Total knee arthroplasty (TKA) is the most common treatment for end-stage knee osteoarthritis [1, 2]. While the number of patients undergoing TKA is increasing annually with the aging global population, the management of postoperative pain has attracted much recent attention. It has been found that postoperative pain can directly affect early functional exercise and reduce the effectiveness of surgical treatment . In recent years, continuous femoral nerve block (CFNB) has become the gold standard for analgesia after TKA . However, the lack of knowledge by patients and their families about CFNB and patient-controlled analgesia (PCA) often leads to under-dosing or over-dosing of medications, a decrease in general analgesic efficacy, an increased workload for ward nurses, and an increased risk of falls . While the importance of preoperative patient education has been emphasized, its role in improving postoperative analgesia and outcomes remains unclear. In this study, we explored whether preoperative CFNB education could improve postoperative analgesic efficacy; also, we evaluated whether preoperative CFNB education could reduce the PCA-related workload for ward nurses after TKA.
Accordingly, a prospective randomized clinical study was undertaken with the primary aim of evaluating the VAS on motion on POD1, and secondarily, to investigate the PCA-related workload for ward nurses postoperatively.
Subjects and methods
This study was approved by the Ethics Committee of Peking University Third Hospital and recorded on October 18, 2018 to chictr.org.cn (ChiCTR1800018957).
In total, 60 patients undergoing unilateral TKA in our hospital at 08:00–12:00 between November 2016 and March 2018 were enrolled and randomly divided into 2 groups (group A and group B; n = 30 each) using a random number table. The inclusion criteria included: a) being American Society of Anesthesiologists (ASA) grade I - II; b) undergoing unilateral TKA (left or right); c) receiving combined spinal-epidural anesthesia. The exclusion criteria were: a) inability to complete VAS scoring; b) lower-extremity muscle strength below grade V; c) intraoperative intraarticular injection of analgesic drugs; d) history of previous TKA; e) contraindications to spinal manipulation or patient insisting on general anesthesia. Before returning to the ward after surgery, patients in group A and their families received unified face-to-face PCA pump operation training by a nurse anesthetist. In contrast, patients in group B and their families received a nurse-led preoperative visit the day before surgery focusing on a PCA educational pamphlet (see the Additional file 1) about postoperative pain management. After surgery, they watched the PCA demonstration with group A patients/families before returning to the ward.
CFNB was performed by the same anesthesiologist in the post anesthesia care unit (PACU) or operating room before or after TKA. Under ultrasound guidance combined with nerve stimulation, the femoral nerve catheter was placed, and a single dose of 0.5% ropivacaine (20 ml) was administered. All of the patients received combined spinal-epidural anesthesia by the same anesthesiologist, along with subarachnoid administration of 2–3 ml of 0.5% bupivacaine. The epidural catheter was routinely placed 4 cm ahead. During the operation, 1% lidocaine was added to maintain the anesthesia via the epidural catheter according to the operative time (3 ml 1% lidocaine for test doses and 5 ml added each time according to the effect to maintain the block level at about T10). The same TKA procedure was performed, during which the lower-extremity tourniquet was routinely applied. After the operation, the femoral nerve tube was connected to the PCA pump (500 ml of 0.2% ropivacaine, with background infusion rate 5 ml·h− 1, lockout interval 15 min, and the patient controlled bolus 2 ml). In addition to the continuous femoral nerve PCA, all of the patients were intravenously administered 40 mg of parecoxib sodium twice daily for two consecutive days.
After the patients returned to the ward, the ward surgeon routinely recorded the knee flexion angle, and the ward nurses recorded the number of PCA-related nurse calls. An appointed nurse anesthetist followed up the 2 groups for postoperative analgesia on the first and second days after the operation and carried out bedside education according to the specific PCA conditions. Meanwhile, the following data were collected: a) age, height, weight, ASA grade, timing and location of CFNB, time of operation, dosage of epidural drug, and level of postoperative anesthesia; b) VAS scores at rest and during movement on the first and second days after the operation; c) a ten-question questionnaire form (see the Additional file 1) on the first and second days after the operation, used to evaluate the patients’ knowledge about CFNB and PCA - the results of which both the nurse anesthetist and patients were blinded to before the end of the two surveys; and d) the satisfaction (satisfied, not satisfied, and indifferent) of the ward nurses to the PCA mode of CFNB.
Power analysis was based on results of preliminary data comparing VAS score on motion on POD 1 between groups. In our preliminary study, 8 patients were recruited in each group. The medians and quartiles of VAS scores during movement were 4.5 ± 1 and 3 ± 1.75 in group A and B, respectively. Sample-size calculations showed that a Wilcoxon and Mann–Whitney U test with a type I error (two-sided) of α = 0.05 would have 95% power to detect the aforementioned difference in VAS scores on motion between the two groups if the sample size in each group was 24. To account for possible loss in the follow-up period, we enrolled 30 cases per group. All statistical analysis was performed with SPSS for Windows (version 14.0, SPSS, Chicago, IL). Data were tested for normal distribution using the Kolmogorov–Smirnov test. Continuous variables were presented using means and standard deviations (SD) or median and interquartile values, and analyzed using Student’s t test or the Wilcoxon-Mann-Whitney test, respectively. In the case of categorical variables, frequencies were used, and the count data were compared with a χ2 test. A p value of < 0.05 was regarded as statistically significant.