Preoperative geriatric nutritional risk index as a predictor of postoperative delirium in revision arthroplasty: a 10-year retrospective cohort study

Abstract

Objective: This study aims to investigate the association between the preoperative geriatric nutritional risk index (GNRI) and postoperative delirium (POD) in patients undergoing hip or knee revision arthroplasty.

Methods: 820 patients who underwent hip or knee revision arthroplasty from January 2014 to September 2024 were included. The exposure variable was preoperative GNRI, and the outcome variable was POD, diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria using the Confusion Assessment Method (CAM). The study considered covariates such as age, sex, body mass index, albumin, and comorbidities, employing multivariate logistic regression analysis to explore the association between preoperative GNRI and POD.

Results: Among 820 patients, 76 (9.27%) developed POD within 7 days postoperatively. Patients with POD had a significantly lower GNRI (97.53 ± 9.54) compared to those without POD (101.05 ± 8.85, p = 0.003). For each 1-unit increase in GNRI, the risk of POD decreased by 4% (OR = 0.96, 95% CI: 0.94-0.99, p = 0.011). Quartile analysis showed that patients in the highest GNRI quartile had a significantly lower POD incidence compared to those in the lowest quartile (OR = 0.43, 95% CI: 0.20-0.92, p for trend = 0.037). A protective threshold of GNRI was identified at 101.96.

Conclusion: A significant association was observed between preoperative GNRI and POD in patients undergoing hip or knee revision arthroplasty. However, due to the retrospective single-center design and potential unmeasured confounding, further multicenter prospective studies are warranted to validate these findings and explore underlying mechanisms.

Keywords: geriatric nutritional risk index; hip; knee; postoperative delirium; revision arthroplasty.

Figure 1

Flow chart of enrollment. RTHA, revisional total hip arthroplasty; RTKA, revisional total knee arthroplasty; GNRI, geriatric nutritional risk index; POD, postoperative delirium.

Figure 2

The column chart illustrates the relationship between POD and GNRI values. (A) The average GNRI value was 101.05 in the Non-POD group and 97.53 in the POD group. **p < 0.01; (B) Highlights the variability in POD incidence across GNRI quartile ranges: 15.61% in the Q1 group, 5.91% in the Q2 group, 8.78% in the Q3 group, and 6.78% in the Q4 group. GNRI, geriatric nutritional risk index; POD, postoperative delirium.

Figure 3

Relationship between the GNRI value and POD in patients of RTHA and RTKA. (A) Adjusted OR for POD according to values of GNRI; (B) Predicted probabilities and the observed rate of POD. The multivariate model was adjusted for BMI, COPD, valvular heart disease, cerebrovascular, solid-tumor, depression, VTE, admission to surgery, age, sex, surgery time, length of stay, and surgical site. OR, odds ratio; CI, confidence interval; GNRI, geriatric nutritional risk index.

Figure 4

Subgroup analysis of the association between GNRI and POD. OR, odds ratio; CI, confidence interval; BMI, body mass index; COPD, chronic obstructive pulmonary disease; VTE, venous thromboembolism.