Liposomal bupivacaine versus ropivacaine thoracic paravertebral block for thoracoscopic atrial septal defect repair: analgesic efficacy and cardiac biomarker outcomes in a single-center, assessor-blinded, randomized trial

Abstract

Background

Thoracoscopic atrial septal defect (ASD) repair, although minimally invasive, can cause substantial postoperative pain. Thoracic paravertebral block (TPVB) provides effective chest wall analgesia, and liposomal bupivacaine (LB), a long-acting local anesthetic with sustained release, may prolong analgesic duration. We investigated whether ultrasound-guided TPVB using long-acting LB improves postoperative analgesia without compromising cardiovascular safety compared with ropivacaine (RP).

Methods

This randomized, assessor-blinded trial enrolled 94 patients undergoing thoracoscopic ASD repair, who were randomly assigned (1:1) to receive ultrasound-guided TPVB with either LB or RP. Eligibility criteria included age 18–55 years, American Society of Anesthesiologists (ASA) II–III, New York Heart Association (NYHA) II–III, and ejection fraction (EF) >50%. Randomization was computer-generated with sealed envelopes. Safety monitoring included continuous electrocardiography (ECG) for arrhythmias, with Clavien-Dindo grading and Naranjo causality assessment. The primary endpoint was the numerical rating scale (NRS, 0–10) pain score during coughing at 24 h postoperatively. Secondary outcomes included NRS scores at other time points, 72-h opioid consumption (morphine milligram equivalents, MME), time to first rescue analgesia, extended analgesia outcomes [postoperative days 4–7 (POD4–POD7)], perioperative hemodynamics, postoperative recovery, cardiac/inflammatory biomarker trajectories, and adverse events (with postoperative arrhythmias analyzed separately).

Results

Baseline demographic, clinical, and intraoperative variables were well-balanced between groups (all P>0.05). Compared with RP, LB was associated with lower NRS pain scores at rest and during coughing from extubation through 72 h (all P<0.05), including the primary endpoint at 24 h. LB significantly reduced 72-h opioid consumption [median interquartile range (IQR) 7.5 (0, 42) mg vs. 45 (30, 67.5) mg; P<0.001] and increased the proportion of patients requiring no rescue analgesia within 72 h (72.4% vs. 27.6%; P=0.003). Time to first rescue analgesia was numerically longer with LB (20.5 h vs. 12.0 h; P=0.10), and opioid-sparing effects persisted on POD4–POD7 (both P<0.001). Mechanical ventilation duration was shorter in the LB group [5 (3, 6) h vs. 6 (4, 7.5) h; P=0.03]. Cardiac and inflammatory biomarker trajectories did not differ between groups. Non-arrhythmic adverse events were comparable, and postoperative arrhythmia incidence within 30 days was similar between groups (13.3% vs. 15.6%).

Conclusions

In adults undergoing thoracoscopic ASD repair, LB-based TPVB provides superior early analgesia with sustained opioid sparing and shorter mechanical ventilation duration, without signals of increased cardiovascular risk, including postoperative arrhythmias. These findings support considering LB-TPVB as part of multimodal analgesia for thoracoscopic ASD repair, pending confirmation in larger multicenter studies.

Trial Registration

ChiCTR2500112278, retrospectively registered at the Chinese Clinical Trial Registry on November 12, 2025.

Highlight box.

Key findings

• In this randomized trial, liposomal bupivacaine-based thoracic paravertebral block (LB-TPVB) significantly improved early postoperative analgesia compared with ropivacaine in patients undergoing thoracoscopic atrial septal defect repair.

• LB-TPVB reduced 72-hour opioid consumption and increased the proportion of patients not requiring rescue analgesia.

• No significant differences were observed in perioperative hemodynamics, cardiac biomarkers, or 30-day arrhythmia incidence between groups.

What is known and what is new?

• TPVB is an established regional anesthesia technique for thoracic procedures, and LB has been shown to prolong analgesic duration in non-cardiac surgery. However, evidence in minimally invasive cardiac surgery remains limited.

• This study provides randomized evidence that LB-TPVB improves postoperative pain control and reduces opioid requirements in thoracoscopic cardiac surgery, while demonstrating comparable short-term safety profiles.

What is the implication, and what should change now?

• LB-TPVB may be considered as part of multimodal, opioid-sparing analgesia strategies in selected patients undergoing minimally invasive cardiac surgery.

• These findings support further evaluation in larger, multicenter studies with standardized perioperative protocols to confirm generalizability and clinical impact on recovery outcomes.

Introduction

Atrial septal defect (ASD) is one of the most common congenital heart diseases, with an estimated global incidence of approximately 1.6 per 1,000 live births. It is characterized by abnormal communication between the atria caused by defective septal development, resulting in left-to-right shunting of blood (1). Thoracoscopic minimally invasive cardiac surgery has been increasingly adopted for ASD repair because of its smaller incision and faster recovery compared with traditional sternotomy (2). However, postoperative pain remains a significant concern, arising from incisional trauma, pericardial manipulation (3), and pleural irritation by drainage tubes. Severe pain can impair respiratory effort, delay extubation and mobilization, increase the risk of atelectasis, pulmonary infection, and chronic pain, thereby prolonging hospitalization and adversely affecting recovery (4). Optimizing perioperative analgesia is therefore essential to attenuate surgical stress and promote enhanced postoperative recovery.

Opioid-based regimens remain the traditional approach for perioperative pain control in cardiac surgery, yet opioid-related adverse effects, such as respiratory depression, nausea, vomiting, and delayed gastrointestinal recovery, have prompted significant interest in alternative strategies. With the growing emphasis on enhanced recovery after surgery (ERAS) concepts, multimodal analgesic techniques incorporating regional anesthesia (nerve blocks) have gained increasing popularity (5). Of these, thoracic paravertebral block (TPVB) provides reliable unilateral chest wall analgesia and effectively reduces opioid requirements (6). The use of ultrasound guidance has further improved the accuracy and safety of TPVB by enabling precise local anesthetic deposition and dosage control, making this a preferred technique for thoracic and cardiac anesthesia (7). Ultrasound guidance enhances the precision of TPVB by real-time visualization of the paravertebral space, transverse process, and pleura, thereby reducing the risks of pneumothorax, vascular puncture, and block failure compared to blind techniques. However, conventional local anesthetics like ropivacaine (RP) provide only 6-8 hours of analgesia, which is insufficient to cover the most painful postoperative period following thoracoscopic cardiac surgery, often necessitating substantial opioid rescue.

LB is a novel long-acting, sustained-release formulation of bupivacaine that was approved by the U.S. Food and Drug Administration in 2011 (Exparel) and by the National Medical Products Administration of China in 2022 (Aihengping) (8). Conventional bupivacaine provides 6–8 hours (h) of analgesia, which may be insufficient for procedures requiring prolonged pain control, although higher doses are known to increase the risk of central nervous system and cardiovascular toxicity (9). LB is formulated using DepoFoam® multivesicular liposome technology, which encapsulates bupivacaine within lipid vesicles containing discrete aqueous chambers, thus allowing gradual and sustained drug release (10). A single injection of LB can provide effective analgesia for up to 72 h (11). Experimental and clinical studies have demonstrated minimal neural toxicity and good cardiac safety for LB in comparison with conventional local anesthetics (12). When applied in thoracic and cardiac surgery, the use of LB for local infiltration or nerve blocks has been shown to significantly reduce postoperative opioid consumption and shorten hospital stay (13). Specifically, a randomized trial in uniportal thoracoscopy patients found that LB-based intercostal nerve blocks reduced total 72-h opioid use by 50% (24 vs. 48 mg oral morphine equivalent, P=0.02) (13). A separate retrospective analysis further confirmed that LB was associated with a significantly shorter median hospital length of stay (4 vs. 5 days, P<0.01) across various thoracic procedures (14).

However, the evidence base for LB in thoracic surgery cannot be directly extrapolated to minimally invasive cardiac surgery. Patients undergoing thoracoscopic ASD repair present several unique challenges: (I) systemic heparinization and cardiopulmonary bypass (CPB) may alter local anesthetic pharmacokinetics and increase the risk of bleeding complications; (II) CPB-induced systemic inflammatory response and myocardial injury could confound the assessment of analgesic efficacy and safety; (III) the imperative for early extubation and enhanced recovery after cardiac surgery places a premium on opioid-sparing strategies; and (IV) theoretical concerns regarding local anesthetic cardiotoxicity, particularly arrhythmogenicity, are amplified in the context of recent cardiac surgery and myocardial manipulation. These factors underscore the need for dedicated evaluation of LB-TPVB in this specific population.

Based on these findings, the present study aimed to evaluate the efficacy and cardiovascular safety of ultrasound-guided TPVB with LB in patients undergoing thoracoscopic ASD repair. We hypothesized that LB-TPVB would enhance postoperative analgesia, reduce opioid consumption, and facilitate early extubation, without compromising hemodynamic stability or increasing the risk of cardiac adverse events, including postoperative arrhythmias and myocardial injury, thereby providing evidence to optimize multimodal analgesia strategies for minimally invasive cardiac surgery. We present this article in accordance with the CONSORT reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1135/rc).

Methods

This prospective, randomized, and controlled trial was conducted at The First Affiliated Hospital of Nanjing Medical University (Jiangsu Province Hospital) between May 2024 and August 2025.The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The ethics approval process in our institution involves a two-stage procedure. The study protocol was submitted to the Ethics Committee of The First Affiliated Hospital of Nanjing Medical University in March 2024 and was reviewed and approved at the Ethics Committee meeting in April 2024, thus allowing preliminary enrollment and data collection to commence. Formal written ethics approval (approval No. 2024-SR-188) was subsequently issued on the 4^th of March 2025 following completion of the administrative procedures and covering the entire study period retrospectively. Written informed consent was obtained from all participants before enrollment. With regards to trial registration, the Chinese Clinical Trial Registry (ChiCTR) requires a formal written ethics approval document to be uploaded to complete registration. Therefore, the study was registered retrospectively after the formal written approval became available (ChiCTR2500112278, November 2025; accessible at https://www.chinadrugtrials.org.cn/). The study protocol, outcome measures, and plan for statistical analysis were finalized prior to the enrollment of the first participant, and no changes to the primary or secondary outcomes were instigated following initiation of the study.

Participants

Patients undergoing elective thoracoscopic ASD repair between May 2024 and August 2025, with different analgesic strategies employed during the perioperative period, were included in this study. The inclusion criteria were as follows: (I) scheduled for elective thoracoscopic ASD repair with or without concomitant tricuspid valve repair; (II) aged 18 to 55 years, regardless of sex; (III) classified as American Society of Anesthesiologists (ASA) physical status II or III; (IV) classified as New York Heart Association (NYHA) functional class II or III; (V) left ventricular ejection fraction (EF) >50%; (VI) planned postoperative intensive care unit (ICU) admission; (VII) anticipated hospital stay >24 h. The exclusion criteria were as follows: (I) refusal to participate in the study; (II) contraindications for TPVB (such as thoracic infections, skin or soft tissue infections or tumors at the puncture site, unclear or abnormal anatomy at the puncture site, coagulation disorders, or allergy to local anesthetics); (III) chronic use of analgesic medications; (IV) a history of previous cardiac surgery or reoperation; (V) emergency surgery cases; (VI) a history of chronic pulmonary disease with pulse oxygen saturation <90% without supplemental oxygen; (VII) severe hepatic dysfunction (liver enzymes >three-fold higher than the upper normal limit and serum albumin <the normal reference range); (VIII) severe renal insufficiency requiring preoperative renal replacement therapy; and (IX) history of stroke, schizophrenia, depression, Parkinson’s disease, epilepsy, or dementia.

Randomization and blinding

A total of 94 eligible patients undergoing elective thoracoscopic ASD repair were enrolled and randomly assigned in a 1:1 ratio to either the LB or RP group. Randomization was performed using SPSS version 21 (IBM Corp., Armonk, NY, USA), with even and odd numbers representing the LB and RP groups, respectively. Due to visible differences between LB (a milky white suspension) and RP (a clear solution), the paravertebral blocks were administered by a dedicated anesthesiologist who was aware of group assignment and therefore not blinded. The syringes for the study were prepared by a non-blinded researcher and labeled only with patient identifiers to maintain group concealment. Blinding was maintained for patients, postoperative care providers, data collectors, and outcome assessors, thus ensuring that group allocation remained unknown to all involved except for the anesthesiologist performing the nerve block and the non-blinded researcher who prepared the syringes. Accordingly, this trial is described as a single-blind (patient-blinded) study. No formal assessment of blinding success (e.g., patient or assessor guessing of group allocation) was performed.

Anesthetic treatment

Upon arrival in the operating room, we established peripheral venous access. Standard monitoring included heart rate (HR), electrocardiography (ECG), pulse oximetry (peripheral capillary oxygen saturation, SpO₂), and invasive arterial blood pressure (IBP). Baseline regional cerebral oxygen saturation (rSO₂) was recorded by non-invasive cerebral oximetry. Prior to induction, the patient was placed in the lateral decubitus position and the T4 and T6 vertebral levels were identified. Under ultrasound guidance (6–15 MHz), a right-sided TPVB was performed with the probe in an oblique parasagittal orientation to visualize the transverse process and pleura and delineate the paravertebral space. Using an in-plane approach from the lateral probe edge, a 22-gauge needle was advanced into the paravertebral space; after negative aspiration for air and blood, the study drug was then injected. Patients in group A received 10 mL of LB (133 mg) at both T4 and T6, while patients in group B received RP (0.375%, 2 mg/kg) at the same levels. Inferior pleural displacement was used to confirm the spread of injectate. Fifteen minutes later, sensory blockade was reassessed by pinprick to verify the success of blockade.

Following cannulation of the left radial artery, general anesthesia was induced with midazolam (0.05 mg/kg), sufentanil (0.5–1.0 μg/kg), etomidate (0.2–0.3 mg/kg), and rocuronium (0.6–1.0 mg/kg). A double-lumen endotracheal tube was inserted, and mechanical ventilation was adjusted to maintain an end-tidal carbon dioxide partial pressure (PETCO₂) of 35–45 mmHg. The right hemithorax was elevated by approximately 30° and a right internal jugular venous catheter was inserted for central venous access. One-lung ventilation was applied during right thoracic and intracardiac maneuvers. SpO₂ was continuously monitored; if SpO₂ fell to <95%, we performed intermittent manual ventilation or brief two-lung ventilation. Anesthesia was maintained with a combination of sedatives, analgesics, and neuromuscular blocking agents, titrated to maintain a bispectral index (BIS) between 40 and 60. Intermittent boluses of sufentanil were administered intraoperatively as needed to maintain adequate analgesia and hemodynamic stability. Cerebral perfusion was adjusted to maintain rSO₂ >55% and ≥80% of baseline.

Following hemodynamic stabilization, CPB was established by cannulation of the femoral artery and bicaval venous drainage following systemic heparinization (3 mg/kg). Thoracoscopic access was obtained through the right thoracic ports as per institutional routine. The ASD was repaired on a beating heart using a pericardial patch with continuous 4-0 prolene sutures; tricuspid annuloplasty (the Kay technique) was performed when indicated. Following the completion of intracardiac repair and adequate de-airing, the right atriotomy was closed, CPB was discontinued, cannulas were removed sequentially, and heparin was neutralized with protamine. Vasoactive support was individualized according to intraoperative transesophageal echocardiography (TEE) and clinical status. Blood products (platelets, cryoprecipitate, and packed red blood cells) were administered as indicated to correct coagulopathy and anemia. Meticulous hemostasis was achieved, chest drains were inserted, and all incisions were closed in layers. Intraoperative TEE was used to confirm the absence of residual shunt and significant valvular regurgitation. The patient remained intubated and was transferred to the cardiac surgical intensive care unit (CTICU).

The postoperative analgesia protocol was applied in accordance with our institutional fast-track anesthesia pathway aimed at early extubation; patient-controlled analgesia (PCA) was not routinely applied. Upon arrival in the CTICU, pain was assessed using the numerical rating scale (NRS, 0–10) at predefined intervals and as clinically indicated. Rescue analgesia was administered by nursing staff as intravenous dezocine (5 mg per dose) when the patient reported a resting NRS score of 4 or higher. The time to the first request for rescue analgesia was recorded. Following transfer to the ward, breakthrough pain (NRS ≥4) was initially treated by oral acetaminophen/oxycodone combination tablets (5 mg/325 mg) as needed. If pain remained inadequately controlled after oral medication, intravenous dezocine (5 mg per dose) was administered as rescue therapy. All postoperative analgesic administrations (drug, dose, route, and timing) were prospectively documented and subsequently converted to morphine milligram equivalents (MME) for analysis.

For pain assessment and the handling of non-assessable observations, NRS pain scores (0–10) at rest and during coughing were scheduled at extubation (0 h) and at 24 h, 48 h, and 72 h postoperatively. In our CTICU and ward practice, pain is routinely documented every 6 h in the ICU and every 8–12 h on the general ward. Scheduled study assessments were aligned with this routine documentation, and an allowable time window of ±3 h around each scheduled time point was prespecified. For each scheduled time point, the value closest to the target time within the allowable window was used for analysis; observations that could not be verified within this window were treated as missing and were not imputed. The 0-h and 24-h time points both coincided with standard ICU nursing documentation nodes (with fixed records at postoperative hours 0, 6, 12, 18, and 24), which accounted for the complete data at these time points. In contrast, 48 h and 72 h were not fixed routine documentation nodes, and most patients had been transferred to the ward by these time points, where the longer documentation interval (8–12 h) made it less likely that a recorded value would fall within the prespecified ±3 h window. Missing rates were low and balanced between groups; detailed data availability is reported in Table S1.

For patients whose chest drain was removed within 72 h, the last NRS assessment within the 0–72 h period was preferentially recorded immediately prior to drain removal whenever feasible; thereafter, pain assessments continued at the prespecified postoperative time points according to the original schedule, with drain status noted.

Study outcomes

The primary outcome was the NRS (0–10) pain score during coughing at 24 h postoperatively. The secondary outcomes included: (I) NRS pain scores at rest and during coughing at extubation (0 h) and at 48 and 72 h postoperatively; (II) time to first request for rescue analgesia; (III) cumulative opioid consumption within the first 72 h, expressed as MME, the proportion of patients requiring no rescue analgesia within 72 h, and extended analgesia outcomes beyond 72 h (proportion requiring analgesics and MME on postoperative days 4–7, POD4–POD7); (IV) postoperative trajectories of cardiac biomarkers [high-sensitivity cardiac troponin T (hs-cTnT), cardiac troponin I (cTnI), N-terminal pro-B-type natriuretic peptide (NT-proBNP)] and inflammatory markers [procalcitonin (PCT), interleukin-6 (IL-6)]; (V) perioperative hemodynamics, including mean arterial pressure (MAP) and heart rate (HR) measured at prespecified time points: arrival at rest (T0), during TPVB (T1), immediately after tracheal intubation (T2), 1 min after intubation (T3), 5 min after skin incision (T4), and before extubation at the end of surgery (T5); (VI) postoperative recovery outcomes, including duration of mechanical ventilation, length of stay in the ICU, postoperative length of stay, and total length of stay in the hospital; and (VII) adverse events and complications within 30 days. Block-related complications were prespecified and defined as follows: (I) bleeding/paravertebral hematoma—new localized swelling or tenderness at the puncture site accompanied by a decline in hemoglobin ≥2 g/dL from baseline, or evidence of hematoma on ultrasound examination when clinically suspected; (II) local anesthetic systemic toxicity (LAST)—neurological symptoms (perioral numbness, tinnitus, confusion, seizures) and/or cardiovascular instability (arrhythmias, hypotension refractory to vasopressors) occurring within 24 h after block performance and temporally related to the block, assessed using the Naranjo adverse drug reaction probability scale; (III) pneumothorax—presence of air in the pleural space on postoperative chest X-ray performed when clinically indicated; and (IV) block failure—absence of sensory loss to pinprick in the corresponding dermatomes at 15 min after TPVB, or requirement of ≥3 rescue analgesic doses within the first 24 h postoperatively. All enrolled patients underwent continuous ECG monitoring from operating room admission through ICU stay. Patients were clinically assessed at regular intervals for neurological symptoms, local swelling, and hemodynamic changes. Postoperative chest X-ray was routinely performed within 24 h after surgery and repeated if clinically indicated. All potential adverse events were prospectively documented by the clinical team and independently adjudicated by two senior anesthesiologists who were blinded to group allocation; any disagreement was resolved by consensus with a third reviewer. Non-arrhythmic adverse events (subcutaneous emphysema, anemia, vomiting, and elevated liver enzymes) were summarized separately, while postoperative arrhythmias were evaluated in detail (overall incidence, severity, phenotype, timing, and causality assessment).

Sample size calculation

This trial focused on postoperative analgesia. The sample size calculation was based on the primary endpoint, the NRS pain score during coughing at 24 h postoperatively. Based on pilot data and/or previous reports, we prespecified a minimal clinically important difference (MCID) of 1.0 point with an assumed within-group standard deviation (SD) of 1.68. With a two-sided α of 0.05, 80% power, and 1:1 allocation, the required sample size was 45 participants per group. Allowing for an approximate 5% attrition rate (protocol deviations or missing outcome data in our institution), the target enrollment was 47 participants per group (total 94). Ultimately, 45 participants per group were included in our primary analysis, thus meeting the prespecified power for the primary endpoint. As a sensitivity check, we also assessed sample size adequacy using 72-h cumulative opioid consumption (MME) as an alternative driver endpoint (assumed between-group difference: 15 mg with an SD of 35 mg), which yielded a similar required sample size of approximately 44–46 participants per group, thus supporting the fact that the planned enrollment was sufficient for clinically meaningful analgesic outcomes.

Statistical analyses

Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 27.0 (IBM Corp., Armonk, NY, USA). All primary and secondary analyses were conducted on an intention-to-treat (ITT) basis, with patients analyzed according to their original randomized group assignment. Per-protocol analyses were performed as sensitivity analyses.

The distribution of continuous variables was assessed using the Shapiro-Wilk test. Normally distributed continuous variables are presented as mean ± SD and were compared between groups using the independent-samples t-test; within-group comparisons were performed using the paired t-test (test statistic, t). Non-normally distributed continuous variables are presented as median (interquartile range, IQR) and were compared between groups using the Mann-Whitney U test (test statistic, Z). Categorical variables are presented as counts (percentages) and were compared using the χ² test or Fisher’s exact test, as appropriate (test statistic, χ²). Fisher’s exact test was used when >20% of cells in a contingency table had an expected count <5.

Baseline demographic, clinical, and intraoperative characteristics were compared between groups using the tests described above. As shown in Tables 1,2, all baseline variables were well-balanced between the two groups (all P>0.05). In our prespecified statistical analysis plan, we predefined that if any baseline characteristic showed an imbalance with P<0.10, we would adjust for that covariate in multivariable regression models for the corresponding outcomes.

Table 1. Baseline demographic and clinical characteristics of the study population.

Variable	LB group (N=45)	RB group (N=45)	t/x²/z	P value
Gender			x²=0.048	0.83
Male	17 (51.52)	16 (48.48)
Female	28 (49.12)	29 (50.88)
Age (years)	36.6±11.214	38.49±14.958	t=−0.678	0.50
Height (cm)	165.99±8.34	165.33±9.369	t=0.351	0.73
Weight (kg)	64.21±11.855	62.11±11.16	t=0.865	0.39
ASA classification			x²=2.900	0.24
2	13 (54.17)	11 (45.83)
3	31 (51.67)	29 (48.33)
4	1 (16.67)	5 (83.33)
NYHA classification			x²=2.291	0.32
1	25 (56.82)	19 (43.18)
2	18 (41.86)	25 (58.14)
3	2 (66.67)	1 (33.33)
Ejection fraction (%)	64.4 (62.5, 66.1)	64.4 (61.9, 66.1)	z=−0.129	0.90
With comorbidities	8 (34.78)	15 (65.22)	x²=2.862	0.09
Intraoperative sufentanil (µg)	250 (200, 250)	200 (200, 250)	z=−0.483	0.63
Operative time (h)	3.25 (3, 3.5)	3.24 (3, 3.6)	z=−0.469	0.64

Data are presented as n (%), median (interquartile range) or mean ± standard deviation. ASA, American Society of Anesthesiologists; LB, liposomal bupivacaine; NYHA, New York Heart Association; RP, ropivacaine.

Table 2. Baseline laboratory values of the study population.

Variable	LB group (N=45)	RB group (N=45)	t/x²/z	P value
High-sensitivity troponin T (ng/L)	5.1 (3.9, 7.3)	5.1 (3.8, 7.5)	Z=−0.230	0.82
Troponin I (ng/L)	0.1 (0.1, 0.1)	0.1 (0.1, 0.1)	Z=−1.422	0.16
Blood glucose (mmol/L)	4.38 (4.18, 4.89)	4.77 (4.15, 5.06)	Z=−1.590	0.11
NT-proBNP (pg/mL)	220.6 (133.6, 285.8)	182.1 (106.5, 267.1)	Z=−1.239	0.22

Data are presented as median (interquartile range). LB, liposomal bupivacaine; NT-proBNP, N-terminal pro-B-type natriuretic peptide; RP, ropivacaine.

For outcomes assessed repeatedly across groups and time, we applied repeated-measures analysis of variance (ANOVA). When Mauchly’s test indicated a violation of sphericity (P<0.05), primary inference relied on multivariate tests (Pillai’s trace), with Greenhouse-Geisser-corrected univariate tests provided for reference. If the sphericity assumption was met, uncorrected univariate tests assuming sphericity were applied.

For scheduled postoperative pain assessments, an allowable time window of ±3 h around each prespecified time point was used. For each scheduled time point (including the primary endpoint at 24 h), the value closest to the target time within the allowable window was used for analysis; observations that could not be verified within the allowable window were treated as missing and were not imputed. The proportion and reasons for missing data are reported in the Results section and Table S1. All tests were two-sided, and P<0.05 was considered statistically significant.

Results

Clinical characteristics of patients

A total of 94 patients were randomized (1:1). After exclusions per protocol, 45 participants in each group (LB; RP) were included in the primary analysis set. Figure 1 shows a Consolidated Standards of Reporting Trials (CONSORT) diagram summarizing screening, randomization, and follow-up. Baseline demographic, clinical, and intraoperative variables were assessed for comparability between the two groups (Tables 1,2). The dataset included key demographic characteristics (e.g., age, sex, height, weight), clinical profiles (e.g., ASA physical status, NYHA functional class, left ventricular EF, comorbidities), relevant preoperative laboratory values (e.g., hs-cTnT, cTnI, NT-proBNP), and intraoperative details (e.g., intraoperative sufentanil consumption, operative time). As shown in Tables 1,2, all these variables were well-balanced at baseline, confirming the success of the randomization process.

Figure 1.

CONSORT flow diagram summarizing patient screening, randomization, and follow-up. CONSORT, Consolidated Standards of Reporting Trials.

NRS pain scores

Detailed statistical comparisons for NRS pain scores at all postoperative time points are provided in Table 3 (resting) and Table 4 (coughing). Between-group comparisons at each time point, as determined by the Mann-Whitney U test, were statistically significant. At extubation and at 24, 48, and 72 h postoperatively, median NRS scores (IQR)—both at rest and during coughing—were significantly lower in the LB group than in the RP group (Figure 2, Tables 3,4, P<0.001).

Table 3. Postoperative numerical rating scale (NRS) pain scores at rest at specified time points.

Status	Group	0 h	Postoperative time
			24 h	48 h	72 h
Resting	Bupivacaine (N=45)	2 (2, 3)	2 (1, 3)	2 (1, 3)	2 (1, 3)
	Ropivacaine (N=45)	3 (2, 4)	3 (2, 4)	3 (3, 4)	3 (2.5, 4)
	Z	−2.499	−4.147	−4.865	−4.556
	P value	0.01	<0.001	<0.001	<0.001

Data are presented as median (interquartile range).

Table 4. Postoperative numerical rating scale (NRS) pain scores during coughing at specified time points.

Status	Group	0 h	Postoperative time
			24 h	48 h	72 h
Movement	Bupivacaine (N=45)	3 (3, 4)	3 (3, 4)	3 (2, 4)	3 (2, 4)
	Ropivacaine (N=45)	4 (3, 5)	4 (3, 5)	5 (4, 5)	5 (3, 5)
	Z	−3.758	−3.567	−5.075	−4.524
	P value	<0.001	<0.001	<0.001	<0.001

Data are presented as median (interquartile range).

Figure 2.

Scores were compared between the LB and RP groups at extubation (0 h), 24, 48, and 72 h postoperatively. (A) Postoperative NRS pain scores (median with interquartile range) during coughing; (B) Postoperative NRS pain scores (median with interquartile range) at rest. *, P<0.05; ***, P<0.001. LB, liposomal bupivacaine; NRS, numerical rating scale; RP, ropivacaine.

The primary endpoint (24 h coughing NRS) and the 0 h assessments were complete for all 90 patients. For secondary time points at 48 and 72 h, missing NRS observations were infrequent and balanced between groups; detailed data availability by time point and assessment type is presented in Table S1.

Pain-related outcome measures

Total 72-h opioid consumption, measured as MME, was significantly lower in the LB group [median (IQR): 7.5 (0, 42) mg] than in the RP group [45 (30, 67.5) mg; Mann-Whitney test, P<0.001; Table 5, Figure 3A]. The time to first request for rescue analgesia was numerically longer, but not statistically significant, in the LB group (20.5 vs. 12.0 h; Mann-Whitney test, P=0.10; Table 5, Figure 3B). A significantly lower proportion of patients in the LB group required rescue analgesia at 72 h (χ² test, P=0.003), while a significantly higher proportion required no rescue analgesia within the first 72 h (Fisher’s exact test, P<0.001; Table 5, Figure 3C,3D). Beyond 72 h postoperatively, both the proportion of patients requiring analgesics and the median MME on postoperative days 4–7 were significantly lower in the LB group (χ² test and Mann-Whitney test, respectively; both P<0.001; Table 5).

Table 5. Postoperative recovery and analgesia-related outcomes.

Project	Bupivacaine (N=45)	Ropivacaine (N=45)	t/x²/z	P value
Postoperative mechanical ventilation time (h)	5 (3, 6)	6 (4, 7.5)	Z=−2.158	0.03
ICU stay time (h)	22 (18.5, 27)	24 (21, 35)	Z=−1.766	0.08
Postoperative hospital stay (days)	7 (6, 8)	7 (6, 8)	Z=−0.712	0.48
Total hospital stay (days)	12.27±3.048	12.47±3.341	t=−0.297	0.77
Whether analgesia was required (event status)			x²=8.598	0.003
No	21 (72.41)	8 (27.59)
Yes	24 (39.34)	37 (60.66)
First request for analgesia time (h)	20.5 (12, 28.5)	12 (8.5, 22)	Z=−1.662	0.10
Total amount of analgesic drugs within 72 h after surgery (MME) (mg)	7.5 (0, 42)	45 (30, 67.5)	Z=−3.547	<0.001
Still requires analgesia 72 h after surgery			x²=11.383	0.001
No	30 (68.18)	14 (31.82)
Yes	15 (32.61)	31 (67.39)
Average MME (mg) from D4 to D7	0 (0, 7.5)	7.5 (0, 15)	Z=−3.232	0.001

Data are presented as n (%), median (interquartile range) or mean ± standard deviation. ICU, intensive care unit; MME, morphine milligram equivalent.

Figure 3.

Analgesia-related outcomes within the first 72 h postoperatively. (A) Total 72-h opioid consumption (MME); (B) Time to first request for rescue analgesia; (C) Proportion of patients requiring rescue analgesia within 72 h; (D) Proportion of patients not requiring any rescue analgesia within 72 h. **, P<0.01; ***, P<0.001. MME, morphine milligram equivalents.

Changes in serum biomarkers

Serial measurements of cardiac and inflammatory biomarkers revealed significant changes over the postoperative period (Table 6). Cardiac biomarkers, including hs-cTnT, cTnI, and NT-proBNP, exhibited a characteristic pattern, peaking within the first 24–48 h after surgery before beginning a gradual decline by 72 h (Figure 4A-4C). Similarly, the inflammatory markers PCT and IL-6 showed significant time-dependent increases, reaching their highest levels in the early postoperative phase (Figure 4D,4E). Critically, despite these pronounced temporal changes, no statistically significant differences were observed between the LB and RP groups at any of the assessed time points (Table 6). Hemodynamic parameters (MAP and HR) measured from baseline (T0) through the end of surgery (T5) also demonstrated significant variation over time (Table 7). However, there were no consistent or clinically meaningful patterns of difference between the two treatment groups across these time points.

Table 6. Postoperative biomarker levels.

Biomarker	Time point	Bupivacaine (N=45)	Ropivacaine (N=45)	P value	95% CI
hs-cTnT (ng/L)	24 h	513.4±325.753a	533.3±315.78a	0.77	−154.306, 114.502
	48 h	534.2±420.917a	580.78±332.917a	0.56	−205.571, 112.398
	72 h	423.18±373.399a	451.72±336.941a	0.70	−177.539, 120.455
cTnI (ng/L)	24 h	1.09±1.334a	0.97±0.918a	0.62	−0.358, 0.601
	48 h	2.29±2.379a	1.85±1.781a	0.32	−0.434, 1.327
	72 h	1.56±2.028a	0.97±1.046a	0.08	−0.082, 1.276
BNP (pg/mL)	24 h	302.34±230.038a	278.24±370.82	0.71	−105.171, 153.380
	48 h	813.19±1139.28a	677.38±606.797a	0.48	−246.589, 518.203
	72 h	1164.1±1548.077a	972±837.095a	0.47	−329.268, 713.468
PCT (ng/mL)	24 h	1.2638±1.71046	1.1489±1.01404	0.70	−0.474, 0.704
	48 h	1.4954±1.98345	1.5344±1.55124a	0.92	−0.785, 0.707
	72 h	1.2062±1.67561	1.1647±1.06954	0.89	−0.547, 0.630
IL-6 (pg/mL)	24 h	59.4637±32.4018	68.729±27.19878	0.15	−21.798, 3.267
	48 h	75.3377±47.33764a	88.3089±39.70269a	0.16	−31.274, 5.332
	72 h	67.0325±67.48218	69.4093±32.58864	0.83	−24.577, 19.824

Data are presented as mean ± standard deviation unless otherwise specified. ^a, P<0.05 vs. baseline (preoperative) value. CI, for the between-group difference in means (LB vs. RP; LB − RP). CI, confidence intervals; cTnI, cardiac troponin I; BNP, B-type natriuretic peptide; hs-cTnT, high-sensitivity cardiac troponin T; IL-6, interleukin-6; LB, liposomal bupivacaine; PCT, procalcitonin; RP, ropivacaine.

Figure 4.

Postoperative trajectories of serum biomarkers. (A) High-sensitivity cardiac troponin T (hs-cTnT); (B) cardiac troponin I (cTnI); (C) N-terminal pro-B-type natriuretic peptide (NT-proBNP); (D) procalcitonin (PCT); (E) interleukin-6 (IL-6).

Table 7. Intraoperative hemodynamic parameters (heart rate and mean arterial pressure).

Time point	HR, beat/min				MAP, mmHg
	Bupivacaine (N=45)	Ropivacaine (N=45)	P value	95% CI	Bupivacaine (N=45)	Ropivacaine (N=45)	P value	95% CI
T0	81.24±10.347	81.69±10.852	0.84	−4.886, 3.997	91.59±9.883	90.43±11.754	0.61	−3.389, 5.709
T1	79.49±10.964	78.22±10.84	0.58	−3.301, 5.834	89.7±9.857	87.11±11.612	0.26	−6.458, 2.656
T2	80.62±11.067	79.04±14.57	0.56	−3.842, 6.998	83.67±9.19	85.57±12.334	0.41	−5.391, 3.082
T3	78.64±10.252	74.62±13.799	0.12	−1.070, 9.115	81.85±8.692	83.53±10.431	0.41	−4.886, 3.997
T4	70.91±10.144	70.67±11.052	0.91	−4.200, 4.689	78.61±8.711	79.76±11.341	0.59	−3.842, 6.998
T5	87.69±10.266	86.78±9.839	0.67	−3.302, 5.124	84.34±10.682	82.46±9.54	0.38	−4.200, 4.689

Data are presented as mean ± standard deviation unless otherwise specified. CI, for the between-group difference in means (LB vs. RP; LB − RP). CI, confidence intervals; HR, heart rate; LB, liposomal bupivacaine; MAP, mean arterial pressure; RP, ropivacaine.

Postoperative outcome measures

With regards to recovery outcomes, the duration of mechanical ventilation was significantly shorter in the LB group [median (IQR): 5 (3, 6) h] than in the RP group [6 (4, 7.5) h; P=0.03]. In contrast, length of stay in the ICU, postoperative length of stay, and total length of stay in hospital, were comparable between groups, with no statistically significant differences observed (Table 5). Non-arrhythmic adverse events within 30 days are summarized in Table 8. The incidences of subcutaneous emphysema, anemia, vomiting, and elevated liver enzymes were low and did not differ significantly between groups (all P>0.05). No prespecified block-related complications—including bleeding, paravertebral hematoma, LAST, pneumothorax, or block failure—were observed in either group. No serious adverse events related to the study drugs were identified.

Table 8. Adverse events.

Adverse event	Bupivacaine (N=45)	Ropivacaine (N=45)	Test statistic	P value
Subcutaneous emphysema	0 (0.0)	1 (2.2)	Fisher’s exact	>0.99
Anemia	4 (8.9)	6 (13.3)	χ²=0.450	0.50
Vomiting	2 (4.4)	3 (6.7)	Fisher’s exact	>0.99
Elevated liver enzymes	4 (8.9)	6 (13.3)	χ²=0.450	0.50
Any adverse event	10 (22.2)	16 (35.6)	χ²=1.947	0.1
No adverse event	35 (77.8)	29 (64.4)	χ²=1.947	0.163

Values are n (%). Fisher’s exact test was used when expected cell counts were <5. Arrhythmia events are reported separately in Tables 9,10.

Arrhythmia events

Postoperative arrhythmias within 30 days were comparable between groups [LB 6/45 (13.3%) vs. RP 7/45 (15.6%); P=0.76] (Table 9). Most events were mild (Clavien-Dindo grade I), and no arrhythmia classified as grade ≥III (requiring electrical cardioversion or life-threatening events) or higher in either group (Table 9). Phenotype and timing distributions were balanced (Tables 10,11); notably, during the 24–48 h postoperative window, considered clinically relevant for systemic exposure to LB, arrhythmia incidence was identical in both groups [1/45 (2.2%) each]. Naranjo causality assessment identified only one ‘probable’ case in each group, with the remaining events classified as ‘possible’ or ‘doubtful’ (Table 12).

Table 9. Postoperative arrhythmia within 30 days (overall incidence and severity).

Outcome	Bupivacaine (N=45)	Ropivacaine (N=45)	Test statistic	P value
Any arrhythmia	6 (13.3)	7 (15.6)	χ²=0.090	0.76
No arrhythmia	39 (86.7)	38 (84.4)	–	–
Severity (Clavien-Dindo)
Grade I (transient; no intervention)	4 (8.9)	5 (11.1)	χ²=0.124	0.73
Grade II (requiring medication)	2 (4.4)	2 (4.4)	Fisher’s exact	>0.99
Grade ≥ III (electrical cardioversion/life-threatening / death)	0 (0)	0 (0)	–	–

Values are n (%). Fisher’s exact test was used for sparse data. Grade ≥ III events were not observed in either group.

Table 10. Arrhythmia phenotype.

Type	Bupivacaine (N=45)	Ropivacaine (N=45)	P value
Atrial arrhythmias
Premature atrial contractions, occasional (<30/h)	2 (4.4)	2 (4.4)	>0.99
Premature atrial contractions, frequent (≥30/h)	0 (0)	1 (2.2)	>0.99
New-onset atrial fibrillation, self-terminating	1 (2.2)	0 (0)	>0.99
New-onset atrial fibrillation, requiring medication	1 (2.2)	1 (2.2)	>0.99
Atrial flutter	0 (0)	1 (2.2)	>0.99
Ventricular arrhythmias
Premature ventricular contractions, occasional	1 (2.2)	2 (4.4)	>0.99
Premature ventricular contractions, frequent	1 (2.2)	0 (0)	>0.99
Non-sustained VT (<30 s)	0 (0)	0 (0)	–
Sustained VT	0 (0)	0 (0)	–
Conduction abnormalities
Transient AV block (I degree)	1 (2.2)	0 (0)	>0.99
AV block (≥ II degree)	0 (0)	0 (0)	–

Values are n (%). AV, atrioventricular; VT, ventricular tachycardia.

Table 11. Timing of arrhythmia onset (attribution-relevant windows).

Time window	Clinical context	Bupivacaine (N=45)	Ropivacaine (N=45)	P value
Intraoperative–<12 h	Anesthesia/intubation/surgical stress	3 (6.7)	3 (6.7)	>0.99
12–<24 h	Early inflammatory response	1 (2.2)	2 (4.4)	>0.99
24–<48 h (key window)	Expected peak exposure of LB	1 (2.2)	1 (2.2)	>0.99
48–<72 h	Sustained-release phase	1 (2.2)	1 (2.2)	>0.99
≥72 h	After substantial elimination	0 (0)	0 (0)	–

Values are n (%). LB, liposomal bupivacaine.

Table 12. Naranjo causality assessment among patients with arrhythmia.

Naranjo category	Score range	Bupivacaine (N=45)	Ropivacaine (N=45)	P value
Definite	≥9	0 (0)	0 (0)	–
Probable	5–8	1 (16.7)	1 (14.3)	>0.99
Possible	1–4	2 (33.3)	3 (42.9)	>0.99
Doubtful	≤0	3 (50.0)	3 (42.9)	>0.99

Values are n (%). P values were calculated using Fisher’s exact test for sparse data.

Discussion

Postoperative pain after minimally invasive cardiac surgery remains clinically important, and optimal analgesic strategies are still evolving. Although TPVB provides effective unilateral chest wall analgesia, the limited duration of conventional local anesthetics often necessitates the use of supplemental opioids. LB, a sustained-release formulation, has demonstrated opioid-sparing effects in thoracic surgery (14,15), yet evidence in thoracoscopic cardiac procedures, particularly with integrated safety assessment, has been limited. In this randomized trial in adults undergoing ASD repair, we evaluated whether ultrasound-guided two-level TPVB with LB could improve analgesia without compromising hemodynamic or cardiovascular safety compared with RP.

Our main findings were that LB-TPVB was associated with improved early postoperative analgesia and opioid sparing. Specifically, LB achieved lower NRS pain scores from extubation to 72 h, reduced cumulative 72 h opioid consumption, and was associated with a shorter duration of mechanical ventilation, while postoperative trajectories of cardiac biomarkers (hs-cTnT, cTnI, NT-proBNP) and inflammatory markers (PCT, IL-6) were similar between groups. Notably, we also prespecified extended analgesia outcomes beyond 72 h (POD4–POD7), demonstrating the persistence of reduced analgesic requirements. Collectively, these results support the role of LB-TPVB as part of multimodal, opioid-sparing analgesia pathways for minimally invasive cardiac surgery.

Our findings are broadly consistent with previous studies reporting improved pain control and reduced opioid exposure with LB-based regional techniques in thoracic (16,17) and cardiac surgery (18-20). Differences in effect size across studies likely reflect heterogeneity in surgical populations, block technique and dosing, and background multimodal analgesia protocols (21-23). In our study, the use of a standardized right-sided two-level TPVB (T4/T6) with a total LB dose of 266 mg and a uniform thoracoscopic ASD cohort may have contributed to the observed magnitude and consistency of opioid-sparing effects. Importantly, in addition to analgesic benefits, our integrated assessment of hemodynamics and perioperative biomarkers provides additional reassurance regarding cardiovascular safety in this specific population.

Beyond the above safety findings, the clinical application of TPVB in cardiac surgery warrants specific consideration in the context of intraoperative systemic heparinization. Although thoracic epidural analgesia and paravertebral blocks have well-documented analgesic benefits in cardiac procedures, the theoretical risk of neuraxial hematoma in anticoagulated patients has limited their widespread use. In recent years, ultrasound-guided chest wall blocks, including TPVB and erector spinae plane block, have gained increasing attention as alternative regional anesthesia techniques in cardiac surgery. Available evidence suggests that TPVB is a reasonably safe option in patients receiving anticoagulant or antiplatelet therapy. A systematic review of 15 relevant articles demonstrated a low risk of bleeding with TPVB in anticoagulated patients, and a retrospective analysis of 138 patients undergoing cardiovascular surgery with coagulation abnormalities or on antithrombotic therapy reported no paravertebral or neuraxial hematoma events after TPVB. Furthermore, preoperative TPVB combined with general anesthesia has been shown to facilitate early extubation and reduce opioid requirements in cardiac surgery requiring full heparinization.

In our center, we selected TPVB over thoracic epidural analgesia based on several considerations. First, TPVB provides unilateral blockade with less hemodynamic impact and a lower incidence of complications such as nausea, vomiting, hypotension, and urinary retention compared with thoracic epidural analgesia. Second, the puncture site of TPVB is remote from the neuraxial structures, theoretically conferring a lower risk of hematoma than epidural blockade in the setting of systemic heparinization. Third, ultrasound guidance enables real-time visualization of the paravertebral space, transverse process, and pleura, thereby enhancing the precision and safety of the procedure. In the present study, all TPVB procedures were performed prior to systemic heparinization, with a sufficient time interval between puncture and heparin administration, and a single-injection technique was used rather than catheter placement to avoid additional bleeding risks associated with catheter retention and removal during postoperative anticoagulation. No block-related bleeding complications or hematomas occurred, supporting the safety profile of TPVB in this patient population. Nevertheless, the risk-benefit ratio should still be weighed on an individual patient basis, particularly in the context of anticoagulation.

With respect to safety, we separately evaluated non-arrhythmic adverse events and postoperative arrhythmias. Non-arrhythmic adverse events (such as subcutaneous emphysema, anemia, vomiting, and elevated liver enzymes) were infrequent and comparable between groups. Given the theoretical concern of local anesthetic-related cardiotoxicity, we specifically examined postoperative arrhythmias in detail. The overall 30-day incidence of arrhythmia did not differ between groups (13.3% vs. 15.6%), most events were mild and transient, and no severe arrhythmias requiring electrical cardioversion or resulting in life-threatening complications were observed. Notably, during the 24–48 h postoperative window, clinically relevant for systemic exposure to LB, the incidence of arrhythmia was identical in both groups. Naranjo causality assessment suggested that most events were unlikely to be directly attributable to the study drug, supporting the fact that when provided at the dose administered in this study, LB-TPVB did not increase the clinically meaningful risk of arrhythmia in this relatively young, low-risk cohort (24).

We selected plain RP as the comparator to reflect our institutional standard for TPVB and to isolate the effect of the sustained-release LB formulation, avoiding potential confounders associated with adjuncts such as dexamethasone. Our findings have practical implications for enhanced recovery pathways. Improved early analgesia and reduced opioid exposure may facilitate respiratory mechanics, mobilization, and earlier readiness for extubation, which is particularly relevant for thoracoscopic cardiac surgery. While the observed shortening of mechanical ventilation time is clinically encouraging, larger pragmatic studies are now needed to determine whether these benefits translate into reductions in the lengths of stay in the ICU or under varying institutional protocols (25,26).

Several limitations merit consideration. First, this was a single-center and randomized trial that was primarily powered for the primary analgesic endpoint; the sample size may have been insufficient to detect small between-group differences in infrequent adverse events or in biomarker trajectories. Second, external generalizability is constrained by the homogeneous surgical setting (thoracoscopic ASD repair) and the standardized unilateral two-level TPVB technique and dosing; thus, results may not extrapolate to open cardiac surgery, different block strategies, or substantially different multimodal analgesia regimens. Third, pain assessment is inherently subjective, and complete blinding of the anesthesiologist performing the block was not feasible due to the visible difference between the two drugs; however, we mitigated potential bias by ensuring that patients, postoperative care providers, data collectors, and outcome assessors remained blinded, and by using concealed allocation, standardized perioperative management, and prespecified outcome definitions. No formal assessment of blinding success was performed, which is a limitation Fourth, the study population was restricted to adults aged 18–55 years. This age range reflects the typical demographic profile of patients undergoing thoracoscopic ASD repair in our practice, as this minimally invasive approach is predominantly indicated for younger adults; therefore, caution is warranted when extrapolating these findings to older populations or higher-risk cardiac cohorts. Future multicenter studies with broader inclusion criteria, standardized co-interventions, and longer follow-up periods (including chronic postsurgical pain and functional recovery) will be valuable to confirm generalizability and the durability of benefit (27-30).

In summary, for adults undergoing thoracoscopic ASD repair, ultrasound-guided two-level TPVB with LB provided superior early postoperative analgesia and opioid sparing, with no signs of increased cardiovascular risk, including postoperative arrhythmias. These data support the consideration of LB-TPVB within multimodal analgesia strategies for minimally invasive cardiac surgery.

Conclusions

In adults undergoing thoracoscopic ASD repair, ultrasound-guided two-level TPVB with LB (266 mg) provided superior early postoperative analgesia, consistently lower median NRS scores from extubation to 72 h, together with reduced 72-h opioid consumption and a shorter duration of mechanical ventilation, compared with RP. Serial cardiac (hs-cTnT, cTnI, NT-proBNP) and inflammatory (PCT, IL-6) biomarkers, as well as perioperative hemodynamics and 30-day adverse events, did not differ significantly between groups, indicating a favorable safety profile. Extended outcomes showed that analgesic requirements beyond 72 h (POD4–POD7) were lower with LB, suggesting persistence of opioid-sparing effects. These findings provide a basis for optimizing multimodal analgesia strategies in minimally invasive cardiac surgery.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee of The First Affiliated Hospital of Nanjing Medical University (approval No. 2024-SR-188). Written informed consent was obtained from all patients.

Data Sharing Statement

Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1135/dss

jtd-18-9-1049-dss.pdf ^{(75.9KB, pdf)}

DOI: 10.21037/jtd-2026-1135