Postoperative cognitive dysfunction (POCD) often occurs in elderly patients and may involve neuroinflammation. This study was to determine whether anesthetic choice (intravenous vs. volatile anesthetics) affects cognitive impairment and neuroinflammation in elderly rat. Total 54 twenty-month old male Fischer 344 rats were assigned randomly to control, right carotid exposure under propofol-buprenorphine or isoflurane-buprenorphine anesthesia groups. They were tested by Barnes maze and fear conditioning from 6 days after the surgery. Their brains were harvested 24 h after the surgery for quantifying interleukin (IL) 1β, tumor necrosis factor (TNF)α and ionized calcium binding adaptor molecule 1 (Iba-1). We showed that the heart rates and mean arterial blood pressure were similar during surgery under propofol-buprenorphine or isoflurane-buprenorphine anesthesia. There was no difference in the surgery-induced increase of the plasma IL-1β and TNFα levels under these two types of anesthesia. Rats subjected to surgery took longer than control rats to identify the target hole 8 days after the completion of training sessions in Barnes maze [32 ± 23 s for control, 118 ± 64 s for propofol group (P < 0.05 vs. control), 107 ± 64 s for isoflurane group (P < 0.05 vs. control)] and had less freezing behavior in the fear conditioning test. Surgery and anesthesia increased IL-1β and Iba-1 but did not affect tau phosphorylated at S199/202 and S396 in the cerebral cortex and hippocampus. Our results suggest that surgery under general anesthesia induces neuroinflammation and cognitive impairment. Anesthetic choice may not be a significant modifiable factor for these effects.