Updated systematic review and meta‑analysis for risk predictors of 30‑day unplanned readmission following gastric cancer surgery.

Readmission to hospital is a common occurrence due to adverse post-operative events. The primary objective of the present study was to examine the possible risk predictors for 30-day unplanned readmissions after the surgical treatment of gastric cancer. The secondary aim was to determine the clinical characteristics that are associated with readmission. Studies that reported the risk factors and clinical characteristics of patients with gastric cancer who had an unplanned 30-day post-operative readmission were identified systematically from PubMed, Cochrane Central Register of Controlled Trials, Web of Science and Embase databases, with a final search date of March 30, 2024. A systematic review and meta-analysis were then performed to estimate the risk predictors and relevant clinical characteristics for readmission. A total of 16,154 patients from 12 studies were included in the present study, with 1,736 patients who were readmitted and 14,418 patients who were not readmitted. A higher proportion of patients with an age ≥70 years, cardiovascular comorbidity, Nutritional Risk Screening (NRS) 2002 score ≥3, respiratory diseases, male sex, American Society of Anesthesiologists score ≥3, combined multi-organ resection, greater depth of invasion (T3-4/T1-2), discharge to home with provision of care services, neoadjuvant therapy, post-operative complications or a blood transfusion were found in the readmission group compared with that in the non-readmission group. A meta-analysis was also performed to calculate risk predictors using the results of multivariate regression analyses from the original literature. This identified cardiovascular comorbidity, NRS 2002 score ≥3, pancreatectomy and post-operative complications as risk predictors for 30-day unplanned readmission following surgery for gastric cancer. Therefore, it is recommended that extra attention and support should be given to patients with these high-risk predictors during the perioperative period.