Background: Postoperative delirium (POD) represents a serious complication following anesthesia and surgical procedures for patients undergoing surgical intervention. At present, there is no effective index to predict the occurrence of POD ,so the aim of this study was to validate whether cerebrospinal fluid (CSF) PGRN could predict the occurrence of POD.Methods: We conducted a prospective nested case-control and 1:1 matched (on age, diagnosis, American Society of Anesthesiologist’ (ASA) physical status, duration of surgery, and intraoperative blood loss) study. A total of 600 Han Chinese patients over the age of 65~ 90 who underwent unilateral total knee arthroplasty were included in the PNDABLE study from June 2020 to November 2020. POD cases and non-POD controls were selected from using Confusion Assessment Method (CAM) on the first, second, third and seventh (or before discharge) postoperative days. Delirium severity was measured by the Memorial Delirium Assessment Scale (MDAS). In the six month, cognitive function was assessed with the modified Telephone Interview for Cognitive Status (TICS-m) and the World Health Organization Quality of Life brief version (WHOQOLBREF). CSF PGRN and core biomarkers were measured by ELISA using the microplate reader. The associations of CSF PGRN levels with POD risk and CSF core biomarkers (Aβ1–42, Aβ1–40 ,T-tau and P-tau)were assessed. The effect of CSF PGRN on predicting POD occurrence was evaluated with the area under the receiver operating characteristic (ROC) curve (AUC).Results: POD incidence was 9.7% (53/545). There were significant differences in preoperative CSF PGRN concentration between POD patients and non-POD (P<0.001), and CSF PGRN levels decreased with age, as demonstrated by a significantly negative correlation. CSF PGRN levels decrease with CAM scores and MDAS scores, as demonstrated by significantly negative correlations(r=-0.692, P=0.001; r=-0.435, P=0.001). There were positive associations of CSF PGRN with Aβ1–42 (β = 0.756, P < 0.001) , Aβ1–40 (β =0.637, P < 0.001) and negative associations of CSF PGRN with T-tau (β = -0.716, P < 0.001) and P-tau (β = -0.739, P < 0.001) in POD patients. The ROC curve analysis of PGRN showed that PGRN concentrations had high diagnostic value for POD.Conclusions: Aβ pathology is associated with increasing in CSF PGRN whereas Tau pathology and neurodegeneration is associated with decreasing in CSF PGRN in POD patients.CSF PGRN can predict the occurrence of POD in elderly patients.Clinical Trial Registration: www.clinicaltrials.gov, identifier ChiCTR2000033439.