We conducted a single-centre, randomized-controlled trial conducted in the Joint Surgery Department ward of the First Affiliated Hospital of Bengbu Medical College. The trial was approved by the ethical committees of the First Affiliated Hospital of Bengbu Medical College. Written informed consent was obtained from all participating patients. A simple random number table generated a random number sequence. Participants were randomly allocated to the Trial or Control Group. This procedure was carried out by a research nurse who did not take part in the outcome assessment.
By communicating well with the patients diagnosed with primary osteoarthritis, we enrolled 80 patients underwent unilateral TKA from September 2016 to September 2018. Because the outcomes were short-term estimations and very few complications occurred in the experiment, patient compliance was good. We prespecified the inclusion criteria as patients with an age younger than 80 years, weight below 80 kg and unilateral knee. The exclusion criteria were as follows: refusal to participate, unable to cooperate, revision TKA, allergy or intolerance to TXA and ropivacaine, major psychological problems, renal insufficiency, hepatic insufficiency, history of previous thromboembolism, history of stroke, and bilateral TKA.
The patients were randomly assigned to two groups, by a computerized randomization list: Trial Group, percutaneous intra-articular injections of ropivacaine, and TXA and simultaneous removal of the drainage tube (0-24 hours after surgery) one day after TKA; Control Group, no intra-articular injections one day after TKA, and the drainage tube was pulled out when the amount of drainage fluid was below 50 mL/24 hours (2-4 days after surgery).
Perioperative identical medications in both groups
Preoperative antibiotic prophylaxis for infection was administered intravenously in both groups with first generation cephalosporin antibiotics half an hour before surgery. Intraoperatively we routinely applied periarticular multisite infiltration for perioperative pain management. A total injection volume of 70 mL of the drug solution included 70 mg ropivacaine (Jia Bo Pharmaceutical Co. Ltd, Qing Yuan city, Guang Dong Province, China) as a local anesthetic, and 0.1 mg epinephrine (Sui Cheng Pharmaceutical Co. Ltd, Xin Zheng city, He Nan Province, China). The drug-containing solution was injected into both side of the posterior joint capsule, quadriceps tendon, medial and lateral capsule, and subcutaneous tissue around the incision just prior to implantation prosthesis. After suturing the deep fascia, a 15 mL solution containing 2 g TXA (Wu Zhou Pharmaceutical Co. Ltd, Wu Zhou city, Guang Xi Province, China) [0.5 mg] was injected into the joint cavity before subcutaneous closure. The drainage tube was opened after being clamped for two hours. Except for additional intra-articular injection of TXA and ropivacaine in trial group one day after TKA, the medications administered were the same in both groups. Starting the day after TKA, 40 mg of the non-steroid anti-inflammatory drug parecoxib sodium was given intravenously two times per day. If the patient's pain remained intolerable, we administered 100 mg tramadol (Meng Di Pharmaceutical Co. Ltd, Tong Zhou District, Bei Jing city, China) as rescue therapy. After surgery, every patient received standard thromboembolism prophylaxis based on the guidelines. The patients were given physiotherapy by means of passive and active physiotherapy, and lower-extremity strength training, which were overseen by the professional physiotherapist after surgery in a progressive manner. A continuous passive motion (CPM) machine was applied to improve the range of motion as tolerated, twice a day. For chemical prophylaxis, subcutaneous injection of low-molecular-weight heparin (Hong Ri pharmaceutical Co. Ltd, Tian Jing city, China, 3200 IU, qd) was administered subcutaneously to prevent deep vein thrombosis (DVT) during hospitalization.
At our institution，all surgical procedures were performed by the corresponding author, who is an experienced arthroplasty surgeon. A pneumatic tourniquet was used around the upper thigh throughout the operation. We designed a standard medial prepatellar skin incision. During the surgical procedure, we expanded the skin incision as needed to fully expose the knee joint. Osteotomy of the femur and tibia were performed with the help of osteotomy instruments. The femoral medullary cavity canal was plugged with a bone block. All patients received posterior stabilized knee prostheses that were fixed with cement and the patella was resurfaced. At the end of the surgical procedure, an intra-articular drainage tube was applied in all patients in our joint surgery ward. Pressure dressing was applied from the ankle to the thigh. An ice pack was placed around the knee joint until 24 hours after surgery.
The additional percutaneous intra-articular multi-drugs injection was administered one day after TKA in the Trial Group, and no injection was provided for the Control Group. An additional intra-articular injection of multi-drug was administered at 08:30 one day after TKA, without considering the end time of TKA. In the Trial Group, Multi-drugs injections included ropivacaine 100 mg (10 mL), TXA 1 g and saline10 mL. We recommended that the patient actively exercise the knee joint to promote blood flow before drainage tube removal in the Trial Group; then we remove the drainage tube. The injection was performed above the patella on either side of the quadriceps tendon using a 20 mL injector during knee extension (Fig. 1). Povidone iodine and ethanol were applied to disinfect the skin, and a total of 20 mL drug solution was injected intra-articularly into the knee. We need to make sure that all of the drug solution goes into the joint cavity. In control group, the patients did not receive additional intra-articular injection. Drainage tube removed according to the amount of drainage fluid below 50 mL/24 hours. Removal of the drainage tube may take 2 to 4 days after surgery.
A research assistant who was not involved in the research collected data. The research assistants were blinded to whether an intra-articular ropivacaine, and TXA injection was used one day after surgery.
The primary outcome variables were length of hospital stay after surgery, and an assessment of the VAS score, which scaled from 0 mm (representing no pain) to 100 mm (representing extreme pain), was performed at rest and during movement. Postoperative pain was assessed via the VAS 1, 2, 3, 4 and 5 days after surgery (7:00, 12:00, and 20:00). The most intense pain that the patients underwent throughout a physical recovery treatment was identified as the VAS score of the activity.
The secondary outcomes of this study were the total analgesic (non-steroidal anti-inflammatory drug) consumption, ROM of the knee joint, TBLV, blood transfusion rate, amount of blood lost through the drainage tube, the circumference of the knee joint (10 cm above the superior border of the patella and patella centre), wound related complications such as infection, and symptomatic DVT within 30 days of surgery.
We calculated the Total blood loss in accordance with the following formula (16). Total blood loss = PBV ´ ( Hctpre - Hctpost ) / Hctave, where PBV, Hctpre, Hctpost and Hctave stand for Patient’ s Blood Volume, Initial Preoperative Haematocrit, Haematocrit level on the morning of the fourth postoperative day & Average of Hctpre and Hctpost respectively. PBV was estimated using Nadler’s formula (17). PBV= [k1´height (m3)] + [ k2 ´ body weight (kg)] + k3. For males, k1 =0.3669, k2 =0.03219, and k3 =0.1833, and for females, it was k1 =0.6041, k2 =0.03308, and k3 =0.1833. Because a tourniquet was used throughout the operation, the blood loss was negligible during operation.
Sample size calculation
Forty participants per group would be required for this trial to illustrate a statistically significant difference with a 10 point reduction in the VAS pain scale score, a 200 mL decrease in blood loss, a 5% significance level, 80 % power and two-sided analysis. (18)
GraphPad Prism 8 software was used for statistical analysis. Missing values were replaced by the average values of the same group at the same time point. We performed all data analyses according to the principle of intention-to-treat. Student's t-test was used to compare continuous variables. The chi-square test was used to compare categorical variables. Continuous variables are represented as the mean ± SD. Statistical significance was defined as p<0.05.