Nomogram for predicting urinary retention after radical hysterectomy for cervical cancer with comparative analysis of surgical methods.

OBJECTIVE

To identify risk factors for urinary retention after radical hysterectomy for cervical cancer (CC) and establish a nomogram model to facilitate individual treatment decisions.

METHODOLOGY: In this retrospective study, records of 438 patients with CC undergoing radical hysterectomy between December 2011 and December 2018 at Fujian Maternity and Child Health Hospital, China were analyzed. Patients were divided into the abdominal operation group (196 cases) and the laparoscopic group (242 cases) based on the surgical method. Short-term outcomes were compared. Independent risk factors for postoperative urinary retention identified by multivariate logistic regression were used to develop a nomogram. The performance of the nomogram was assessed using the receiver operating characteristic (ROC) curve and area under the curve (AUC) analyses.

RESULTS

The mean age of patients in the abdominal group was lower than that in the laparoscopic group (P < 0.05). Operation time and the number of dissected lymph nodes were higher in the laparoscopic group, whereas postoperative hospital stay and blood loss were lower (P < 0.05). The incidence of urinary retention in the laparoscopic group (53.7%) was significantly higher than in the abdominal group (38.2%; P < 0.05). Multivariate logistic regression results identified surgical approach, operation time, lympho-vascular invasion, and age as independent risk factors for postoperative urinary retention (P < 0.05). Calibration curve for the urinary retention nomogram aligned with the ideal curve. AUCs of the training and the validation sets were 0.819 (95% CI, 0.798-0.895) and 0.795 (95% CI, 0.782-0.824), respectively.

CONCLUSION

Surgical approach, operative time, lymphovascular invasion, and age were identified as independent risk factors for urinary retention after hysterectomy in CC patients. The developed nomogram based on these independent risk factors had good accuracy and predictive ability.