Visceral adiposity rather than BMI predicts overall survival after curative gastrectomy for gastric cancer: a retrospective cohort study with age-stratified analyses.

[OBJECTIVE] The prognostic value of body mass index (BMI) remains contested in gastric cancer, in part because BMI conflates distinct compartments of body composition. This study examined the respective associations of BMI and computed tomography-derived visceral fat area (VFA) with postoperative morbidity and overall survival (OS) after curative gastrectomy, while also probing whether age modifies these relationships.

[METHODS] A retrospective cohort was assembled from patients undergoing curative-intent gastrectomy for gastric cancer at the First Affiliated Hospital of Henan Medical University between 2015 and 2019. BMI was grouped using WHO Asia-Pacific thresholds, whereas VFA was quantified on preoperative CT images at the L3 level. OS was prespecified as the primary endpoint, with postoperative complications treated as a secondary endpoint. Survival was characterized using Kaplan-Meier estimation and tested with Cox proportional hazards modeling under multivariable adjustment. Age-stratified analyses were implemented to evaluate effect modification.

[RESULTS] Among 868 eligible patients, underweight status was associated with significantly worse OS, whereas higher BMI categories were not accompanied by an increased mortality risk. In contrast, elevated VFA independently predicted inferior OS irrespective of BMI. The BMI-OS association differed across age strata, suggesting age-dependent heterogeneity in the prognostic meaning of BMI, while the adverse association between visceral adiposity and survival was comparatively consistent across age groups. Correlation between BMI and VFA was weak, reinforcing the premise that these metrics encode non-overlapping biological information.

[CONCLUSION] In gastric cancer patients treated with curative gastrectomy, BMI and fat distribution appear to convey divergent prognostic signals. Visceral adiposity, as indexed by CT-derived VFA, provides survival-related information beyond BMI and may more effectively flag patients at heightened risk for adverse outcomes. Routine incorporation of CT-based body composition profiling into preoperative evaluation could refine risk stratification and support more individualized perioperative management.