A prediction model for potential intraoperative laparoscopic hemostasis in spleen-preserving No. 10 lymphadenectomy for proximal gastric cancer

Abstract

To identify the risk factors for intraoperative laparoscopic hemostasis during laparoscopic spleen-preserving splenic hilar lymph node dissection (LSPSD) for proximal gastric cancer (GC) and to develop and validate a model to estimate the risk of intraoperative laparoscopic hemostasis. Between January 2011 and December 2014, we prospectively collected and retrospectively analyzed the medical records of 398 patients with proximal GC who underwent LSPSD. The data were split 75/25, with one group used for model development and the other for validation testing. Of the 398 patients enrolled in this study, 174 (43.7%) required laparoscopic hemostasis treatment. A multivariate analysis determined that the risk factors for the model group were gender, preoperative N stage, and terminal branches of the splenic artery (SpA), and each factor contributed 1 point to the risk score. The intraoperative laparoscopic hemostasis rates were 11.5%, 33.6%, 58.5%, and 73.5% for the low-, intermediate-, high-, and extremely high-risk categories, respectively (p < 0.001). Blood loss volume (BLV) and operative time (in min) for LSPSD increased significantly (p < 0.001) as the risk increased. The area under the receiver operating characteristic curve for the intraoperative laparoscopic hemostasis score was 0.700. The observed and predicted incidence rates were parallel for intraoperative laparoscopic hemostasis in the validation set. This simple, efficient scoring system using the factors for gender, preoperative N stage, and terminal SpA branches can accurately predict the risk of intraoperative laparoscopic hemostasis during LSPSD to improve surgical safety.

Keywords: Laparoscopic hemostasis; Laparoscopy; Scoring system; Spleen-preserving No. 10 lymph node dissection; Stomach cancer.