Objective The aim of this study is to observe the influence of sevoflurane or propofol anesthesia on post-operative cognitive functions of elderly patients and their serum homocysteine （Hcy）.　Methods　Forty elderly patients，ASA Ⅰ-Ⅱ，undergoing gastrectomy surgery， were randomly divided into sevoflurane group （group A） and propofol group （group B）. The venous blood sample was collected to test the homocysteine concentration by enzyme-linked immunosorbent assay （ELISA） at the begining of surgery（T1）， 2 h during the surgery（T2）， 24 h（T3） and 72 h after the surgery （T4）. Meantime， the mini-mental state examination table（MMSE） was assessed. At 1 d before the surgery（S1）， 6 h（S2）， 24 h（S3） and 72 h after the surgery（S4）.　Results　Compared with the preoperative values， the MMSE score of group B significantly decreased（P<0.05） at S2， and the MMSE score of group A were lower than those of group B（P<0.05）. There was no significant difference between group in preoperative serum Hcy［group A：（5.8±1.4） μmol/L； group B:（5.9±1.6） μmol/L］. Compared with the preoperative values， serum Hcy of group A（6.7±1.5） μmol/L were significantly higher at T2 （P<0.05）. However， serum Hcy decrease to the preoperative value level（P>0.05） at T3 and T4. While in group B， the serum Hcy keep stable at T2， T3 and T4（P>0.05）.　Conclusions　Both propofol and sevoflurane can cause short-term postoperative cognitive functions in the elderly patients. Continuous inhalation of 1.3 MAC sevoflurane for 3.5 h has significant impact on postoperative cognitive dysfunctions than infusion of propofol at the rate of 3 μg·kg-1·h-1-5 μg·kg-1·h-1 for 3.5 h. One of its mechanisms may be associated with increased serum Hcy concentration.