RESUMO
Objective:To investigate the association between Onodera's prognostic nutritional index (OPNI) and postoperative adverse outcomes in elderly gastrointestinal surgery patients and assess the predictive value.Methods:A total of 230 elderly patients who received gastrointestinal surgery were prospectively enrolled. Clinical data, including age, sex, preoperative laboratory parameters, surgery process and clinical outcomes, were collected. The optimal cut-off value of OPNI was obtained using NRS 2002, a well-recognized nutritional risk screening tool, as the standard. The associations of OPNI, geriatric nutritional risk index (GNRI) and albumin with in-hospital mortality, complication incidence and duration of postoperative hospital stay were evaluated using Chi-square test or nonparametric test as appropriate. Confounders were identified through univariate analysis and logistic and linear regression models were developed to validate the correlation and assess the predictive value of OPNI for postoperative clinical outcomes.Results:The optimal cut-off value for the OPNI was 41.25, which yielded a sensitivity of 72.7% and a specificity of 59.9% with area under the curve (AUC) at 0.682. The incidence of OPNI-based malnutrition (defined as OPNI < 41.25) was 50% (115/230). Univariate analysis indicated that patients with OPNI < 41.25 had a significantly higher mortality (8.70% versus 2.61%, P = 0.046) and complication incidence (20.00% versus 9.57%, P = 0.026) and significantly longer postoperative hospital stay (11.17 d versus 8.49 d, P = 0.009) than patients with OPNI ≥ 41.25. Patients with GNRI < 98 had a longer postoperative hospital stay than those with GNRI ≥ 98 (10.71 d versus 7.55 d, P = 0.001) while there was no significant difference in mortality or complication incidence between the two groups ( P > 0.05). As for subgroups divided according to albumin levels (< 35 g/L or ≥35 g/L), no significant differences in mortality, postoperative complications incidence, or duration of postoperative hospital stay were observed ( P > 0.05). Multivariate analysis verified that OPNI < 41.25 was an independent risk factor for the development of postoperative complications ( OR: 2.660, 95% CI: 1.079-6.557, P = 0.034) and prolonged postoperative hospital stay ( R2 = 0.135, regression coefficient = 2.73, P = 0.047), where the AUC of the regression model for complications was 0.812 (95% CI: 0.741-0.882). GNRI < 98 was the independent risk factor for prolonged postoperative hospital stay ( R2 = 0.134, regression coefficient = 2.797, P = 0.049). Conclusion:OPNI is an independent risk factor for adverse clinical outcomes after gastrointestinal surgery in elderly patients and demonstrates good predictive value with the cut-off value of 41.25.