hosmer-Lemeshow goodness of fit test, and decision curve analysis.

Results: Through Lasso regression and multifactor logistic regression, 7 predictors were identified, including FSH, E2, depression mood (moderate, severe), daily exercise time, sun exposure, caffeine intake, and shift work (>16h/w) for constructing the nomogram model. The AUC for the modeling set was 0.971 (95%CI:0.952∼0.989), for the internal validation set was 0.960 (95%CI:0.937∼0.979), and for the external validation set was 0.850 (95%CI:0.739∼0.960), indicating good predictive performance of the model.

Conclusion: This study established and validated a nomogram model composed of 7 clinical indicators for sleep disorders in RIF patients. The predictors include both physiological indicators and daily lifestyle habits, demonstrating significant predictive value and clinical application efficiency. It can be used for early identification of potential sleep disorders in RIF patients, providing certain reference significance for clinical work.