Sanne K Stuart, Julie F M Geerts, Bastiaan R Klarenbeek, Merlijn Hutteman, Misha D P Luyer, Sjoerd M Lagarde, Bas P L Wijnhoven, Grard A P Nieuwenhuijzen
While postoperative complications are established predictors of readmission after esophagectomy, reasons for readmission are still elusive, specifically in patients without complications during the index admission. This study assessed reasons for readmission and explored potential predictors in this lower-risk subgroup. A retrospective multicenter cohort study included all patients ≥18 years who underwent curative esophagectomy between 2020 and 2023 across three Dutch hospitals. Patients with in-hospital mortality or missing readmission data were excluded. The primary outcome was the reason for 30-day readmission. Secondary outcome was the identification of risk factors for readmission using uni- and multivariable logistic regression. Of 663 patients, 350 (52.8%) had an uncomplicated index admission, with a 30-day readmission rate of 12.6% (44/350). The most common causes for readmission were anastomotic leakage (32.6%), pulmonary complications (20.9%), insufficient intake (14.0%), and infections (14.0%). No significant associations were found with age, sex, the American Society of Anesthesiologists (ASA) classification, Charlson Comorbidity Index, tube feeding, length of stay, or surgical approach. Among patients readmitted with anastomotic leakage, 11/14 had met functional recovery criteria at discharge. Three patients had unexplained subtle increases in C-reactive protein (CRP) or leukocyte count. Three patients underwent a computed tomography (CT) scan during the index admission. Readmission remains common, even in patients with an initially uneventful recovery, and is frequently caused by undetected or delayed complications including anastomotic leakage. Traditional predictors did not identify patients at risk for readmission in this cohort. Subtle deviations before discharge may represent early warning signs and require validation in future prospective research.