To investigate the correlation between preoperative sleep monitoring parameters and serum neurofilament light polypeptide (NfL) with postoperative delirium (POD) in patients with chronic obstructive pulmonary disease (COPD) complicated with lung cancer.

A total of 87 patients with COPD and lung cancer who underwent thoracoscopic lung resection in our hospital from January 2021 to August 2025 were enrolled. Among them, 19 patients who developed POD were assigned to the observation group, and 68 patients without POD were assigned to the control group. Clinical data, laboratory indicators, preoperative polysomnographic parameters, and intraoperative parameters were collected. Multivariate logistic regression was used to identify independent risk factors for POD, and receiver operating characteristic (ROC) curves were constructed to evaluate the predictive value of each parameter and their combination.

The observation group had significantly higher age, proportion of American Society of Anesthesiologists (ASA) grade Ⅲ, intraoperative onelung ventilation time [(97.85±16.37) min vs. (88.14±16.05) min, P=0.001], propofol dosage [(790.34±68.34) mg vs. (648.14±68.91) mg, P<0.001], and serum NfL level [(77.45±22.36) pg/ml vs. (61.34±10.02) pg/ml, P<0.001] than the control group. The rates of REM sleep absence (73.68% vs. 47.06%, P<0.05) and atypical sleep (47.37% vs. 17.65%, P<0.05) were also significantly higher in the observation group. Multivariate logistic regression showed that intraoperative onelung ventilation time (OR=1.063, 95%CI: 1.035~1.093, P<0.001), propofol dosage (OR=1.006, 95%CI: 1.003~1.009, P<0.001), serum NfL (OR=1.397, 95%CI: 1.182~1.652, P<0.001), rapid eye movement (REM )sleep absence (OR=2.294, 95%CI: 1.164~4.519, P=0.016), and atypical sleep (OR=1.178, 95%CI: 1.050~1.635, P=0.008) were independent risk factors for POD. ROC curve analysis revealed that the area under the curve (AUC) values for serum NfL, REM sleep absence, propofol dosage, atypical sleep, and onelung ventilation time were 0.757, 0.602, 0.762, 0.778, and 0.763, respectively. The combined model incorporating sleep parameters, NfL, onelung ventilation time, and propofol dosage yielded an AUC of 0.939 (95%CI: 0.898~0.981) with a Youden index of 0.758.

Elevated preoperative serum NfL, REM sleep absence, atypical sleep, prolonged intraoperative onelung ventilation time, and increased propofol dosage are independent risk factors for POD in elderly COPD patients undergoing thoracoscopic lung cancer surgery. The combined preoperative assessment of sleep parameters and serum NfL demonstrates good predictive value for POD and may facilitate early identification of highrisk patients to guide perioperative brainprotective strategies.