We obtained ethical approval for this study (Registration No. 2020 − 0224) from the institutional review board in March 2020, and informed written consent was obtained from all participants. Our study was also listed on the clinicaltrials.gov (NCT04307290) at March 13th, 2020.
A total of 80 patients aged between 20 and 80 years, with American Society of Anesthesiologists (ASA) physical status class I–III, who were scheduled to undergo elective TKA were enrolled in this study.
Patients with history of rheumatoid arthritis, diabetes mellitus, hepatic or renal disease, allergy to the drug being studied, heart block greater than first degree, left ventricular ejection fraction < 55%, or chronic administration of anti-inflammatory drugs or opioids were excluded. Patients with tourniquet duration of < 60 minutes or > 150 minutes or those with conversion to general anesthesia during surgery were also excluded.
Randomization sequence was created using the Stata 9.0 (Stata Corp, College Station, TX, USA) statistical software and was stratified by center with a 1:1 allocation using random block sizes of four. Assignments were concealed in sealed envelopes. Participants were randomly assigned following simple randomization procedures (computerized random numbers) to the DEX group (n = 40), which received intravenous DEX (0.5 ㎍/kg bolus over 10 minutes, followed by 0.5 ㎍/kg/h infusion from 10 minutes before the start of surgery to the end of surgery); or to the CON group (n = 40), which received an equivalent volume of normal saline bolus over 10 minutes and infusion of placebo from 10 minutes before the start of surgery until the end of the surgery.
All patients, attending anesthesiologists responsible for patient care, and nurses were blinded to the anesthetic agent during the study period.
Anesthesia and perioperative care
Standard monitoring included electrocardiography, noninvasive arterial blood pressure monitoring, and pulse oximetry. All patients received spinal anesthesia administered by an attending anesthesiologist and 500 mL Ringer’s lactate solution was administered to ensure hydration before spinal anesthesia.
Spinal anesthesia was induced in the lateral decubitus position. These patients received an intrathecal injection of 15 mg 0.5% bupivacaine (in 5% glucose) at the level of L4-5 via a 25-gauge needle. Patients were then turned to the supine position and the level of sensory block was evaluated by pinprick.
Peak level of sensory block, sensory dermatome at tourniquet pain, time between tourniquet application and onset of pain, and tourniquet pain were determined by an independent researcher at 5, 10, 15, and 20 minutes after the spinal injection and then at 10-minute intervals until the complete resolution of the sensory block. The affected extremity in all patients was exsanguinated with an Esmarch bandage, and a tourniquet was applied at a pressure of 300 mmHg during the surgery.
When the patient felt a poorly localized and diffuse pain after tourniquet inflation, despite adequate sensory blockade during surgery, the pain was considered to be induced by the tourniquet. Intravenous fentanyl was administered at a dose of 100 ㎍ as supplemental analgesia if the patients experienced tourniquet pain at any time during the procedure. General anesthesia was induced if intravenous fentanyl was ineffective. The need for supplementary intravenous fentanyl and any conversion to general anesthesia were recorded. In case of intraoperative bleeding, volume replacement with Ringer’s lactate solution was performed, according to the decision of an attending anesthesiologist instead of blood products.
Systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial blood pressure (MAP), and heart rate (HR) were monitored every 5 minutes until the end of surgery. Hypotension was defined as a 30% reduction in basal MAP, which was treated with 5 mg intravenous ephedrine administration. When the HR was < 50 beats/min (bradycardia), 0.2 mg glycopyrrolate was injected intravenously.
Before the start of surgery, femoral nerve block was performed to reduce postoperative pain using a bolus of 20 mL 0.75% ropivacaine. Concurrently, patient-controlled analgesia (PCA) pump containing 1000 ㎍ fentanyl, 150 mg ketorolac, and 0.6 mg ramosetron in 150 mL saline was set to deliver a basal infusion of 2 mL/h and bolus doses of 1 mL, with a 15-minute lockout period for postoperative analgesia. PCA volume consumed at 24 and 48 hours after surgery and total PCA volume consumed in 48 hours after surgery were recorded. Postoperative pain intensity at rest was measured using a 100-mm linear visual analog scale (VAS). The VAS score for pain at rest was measured at 24 and 48 hours after surgery. Fentanyl in a dose of 100 ㎍ was administered for pain with VAS score ≥ 50 or 30 mg ketorolac was administered for pain with VAS score < 50 or on patient request.
Measurements of metabolic, coagulative, and temperature changes
Arterial blood samples were obtained from the radial artery 1 minute before the start of spinal anesthesia as a baseline value and at 10 minutes after tourniquet release.
Five milliliters of blood samples were extracted through the radial artery, and 1 mL was collected in a heparinized syringe; following this, the remaining blood samples were immediately sent through the pneumatic tube system, guaranteeing their processing in the least amount of time possible, not > 5 minutes by institutional standards. The metabolic variables included pH, lactate, calcium, potassium, bicarbonate, CO2, base excess and coagulative change included antithrombin III activity.
Tympanic temperature measurements were made using the Braun Thermoscan (Thermoscan Inc., San Diego, CA) inserted into the auditory canal at 1 minute just before the start of spinal anesthesia as a baseline and at 10 minutes after tourniquet release.
The sample size was calculated using the PASS 2008 (NCSS, LLC. Kaysville, Utah, USA) software. A preliminary investigation had shown that mean ± standard deviation (SD) of the two treatment groups for total PCA volume used in 48 hours after surgery as a primary outcome were 125.67 ± 8.2 and 120.90 ± 5.8, respectively. The investigation revealed that a sample size of 36 patients per group would enable the detection of a significant difference with a power of 80% and an α-coefficient of 0.05. The final sample size of this study was determined to be 40 patients per group, with a dropout rate of 10%. The SPSS version 18.0 (SPSS Inc., Chicago, IL, USA) was used for the statistical analysis. The data are presented as mean ± SD, median (interquartile range), or number (%) of patients. The groups were compared using an independent t-test or the Mann–Whiney U test according to normality test and categorical variables with the χ2 test or Fisher exact test as appropriate.
The group-by-time interaction assesses whether the change over time differs between groups. Repeated measure analysis of variance was used for intra-group comparisons with P values that were adjusted with the Bonferroni correction for multiple comparisons. A P value < 0.05 was considered statistically significant.