Effect of multidimensional perioperative factors on postoperative delayed neurocognitive recovery in elderly patients undergoing single-port video-assisted thoracoscopic lobectomy: A prospective cohort study
Effect of multidimensional perioperative factors on postoperative delayed neurocognitive recovery in elderly patients undergoing single-port video-assisted thoracoscopic lobectomy: A prospective cohort study

Effect of multidimensional perioperative factors on postoperative delayed neurocognitive recovery in elderly patients undergoing single-port video-assisted thoracoscopic lobectomy: A prospective cohort study

Curr Probl Surg. 2026 Sep;82:102101. doi: 10.1016/j.cpsurg.2026.102101. Epub 2026 Jul 6.

ABSTRACT

OBJECTIVE: Delayed neurocognitive recovery (DNR) is a frequent postoperative complication in elderly patients, yet predictors in single-port video-assisted thoracoscopic (SP-VATS) lobectomy remain unclear, especially regarding psychosocial and frailty factors. We aimed to identify comprehensive preoperative and intraoperative predictors of DNR in this population.

METHODS: This prospective study enrolled 309 patients aged ≥60 years undergoing SP-VATS lobectomy. Preoperative psychological status (anxiety, depression, sleep quality), physiological condition (frailty, oral health), and social support were assessed. Cognition was evaluated with the Montreal Cognitive Assessment (MoCA) at baseline, postoperative day (POD) 7, and POD 30. LASSO-penalized logistic regression and conventional logistic regression identified independent predictors.

RESULTS: Overall DNR incidence was 24.60% (76/309), with 16.50% (51/309) at POD7 and 20.39% (63/309) at POD30. Univariate analysis yielded 20 significant variables. LASSO retained four candidates: Pittsburgh Sleep Quality Index (PSQI), Perceived Social Support Scale (PSSS), Surgical Apgar Score (SAS), and Hospital Anxiety and Depression Scale (HAD). Multivariable logistic regression showed higher SAS (OR=0.582, 95% CI: 0.447-0.756, P<0.001) as an independent protective factor, whereas higher PSQI (OR=2.059, 95% CI: 1.643-2.580, P<0.001) and HAD scores (OR=0.362, 95% CI: 0.161-0.813, P=0.014) were independent risk factors. The model exhibited excellent discrimination (AUC=0.919, 95% CI: 0.868-0.970), and sensitivity analyses using three alternative DNR definitions confirmed robustness (kappa: 0.61-0.82).

CONCLUSIONS: Cognitive decline is common after SP-VATS lobectomy in elderly patients. DNR is influenced by psychological, physiological, social, and intraoperative hemodynamic factors. Preoperative screening for modifiable risks may reduce postoperative cognitive impairment and improve outcomes.

PMID:42700994 | DOI:10.1016/j.cpsurg.2026.102101