Complications with laparoscopically assisted gastrectomy: multivariate analysis of 300 consecutive cases.

Background: Complications associated with laparoscopically assisted gastrectomy (LAG) are not significantly different from those associated with open gastrectomy. However, additional risks related to abdominal access, pneumoperitoneum, and special electrosurgical instruments result in an increased incidence of complications with LAG. This study analyzed the causes and risk factors linked to postoperative morbidity.

Methods: A retrospective review analyzed the data of 300 patients who underwent consecutive LAG for gastric cancer in our department from May 2003 to October 2006. Among the 300 patients, total gastrectomy was performed for 42 patients, distal gastrectomy for 258 patients, and proximal gastrectomy for 3 patients. The clinical and operative data obtained included body mass index, medical comorbidities, history of previous abdominal surgery, operative time, type of surgery, extent of lymph node dissection according to the Japanese Guideline, number of retrieved lymph nodes and lymph node metastases, additional operative procedure, depth of tumor invasion, and disease stage. The outcome data consisted of mortality, major morbidities, and postoperative hospital stay. The 300 cases were divided into two periods: 50 cases in the first period and 250 cases in the second period.

Results: Postoperative complications developed in 61 cases (20.3%), wound infection in 21 cases (7%), intraabdominal abscess in 3 cases (1%), bleeding in 12 cases (4%), stenosis in 13 cases (4.3%), leakage in 3 cases (1%), acute pancreatitis in 2 cases (0.7%), pulmonary complication in 4 cases (1.3%), renal complication in 4 cases (1.3%), and cardiac complication in 2 cases (0.7%). The 30-day mortality rate was 0.7% (n=2). Univariate analysis proved that gender, operative period, comorbidity, and operative times were important risk factors. Multivariate analysis proved that cormobidity and operative period were important risk factors.

Conclusion: The data suggest that LAG can be performed with acceptable perioperative complication rates. The surgeon's experience and careful patient selection determined optimal patient outcomes.