Laparoscopic splenectomy： color Doppler flow imaging for preoperative evaluation

Background Laparoscopic splenectomy （LS） is currently the standard approach for resection of a normal-sized spleen. However, this method becomes technical challenge in cases of splenomegaly due to intraoperative hemorrhage. A complete understanding of the splenic vessel anatomy is important to facilitate the difficult laparoscopic procedure. In this retrospective study, we examined the role of color Doppler flow imaging （CDFI） in splenic vessel anatomy and evaluated its value for LS. Methods Forty-eight patients who underwent splenectomy for various hematologic and autoimmune disorders from May 2004 to December 2007 were enrolled in this study. Twenty-three patients underwent preoperative CDFI examination that included examination of the anatomic type of splenic pedicle, the adjacent relationship between the splenic vessel and pancreas, and spleen size （CDFI group）. In the remaining 25 patients, ultrasonic inspections of the splenic vessel were not performed （non-CDFI group）. Laparoscopic splenectomies in the CDFI group were performed in accordance with the information provided by the preoperative CDFI in each patient. In the non-CDFI group, LS was performed according to the conventional method. In the CDFI group, the constituent ratios of the above-mentioned parameters by CDFI were compared with those recorded during LS using the chi square test. The effectiveness of the technique on surgery in both groups was compared with an independent sample Student＇s ttest. Results All laparoscopic splenectomies in both groups were performed successfully. However, 2 cases in the non-CDFI group were converted to LS with the assistance of micro-incision because the branches of the splenic vein were inadvertently torn. Two anatomic types of splenic pedicle and four different adjacent relationships between the splenic vessel and pancreas were detected by CDFI. About 80% of spleens fit the criteria of megalosplenia. There were no statistically significant differences between the constituent ratios of the parameters by CDFI and those by intraoperative telerecording in the CDFI group （χ^2=0.383, 1.072, 0.119, P=0.536, 0.784, 0.730）. However, statistically significant differences were observed in the operative time （（158.70±42.51） minutes vs （200.65±47.89） minutes, P=0.003）, intraoperative blood loss （（55.87±17.36） ml vs （101.83±62.21） ml, P=0.001）, and recovery time of gastrointestinal function （（24.39±8.88） hours vs （30.60±9.45） hours, P=0.024） between the groups. Conclusi