Abstract
AIM: Enhanced recovery after surgery (ERAS) gradually shortens the length of stay but increases the rate of unplanned readmission after discharge. Currently, objective discharge criteria for patients after radical gastrectomy is lacking. This study aimed to construct and validate a nomogram for estimation of the possibility of safe discharge on the fifth-day post radical gastrectomy. METHODS: We enrolled 496 consecutive patients undergoing radical gastrectomy as the development cohort. After the fifth day of surgery, patients were assigned to the postoperative complication group and no postoperative complication group. Multivariate logistic regression analyses were performed for both groups. Then, we constructed the risk prediction model of postoperative severe complications (PSCs) and applied it to evaluate whether the patient could be discharged safely. The external validation cohort comprised 245 patients, whom we used to evaluate the capability of our model to predict the risk of PSCs. The primary measure was the negative predictive rate (NPR) and the area under the curve (AUC). RESULTS: Through multivariate analysis, gender, maximum body temperature on the 4th postoperative day (POD4), oral intake and ambulatory duration on POD4, the proportion of neutrophils (≥75% or <75%) and pain score (≥4 or <4) on POD5, and defecation with 5 days after the procedure (yes or no) were identified as independent predictors for PSCs. Upon incorporation of these variables, the nomogram demonstrated a good NPR of 0.957 and 0.916 and AUC of 0.918 and 0.719 in the two cohorts, respectively. With a nomogram score of 110, patients were stratified into low and high risk of PSCs. CONCLUSION: The nomogram demonstrated good predictive potential for low-risk patients. It could serve as an objective safe discharge approach for patients after the fifth day of radical gastrectomy.