Predictive value of geriatric nutritional risk index in cardiac and cerebrovascular events after endovascular aortic aneurysm repair.

Objective: To evaluate the effect of malnutrition assessed by the Geriatric Nutritional Risk Index (GNRI) on major adverse cardiac and cerebrovascular events (MACCE) in the elderly patients after endovascular aortic aneurysm repair (EVAR).

Materials and methods: This was a retrospective cohort study of elderly patients who underwent EVAR in a tertiary hospital. Malnutrition status was assessed by the GNRI. The primary outcome was MACCE. The predictive ability of the GNRI was compared with both the Revised Cardiac Risk Index (RCRI) and the modified Frailty Index (mFI) using Receiver operating characteristic (ROC) curve.

Result: A total of 453 patients underwent EVAR November 2015 and January 2020 was retrospectively analyzed, equally divided into three (low/medium/high) groups according to GNRI values which ranked from low to high. Five (1.10%) patients were lost in follow-up after surgery, and the median length of follow-up was 28.00 (15.00-47.00) months. The high GNRI values reduced length of hospital stay following EVAR in comparison to patients in low GNRI values group (β 9.67, 95% CI 4.01-23.32, p = 0.0113; adjusted β -1.96, 95% CI -3.88, -0.05, p = 0.0454). GNRI status was associated with a significantly increased risk of long-term mortality after EVAR (Medium GNRI, unadjusted HR 0.40, 95%CI 0.23-0.70, p = 0.0014; adjusted HR 0.47, 95%CI 0.26-0.84, p = 0.0107; high GNRI, 0.27 95%CI 0.14-0.55; p = 0.0003; adjusted HR 0.32 95%CI 0.15-0.68, p = 0.0029). Both medium and high GNRI values were linked to significantly reduced risks of MACCE compared to low GNRI score patients (Medium GNRI, unadjusted HR 0.34, 95%CI 0.13-0.88, p = 0.00265; adjusted HR 0.37, 95%CI 0.14-0.96, p = 0.0408; High GNRI, 0.26 95%CI 0.09-0.78; p = 0.0168; adjusted HR 0.21 95%CI 0.06-0.73, p = 0.0029). Compared with the RCRI and mFI, the GNRI had better discrimination in predicting long-term MACCE. An area under the curve (AUC) for GNRI mFI, and RCRI is 0.707, 0.614 and 0.588, respectively. (Z statistic, GNRI vs. mFI, p = 0.0475; GNRI vs. RCRI, p = 0.0017).

Conclusion: Malnutrition assessed by the GNRI may serve as a useful predictor of long-term MACCE in elderly patients after EVAR, with preferable discrimination abilities compared with both RCRI and mFI.