Gastric Cancer. 2026 Jul 29. doi: 10.1007/s10120-026-01783-2. Online ahead of print.
ABSTRACT
BACKGROUND: Surgical complications after gastrectomy significantly impact recovery and long-term prognosis. While iron status influences outcomes, the conventional 20% transferrin saturation (TSAT) threshold ignores sex-specific physiological differences. We developed a modified TSAT (mTSAT) based on sex and anemia status and evaluated its value in predicting complications.
METHODS: We retrospectively analyzed 243 patients who underwent curative gastrectomy. Optimal TSAT cut-off values for predicting complications classified as Clavien-Dindo grade II or higher were determined using receiver operating characteristic curves stratified by sex and anemia status. Predictive performance was compared between mTSAT and the conventional 20% threshold using the corrected Akaike Information Criterion.
RESULTS: The optimal mTSAT cut-offs were 22.2% for men and 14.2% for women. The mTSAT model demonstrated an improved fit compared with the conventional threshold. The low mTSAT group (mTSAT below the sex-specific cut-off) comprised 78 patients and exhibited a significantly higher incidence of complications than the normal mTSAT group, particularly anastomotic leakage and delayed gastric emptying/ileus. Moreover, postoperative hospital stay was significantly longer in this group. In the primary multivariable analysis, low mTSAT was significantly associated with postoperative complications, whereas the conventional 20% threshold was not. After additional adjustment for clinical stage and surgical approach, the association was attenuated and no longer reached statistical significance.
CONCLUSIONS: Preoperative low mTSAT was associated with an increased risk of postoperative complications. Rather than a standalone predictor, mTSAT functions as a valuable composite clinical indicator reflecting stage-related systemic vulnerability, surgical invasiveness, and underlying malnutrition or iron deficiency.
PMID:42525320 | DOI:10.1007/s10120-026-01783-2
