Lipidomics reveals phosphatidylserine (20:4/20:4) as a novel biomarker for delayed neurocognitive recovery in clinical and animal aging models

The accumulation of various metabolic substances in elderly patients following surgery and anesthesia contributes to delayed neurocognitive recovery (dNCR). Progress in clinical diagnosis and therapeutic interventions is hindered by the lack of reliable biomarkers. This study aims to identify lipid markers that can predict dNCR during the perioperative management for elderly patients. A systematic lipidomics approach was employed to examine the impact of anesthesia and surgery on metabolism and to identify potential biomarkers for predicting and diagnosing dNCR. The study included 67 patients (25 with dNCR, 42 without). Blood samples were collected pre- and post-surgery for metabolite analysis. Differential metabolites were identified using stringent criteria (variable importance in projection [VIP] > 1, fold change [FC] > 1.5 or < 0.67, adjusted P-value < 0.05, area under the curve [AUC] > 0.75). In a mouse model, lipid metabolism alterations in the brain after anesthesia and surgery were analyzed in relation to cognitive dysfunction to identify blood biomarkers linked to dNCR. Thirteen lipid molecules in preoperative blood samples showed significant associations with dNCR. Phosphatidylserine(PS) (20:4/20:4) levels in the prefrontal cortex of dNCR mice were markedly elevated. Receiver operating characteristic (ROC) analysis demonstrated strong correlations between pre- and postoperative PS (20:4/20:4) levels and dNCR. Logistic regression confirmed that preoperative PS (20:4/20:4) levels were significantly associated with postoperative dNCR risk. Dynamic monitoring of PS (20:4/20:4) levels offers an early indicator of dNCR risk. However, whether lowering PS (20:4/20:4) levels would reduce dNCR incidence requires further mechanistic studies. The trial was registered at ClinicalTrials.gov (NCT05105451, registered 2021–11–03).