Colorectal cancer is the fourth most common cancer in China. Up to 30% of patients with colorectal cancer present with an emergency obstruction of the large bowel at the time of diagnosis, and 70% of all malignant obstruction occurs in the left-sided colon. Patients with obstruction are associated with worse oncologic outcomes compared with those having nonobstructive tumors. Conventionally, patients with malignant large bowel obstruction receive emergency surgery, with morbidity rates of 30%-60% and mortality rates of 7-22%, and about two-thirds of such patients end up with a permanent stoma.

Self-expanding metallic stents (SEMS) haven been used as a bridge to surgery (to relieve obstruction prior to elective surgery) in patients with potentially resectable colorectal cancer. Several clinical trials demonstrate that SEMS as a bridge to surgery may be superior to emergency surgery considering the short-term outcomes. SEMS is associated with lower morbidity and mortality rate, increased primary anastomosis rate, and decreased stoma creation rate. Although about half of patients can achieve primary anastomosis after stent placement, the primary anastomosis rate is still significantly lower compared with nonobstructing elective surgery. The interval between stent placement and surgery may be not long enough that bowel decompression is insufficient at the time of operation. Furthermore，the long-term oncologic results regarding SEMS as a bridge to surgery are still limited and contradictory. Sabbagh et al. suggest worse overall survival of patients with SEMS insertion compared with emergency surgery, the 5-year cancer-specific mortality was significantly higher in the SEMS group (48% vs 21%, respectively, P=0.02). One interpretation is that tumor cells may disseminate during the procedure of colonic stenting placement. We hypothesis that immediate chemotherapy after stenting may improve overall survival by eradicating micrometastasis. Moreover, neoadjuvant chemotherapy prolongs the interval between stent placement and surgery, and the time for bowel decompression is more sufficient, which may increase the success rate of primary anastomosis and decrease risk of stoma formation.

Inclusion Criteria:

• Radiologically proven colonic obstruction of the left colon/upper rectum presumed secondary to a carcinoma
• Able to give written, informed consent
• Primary tumor was resectable
• ECOG score 0 or 1
• Haemoglobin greater than 100 g/L after transfusion before chemotherapy,
• White blood cells greater than 3.0×10⁹ /L
• Platelets greater than 100×10⁹ / L;
• Glomerular filtration rate greater than 50 mL per minute as calculated by the Wright or Cockroft formula
• Bilirubin less than 1.5×Upper Limit of Normal(ULN)
• ALT and AST less than 2.5×ULN

Exclusion Criteria:

• Distal rectal cancers(equal or less than 10cm from the anal verge)
• Patients with signs of peritonitis and/or bowel perforation
• Patients who did not give informed consent
• Patients who were considered unfit for operative treatment or refuse surgery.
• Patients with suspected or proven metastatic adenocarcinoma;
• Patients with unresectable colorectal cancer, or planning for palliative treatment.