Backgrounds： With the continuous improvement of sterilization and endoscopic structure, the infection caused by endoscopy has gradually declined.With the rapid development of digestive endoscopic therapy in the past decade, therapeutic endoscopy has been widely carried out worldwide. These techniques have caused the mucous membrane or deeper damage to achieve the goal of curing the disease. During therapeutic endoscopic procedures, endogenous bacteria may be ectopic to the blood circulation due to mucosal or deeper damage. The endoscope is used to in and out lumens multiple times, and the injections are injected into the tissues through the accessories. These processes may bring the pathogenic bacteria from the patient's mouth into the digestive tract through the endoscope and enter the blood through the damaged mucosa. In addition, bacteremia associated with endoscopic procedures may cause bacterial infections in distant organs (eg infective endocarditis).

Postoperative infection rates can reach 12-22% . The results of the etiological culture show that it is consistent with the bacteria of oral bacteria. It is possibly related to multiple passage into the digestive tract of endoscopic and accessory . However, the endoscopic operation process will inevitably lead to subsequent infections.

ESD treatment involves endoscopic multiple access to the upper digestive tract through the mouth, attachments and injection needles and other multiple exposure to the wound, so the probability of postoperative infection is significantly higher than the average endoscope.

Investigators proposed to gargle patients with chlorhexidine acetate before ESD to improve the oral microenvironment and reduce the pathogenic bacteria in the oral cavity, so as to observe whether it can achieve the effect of preventing postoperative infection.

Methods and patients

1. Objectives: This study is a prospective randomized controlled study in single center of Peking University Third Hospital. The purpose of the study is to evaluate the preventive effect of chlorhexidine acetate gargle on the infection of early upper gastrointestinal cancer after endoscopic ESD therapy.
2. Calculation of sample size: According to the postoperative infection rate of 10%, 25% improvement is given after the gargle is administered, and the error range is calculated at 2%. The sample size needs 306 cases.

The investigator proposed to gargle patients with chlorhexidine acetate before ESD to improve the oral microenvironment and reduce the pathogenic bacteria in the oral cavity. The main purpose of the study is to evaluate the preventive effect of chlorhexidine acetate gargle on the infection of early upper gastrointestinal cancer after endoscopic ESD therapy.

Patients and Methods:

Patients:

Inclusion criteria:

1. Diagnosis of early gastrointestinal cancer
2. in line with endoscopic ESD treatment indications
3. no limitation of age

Exclusion criteria:

1. patients do not agree with the mouthwash
2. The patient requested surgery

Methods

Screening and enrollment:

1. Patients with early-stage upper gastrointestinal cancer diagnosed endoscopically in the ivestigator's hospital.
2. ESD indications evaluation: endoscopic ultrasonography was used to assess the depth of lesion infiltration and surrounding lymph nodes. Abdominal or chest CT was used to further evaluate local lymph nodes and distant metastases.
3. Patients with indications for treatment sign informed consent

Treatment programs:

1. After the patient signed the informed consent, they will be divided into the experimental group and control group according to random envelopes in accordance with the sequence of patients were admitted. Patients in the treatment group are given twice daily three days before endoscopic treatment.Patients in the controlled group will not receive gargle.
2. Treatment of second-generation cephalosporins or levofloxacin (allergic cephalosporins) intravenously half an hour before ESD procedure to prevent infection.
3. The main steps of ESD include: marking around the lesion, submucosal injection to fully lift of the lesion, cutting the mucosa around the circumference, submucosal peeling to completely separate the submucosa and the muscularis propria to achieve a complete resection of the lesion, and checking the wound surface vessels and lesions.
4. Monitor blood routine and body temperature changes on the first and third days after treatment. If there are signs of infection in the clinic, monitor PCT and add appropriate antibiotics.

Calculation of sample size: According to the postoperative infection rate of 10%, 25% improvement is given after the gargle is administered, and the error range is calculated at 2%. The sample size needs 306 cases.

Inclusion Criteria:

1. Diagnosis of early gastrointestinal cancer or high grade intraepithelial neoplasia
2. In line with endoscopic ESD treatment indications
3. No limitation of age

Exclusion Criteria:

1. Patients do not agree with gargling
2. The patient requested surgery