Liposomal Bupivacaine versus Bupivacaine Hydrochloride Plus Dexamethasone for Combined Adductor Canal and iPACK Blocks in Total Knee Arthroplasty: A Randomized Controlled Trial

Abstract

Purpose

To compare postoperative analgesia and early quadriceps recovery between liposomal bupivacaine and bupivacaine hydrochloride plus dexamethasone administered through combined adductor canal block (ACB) and infiltration between the popliteal artery and capsule of the posterior knee (iPACK) in patients undergoing total knee arthroplasty (TKA).

Methods

This single-center, randomized, controlled, outcome-assessor-blinded trial included 60 patients in the final analysis (30 per group). The BL group received liposomal bupivacaine plus bupivacaine hydrochloride, whereas the Dex group received bupivacaine hydrochloride plus dexamethasone; both regimens were administered through combined ACB and iPACK. The primary outcome was the visual analog scale (VAS) pain score at rest 24 hours after surgery. Secondary outcomes included pain at other time points, intraoperative hemodynamics and opioid use, quadriceps muscle strength, patient-controlled intravenous analgesia (PCIA) consumption, rescue analgesia, and opioid-related adverse events. The trial was registered with the Chinese Clinical Trial Registry (ChiCTR2500115826; December 31, 2025).

Results

At 24 hours, resting VAS scores were 2.4±0.8 in the BL group and 2.9±1.1 in the Dex group, with a mean difference of −0.52 points (95% CI, −1.01 to −0.02; P=0.040). Repeated-measures analysis showed significant group, time, and group-by-time effects for both resting and movement VAS, and quadriceps strength favored the BL group at 12 hours. The 48-hour cumulative PCIA volume was 99.75 (96.00, 100.00) mL versus 100.00 (100.00, 100.00) mL (P<0.001). Rescue analgesia was more frequent in the Dex group overall (RR=2.40, 95% CI 1.12–5.14; P=0.024), including at 48 hours (3/30 vs 11/30; P=0.030).

Conclusion

In patients undergoing TKA, the liposomal bupivacaine-containing ACB+iPACK regimen was associated with lower postoperative pain at selected time points and better quadriceps strength at 12 hours than the bupivacaine hydrochloride plus dexamethasone regimen. These findings suggest a potential early analgesic benefit during the first 72 postoperative hours, although the clinical importance of the pain difference warrants cautious interpretation.

Introduction

Although total knee arthroplasty (TKA) can relieve pain and improve joint function in patients with end-stage knee osteoarthritis, moderate-to-severe pain during the early postoperative period remains a major barrier to enhanced recovery. Inadequate pain control can impede knee flexion and extension, ambulation, and rehabilitation; increase opioid exposure and opioid-related adverse effects; delay functional recovery; and reduce patient satisfaction.1–5 Perioperative analgesia for TKA should therefore address pain during movement, preserve motor function, and minimize opioid use rather than focus solely on pain at rest. Adductor canal block (ACB) primarily targets sensory input from the anteromedial knee, whereas infiltration between the popliteal artery and capsule of the posterior knee (iPACK) targets sensory branches supplying the posterior capsule. Their combination can cover pain arising from the anterior and posterior aspects of the knee while limiting the involvement of major motor nerves, making it a function-oriented regional analgesic strategy.2,6 However, a single injection of a conventional local anesthetic may not adequately cover pain beyond the first 24 postoperative hours or the continuing demands of rehabilitation.7,8 Extending analgesia without increasing motor blockade therefore remains an important goal of multimodal analgesia for TKA.

Liposomal bupivacaine may prolong analgesia after a single peripheral nerve block, whereas dexamethasone can prolong the effect of conventional local anesthetics.9–14 Because the regimens differ in formulation, total local-anesthetic exposure, and adjuvant use, this study was designed as a comparison of complete clinical regimens rather than formulations alone. Assessments at 12, 24, 48, and 72 hours characterized early postoperative analgesia and quadriceps-strength recovery. We compared postoperative pain, PCIA use, rescue analgesia, and quadriceps strength between the two ACB+iPACK regimens.

Materials and Methods

This report was prepared in accordance with the CONSORT 2025 statement. This single-center, randomized, controlled, outcome-assessor-blinded, two-group parallel clinical trial used a 1:1 allocation ratio. It was approved by the Ethics Committee of Beijing Anzhen Nanchong Hospital of Capital Medical University and Nanchong Central Hospital (approval No. 2025–220) and was conducted in accordance with the Declaration of Helsinki. The trial was registered with the Chinese Clinical Trial Registry (ChiCTR2500115826; registration date: December 31, 2025). All participants provided written informed consent. Patients and the public were not involved in the design, conduct, reporting, or dissemination planning of the trial.

Participants

Patients scheduled to undergo primary, elective, unilateral TKA at Beijing Anzhen Nanchong Hospital of Capital Medical University and Nanchong Central Hospital between January and June 2026 were screened for eligibility. All analyzed participants underwent primary unilateral TKA; none had previous ipsilateral knee surgery, revision TKA, or prior contralateral TKA.

The inclusion criteria were American Society of Anesthesiologists (ASA) physical status II or III, age 55–80 years, a diagnosis of knee osteoarthritis with an indication for TKA, and voluntary participation with written informed consent.

The exclusion criteria were a history of allergy to bupivacaine or another local anesthetic; clinically significant hepatic or renal dysfunction, particularly cirrhosis with abnormal liver enzyme levels; lumbar disc herniation with lower-limb neurologic symptoms or a history of peripheral nerve injury in a lower limb; electrocardiographic evidence of conduction block or ventricular arrhythmia; inability to complete study assessments; a history of psychiatric or neurologic disease, alcohol or drug misuse; or long-term use of opioid analgesics.

Participants were withdrawn or the study intervention was discontinued if consent was withdrawn, the nerve block failed to meet the prespecified effectiveness criterion, a serious anesthesia-related adverse event occurred, follow-up was lost, adherence was poor, a comorbidity worsened substantially during hospitalization, a life-threatening condition developed, or a serious surgery-related adverse event occurred. All withdrawals and their reasons were recorded in the participant flow diagram.

Randomization and Blinding

A random allocation sequence was generated using SPSS version 25.0 by an investigator who was not involved in clinical procedures or data analysis. Allocation was concealed using sequentially numbered, sealed, opaque envelopes. A separate investigator prepared the study solutions according to group allocation and provided them to the anesthesiologist performing the ACB and iPACK. The solution-preparation investigator and the anesthesiologist performing the blocks were aware of group assignment but did not participate in postoperative follow-up or statistical analysis. The anesthesiologists responsible for general anesthesia and perioperative monitoring and the trained outcome assessors were blinded to allocation. Participants were not blinded because the study solutions differed in appearance. The trial was therefore randomized and outcome-assessor blinded.

Nerve Block Procedures

Routine vital-sign monitoring was established in the preanesthesia area. Maximum active knee-flexion angle and pain score were recorded before the blocks, after which ultrasound-guided iPACK and ACB were performed sequentially. The BL solution comprised 20 mL (266 mg; 13.3 mg/mL) of bupivacaine liposome injection (Hunan Kelun Pharmaceutical Co., Ltd., Yueyang, Hunan, China) and 10 mL of 0.75% bupivacaine hydrochloride (total volume, 30 mL). The Dex solution comprised 10 mL of 0.75% bupivacaine hydrochloride, 5 mg (1 mL) of dexamethasone, and 19 mL of 0.9% sodium chloride (total volume, 30 mL). The 20-mL/266-mg liposomal preparation corresponded to one complete commercial unit; both study regimens were prepared to the same total injection volume of 30 mL. For the iPACK block, the patient was placed in the lateral decubitus position. A low-frequency transducer was positioned in the popliteal fossa and moved proximally to identify the popliteal artery and the distal femoral surface. The needle was advanced laterally into the tissue plane between the popliteal artery and the femur. After negative aspiration, 10 mL of study solution was injected medially and another 10 mL laterally. The patient was then placed supine with the thigh slightly abducted. A high-frequency transducer was positioned on the medial thigh to identify the femoral artery within the adductor canal. After the needle tip reached the plane between the lateral aspect of the femoral artery and the vastus medialis muscle and negative aspiration was confirmed, the remaining 10 mL was injected. All blocks were performed by the same anesthesiologist with relevant experience. Representative ultrasound images showing the needle trajectories and target regions for the iPACK and adductor canal blocks are provided in Figure 1.

Figure 1.

Representative ultrasound images of the iPACK and adductor canal blocks. (A) iPACK block: 1, popliteal artery; 2, femur; 3, sciatic nerve; BF, needle trajectory; red dashed outline, target region. (B) Adductor canal block: 1, femoral artery; 2, saphenous nerve; SM, sartorius muscle; VM, vastus medialis muscle; AM, adductor magnus muscle; red dashed outline, target region. Medial and lateral orientation is indicated on each panel.

Both groups received 75 mg of bupivacaine hydrochloride; the BL regimen additionally contained 266 mg of liposomal bupivacaine, whereas the Dex regimen contained 5 mg of dexamethasone. Thus, the trial compared complete regimens rather than pharmacologically equivalent formulations.

Fifteen minutes after completion of the blocks, maximum active knee-flexion angle and pain intensity were reassessed. A block was considered effective if knee flexion increased by at least 5 degrees from baseline or the pain score decreased by at least 3 points. These prespecified operational thresholds were informed by prior literature on 15-minute block assessment, clinically meaningful acute-pain reduction, and clinically relevant flexion improvement.15–17 The individual components of the composite criterion were not recorded separately.

Anesthesia Protocol

Participants fasted for 8 hours and abstained from clear fluids for 4 hours before surgery. Electrocardiography, pulse oximetry, noninvasive blood pressure, and end-tidal carbon dioxide were monitored after arrival in the operating room. General anesthesia was induced with sufentanil 0.5 µg/kg, propofol 1–2 mg/kg, etomidate 0.2 mg/kg, and cisatracurium 0.2 mg/kg. After adequate neuromuscular blockade, the trachea was intubated and mechanical ventilation was initiated. Volume-controlled ventilation was used with a tidal volume of 6–8 mL/kg, a respiratory rate of 12 breaths/min, and an inspiratory-to-expiratory ratio of 1:2. Anesthesia was maintained with remifentanil 5 µg/kg/h and 1–2% sevoflurane. The sevoflurane concentration was adjusted to maintain a bispectral index of 50–60. Remifentanil was discontinued after prosthesis implantation, and sevoflurane was discontinued during skin closure. At the end of surgery, a PCIA pump containing sufentanil 150 µg, butorphanol 10 mg, and palonosetron 0.25 mg diluted with 0.9% sodium chloride to 100 mL was connected. The settings were a 2.0-mL loading dose, a 2.0-mL/h background infusion, a 0.5-mL patient-controlled bolus, and a 15-minute lockout interval. Participants were transferred to the postanesthesia care unit, extubated after meeting extubation criteria, and returned to the ward after their vital signs stabilized. If the postoperative VAS score exceeded 5 and pain was not substantially relieved after two consecutive patient-controlled boluses, tramadol 100 mg was administered intramuscularly as rescue analgesia.

No additional scheduled non-opioid analgesia or periarticular local infiltration analgesia was used after return to the ward; both groups received only the PCIA regimen and rescue tramadol described above.

Outcome Measures

The primary outcome was the VAS pain score at rest 24 hours after surgery. Pain was rated on a 0–10 scale, where 0 indicated no pain and 10 indicated the worst imaginable pain; higher scores indicated greater pain intensity.

Secondary outcomes included VAS scores during movement at 12, 24, 48, and 72 hours and at rest at 12, 48, and 72 hours; heart rate and blood pressure before anesthesia induction (T1), 10 minutes after skin incision (T2), and at completion of prosthesis implantation (T3), as well as intraoperative opioid use; quadriceps strength at 12, 24, 48, and 72 hours; cumulative PCIA volume and effective button presses at 12, 24, and 48 hours; opioid-related adverse events; and rescue analgesia at each time point. Intraoperative opioid doses were converted to morphine equivalents using published conversion factors.18 Quadriceps strength was assessed by trained blinded outcome assessors using a 0–5 manual scale; formal inter-rater or intra-rater reliability was not measured.

Routine perioperative safety monitoring was performed. The specific block-related complications listed by the reviewer were not collected as separate prespecified study variables.

Sample Size Calculation

The sample size was calculated using PASS version 25.0, with 24-hour resting VAS as the primary outcome. The calculation was based on a 10-patient pilot study conducted at our institution before the randomized trial; none of these patients were included in the final analysis. Pilot mean VAS scores were 2.35±0.89 and 3.32±1.19 in the liposomal bupivacaine and bupivacaine hydrochloride plus dexamethasone groups, respectively. With a two-sided α of 0.05, 90% power, and 1:1 allocation, the required sample size was calculated using the following formula:

In the formula, Z(α/2)=1.96, Z(β)=1.28, μ1-μ2=0.97, σ1=0.89, and σ2=1.19. Substitution yielded an estimated sample size of 24.66 participants per group, which was rounded to 25. Allowing for an anticipated attrition or exclusion rate of approximately 15%, the target evaluable sample size was 30 participants per group (60 in total). VAS scores at rest at 48 and 72 hours were secondary exploratory outcomes and were not included in the sample-size calculation.

Statistical Analysis

Statistical analyses were performed using SPSS version 25.0. Normality and homogeneity of variance were assessed using the Shapiro–Wilk and Levene tests. Continuous variables were compared using independent-samples t tests or Mann–Whitney U-tests, as appropriate; categorical variables were compared using Pearson chi-squared or exact tests. Repeated VAS measurements were analyzed by repeated-measures ANOVA for group, time, and group-by-time effects, with partial eta squared (partial η²) and 95% CIs reported as effect sizes. Rescue analgesia was analyzed using modified Poisson GEE with robust standard errors to estimate RRs and 95% CIs; when the group-by-time interaction was nonsignificant, a main-effects model was reported. Time-specific comparisons used two-sided Fisher exact tests. No multiplicity adjustment was prespecified for secondary outcomes; these analyses were therefore considered exploratory and their P values nominal. All tests were two-sided, with P<0.05 considered statistically significant.

Results

Of 68 randomized participants (34 per group), 60 (30 per group) were included in the final analysis; postrandomization exclusions are shown in Figure 2. Baseline demographic and clinical characteristics are summarized in Table 1.

Figure 2.

Participant flow diagram.

Table 1.

Demographic and Clinical Characteristics

	Dex Group (n=30)	BL Group (n=30)	P value
Age, years, mean±SD	71.0±4.7	71.0±4.5	0.61
Sex, female/male, n	22/8	22/8	1.00
Height, cm, mean±SD	156.0±5.9	156.0±5.4	0.71
Weight, kg, mean±SD	60.43±9.84	62.7±8.93	0.45
BMI, kg/m2, mean±SD	25.0±2.9	26.0±3.2	0.20
ASA physical status, n (%)
II	17 (56.7)	14 (46.7)
III	13 (43.3)	16 (53.3)	0.44
Surgical duration, min, mean±SD	128.0±14.0	129.0±31.0	0.81

Note: Preoperative VAS, knee range of motion, anxiety/depression status, and tourniquet time were not collected. Neither group received periarticular local infiltration analgesia.

Postoperative Pain

Postoperative VAS scores at rest and during movement are shown in Figure 3.

Figure 3.

Postoperative visual analog scale (VAS) pain scores at rest and during movement. (A) VAS scores at rest. (B) VAS scores during movement. Data are presented as mean±SD. *P<0.05; **P<0.01; ***P<0.001.

At 24 hours, resting VAS was 2.4±0.8 in the BL group and 2.9±1.1 in the Dex group (mean difference, −0.52; 95% CI, −1.01 to −0.02; P=0.040). Resting and movement VAS scores were lower in the BL group at 24, 48, and 72 hours, with no between-group difference at 12 hours. Repeated-measures ANOVA showed significant group, time, and group-by-time effects for resting VAS [F(1,58)=7.49, P=0.008, partial η²=0.114 (95% CI, 0.008–0.273); F(3174)=20.40, P<0.001, 0.260 (0.147–0.352); F(3174)=4.03, P=0.008, 0.065 (0.005–0.133), respectively] and movement VAS [F(1,58)=14.12, P<0.001, 0.196 (0.044–0.361); F(3174)=20.94, P<0.001, 0.265 (0.151–0.357); F(3174)=4.93, P=0.003, 0.078 (0.011–0.151), respectively].

Intraoperative Outcomes

As shown in Figure 4, heart rate, systolic blood pressure, and diastolic blood pressure did not differ significantly between groups at T1, T2, or T3 (all P>0.05). T1 was before anesthesia induction, T2 was 10 minutes after skin incision, and T3 was at completion of prosthesis implantation. Intraoperative opioid use, expressed as morphine equivalents, also did not differ significantly between groups [36.03 (30.95, 38.73) vs 35.90 (30.97, 40.55); Z=−0.133; P>0.05].

Figure 4.

Intraoperative vital signs in the two groups. (A) Heart rate. (B) Systolic blood pressure. (C) Diastolic blood pressure. T1, before anesthesia induction; T2, 10 minutes after skin incision; T3, completion of prosthesis implantation.

As shown in Table 2, quadriceps strength at 12 hours favored the BL group (P=0.01), with no between-group differences at later time points. At 48 hours, cumulative PCIA volume was 99.75 (96.00, 100.00) mL in the BL group and 100.00 (100.00, 100.00) mL in the Dex group (P<0.001); 12- and 24-hour PCIA values did not differ significantly.

Table 2.

Quadriceps Muscle Strength and PCIA Consumption

	Dex Group (n=30)	BL Group (n=30)	P value
Quadriceps strength at 12 h, n (%)			0.01
Grade 3	6 (20.0)	0 (0.0)
Grade 4	14 (46.7)	11 (36.7)
Grade 5	10 (33.3)	19 (63.3)
Quadriceps strength at 24 h, n (%)			1.00
Grade 4	3 (10.0)	3 (10.0)
Grade 5	27 (90.0)	27 (90.0)
Quadriceps strength at 48 h, n (%)			1.00
Grade 5	30 (100.0)	30 (100.0)
Quadriceps strength at 72 h, n (%)			1.00
Grade 4	0 (0.0)	0 (0.0)
Grade 5	30 (100.0)	30 (100.0)
PCIA consumption at 12 h, mL, mean±SD	25.0±2.7	24.0±1.7	0.44
PCIA consumption at 24 h, mL, mean±SD	51.0±3.4	50.0±3.1	0.17
PCIA consumption at 48 h, mL, median (IQR)	100.00 (100.00–100.00)	99.75 (96.00–100.00)	<0.001

Adverse Events and Rescue Analgesia

Nausea or vomiting occurred in 12 participants in the Dex group and 8 in the BL group; the difference was not significant. No dizziness or pruritus was reported in either group.

Rescue analgesia occurred in 3/30 versus 7/30 participants at 12 hours, 5/30 versus 6/30 at 24 hours, 3/30 versus 11/30 at 48 hours, and 4/30 versus 11/30 at 72 hours (BL vs Dex; Table 3). The group-by-time interaction was not significant (P=0.284); in the main-effects GEE model, the Dex group had a higher overall risk of rescue analgesia (RR=2.40; 95% CI, 1.12–5.14; P=0.024). At 48 hours, rescue analgesia was more frequent in the Dex group (36.7% vs 10.0%; RR=3.67, 95% CI 1.14–11.84; P=0.030); other time-point comparisons were not significant.

Table 3.

Rescue Analgesia by Time Point, Absolute Risk Differences, and Risk Ratios

Time Point	BL Group, n/N (%)	Dex Group, n/N (%)	Absolute Risk Difference, Percentage Points	RR (95% CI)	P value
12 h	3/30 (10.0)	7/30 (23.3)	13.3	2.33 (0.67–8.18)	0.299
24 h	5/30 (16.7)	6/30 (20.0)	3.3	1.20 (0.41–3.51)	1.000
48 h	3/30 (10.0)	11/30 (36.7)	26.7	3.67 (1.14–11.84)	0.030
72 h	4/30 (13.3)	11/30 (36.7)	23.3	2.75 (0.99–7.68)	0.072

Notes: Risk differences are Dex minus BL and RRs use BL as the reference. Time-specific P values are two-sided Fisher exact tests and were not adjusted for multiplicity. The modified Poisson GEE overall group effect was RR=2.40 (95% CI, 1.12–5.14; P=0.024), with a group-by-time interaction P=0.284.

Discussion

At 24 hours, the between-group difference in resting VAS was −0.52 points (95% CI, −1.01 to −0.02). Although statistically significant, this was smaller than commonly cited clinically important differences of approximately 1 point on a 0–10 VAS and 15 mm in arthroplasty analgesia trials.19,20 Thus, the magnitude of the analgesic difference appears modest.

The absence of a between-group VAS difference at 12 hours may reflect residual conventional local-anesthetic activity and the prolonging effect of dexamethasone. In a recent TKA trial, perineural dexamethasone prolonged analgesia after combined ACB and iPACK from 8.8 to 15.9 hours.21

Differences emerging after 24 hours are pharmacologically compatible with prolonged-release liposomal bupivacaine.13,22 However, block duration and drug concentrations were not measured, and total local-anesthetic exposure differed between groups; this remains a plausible explanation rather than a demonstrated mechanism.

Previous ACB trials have reported heterogeneous effects of liposomal bupivacaine. Hubler et al found lower activity pain and opioid use, Malige et al reported lower inpatient pain and opioid use, whereas Quaye et al found lower opioid use without lower pain scores.23–25 Differences in comparator regimens and concomitant analgesia may explain this variability.

Intraoperative vital signs and opioid use did not differ between groups. Because these were secondary measurements and the trial was not powered for safety, no safety advantage can be inferred.

ACB+iPACK is intended to limit motor involvement,26,27 but the 12-hour quadriceps-strength difference should be interpreted cautiously. Strength was measured manually by trained blinded assessors without formal reliability testing, and pain or patient effort may influence scores. Thus, this finding does not prove differential motor-nerve preservation.

Because ambulation distance, time to first mobilization, postoperative knee range of motion, rehabilitation completion, and patient-reported function were not assessed, the pain and manual-strength findings should not be interpreted as evidence of improved rehabilitation.

Rescue analgesia was more frequent in the Dex group overall (RR=2.40; 95% CI, 1.12–5.14; P=0.024), and the 48-hour comparison was also significant (P=0.030). Because these secondary time-point analyses were not adjusted for multiplicity, they should be interpreted as exploratory. Postoperative adverse-event rates did not differ significantly between groups.

Because the BL regimen included additional liposomal bupivacaine whereas the Dex regimen included dexamethasone, equal injection volume did not imply equivalent local-anesthetic exposure. The findings therefore apply to the complete regimens rather than to formulation alone.

Key limitations include the single-center, modest sample; incomplete blinding; unmatched local-anesthetic exposure; use of an operational composite block-success criterion without separate component counts; absence of formal reliability testing for manual strength; unadjusted multiple secondary comparisons; lack of structured collection of all block-related complications; and follow-up limited to 72 hours. These factors limit causal, safety, and longer-term functional inferences.

Conclusion

In patients undergoing TKA, the liposomal bupivacaine-containing ACB+iPACK regimen was associated with lower postoperative pain at selected time points and better quadriceps strength at 12 hours than the bupivacaine hydrochloride plus dexamethasone regimen. These findings suggest a potential early analgesic benefit during the first 72 postoperative hours, although the clinical importance of the pain difference warrants cautious interpretation.

Funding Statement

This study was supported by the Medical Empowerment Public Welfare Special Fund of the China Red Cross Foundation (Grant No. 2023HXKT011).

Data Sharing Statement

Deidentified participant data may be obtained from the corresponding author, Ying Xie (475153895@qq.com), upon reasonable request and subject to ethics approval and a data-use agreement. Data will be available for 5 years after publication and will not be deposited in a public repository.

Ethics Approval and Informed Consent

The study was approved by the Ethics Committee of Beijing Anzhen Nanchong Hospital of Capital Medical University and Nanchong Central Hospital (approval No. 2025-220) and was conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent before enrollment.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors report no financial or non-financial conflicts of interest in relation to this work.