Preoperative Plasma and Cerebrospinal Fluid SNAP-25 Levels Predict Delayed Neurocognitive Recovery in Elderly Patients: A Prospective Observational Study.

OBJECTIVE

Delayed neurocognitive recovery (DNR) is a common postoperative complication in elderly patients, and its prevention and treatment remain challenging. Synaptosomal-associated protein 25 (SNAP-25), a potential biomarker for several neurodegenerative diseases, has been reported to be associated with the occurrence of DNR. However, whether SNAP-25 can predict the development of DNR remains unclear. Therefore, this study aimed to investigate the predictive value of preoperative SNAP-25 levels in cerebrospinal fluid (CSF) and plasma for DNR in elderly patients.

METHODS

In this prospective observational study, elderly patients aged 65-75 years undergoing total knee arthroplasty under spinal anesthesia were enrolled. 3 mL of venous blood was collected upon entry into the operating room, and 1 mL of CSF was obtained after successful lumbar puncture and prior to intrathecal anesthetic injection. Neurocognitive function was assessed 1 day before surgery and 7 days after surgery. The primary outcome was the incidence of DNR. The predictors were preoperative CSF and plasma SNAP-25 levels. The association between SNAP-25 levels and DNR was further evaluated.

RESULTS

A total of 86 patients completed the study, among whom 21 developed DNR, yielding an incidence of 24.4%. Preoperative SNAP-25 levels in both CSF and plasma were significantly lower in the DNR group than in the non-DNR group. Given the correlation between CSF and plasma SNAP-25 levels (r = 0.89), two separate multivariable logistic regression models were constructed to avoid multicollinearity, higher SNAP-25 levels were associated with a reduced risk of DNR. ROC analysis showed that CSF SNAP-25 had an AUC of 0.805 (95% CI: 0.706-0.905), with an optimal cutoff value of 737.5 pg/mL (sensitivity: 0.905; specificity: 0.569), whereas plasma SNAP-25 had an AUC of 0.727 (95% CI: 0.618-0.837), with an optimal cutoff value of 543.5 pg/mL (sensitivity: 0.714; specificity: 0.708).

CONCLUSION

Preoperative SNAP-25 in both CSF and plasma is associated with DNR; although CSF SNAP-25 shows better discrimination, plasma SNAP-25 offers more balanced performance and greater clinical feasibility as a practical biomarker for identifying at-risk patients.

TRIAL REGISTRATION: ChiCTR2500113164.