Validation of a Model for Predicting Duodenal Stump Leakage After Gastrectomy

A Multicenter Prospective Study of Artificial Intelligence Predicting Duodenal Stump Leakage After Laparoscopic Radical Gastrectomy for Gastric Cancer

This study aims to validate a machine learning model for predicting duodenal stump leakage after laparoscopic radical gastrectomy for gastric cancer.

Gastrectomy is an essential procedure in radical surgery for gastric cancer. Duodenal stump leakage (DSL) is one of the critical short-term complications after distal and total gastrectomy in gastric cancer patients. Identifying patients with high-risk of DSL will assist the surgeons' decision making to give efficient previous intervention, such as a more rigorous operation, placing dual-lumen flushable drainage catheter and decompression tube in afferent loop. Investigators have developed a high-performance machine learning model based on 4070 gastric cancer patients, which showed good discrimination of DSL. Hence, this multi-center prospective study will validate the reliability of this model for predicting DSL in gastric cancer patients who receive laparoscopic distal or total gastrectomy.

Inclusion Criteria:

1. Aged older than 18 years and younger than 85 years
2. Primary gastric carcinoma confirmed by preoperative pathology result
3. Expected curative resection via laparoscopic distal or total gastrectomy and reconstruction via Billroth-II or Roux-en-Y anastomosis
4. American Society of Anesthesiologists (ASA) class I, II, or III
5. With full documents of preoperative examinations such as blood test and abdominal CT scanning
6. Written informed consent

Exclusion Criteria:

1. Pregnant or breastfeeding women.
2. Severe mental disorder or language communication disorder.
3. Other surgical procedures of gastrectomy is performed.
4. Interrupted of surgery for more than 30 minutes due to any cause.
5. Malignant tumors with other organs
6. Performed gastrectomy in the past