Abstract: BACKGROUND: Total knee arthroplasty is the main therapeutic regimen for end-stage osteoarthritis. However, diabetes mellitus can affect the treatment effect and prognosis.  
OBJECTIVE: To explore the effect of diabetes mellitus on perioperative blood loss and postoperative pain in patients undergoing primary total knee arthroplasty.
METHODS: A retrospective study was conducted on 154 patients who underwent primary total knee arthroplasty and met the inclusion criteria in the Osteonecrosis and Joint Reconstruction Ward of Xi’an Honghui Hospital Affiliated to Xi’an Jiaotong University from January to April 2021. Patients were divided into a non-diabetic group and a diabetic group according to their diagnosis, with 32 cases in the diabetic group, 9 males and 23 females, aged 55 to 80 (66.58±7.16) years and 122 cases in the non-diabetic group, 34 males and 88 females, aged 44 to 83 (66.69±6.63) years. Perioperative blood loss (including total blood loss, hidden blood loss, the falling value of hemoglobin and hematocrit) was calculated for both groups. Visual analog scale scores, hospital for special surgery knee score, and Caprini scores were recorded preoperatively and postoperatively.  
RESULTS AND CONCLUSION: (1) Total blood loss was significantly lower in the non-diabetic group (729.93±233.83 mL) than that in the diabetic group (853.69±184.91 mL) (P < 0.05). Latent hidden blood loss was also significantly lower in the non-diabetic group (624.40±233.19 mL) than that in the diabetic group (749.08±179.49 mL) (P < 0.05). (2) In the non-diabetic group, the visual analog scale scores preoperatively and 1 month postoperatively were significantly lower than those in the diabetic group (P < 0.05). The differences in visual analog scale scores at 3 days and 3 months postoperatively between the non-diabetic group and the diabetic group were not statistically significant (P > 0.05). (3) The hospital for special surgery knee score at 1 month postoperatively was significantly higher in the non-diabetic group than that in the diabetic group (P < 0.05). There was no significant difference in hospital for special surgery knee score between the two groups at 3 months postoperatively (P > 0.05). (4) There was no statistically significant difference in preoperative and postoperative Caprini scores between the two groups (P > 0.05). (5) It is concluded that having diabetes increases total and occult hidden blood loss in primary total knee arthroplasty. In the short term after total knee arthroplasty, diabetes increases the patient’s pain and affects the recovery of joint function, but the negative effects fade with time.

Key words: total knee arthroplasty, diabetes mellitus, osteoarthritis, total blood loss, hidden blood loss, visual analog scale score, rehabilitation