Abdominal aortic calcification predicts anastomotic leakage after gastrectomy: a multicenter retrospective cohort study from the Hiroshima Surgical study group of Clinical Oncology

Anastomotic leakage (AL) is a critical complication following gastrectomy for gastric cancer (GC). Abdominal aortic calcification (AAC), a surrogate marker of systemic atherosclerosis and frailty, has been associated with AL only in single-center studies. This multicenter study evaluated the impact of preoperative AAC on the incidence of AL. We retrospectively analyzed 750 patients who underwent radical gastrectomy for pStage I–III GC between 2021 and 2023 at six institutions affiliated with the Hiroshima Surgical study group of Clinical Oncology (HiSCO). AAC was quantified using the Agatston score on preoperative computed tomography scans. Patients were divided into high- and low-AAC groups using an exploratory cutoff value of 540 derived from receiver operating characteristic analysis. Multivariable and propensity score matching (PSM) analyses were performed. A high Agatston score was independently associated with AL in a multivariable model restricted to preoperatively available variables (odds ratio [OR], 2.02; 95% confidence interval [CI], 1.03–3.98; p = 0.041) and in a comprehensive model additionally adjusted for intraoperative factors (OR, 2.04; 95% CI, 1.02–4.08; p = 0.044). When analyzed as a continuous variable, the strength of this association was attenuated. Conversely, when analyzed as a binary variable after adjusting for institutional clustering, statistical significance was maintained (OR, 2.20). After PSM, the high-AAC group still showed a significantly higher incidence of AL. Among patients undergoing Billroth I reconstruction, AL was significantly more frequent in the high-AAC group (7.6% vs. 0.9%, p = 0.002), particularly after laparoscopic or robotic surgery; no significant differences were observed for other reconstruction methods. Preoperative severe AAC, quantified by the Agatston score, was associated with an increased risk of AL after gastrectomy, particularly among patients undergoing Billroth I reconstruction. However, considering the borderline statistical significance, the exploratory nature of the data-derived cutoff value, and the retrospective design, these findings should be interpreted cautiously as hypothesis-generating. The clinical utility and generalizability of AAC for risk assessment require rigorous validation in independent prospective cohorts.