Objective To evaluate the prophylactic effect of perioperative use of moderate-dose intravenous dexamethasone on postoperative nausea and vomiting, and pain after total knee arthroplasty in the elderly patients. Methods This study, a prospective, randomized, controlled trial, selected a total of 160 patients of American Society of Anesthesiologists (ASA) grade Ⅰ or Ⅱ, who were admitted to the Joint Surgery Department of the Sixth People′s Hospital Affiliated to Shanghai Jiao Tong University from January to May 2021 and scheduled for total knee arthroplasty under general anesthesia. They were randomly divided into dexamethasone group (n=80) and saline group (n=80). Intravenously, 2 ml dexamethasone (10 mg) or 2 ml saline was injected 20 min before the induction of general anesthesia. Six patients whose operation exceeded 200 min were excluded, leaving 78 cases in the dexamethasone group and 76 cases in the saline group for statistical analysis. The primary outcome measures were a visual analogue scale (VAS) at postopera-tive 4,12, 24 and 36 h to assess the pain and a verbal rating scale (VRS) to assess the severity and incidence of nausea and vomiting. The secondary indicator were the use of analgesics and antiemetics, blood glucose levels, length of hospital stay, and infection rate at postoperative 30 days. SPSS statistics 22.0 was used for data analysis. Depending on the type of data, analysis of variance, t test, χ² test, or Fisher′s exact test was used for intergroup comparison. Results Nausea, vomiting, and pain occurred mostly within 24 hours after total knee arthroplasty. The VRS scores, incidence of postoperative nausea and vomiting (PONV) and VAS scores in the dexamethasone group were significantly lower than those in the saline group at postoperative 4,12, 24 h (P<0.05), but the difference was not significant at postoperative 36 hours (P>0.05). The blood glucose in dexamethasone group was significantly higher than that in saline group at postoperative 1 day [(8.5±1.7) vs (5.9±1.4) mmol/L; P<0.05], but the difference at postoperative 3 days was not significant (P>0.05). The mean postoperative hospitalization stay was significantly shorter in the dexamethasone group than in the saline group [(3.0±0.8) d vs (4.8±1.5) d; P<0.05]. There was no significant difference between the two groups in the infection rate at postoperative 30 d (P>0.05). Conclusion Preoperative prophylactic intravenous injection of 10 mg dexamethasone can effectively reduce the severity and incidence of nausea and vomiting in the elderly patients after knee arthroplasty, significantly alleviate postoperative pain, reduce postoperative use of analgesics and antiemetics, and shorten hospital stay.