A Randomized Controlled Study on the Safety and Effectiveness of Total Knee Arthroplasty Operated At Different Times under ERASA.

Background Total knee replacement (TKA) is the most effective surgical method for end-stage knee joint disease. However, around (cid:0) 75%-92% (cid:0) of patients after TKA are not satised. In the earlier days, TKA done at different times had different outcomes. Till this date, Enhanced Recovery After TKA has upgraded to a higher level. However, its safety and effectiveness concerning time have been remained unclear. So this study, we design to compare the effectiveness and safety of TKA surgery operated at different times, under ERAS. Material and method In this study, 80 patients were randomly included for primary unilateral TKA who fullled the study criteria, between May 1st, 2019, and November 31th, 2019 in the Department of Orthopedics, West China Hospital of Sichuan University. All patients operated and peri-operatively managed according to standard ERAS protocol. The measuring indicators include VAS score, Range of knee motion, PSQI sleep quality score, HSS knee score, operation duration, total hospital stay, anesthesia time, rst urination time, hemoglobin level, hematocrit, blood loss, blood glucose, inammatory indicators (CRP and IL-6), radiological evaluation, gastrointestinal complications, appetite, total hospitalization costs, and satisfaction. Result Out of 80 patients, 40 were operated before 2 pm and 40 were operated after 2 pm. The VAS score for knee pain on day 1 after surgery in the group before 2 pm / after 2 pm, resting (P= 0.831); motion (P = 0.877). On discharge, resting VAS (P= 0.681); motion, (P = 0.912). At 21 st days VAS Score at rest (P= 0.679), At motion (P= 0.979). Before 2 pm/after 2pm group ,Blood loss (cid:0) P=0.328 (cid:0) ,ROM at the 21 st day after surgery (p = 0.476). PSQI at 21 st days after surgery (P = 0.188), HSS knee score on the 21 st day after surgery (P = 0.762), respectively. standard ERAS protocol, total knee replacements at different times do not its and effectiveness.


Introduction
Total knee replacement (TKA) is the most effective way to treat arthritis. Since the early 1970s, TKA began to develop in the United States and gradually spread to the world (1). Till this date, TKA has become a very common, low-cost, and e cient orthopedic surgery. TKA currently has up to 200,000 TKA operations per year in China (2). Considering the aging of the population and the extension of life expectancy, the number of TKA operations will increase dramatically in the future (3). Reasons are TKA improves the quality of life, relieves joint pain, and improves joint function. Moreover, the long-term effects of patients with end-stage knee joint disease are better (4). However, the satisfaction of patients after TKA is in the range of 75-92%. Low satisfaction directly affects the quality of life (5).
In the traditional rehabilitation model, surgery at different time points may affect the effectiveness of surgery. Muppavarapu et.al. Found that the outcome of TKA after Monday or Tuesday is better than Thursday or Friday (6). The results of the rst operated patient and the last may be different on the same surgery day. A study on orthopedic surgery found that the mortality rate of patients who performed surgery in the morning was lower, while the mortality rate of cases performed at night was the highest (7). In the afternoon, the work e ciency of surgeons or surgical staff also declined. In two independent studies, Kelz et al. found an increased incidence of complications during general emergency surgery and vascular surgery after work (8,9). The quality of care, such as fatigue of hospital staff, decreased quali cations in late shifts, and biological patient factors, such as longer fasting, lead to an increased risk of postoperative complications later in the day. Surgery at different time nodes, different duration of fasting, patients has different waiting times, worrying that the energy of the surgical team will decline, affecting the effectiveness and satisfaction of the surgery (9,10).
Danish surgeon Kehlet proposed the concept of Enhanced Recovery After Surgery (ERAS) in the year 1997 (11). According to the ERAS model, shortening the fasting time, proper patient counseling, strengthening preoperative nutrition, early diet pan after the operation, and exercising and mobilizing early. There is a lot of evidence that it is safe to fast for 6-8 hours for solid foods and 2-3 hours for carbohydrate beverages. Fasting too long before surgery will trigger metabolic reactions, which may increase the body's response to surgical wounds, resulting in increased insulin levels (12). The metabolic effects of prolonged fasting and surgery may impair the patient's postoperative recovery. Further, some surgical studies have shown that shortening preoperative fasting and it can reduce postoperative Length Of Stay LOS (13).
Objective of this study is to compare of outcomes among TKA candidates on the aspect of time, Under the ERAS model. The comparision parameters could be summarized as safety and effectivess. Thus, study comapares safety and effectiveness of similar patients with same treatment methodology but different operating time among TKA candidates under ERAS.

Methods
This is a single-centered study and conducted in the Department of Orthopedic Surgery, West China Patients were randomly divided in to two groups. Patients who underwent surgery before 2 pm were in Group A, and those who performed surgery after 2 pm were in Group B ,according to a computerised random sequence generator. The sequence was concealed until the interventions were assigned by a sealed envelope method in the operating room. The observers collecting the data after the surgeries were uninvolved in the experimental operations and were unaware of the intervention assignments.
All patients were randomly taken to the operation room and received general anesthesia. All operations were performed by an experienced senior surgeon with three assistants. An anterior midline skin incision was made by the medial para-patellar approach, using intramedullary guides for the femur and extramedullary guides for the tibia and prosthesis xed with bone cement. All patients neither used tourniquets nor postoperative drainage tubes. At the end of the operation, it was wrapped with an elastic bandage and the wound was covered with the ice pack. All patients were managed according to the same perioperative treatment plan. Patients are allowed to eat fat and solid food 8 hours before anesthesia, 6 hours can eat milk and semi-liquid diet, 4 hours can eat a semi-liquid diet without fat and milk, 2 hours can drink 5 ml / kg of fresh drinks, including water, sugar, electrolytes, coffee, and tea. Our department prepared nutrition supplement powder for all patients, each packet contains energy 193 kcal, CHO 48 g, Na 3 mg. 0 g protein, fat, dietary ber, should take with 200 ml warm water 2 hours before surgery. All patients got written consent, counseling and patients education, watched the operation introduction video before the operation, and post-operative patients shared the surgical experience with other patients, friends and families, and postoperative exercise and precautions. All patients were managed according to the same perioperative treatment plan. Both groups of patients received the same pre-operative consultation, the same anesthesia, the same surgical procedures, and pre-and post-operative care. After the operation, mechanical thrombosis is prevented by intermittent in ation of lower limb pumps. Postoperative functional exercises include lower limb strength exercise, ankle back extension exercises, and knee joint exion and extension exercises. Both groups of patients received the same thrombolytic drug, anti-infection, and analgesic regimens. The same senior professor supervised before, after, and during the follow-up.
Basic data were collected and compared the demographic characteristics of the two groups, the main indicators are the patient's height, weight, gender, sex, BMI, diagnosis, and ASA. BMI calculated based on height and weight. Pain evaluation is the main result of our research, assessed by a visual analog scale (VAS), where the lowest score is 0 and the highest score is 10. Motion and resting VAS were measured at the time of admission, the rst day after surgery, the day of discharge, and the follow-up. The resting VAS was measured when the knee exed at 45 degrees. Sleep quality was assessed by the Pittsburgh Sleep Quality Index (PSQI) at admission and the last follow-up. Knee function was evaluated by the Range Of Motion (ROM) and the HSS knee score, which were evaluated during admission, discharge, and follow-up.
Hematocrit (Hct), hemoglobin (Hb), and blood glucose at admission and discharge were recorded.

Candidates ow and baseline data
84 patients meet the inclusive and exclusive criteria among then 2 were unwilling to participate and 2 were lost during follow up, 80 patients remained untill the study over. 40 patients were operated morning and remaing 40 were operated at evening. patients ow chart is shown in table1. There are 40 patients on both groups and no statistical difference in the ages of groups A and B (p=0.421). The mean height of both groups are similar and no statistical signi cance between the groups (p=0.874). The mean weight of both groups also similar and no statistical difference between the groups (p=0.917). BMI between the two groups shows no signi cant difference between the groups (P = 0.460). Group A contains only 3 males, while group B contains 4 males and shows no signi cant difference (p=1.000). Similarly, 1 RA patient in group A, 2 RA patients in group B, and the rest were OA and was no statistical difference between the two groups (P= 1.000). There was 1 patient in group A of ASA grade 1 and 2 patients in group B. There were In ammation indicators (interleukin 6 and C-reactive protein) on admission, on the rst day after surgery and upon discharge were recorded.

TBL(mL) = PBV×(Hctpre − Hctpost)/Hctave
Among them, total blood loss (TBL), patient's blood volume (Patient's blood volume, PBV), initial hematocrit level before Hct pre operation, minimum hematocrit level after Hctpost operation, Hct before and after Hctave operation The average of the sum of Hct.k1 = 0.3669, Male k2 = 0.3219, k3 = 0.6041; Female k1 = 0.3561, k2 = 0.3308, k3 = 0.1833 Furthermore, Comparison was made between the group on total anesthesia time, total operation time, total hospital stay, urine output for the rst time after surgery, gastrointestinal adverse effect, appetite, and total cost. Gastrointestinal complaints mainly focus on postoperative diarrhea, constipation, dry mouth, nausea/vomiting. Appetite conditions classi ed by not good, normal, good, and very good. patient Satisfaction where satisfaction classi ed as very satis ed, satis ed, normal, and not satis ed. X-ray lms of two groups of patients were taken for imaging evaluation. The comparison made between prosthesis loosening, light transmission where light transmission line > 2 mm counted as positive for transmission.
The evaluation of the lower limb force line used the femoral tibial mechanical axis angle (MFT), which is the angle of the tibial mechanical axis.
Statistical analysis for quantitative data expressed as mean and standard deviation (SD); categorical variables are expressed as propertions. According to the distribution characteristics of the data, the t-test or Mann-Whitney U test was used to evaluate the difference of continuous variables between groups. A Chi-square test or Fisher's exact probability test is used to compare categorical Varibles. P < 0.05 was considered statistically signi cant and used SPSS 19.0 to complete the statistical analysis.
18 cases in ASA grade 2 group A and 15 cases in group B. There were 22 cases in ASA grade 3 group A and 23 cases in group B. There were no ASA grade 4 patients in both groups. American Society of Anesthesiologists (ASA) were collected preoperatively, with no statistical difference (p> 0.05).

Pre-operative data
As shown in Table 2: Two knee VAS scores were measured before surgery, at rest and motion. Compared with group A, the average knee pain VAS score during rest and motion is similar in group B, but there is no statistical difference between the two groups (P = 0.602), (P=0.689) respectively. Similarly, there is no statistical difference in preoperative ROM between the two groups (P = 0.646). In terms of the HSS score, there is no statistical difference between group A and group B (P = 0.505). In the sleep score, there is no statistical difference between the two groups (P = 0.385). There is no statistical difference in HCT content between the two groups (P = 0.456). In terms of hemoglobin level, there is no statistical difference between the two groups (P = 0.203). There were no signi cant differences in blood glucose levels between the groups (P=0.203), CRP, and IL-6 between groups A and B (P> 0.05). In summary, there was no statistical difference in baseline data between the two groups pre-operatively.
3. Post-operative data 3.1 pain score As shown in Table 3, resting VAS and motion VAS were measured three times after surgery and compared between groups. The resting VAS on the rst postoperative day: no signi cant difference between the groups (P = 0.831); Motion VAS: no signi cant difference between the groups (P = 0.877). Resting VAS at discharge: no signi cant difference between the groups (P = 0.681); Motion VAS: no signi cant difference between the groups (P = 0.912). Resting VAS at follow-up, the resting no signi cant difference between the groups (P = 0.679); Motion VAS: no signi cant difference between the groups (P = 0.979). From the above data, it can be seen that there is no statistical difference in the VAS score between the two groups during the perioperative period and the follow-up period.

Function Scale
HSS knee score: during follow-up no signi cant difference between the groups (P = 0.762). ROM: at discharge; during discharge no signi cant difference between the groups (P = 0.546) and at follow-up no signi cant difference between the groups (P = 0.476). There was no signi cant difference in the HSS knee score and ROM score between the two groups during the perioperative period and follow-up.

Sleep Quality
The PSQI comparison of sleep scores between the two groups is (P = 0.188). There was no statistical difference between the two groups in preoperative and postoperative sleep scores.

Blood Index
At discharge, HCT (%): no statistical difference between the groups (P = 0.987). At discharge, HGB no statistical difference between the groups (P = 0.545). At discharge blood sugar no statistical difference between the groups (P = 0.564). rst day after surgery and discharge CRP no statistical difference between the groups (P = 0.968) ,(P = 0.108) respectively. Interleukin -6 during the rst day after surgery and discharge, no statistical difference between the groups (P=0.153), (P=0.280) respectively.

TimeDifferences, Total Blood Loss and Radiological Assessment
As shown in Table 4, total hospital stay (days) (P = 0.988). The rst postoperative urine output (hours) (P = 0.890), Operation time (minutes); (P = 0.882). Anesthesia time (minutes): (P = 0.772), Blood loss (ml) ; (P = 0.328) where data shows no signi cant different the groups. The prostheses of both groups were in place, and there was no loosening, no light transmission line> 2mm. The evaluation of the lower limb force line used the femoral tibial mechanical axis angle (MFT), which is the angle of the tibial mechanical axis. There were 35 cases (87.5%) within 3 °of MFT in group A and 36 cases (90%) in group B. There was no statistical difference between the two groups.

Appetite, gastrointestinal complications, Total Hospital Cost and Satisfaction
As shown in Table 5, A and B groups have 8 cases and 7 cases of "Bad", 15 cases and 16 cases of "Normal ", 16 cases and 15 cases of "Good", "Very Good" for 1 case, 2 cases respectively. There was no statistical difference between the two groups (P = 0.837). Gastrointestinal complications: Constipation: 12 cases (30%) in group A, 15 cases (37.5%) in group B (P = 0.478); 2 cases (5%) in group A with dry mouth, and 3 cases (7.5%) in group B (P = 1.000); diarrhea group 3 cases (7.5%), group B 4 cases (10%) (P = 1.000), nausea / vomiting group 4 cases (10%), group B 4 cases (10%) (P = 1.000), there is no statistical difference between the two groups of gastrointestinal complications. The total cost is calculated in RMB, including the cost of the prosthesis, there was no statistical difference between the total cost of the two groups (P = 0.808). Comparing the satisfaction rate, 27 cases (67.5%) were "very satis ed" in group A, 24 cases (60%) in group B; 13 cases (32.5%) in "satis ed" group A and 16 cases (40%) in group B; No "general" and "unsatisfactory". There was no statistical difference between the two groups (P = 0.489).

Discussion
In the earlier days, surgery at different time points may affect the effectiveness of surgery. Muppavarapu et. al found that TKA performed on Monday or Tuesday was better than Thursday or Friday(6). Halvachizaden et. al. found that night surgery is less effective than day surgery and morning is better than an afternoon (16). But Montaigne et al found that in cardiac surgery, afternoon surgery is more effective (7).In two independent studies, Kelz et al. Found (7)(8)(9). Early day's surgical model, fasting is usually required overnight before surgery. Fasting too long before surgery will trigger metabolic reactions, which may increase the body's response to surgery, resulting in increased insulin resistance and the surgical wound may impair the patient's postoperative recovery (12,13).
The ERAS model focus on reducing surgical wound through the use of minimally invasive surgical procedures, thereby reducing postoperative complications, saving costs, shortening the length of hospital stay (LOS), improving patient satisfaction and promoting faster recovery (17). The nutrition mode under the ERAS model requires shortening the fasting time, strengthening preoperative nutrition, early postoperative food supplements, exercising, and mobilizing early (18). The Society of Anesthesiology recommends fasting for 2-3 hours before surgery (19). Some studies have shown that shortening preoperative fasting can reduce postoperative LOS (13). Preoperative education helps to increase patient con dence, improve patient Preoperative education has a modest bene cial effect on perioperative anxiety (9).
The main goal of postoperative analgesic treatment is to reduce pain, reduce the need for opioids and reduce the adverse reactions associated with opioids, and accelerate patient recovery. This directly affects patient satisfaction, postoperative care level, nancial burden, and anxiety (20,21). Changes in the degree of pain can be assessed by VAS. Karlsen et al. conducted a meta-analysis of pain management after TKA and found that there is no best strategy for pain management after TKA, in the literature (22). In our study, the improvement of patients' postoperative pain in this study is similar to that of Wang et. al. (23). As can be seen from the results of this study, the pain during the follow-up period has been greatly reduced compared with before the operation. Patients may suffer from sleep disorders due to pain, socioeconomic burden, and even mental disorders before receiving TKA, and sleep disorders are also prone to occur after TKA (24). Research conducted by Brien suggests that lack of sleep (4 hours or less) may result in decreased daily performance and persistent inattention. Also, in healthy adults, poor sleep quality directly affects the quality of life and mental state, which is even worse in perioperative patients who need more sleep (25,26). But there seems to be a vicious circle: pain-poor sleep quality-increased pain-poor sleep quality. Studies have shown that there is a statistically signi cant correlation between pain intensity and sleep quality. Besides, nighttime pain is signi cantly associated with joint swelling, elevated ESR, and CRP, all of which are consistent with the characteristics of local and systemic reactions in TKA surgery (24)(25)(26). Our result shows there is no signi cant difference between the groups on the above indicators and found improved after surgery.
HSS and ROM represent the patient's knee function. Miao et al found that 6 weeks after TKA, the joint function of patients has exceeded that of pre-operation (27). Canovas and others found that TKA can greatly improve the quality of life scores. They found that pain and / or stiffness during exercise is associated with the patient's quality of life and patient satisfaction (5). In addition, Zhou et al found at 6 weeks after surgery, the patient showed signi cant improvement in walking and bending activities (28). Our study found that 3 weeks after surgery, ROM and HSS scores were signi cantly. improved compared with before surgery, and the results of the two groups were not statistically different. Compared with patients guided by surgeons and anesthesiologists for perioperative care, patients with TKA through ERAS have signi cantly lower LOS (29,30). Urinary retention after TKA has greatly hindered early discharge(31). Stowers et. al. found no signi cant difference in LOS between ERAS and non-ERAS(32).LOS is not only affected by a single independent factor, it is the result of a combination of various factors. In this study, LOS, time of urination after surgery is shorter and there was no statistical difference between the two groups.
Various factors will affect TKA post-operative outcomes, Shortening of anesthesia time is directly related to early postoperative rehabilitation (32). So n et.al. Found that the prolongation of general anesthesia time was signi cantly associated with postoperative nausea and vomiting (33). In our study, the anesthesia time and the operation time of the two groups of patients were the same, found to be no statistical difference in the anesthetic time and proportion of postoperative gastrointestinal complications between the two groups. It is estimated that by 2030, the global prevalence of Diabetes Mellitus(DM)disease will increase from 6.4% in 2010 to 7.7% (34). According to reports, compared with patients without DM, TKA patients with DM have more perioperative complications (35). Also, DM can impair the healing of wounds by delaying collagen synthesis and adversely affect the musculoskeletal system (34). Therefore, the blood glucose level before and after controlling TKA is very important (34,36). Due to the use of advanced technology to minimize the risk of DM-related adverse effects, the success rate of TKA continues to increase (37). In our study, there was no statistical difference in blood glucose between the two groups, suggesting that surgery at different surgical time has no signi cant effect on blood glucose changes.
After a major surgery, patients often suffer from loss of appetite, but no literature has speci cally studied the relationship between TKA patients, loss of appetite, and its relation to the surgery start time. Prodger et al found that after TKA, patients usually suffer from loss of appetite and disappear within 6 weeks (38). Besides, patients with gastrointestinal disease (GI) are at risk of osteopenia or osteoporosis, such as post-gastrectomy status, in ammatory bowel disease (IBD), and abdominal disease, which may cause fractures around the prosthesis and require revision (39). Because most TKA patients are elderly osteoporosis patients, the incidence of fractures around the prosthesis is higher (39). Other causes of gastrointestinal symptoms include C. di cile and pseudomembranous colitis (40,41). It is important to observe gastrointestinal symptoms closely, especially intestinal dysfunction caused by analgesic. This study explored the relationship between different surgical start times and postoperative gastrointestinal reactions and appetite. Postoperative gastrointestinal reactions were not statistically different between the two groups, and appetite was not statistically different. It is suggested that under the ERAS, surgery at different time points does not affect appetite and does not increase the risk of gastrointestinal complications.
The total cost of hospitalization and patient satisfaction are other evaluation indicators of TKA's success. Despite the continuous improvement, TKA satisfaction should be higher. It has been reported in the literature that about 20% of TKA patients are dissatis ed after the operation and their functional score is not high (20,21,42,43). However, this study shows that postoperative satisfaction is higher, which may be due to the different psychosocial behavior of patients in different places. The higher satisfaction rate may be due to the psychosocial behavior of the patients. There was no signi cant difference in hospitalization costs between the two groups,in this study.
In this study, the VAS pain score, joint function score, joint mobility, blood in ammatory indicators, blood loss, hemoglobin level, blood sugar, gastrointestinal complications, PSQI sleep score, operation duration, anesthesia duration, rst postoperative urine output, Patient satisfaction, radiological evaluation, total hospitalization costs, and differences between the two groups were not statistically signi cant. Every parameter of the two groups of patients was better than those before the operation, suggesting that TKA signi cantly improved the patient's condition which is safe and effective. In summary, under the ERAS model, TKA patients' personalized dietary management and full communication and pre-operation before surgery eliminate concerns, and different time nodes have no difference in the safety and effect of TKA surgery.
This study also has some limitations. The study was conducted at West China Hospital of Sichuan University and is a single-center prospective study. The research study sample size is small, and the preoperative diagnosis is limited to OA and RA patients, the follow-up time is short, and the methods used to measure various outcome indicators are relatively limited. In the future, a multi-center large sample long follow-up study may be needed to further depth of the study.

Conclusion
This prospective study aimed to compare the difference between the results of TKA before and after 2 pm under ERAS. We found that there was signi cantly safe and effective of the procedure until three weeks follow up. Furthermore, there was no statistically signi cant difference between the two groups on postoperative, discharge, and 3 weeks of postoperative follow-up. In the ERAS, TKA patient-oriented diet management, patient education, and counseling before surgery success made no difference in the impact of different time on the aspect of safety and effectiveness.

Declarations
Ethics approval and consent to participate This study was approved by the Ethics Committee and Institutional Review Board of West China Hospital, Sichuan University (Registration: ChiCTR1900022256). This is observational study, only verbal consent was obtained from all individual participants included in the study.  Data shown as mean ± standard deviation include visual analog score VAS, activity ROM, HSS knee score, and Pittsburgh sleep quality score PSQI. Hematocrit HCT is measured in%, hemoglobin HGB is measured in g / L, blood glucose mmol / L is measured in L, C-reactive protein CRP, unit mg / L, interleukin-6 IL-6 pg / dL. The data are shown as mean ± standard deviation: visual analog score-VAS, rst day-D1, discharge-Discharge, follow-up-F / U, Pittsburgh sleep quality score PSQI. Hematocrit-HCT (%). Hemoglobin-HGB (g / L), Glucose Mmol/L, C-reactive protein CRP (mg / L), interleukin-6 IL-6 (pg / dL), ROM (°), 1.000 The data are shown as mean ± standard deviation LOS hospitalization days (in days) UFT rst micturition time (in hours), operation time and anesthesia time in minutes, blood loss in ml, MFT femoral tibial mechanical axis angle. 1.000 The total cost data are expressed as mean ± standard deviation, percentage and the satisfaction is divided into four categories. Gastrointestinal complication; Constipation, Dry mouth, Diarrhea, Nausea/Vomiting where appetite includes Not good, Normal, Good, Very Good.