Association of preoperative prognostic nutritional index with risk of postoperative delirium: A systematic review and meta-analysis

Study objective To assess the association between prognostic nutritional index (PNI) and risk of postoperative delirium (POD) in adult patients. Methods MEDLINE, Google scholar, EMBASE, and Cochrane library databases were searched from inception till April 2022. The primary outcome was the association between PNI and the risk of POD, while the secondary outcomes were correlations of other prognostic factors with POD risk. The correlation between PNI and the incidence of POD was assessed with three approaches: Difference in preoperative PNI between POD and non-POD groups (Model 1) as well as the association of PNI as a continuous parameter (Model 2) or as a binary variable (i.e., low vs. high using a PNI cut-off value of 50) (Model 3) with POD risk. Results Analysis of nine observational studies published from 2010 to 2021 recruiting 3,743 patients showed a POD incidence of 6.4–35%. Our meta-analysis demonstrated a lower PNI among patients in the POD group (MD: −3.78, 95% CI: −4.85 to −2.71, p < 0.0001, I2 = 54.2%) compared to the non-POD group (Model 1). Pooled results revealed a negative association between PNI and POD risk for both Model 2 (OR: 0.91, 95% CI: 0.86–0.97, p = 0.002, I2 = 71%) and Model 3 (OR: 1.68, 95% CI: 1.26–2.23, p < 0.0001, I2 = 0%). Besides, while our results supported an age-dependent increase in POD risk, other factors including body-mass index, surgical time, health status, hypertension, diabetes mellitus, and male gender were non-significant predictors of POD. Conclusion Our results demonstrated a negative association between PNI and POD, which warrant further large-scale studies for validation. Systematic review registration https://www.crd.york.ac.uk/prospero/ , identifier CRD42022323809.