Predictive value of the nutritional risk index for postoperative complications in individuals with pancreatic cancer undergoing pancreaticoduodenectomy.

Purpose: To explore the predictive value of the geriatric nutritional risk index (GNRI) for postoperative complications and their severity in older adults with pancreatic cancer undergoing pancreaticoduodenectomy (PD).

Methods: This study conducted a retrospective analysis of 109 older adults with pancreatic cancer undergoing PD at the Department of Hepatobiliary and Pancreatic Surgery, Affiliated Hospital of Quzhou Medical University, between May 2019 and May 2022. Statistical analysis of clinical data was performed to assess the predictive value of the GNRI for postoperative complications and their severity in older adults with pancreatic cancer undergoing PD.

Results: Among the 109 individuals in this study, a total of 41 older adults undergoing PD experienced postoperative complications. The outcomes of the regression analysis showed that preoperative GNRI (OR = 0.184, 95 % CI = 0.023-0.457, P < 0.001), albumin levels (OR = 0.897, 95 % CI = 0.812-0.912, P < 0.001) and haemoglobin levels (OR = 1.231, 95 % CI = 1.043-1.451, P = 0.034) significantly influence the incidence of postoperative complications in older adults with pancreatic cancer. The prognostic value of the GNRI in predicting overall postoperative complications boasts a sensitivity of 83.2 %, specificity of 71.2 %, positive predictive value of 81.1 %, negative predictive value of 65.9 %, accuracy rate of 73.1 %, area under the curve (AUC) of 0.756 (P < 0.001) and a discriminative threshold of 97.0. Furthermore, the predictive efficacy of the GNRI in gauging the severity of postoperative complications demonstrates a sensitivity of 85.32 % and specificity of 79.54 %.

Conclusion: The GNRI can offer a faster, simpler and more effective method for evaluating nutritional risk in individuals with pancreatic tumours. Moreover, for older adults undergoing PD, it can serve as a convenient and efficient nutritional predictive indicator for postoperative complications and their severity.