Association between BDNF and postoperative cognitive dysfunction across surgical stages: a systematic review and meta-analysis
Postoperative cognitive dysfunction (POCD) is a frequent complication after surgery, particularly among older adults, and its underlying mechanisms remain incompletely understood. Brain-derived neurotrophic factor (BDNF) is a key regulator of neuronal survival and synaptic plasticity, and has been implicated in POCD but published findings are inconsistent. We therefore performed a systematic review and meta-analysis to evaluate the association between perioperative BDNF levels and POCD across surgical time points. Following PRISMA guidelines, we searched PubMed, EMBASE, Web of Science, and Scopus were searched for observational studies that reported perioperative BDNF levels in patients with and without POCD. Two reviewers independently extraction data and assessed study quality using the Newcastle-Ottawa Scale. Statistical analyses were conducted in RevMan 5.3, with mean differences (MD) with 95% confidence intervals (CI) were calculated for absolute BDNF levels and perioperative changes (ΔBDNF) at preoperative, intraoperative, and postoperative stages. Six studies including 135 patients with POCD and 247 without POCD met the inclusion criteria. The fixed-effects meta-analysis showed lower intraoperative BDNF levels in the POCD group (MD = −5.62, 95% CI[−8.95, −2.29], P = 0.0009; I2 = 0%). Similarly, intraoperative ΔBDNF was reduced in the POCD group under a fixed-effects meta-model (MD = −9.73, 95% CI[−16.08, −3.38], P = 0.003; I2 = 0%). The subgroup analysis indicated that the type of surgery may contribute to the high heterogeneity observed in 24-h postoperative BDNF and in 24-h postoperative ΔBDNF. These findings suggest that intraoperative BDNF levels and their dynamic changes may serve as preliminary exploratory biomarker for POCD. The intraoperative phase appears to be a critical window for BDNF monitoring, yet prospective cohort studies are urgently needed to externally validate its uesfulness for screening patients at high risk of POCD before clinical implementation. Current evidence remains preliminary, since only six small observational studies were included and substantial statistical heterogeneity was present in multiple postoperative subgroup analyses. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD420251265921.