To investigate determinants of respiratory support duration and construct an interpretable weaning prediction model for late preterm infants, highlighting the clinical significance of accurate weaning predictions.
Key Findings:
Multiple pregnancy (HR = 1.289, 95% CI: 1.038–1.601, P = 0.022) and elevated superoxide dismutase (SOD) (HR = 1.014, 95% CI: 1.009–1.019, P < 0.001) were identified as risk factors for prolonged respiratory support.
Protective factors included nasal continuous positive airway pressure (NCPAP) (HR = 0.703, 95% CI: 0.564–0.878, P = 0.002) and lymphocyte percentage (LYM_PC) (HR = 0.987, 95% CI: 0.978–0.997, P = 0.008).
The model achieved a C-index of 0.677 (optimism-corrected 0.661) and time-dependent AUC values of 0.732, 0.769, and 0.782 at days 3, 5, and 8.
Interpretation:
The LASSO-Cox nomogram demonstrated moderate-to-good discrimination, indicating its potential utility in predicting respiratory support duration and improving clinical decision-making.
Limitations:
The study was retrospective and conducted at a single center, which may limit generalizability and introduce biases.
The model's performance may vary in different populations or settings.
Conclusion:
The LASSO-Cox nomogram provides a transparent basis for individualized weaning assessment in late preterm infants, with potential implications for clinical practice.
