Development and validation of a nomogram based on LASSO regression for predicting early postoperative polyp
Hanshuang Zhang1, Huizhen Zheng2, Siqi Wang1
1Department of Otorhinolaryngology Head and Neck Surgery, Wenzhou Central Hospital, Wenzhou, Zhejiang, China.
Background:
Chronic rhinosinusitis with nasal polyps (CRSwNP) has a high early postoperative recurrence rate, and precise individualized prediction tools are currently lacking.
Objective:
To construct and validate a nomogram model based on commonly used clinical indicators to predict the risk of early postoperative polyp recurrence in patients with CRSwNP.
Methods:
A retrospective cohort study design was adopted. A total of 260 and 114 patients with CRSwNP who underwent endoscopic sinus surgery in our hospital from January 2021 to April 2025 were selected as the training cohort and internal validation cohort, respectively. The training cohort was divided into a recurrence group and a non-recurrence group according to recurrence status within 6 months. In addition, 242 patients from Wenzhou People's Hospital were included as the external validation cohort. Independent influencing factors were screened using LASSO regression and multivariate logistic regression, and a nomogram prediction model was constructed. The model performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis.
Results:
Asthma, stage 3 sinusitis, Lund-Mackay score, eosinophils (EOS), and immunoglobulin E (IgE) were independent risk factors for early postoperative recurrence (P < 0.05). The AUCs (95% CIs) of the training cohort, internal validation cohort, and external validation cohort were 0.923 (0.882-0.963), 0.881 (0.827-0.936), and 0.890 (0.841-0.939), respectively. The calibration curves and Hosmer-Lemeshow test showed that the model had good calibration, and decision curve analysis indicated that it provided clinical net benefit.
Conclusion:
The nomogram model has certain predictive performance, can identify patients at high risk of early postoperative polyp recurrence in CRSwNP, and has the potential to serve as an important auxiliary tool for individualized postoperative clinical management.