Prognostic value of preoperative nutritional status in patients with gastric cancer undergoing radical gastrectomy: a comprehensive meta-analysis and multidimensional risk assessment.

BACKGROUND

Gastric cancer (GC) is the fifth leading cause of cancer-related death globally and imposes significant clinical and economic burdens, particularly in high-incidence regions, such as East Asia. Despite advancements in treatment, postoperative complications persist in 12.8% to 14.0% of patients, with malnutrition being a crucial risk factor. Geographic variations in treatment paradigms (eg, predominant neoadjuvant therapy in the West vs upfront surgery in East Asia for selected patients with stage III disease) necessitate region-specific prognostic analyses of the same.

METHODS

This study aimed to evaluate the prognostic role of preoperative nutritional indices, the Prognostic Nutritional Index (PNI) and Nutritional Risk Screening 2002 (NRS-2002), using a dual approach: a meta-analysis of 27 global cohorts (2016-2023: 19,285 patients) and a retrospective cohort study (2021-2024: 203 patients).

RESULTS

Pooled meta-analysis revealed that low preoperative PNI was significantly correlated with reduced overall survival (OS; multifactorial study hazard ratio [HR], 1.344 [95% CI, 1.111-1.625]; P <.001; univariate study HR, 3.140 [95% CI, 1.564-6.302]; P <.001) and disease-free survival (DFS; multifactorial study HR, 0.977 [95% CI, 0.667-1.432]; P <.001; univariate study HR, 2.890 [95% CI, 2.236-3.735]; P =.185), alongside increased risks of infectious complications (risk ratio, 0.665 [95% CI, 0.544-0.814]; P <.001). Retrospective data demonstrated that the combined PNI and NRS-2002 assessment improved the predictive accuracy for early postoperative complications (area under the curve: 0.657 for PNI + NRS-2002 vs 0.541 for PNI alone). Patients with complications experienced prolonged hospitalization (Δ = 6.31 days; P <.001) and higher costs. Multivariate analysis identified low PNI (odds ratio [OR], 2.24) and high NRS-2002 (OR, 2.49) as independent predictors of adverse outcomes.

CONCLUSION

Our findings underscore the necessity of integrating preoperative nutritional screening into clinical protocols to stratify high-risk patients, guide targeted interventions, and mitigate the economic burden associated with postoperative morbidity. Our multidimensional model highlights the synergistic value of combining immunonutritional biomarkers with conventional risk tools, offering a translatable framework for optimizing perioperative care in GC. Although our findings reflect GC management in high-incidence regions, validation in Western cohorts undergoing neoadjuvant therapy is required.