Objective To evaluate the incidence and the causes of the intra-and postoperative cerebrospinal fluid(CSF)leakage during transsphenoidal surgical removal of pituitary adenomas,and discuss the sella closure methods.Method During the period from January 1998 to May 2008,118 patients underwent 134 transsphenoidal operations for pituitary adenomas.Two different methods to close the sella were used.The first one consisted packing the sella turcica and sphenoidal sinus with autologous fat and restoring the defect of sella turcica with autologous bone.In the second method,tht regenerated oxidized cellulose and collagen sponge with human fibrin were used to cover the seUa membrane defect and the dural defect,followed by packing the seHa with autologous fat.Proportions of each type of pituitary adenomas were analyzed.The frequency of the intra-and postoperative CSF leakage were evaluated in different type tumors and different operations.Data had been studied by statistical analysis.Results The incidence rate of intraoperative CSF leakage(26.5%,9/34)Was obviously increased in patients with growth homone adenoma compared with other type of adenoma(13.0%,13/100).Patients with invasive adenoma had an increased incidence rate of intraoperative CSF leakage(21.1%,8/38)compared with noninvasive adenoma(14.6%,14/96)(P＜0.05).When the descent of the sella turcica membrane Was visualized with tumor being totally removed,the incidence rate of intraoperative CSF leakage increased(18.7%,20/107)compared with when membrane descent was not seen with tumor remaining(7.4%,2/27)(P＜0.05).There were 3 cases postoperative CSF leakage happened in all 12 cases which adopted the first method. By the second method,there was no postoperative CSF leakage,and the number of postoperative complications was significantly smaller than the first method (10.0% and 33.3%,respectively). Conclusions The technique of covering the sella membrane and dural defects with oxidized cellulose and collagen sponge with human fibrin in the presence of intraoperative CSF leakage appears to be the most reliable,as no postoperative CSF leakage after applying this technique has been observed.

0.05),while there was difference between length of hospital stay,expense of antibacterial for prevention and total expense of hospitalization(P