Short-term outcomes of posterior-to-greater curvature side-to-side linear stapled anastomosis (PG-LS) versus end-to-end hand-sewn anastomosis in totally minimally invasive Ivor-Lewis esophagectomy for esophagogastric junction cancer: a retrospective cohort study

To compare the short-term clinical outcomes of posterior-to-greater curvature side-to-side linear stapled anastomosis (PG-LS) versus end-to-end hand-sewn anastomosis (HS) in totally minimally invasive Ivor-Lewis esophagectomy for esophagogastric junction (EGJ) cancer. This single-center retrospective study enrolled 138 patients with EGJ cancer who underwent totally minimally invasive Ivor-Lewis esophagectomy, comprising 75 patients in the PG-LS group and 63 in the HS group. The primary endpoints were the incidence of anastomotic leakage and early 30-day anastomotic stenosis. The incidences of anastomotic leakage (PG-LS 6.7% vs. HS 11.1%; RD − 4.4%, 95% CI − 14.0% to 5.2%; P = 0.381) and early 30-day anastomotic stenosis (1.3% vs. 3.2%; RD − 1.8%, 95% CI − 6.9% to 3.2%; P = 0.592) did not differ significantly between the PG-LS and HS groups. After adjustment for age, sex, ECOG performance status, and proximal margin, the OR for anastomotic leakage was 0.48 (95% CI 0.13–1.62; P = 0.236); early 30-day anastomotic stenosis was not adjusted due to the very small number of events. Intrathoracic operative time (90.0 min vs. 115.0 min, P < 0.001) and intraoperative blood loss (50.0 mL vs. 100.0 mL, P < 0.001) were significantly reduced in the PG-LS group. The proportions of patients with postoperative leukocytosis (WBC ≥ 10 × 10⁹/L: 40.0% vs. 61.9%, P = 0.016), elevated procalcitonin (PCT ≥ 0.05 ng/mL: 46.7% vs. 66.7%, P = 0.025), and elevated IL-6 (IL-6 ≥ 7 pg/mL: 41.3% vs. 65.1%, P = 0.006) were significantly lower in the PG-LS group than in the HS group. The rates of pulmonary infection (5.3% vs. 17.5%, P = 0.029) and pleural effusion (6.7% vs. 20.6%, P = 0.021) were also significantly lower in the PG-LS group. No statistically significant difference was observed in the incidence of anastomotic leakage or early 30-day anastomotic stenosis between PG-LS and HS. No significant differences were observed between PG-LS and HS in hospital stay, drain removal, ICU stay, or gastrointestinal recovery time; PG-LS may be non-inferior to HS in postoperative recovery. The PG-LS group had shorter intrathoracic operative time, less blood loss, milder postoperative inflammatory response, and lower pulmonary complications. Clinical decision-making should be individualized, taking into account patient preference, individual anatomical conditions, and surgeon experience.