Article Text

Download PDFPDF
Analgesic efficacy of infiltration between the popliteal artery and capsule of the knee (iPACK) block added to local infiltration analgesia and continuous adductor canal block after total knee arthroplasty: a randomized clinical trial
  1. Chutikant Vichainarong1,
  2. Wirinaree Kampitak1,
  3. Aree Tanavalee2,
  4. Srihatach Ngarmukos2 and
  5. Nattaporn Songborassamee1
  1. 1 Anesthesiology, King Chulalongkorn Memorial Hospital, The Thai Red Cross Society and Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
  2. 2 Orthopedics, King Chulalongkorn Memorial Hospital and Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
  1. Correspondence to Dr Wirinaree Kampitak, Anesthesiology, King Chulalongkorn Memorial Hospital, Bangkok 10330, Thailand; nutong127{at}


Background A combination of motor-sparing analgesia with local infiltration analgesia (LIA) and continuous adductor canal block (CACB) may improve postoperative pain and functional recovery for total knee arthroplasty (TKA). We hypothesized that the addition of a novel technique for posterior knee block, known as the infiltration between the popliteal artery and capsule of the knee (iPACK) block, to LIA with CACB would reduce opioid requirements.

Methods In this double-blinded randomized controlled trial, 72 patients were assigned to receive either LIA with CACB (LIA+CACB group) or iPACK block with LIA and CACB (iPACK+LIA+CACB group). The primary outcome was cumulative postoperative intravenous morphine consumption within 24 hours. The secondary outcomes included numerical rating scale pain scores, incidence of posterior knee pain, performance test results, patient satisfaction, length of stay, and adverse events.

Results Morphine consumption within 24 hours postoperatively showed no significant intergroup difference (LIA+CACB; 1.31±1.85 mg vs iPACK+LIA+CACB; 0.61±1.25 mg, p=0.08). There were no clinically significant differences in the overall pain scores between the groups. The lower Timed Up and Go test scores on postoperative days 1 and 2, along with a shorter duration of hospitalization, were found in the iPACK+LIA+CACB group (p<0.05).

Conclusion The addition of an iPACK block to the LIA and CACB does not reduce the postoperative opioid consumption nor improve analgesia. However, it may improve immediate functional performance and reduce the length of hospitalization after TKA.

Trial registration number TCTR20180702001.

  • continuous peripheral techniques
  • lower extremity
  • acute pain
  • pain measurement
  • ultrasound in pain medicine

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Immediate, moderate to severe postoperative pain is the most commonly reported adverse effect of a total knee arthroplasty (TKA),1 and inadequate analgesia and the side effects of opioids can delay rehabilitation2 and cause patient dissatisfaction.3 However, rapid return to mobility is essential to reduce significant complications, hasten recovery, and achieve discharge criteria.4 Thus, the use of combinations of different kinds of analgesic drugs, which is referred to as multimodal analgesia, has received interest due to the resultant minimization of opioid usage and improvement in postoperative pain, which can promote early rehabilitation and shorten the length of hospitalization.4 5 Recently, the opioid-sparing and motor-sparing effects of ultrasound-guided adductor canal block (ACB) and intraoperative local infiltration analgesia (LIA) have made them popular as a part of postoperative multimodal analgesic protocols in TKA.6 7

Ultrasound-guided ACB, which involves the injection of local anesthesia into the fascial compartment covered by the sartorius muscle from the apex of the femoral triangle to the adductor hiatus, is known to cover the anteromedial aspect of the knee joint.8 The local anesthetic distribution achieved by injecting at the level of the apex of the femoral triangle is called the proximal ACB8 or the femoral triangle block,9 and mostly diffuses among the nerves that cover the entire anteromedial aspect of the knee joint including the saphenous nerve, the medial vastus nerve, the infrapatellar branch of the saphenous nerve, and possibly the medial femoral cutaneous nerve.10 Single injections and continuous technique have been shown to provide excellent pain relief with quadriceps sparing in patients undergoing TKA.6 11 However, these techniques do not relieve posterior knee pain since the posterior knee capsule is innervated by the popliteal plexus, which is formed by contributions from the tibial and posterior obturator nerve.12 LIA involves intraoperative administration of diluted local anesthetics with adjuvants within multiple layers surrounding the knee joints—such as the medial and lateral meniscus remnants, posterior capsule, and subcutaneous tissue using a blind technique by the surgeon.13 The use of a careful injection technique to avoid possible neurovascular complications of blind LIAs, particularly common peroneal nerve injury, may lead to inconsistent analgesia.7 In comparison with ACB or LIA alone, a combination of the two techniques has demonstrated better pain relief and may improve functional recovery in the immediate postoperative phase.14 15

The infiltration between the popliteal artery and capsule of the knee (iPACK) block is a promising novel ultrasound-guided motor-sparing posterior knee block for controlling posterior knee pain after TKA.16 The addition of the iPACK block to ACB was shown to improve analgesia and physical performance in the immediate postoperative period without compromising motor strength in several prospective studies.17–19

The present study aimed to investigate the analgesic efficacy of adding an iPACK block to LIA with continuous ACB (CACB) for patients undergoing TKA. We hypothesized that the opioid requirement over the first postoperative day (POD) (primary outcome) and pain scores (secondary outcomes) would be reduced by the addition of the iPACK block to LIA with CACB.

Materials and methods

Between July 2018 and May 2019, adult patients with American Society of Anesthesiologists classification status I–III scheduled for elective primary TKA using standard spinal anesthesia were actively recruited for participation in this trial. Eligible patients were interviewed and provided with a printed information pamphlet outlining the purpose of the study on the day before surgery. Exclusion criteria included: an age of less than 18 or more than 80 years; body mass index greater than 40 kg/m2; inability to provide informed consent; cognitive or psychiatric history that may interfere with assessment; a varus-valgus knee deformity >20°; knee flexion deformity >30°; contraindication for spinal anesthesia or peripheral nerve block; allergy or intolerance to local anesthetic drugs or any component of the multimodal analgesic regimen; pre-existing chronic pain or opioid drug use (daily or almost daily use of opioid drugs for ≥3 months or morphine use ≥60 mg/day for ≥1 month); pre-existing neuropathy or neurological deficit in the lower extremities.

After obtaining written informed consent, patients were divided into two groups—sham iPACK block with LIA and CACB (LIA+CACB; control group) and iPACK block with LIA and CACB (iPACK+LIA+CACB; experimental group)—based on a computer-generated 1:1 ratio randomization schedule with block sizes of 4 or 6 by a statistician not otherwise involved in the study. Group assignment was sealed in an opaque envelope that was opened by an anesthesiologist nurse in a block room after enrollment. All surgeons, research assistants, operating room and floor nurses, patients, and statisticians were blinded to group allocation. Only the anesthesiologist performing blocks and the anesthesiologist nurse were not blinded and had no further role in the study. Outcome assessors and clinical personnel were also blinded to treatment group randomization.

Multimodal pathway


All patients received 650 mg of acetaminophen and 400 mg of celecoxib orally 30 min before surgery. After patients arrived at the block room and standard monitoring, including pulse oximetry, non-invasive blood pressure measurement, and electrocardiography, was applied, intravenous access was established, and midazolam 1–2 mg intravenous was given as needed for anxiolysis.

The iPACK block was performed in the prone position, with the operative knee slightly flexed using a pillow. Under ultrasound scanning with a 13–6 MHz linear probe (X-Porte; SonoSite, Bothell, WA), the ultrasound probe was placed in the popliteal crease to identify the popliteal artery and femoral condyle.20 The tibial nerve was visualized superficial to the popliteal artery, and the common peroneal nerve was observed to be completely separated from the tibial nerve. After administration of 1–2 mL of 1% lidocaine to anesthetize the skin, a 22 G 10 cm stimulating needle (Stimuplex A100; B Braun, Germany) was inserted using an in-plane approach from the lateral to medial direction between the popliteal artery and the femoral condyles until the needle tip was anterior to the medial edge of the popliteal artery. In the experimental group, 5 mL of 0.25% levobupivacaine with 1:200 000 epinephrine was injected in aliquots to ensure proper spread to the medial aspect of the condyles then the needle was slowly withdrawn, while simultaneously another 15 mL of local anesthetic was sterilely injected until completion at the lateral aspect of the condyle. In the control group, only 5 mL of normal saline was injected.


Spinal anesthesia was performed in the lateral position at the third to fourth or fourth to fifth lumbar levels using 15 mg of 0.5% hyperbaric bupivacaine. The patients received 10 mg of dexamethasone and 4 mg of ondansetron intravenous for postoperative nausea and vomiting prophylaxis. Intraoperative fluid administration and sedation by propofol intravenous were performed at the discretion of the anesthesiologist who was blind to the study.

A minimally invasive mini-midvastus approach with tourniquet was performed. The LIA mixture (composed of levobupivacaine 100 mg, ketorolac 30 mg, epinephrine 0.3 mg diluted with isotonic sodium chloride solution to a total volume of 80 mL) was injected around the medial collateral ligament, anterior capsule, fat pad, medial and lateral meniscus remnants, and the posterior capsule with care taken to avoid inadvertent peroneal nerve injection.


After postanesthesia care unit (PACU) arrival, ultrasound-guided CACB was performed by the same anesthesiologist and the linear transducer was moved slowly either cephalad or caudad from mid-thigh to identify the superficial femoral artery inferior to the sartorius muscle in the area of the apex of the femoral triangle.8 9 Using a sterile, in-plane technique, an 80 mm Tuohy needle with a Perifix 18 G epidural catheter (B Braun Medical, Bethlehem, PA) was inserted in a lateral to medial direction, with the needle tip located between the femoral artery and sartorius muscle. The needle tip was placed superior to the femoral artery, and normal saline solution was then injected to confirm proper positioning and assess catheter patency. In both groups, an epidural catheter was inserted and secured to the skin after confirming, via hydrodissection with sterile saline, that the tip of the catheter was resting anteromedial to the artery. Then 20 mL of 0.25% levobupivacaine was administered into the canal under negative blood aspiration. Levobupivacaine 0.15% was continuously dripped at 5 mL/hour via a disposable infusion pump (COOPDECH Syrinjector, Daiken Medical, Japan) for 60 hours postoperatively. During the PACU stay, in cases with numerical rating scale (NRS) pain scores ≥4, 2 mg of intravenous morphine was administered every 30 min. If a patient continued to exhibit NRS ≥4 up to 1 hour postoperatively, patient-controlled anesthesia (PCA) was ordered using morphine (no basal rate; PCA dose, 2 mg; lockout, 10 min) as a rescue drug.

A postoperative multimodal regimen was prescribed: two consecutive doses of 15 mg ketorolac intravenous, 650 mg oral acetaminophen every 6 hours, and 75 mg oral pregabalin (Lyrica) daily. After the last dose of ketorolac intravenous, 400 mg oral celecoxib (Celebrex) daily and half a tablet of tramadol hydrochloride/acetaminophen (Ultracet) were administered every 8 hours. If patients presented with persisting pain and NRS ≥4, the patient would receive 2 mg of intravenous morphine as rescue therapy. Other medications included 40 mg intravenous esomeprazole daily for preventing upper gastrointestinal bleeding and 4 mg intravenous ondansetron every 6 hours to prevent nausea and vomiting.

Outcome measures

All outcome data were collected by a blinded research assessor. The primary outcome was the cumulative 24-hour postoperative intravenous morphine consumption. Secondary outcomes were postoperative knee pain at rest and during movement, the incidence of moderate to severe posterior knee pain while at rest and during movement, total intravenous morphine consumption, the first time to the end of analgesia, Timed Up and Go (TUG) test scores, quadriceps muscle strength (QDS), the degree of maximum active flexion and extension of the operated knee, and the incidence of adverse events following LIA, CACB and iPACK block—including neural motor blockade, intravascular injection or vascular puncture, and local anesthetic systemic toxicity (LAST). Other parameters, including nausea/vomiting, dizziness, sleep disturbances, patient satisfaction, and lengths of hospital stays, were also recorded.

Outcome assessment included intravenous morphine consumption (administered when patients were complaining of NRS ≥4) during the first 12, 24, and 48 hours. The NRS pain scores (0=no pain, 10=worst pain imaginable) at rest and during movement at 0, 4, 8, 12, 24, 36 and 48 hours postoperatively and at 5 days, 2 weeks, 6 weeks, and 2 months were documented by using telephone interviews. The first time to the end of analgesia was defined as an hour from the end of surgery to the first point where the NRS score was ≥4. Moderate to severe postoperative posterior knee pain at rest and during movement was defined as pain involving the back of the knee with NRS ≥4 and was assessed until 2 month postoperatively. The TUG test, QDS test, and active range of motion (ROM) were evaluated by a blinded physiotherapist until POD 2. The TUG test, measured as time in seconds, required the patient to stand up from an armchair, walk 3 m, turn, walk back to the chair, and sit down.21 Quadriceps muscle strength was measured in the sitting position with full extension, and at 45° and 90° positioning of the knee joint using a digital dynamometer (MicroFET2, Hoggan Health Industries, Salt Lake City, UT, USA).22 Patient satisfaction was assessed 48 hours postoperatively on a visual analog scale (0=least satisfied, 10=most satisfied). Nausea and vomiting scores and dizziness scores were recorded on a visual analog scale (0=none, 10=severe). The incidence of sleep disturbances (excluding disturbances attributable to nursing care) was assessed over 48 hours postoperatively, and the length of hospital stay was defined as the number of hours between the TKA procedure and discharge (discharge criteria23 were assessed by the blinded surgeon). Other adverse events, including LAST and common peroneal nerve motor weakness, were also evaluated.

Statistical analysis

The primary outcome of this study was the cumulative intravenous morphine consumption over 24 hours postoperatively. In a pilot study of 10 patients who received spinal anesthesia without intrathecal morphine combined with intraoperative LIA and CACB, the mean and SD values for the consumption of morphine over 24 hours postoperatively were 4.22 and 2.87, respectively. We considered that a 50% reduction in intravenous morphine consumption would be clinically significant, which corresponds to an effect size of 0.73. Therefore, we determined that a sample size of 31 patients per group would provide 80% power at a two-sided α of 0.05. Accounting for a potential 15% attrition rate, a sample of 36 participants was required for each group. Thus, a total of 72 patients were recruited into this trial.

All analyses were performed in STATA V.14.0 (StataCorp, College Station, TX, USA). Continuous variables are expressed as the mean±SDs, mean (SEM) with a 95% CI or median (IQR) as appropriate. Categorical variables are expressed as numbers and percentages. A Student’s t-test or a Mann-Whitney U test was performed to compare continuous variables between groups. Categorical data were analyzed using the χ2 test or Fisher’s exact test. We applied an autoregressive correlation structure for cumulative intravenous morphine consumption. Given the repeated measures study design, the likelihood of missing data, correlated errors within individuals, and heterogeneity among occasion variances (ie, over time), we applied generalized estimating equations (GEE) to test for differences in the NRS pain score, TUG test, QDS, and ROM. Data collected at baseline and at other time points were included in the GEE analysis. We did not impute values when data were missing and performed intention-to-treat analyses. An effect was considered statistically significant at p<0.05 (95% CI excluded zero).


A total of 113 patients were assessed initially, of whom 41 did not meet the eligibility criteria and were declined participation (figure 1). The remaining 72 patients were enrolled and randomly assigned to one of the two groups. Three patients in each group were excluded after allocation for surgical reasons. Two patients in the LIA+CACB group and one patient in the iPACK+LIA+CACB group were converted from spinal to general anesthesia, and one patient in the iPACK+CACB group had a postoperative complication. Thus, 32 patients in the LIA+CACB group and 33 patients in the iPACK+LIA+CACB group completed the study. Patients were not excluded for protocol violation, according to intention-to-treat analysis. The overall demographics and baseline characteristics in both groups were similar (table 1).

Table 1

Demographics data

Figure 1

Consolidated Standards of Reporting Trials statement flow diagram. CACB, continuous adductor canal block; iPACK, infiltration between the popliteal artery and capsule of the knee block; LIA, local infiltration analgesia.

Assessment of cumulative postoperative 24 hours of intravenous morphine consumption, the primary outcome, indicated no statistically and clinically significant differences were observed between the groups (LIA+CACB, 1.3±1.9 mg vs iPACK+LIA+CACB, 0.6±1.3 mg; p=0.08), nor any difference observed at 12 (LIA+CACB, 0.4±1 mg vs iPACK+LIA+CACB, 0.1±0.5 mg; p=0.11) and 48 hours (LIA+CACB, 1.4±1.9 mg vs iPACK+LIA+CACB, 0.7±1.4 mg; p=0.14) postoperatively.

When comparing the groups based on the mean difference in NRS scores after treatment, the iPACK+LIA+CACB group had a statistically significant reduction in the mean NRS scores regarding knee pain during movement compared with those in the LIA+CACB group within 48 hours postoperatively (p<0.05). However, those scores were consistently below 3 in both groups at all time points and the difference of maximum score was minimal (2.3 vs 2.2), indicating no clinical significance (table 2). Moreover, there were no significant intergroup differences in the first time to the end of analgesia (LIA+CACB, 3.1±7.8 hours vs iPACK+LIA+CACB, 5.9±9.6 hours; p=0.19) and the incidence of moderate to severe posterior knee pain (defined as NRS ≥4) (p>0.05) (figure 2).

Table 2

Numerical pain rating scale at postoperative 0–48 hours

Figure 2

A line graph representing the percentage of the incidence of posterior knee pain (NRS ≥4) at rest and during movement from preoperative until 2 months after total knee arthroplasty. CACB, continuous adductor canal block; iPACK, infiltration between the popliteal artery and capsule of the knee block; LIA, local infiltration analgesia; NRS, numerical rating scale.

The mean changes for TUG test results were significantly lower in the iPACK+LIA+CACB group compared with those in the LIA+CACB group on POD 1 (−30.3 s (−52.6 to −8), p=0.008) and POD 2 (−38.9 s (−61.1, −16.8), p=0.001) (figure 3). Regarding the muscle strength test, the mean QDS test in the iPACK+LIA+CACB group was greater on POD 0–2 but had significantly reduced from baseline at 45° and 90° of the knee compared with the LIA+CACB group on POD 0 (p<0.05) (figure 3). However, the number of patients in the iPACK+LIA+CACB group (30 patients) that could perform the test was higher in POD 0 compared with the LIA+CACB group (24 patients). The active ROM in the knee extension on POD 1 and flexion on POD 0 in the iPACK+LIA+CACB group had significantly changed from baseline compared with the LIA+CACB group (p=0.003 and p=0.02, respectively). However, no significant difference was found in other time points (figure 3).

Figure 3

Timed Up and Go test, quadriceps muscle strength test and active range of motion at postoperative days 0–2. *P value <0.05. CACB, continuous adductor canal block; iPACK, infiltration between the popliteal artery and capsule of the knee block; LIA, local infiltration analgesia; QDS, quadriceps muscle strength test; ROM, range of motion; TUG, Timed Up and Go test.

Patients in the iPACK+LIA+CACB group showed a significantly lower incidence of sleep disturbance on POD 0 (p=0.02) and POD 1 (p=0.021) and significantly shorter mean hospitalization than those in the LIA+CACB group (p=0.007). There was no intergroup difference in postoperative nausea and the incidence of vomiting or dizziness, nor in the satisfaction scores (table 3). Procedure-related complications were not observed in any of the patients during ambulation and none of the patients reported postoperative motor blockade.

Table 3

Side effects, hospital stay and satisfaction score

After being discharged, patients in the iPACK+LIA+CACB group showed a statistically significant but clinically unimportant reduction in the NRS knee pain scores during movement for 5 days postoperatively. Moreover, no significant intergroup difference was observed in the pain scores at the other time points (table 4).

Table 4



This is the first randomized controlled trial to demonstrate that the addition of the iPACK block to a multimodal regimen that includes LIA and CACB does not decrease the postoperative intravenous opioid consumption or improve analgesia after TKA. Although there was a statistically significant decrease in postoperative knee analgesia during movement during the first 5 days after TKA in the iPACK+LIA+CACB group, the difference was marginal and not clinically meaningful. These results are likely due to the synergistic analgesic effect from sufficient motor-sparing blockade in each part of the knee joint in the current multimodal analgesia protocol, similar to those outlined in previous studies where the LIA combined with ACB or the iPACK block was combined with single-shot ACB and LIA.14 15 24

The overall low cumulative intravenous morphine consumption (primary outcome) may cause this study to be underpowered for the detection of statistical differences between the groups. However, we did find that the number of patients in the intervention group (27.3%) requiring intravenous opioids was lower than in the control group (40.6%). Moreover, even if the positive finding in secondary outcomes had been random rather than statistically significant, both primary and secondary outcomes (lower pain scores and improved functional outcome results as well as a shorter hospital stay), may still demonstrate the effectiveness of the iPACK block when combined with LIA and CACB in TKA. Consequently, this may reinforce the importance of motor-sparing blocks in the anterior and posterior aspects of the knee, especially using an unblind technique such as ultrasound-guided CACB and iPACK block, in order to improve the ‘opioid-sparing effect’ for postoperative analgesia in patients undergoing TKA.

Our results demonstrated that adding the iPACK block provided a similar incidences of posterior knee pain even though it has its main effect on the posterior aspect of the knee. These results were likely due to the similar effect of the LIA technique in both groups and the fact that the posterior aspect has less nerve supply than the anterior aspect of the knee.25 The iPACK block may have an additional effect on the other parts of the knee that the ACB and LIA may not adequately cover, possibly via the spreading of local anesthetic to the medial or lateral genicular nerves.26 This might have brought about an improvement of physical function outcomes and increase in patient’s ability to perform—including repeatedly measured characteristics (TUG and quadriceps strength tests),27 28 and the greater proportion of patients in the intervention group who were able to sleep without interference. Nevertheless, it is difficult to directly attribute these improvements to the iPACK block alone since the study was not specifically powered for them.

Over several years, the postoperative analgesic protocol in our institution for TKA has evolved from the prior standard of intrathecal morphine to intrathecal morphine combined with single-shot femoral nerve block or ACB; LIA combined with single-shot ACB,23 and finally to LIA combined with CACB29 when the patients received spinal anesthesia and underwent a perioperative multimodal regimen. These gradual improvements in postoperative analgesia and immediate rehabilitation have led to a significant reduction in the length of hospital stay (from 7 days to 2–3 days). In our study, although the addition of iPACK block to LIA with CACB could reduce the length of hospital stay, the difference was only about 9 hours and therefore was unlikely to affect the hospital stay cost, and also had a minimal impact on early discharge for the patients undergoing TKA.

The differential synergistic effects of these combinations of motor-sparing blocks with continuous catheter analgesia seemed to resolve 5 days postoperatively because no significant difference in the knee pain scores was observed between the groups thereafter. However, the patients in both groups continued to exhibit low pain scores. Therefore, these results likely indicate the importance of the continuous anesthetic infusion technique,30 in which the catheter was removed prior to hospital discharge in our study. Further studies may be required to explore the effect of continuous anesthesia in a disposable pump for home infusion when the patient is discharged early from the hospital.

This study had some limitations. First, due to the concerning dilution effect of 20 mL of the sham iPACK block on the efficacy of intraoperative LIA (the part of posterior capsule injection) in the LIA+CACB group, we decided to used only 5 mL of the sham iPACK block in the LIA+CACB group. Therefore, there is a potential bias due to the unblinding of the anesthesiologist during this procedure. However, all patients in this study were undergoing TKA for the first time and therefore were unlikely to be aware of the difference in the amount of the iPACK block. Second, a combination of iPACK, LIA, and ACB may have resulted in a high dosage of local anesthetic with the potential for LAST. Although this study did not evaluate the plasma levels of local anesthetic to determine the concentrations of local anesthetics in serum, and thus could not exclude the risk of LAST, the effective dosage of LIA usually ranges from 100 to 200 mg of bupivacaine or levobupivacaine, of which the lowest dose was used in this study.13–15 23 24 Moreover, the total dosage (200 mg) of local anesthetic in the combination of iPACK (50 mg), ACB (50 mg), and LIA (100 mg) in this study was less than that in a prior study24 that reported no neurological complications or clinical findings of LAST in patients. Last, the optimal dose for the iPACK block is still undetermined, and further studies are warranted to explore the optimal dose and volume of the iPACK block.

In conclusion, this study demonstrates that the addition of the iPACK block to LIA and CACB does not reduce postoperative opioid consumption and pain in patients undergoing TKA which may, nonetheless, improve immediate functional performance and reduce the duration of hospitalization. Further studies are needed to evaluate the optimal dosage when using this combination of motor-sparing blocks in opioid-sparing multimodal analgesic approaches.


The authors gratefully thank the nurses, anesthetists, physiotherapists, and orthopedic surgeons involved in this trial at King Chulalongkorn Memorial Hospital for their assistance and support.



  • Contributors WK and CV: planning, conception, and design of the study, analyzed the data, interpretation of data, wrote the manuscript, and revised the manuscript. WK, AT and NS: planning and conducting the study, reporting, and acquisition of data, analyzed the data, and gave constructive criticism. AT and SN: performed all surgeries, postoperative management, interpretation of data, and gave constructive criticism.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval University of Missouri–Kansas City Office of Research Compliance (FWA 00005427). This single-center, prospective, double-blinded randomized controlled trial was approved by the Institutional Review Board of the Chulalongkorn University (Ref 327/61), and the trial was registered on the Thai Clinical Trials Registry before patient enrollment (principle investigator: WK; registration date: 2 July 2018; first enrollment date: 6 July 2018).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available upon reasonable request. ORCID iD: CV:; WK:; AT:; SN:; NS: