Article Text

Download PDFPDF
Comparison of postoperative analgesic effects of posterior quadratus lumborum block and intrathecal morphine in laparoscopic donor hepatectomy: a prospective randomized non-inferiority clinical trial
  1. Seungwon Lee1,
  2. Ryung A Kang1,
  3. Gaab Soo Kim1,
  4. Mi Sook Gwak1,
  5. Gyu-Seong Choi2,
  6. Jong Man Kim2 and
  7. Justin Sangwook Ko1
  1. 1Anesthesiology and Pain Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Gang-nam gu, Seoul, Korea
  2. 2Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Gang-nam gu, Seoul, Korea
  1. Correspondence to Professor Justin Sangwook Ko, Anesthesiology and Pain Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Gang-nam gu, Seoul, Korea (the Republic of); justinswko{at}


Background Posterior quadratus lumborum block (QLB) and intrathecal morphine are accepted analgesic strategies in laparoscopic liver resection, but their effects have not been compared after laparoscopic donor hepatectomy. This study was planned to perform this comparison.

Methods Fifty-six donors were randomized to receive bilateral posterior (QLB2, 20 mL of 0.375% ropivacaine on each side, 150 mg total) or preoperative injection of 0.4 mg morphine sulfate intrathecally. Primary outcome was resting pain score at 24 hour postsurgery. Secondary outcomes included cumulative opioid consumption and recovery parameters. Serial plasma ropivacaine concentrations were measured in QLB group. Only the outcome assessor was properly blinded.

Results Mean resting pain score at 24-hour postsurgery was 4.19±1.66 in QLB group (n=27) and 3.07±1.41 in intrathecal morphine group (n=27, p=0.04). Mean difference (QLB group-intrathecal morphine group) was 1.11 (95% CI 0.27 to 1.95), and the upper limit of CI was higher than prespecified non-inferiority margin (δ=1), indicating an inferior effect of QLB. Cumulative opioid consumption was significantly higher in QLB group at 24 hours and 48 hours postsurgery. QLB group exhibited lower incidence of postoperative pruritus at all time points, and there were no differences in other recovery outcomes. All measured ropivacaine concentrations were below the threshold for systemic toxicity (4.3 µg/mL).

Conclusions Bilateral posterior QLB elicited higher resting pain scores at 24-hour after laparoscopic donor hepatectomy than intrathecal morphine and did not meet the definition of non-inferiority.

Trial registration number KCT0005360.

  • regional anesthesia
  • analgesia
  • pain, postoperative

Data availability statement

Data are available on reasonable request.

Statistics from


  • Intrathecal morphine injection has a strong analgesic effect after surgery, but it has several adverse effects.

  • Quadratus lumborum block reduces postoperative pain in various abdominal surgeries.


  • The analgesia provided by quadratus lumborum block compared with intrathecal morphine injection did not meet non-inferiority criteria.

  • Even in the context of extensive hepatectomy, quadratus lumborum block can be safely performed.


  • Intrathecal morphine may be preferred analgesic modality in laparoscopic donor hepatectomy.

  • Further studies between quadratus lumborum block with an adjuvant (ie, dexamethasone) and intrathecal morphine injection are needed.


Laparoscopic donor hepatectomy is increasingly used for living donor liver transplantation.1 Although this surgical improvement meets the cosmetic and functional demands of the donor, severe pain after surgery is still frequently observed in living donors.2 Postoperative pain after laparoscopic donor hepatectomy is multifactorial due to port placement, gas insufflation, and abdominal dissection.3–5 Therefore, effective postoperative pain management is mandated that provides both somatic and visceral pain control of the entire abdomen.6

Our institution has evaluated several analgesic strategies for living liver donors, including thoracic epidural anesthesia,7 local anesthetic wound infiltration,8 intrathecal morphine (ITM) injection combined with intravenous patient-controlled analgesia (PCA),9 and bilateral single injection or continuous erector spinae plane block.4 5 Of these, intrathecal injection of 0.4 mg of morphine sulfate is the current standard at our institution. Although ITM provides prolonged analgesic effects for up to 30 hours,9 the high incidences of pruritus and postoperative nausea and vomiting4 5 have led us to try to find an alternative analgesic method with a comparable analgesic effect but with fewer complications.

The quadratus lumborum block (QLB) is a fascial plane block in which a local anesthetic is injected adjacent to the quadratus lumborum muscle.10 Of the three approaches of QLB, anterior11 or posterior6 QLB provides abdominal analgesia in patients undergoing hepatectomy. However, the analgesic effects of posterior QLB have not been compared with those of ITM in living liver donors. We chose to perform posterior QLB (QLB2) because the local anesthetics injected into the lumbar interfascial triangle,10 where the fascial plane between the posterior aspect of the quadratus lumborum muscle and the middle layer of the thoracolumbar fascia, was expected to reach the thoracic paravertebral space.12 Therefore, we performed this randomized, non-inferiority clinical trial to investigate whether bilateral posterior QLB would provide non-inferior postoperative abdominal analgesia to ITM. In addition, we measured the sequential changes in arterial ropivacaine concentrations in the QLB group to confirm the safety of the local anesthetic doses in living liver donors.


Study participants

The study was conducted at the Samsung Medical Center from September 2020 to May 2021. All participants provided written informed consent. Fifty-six adult donors with an American Society of Anesthesiologists (ASA) Physical Status classification I–II who underwent elective laparoscopic right hepatectomy were enrolled. Patients with a coronary disease, congestive heart disease, arrhythmia or cardiomyopathy; pre-existing cerebrovascular disease or chronic pain; contraindications to block procedures (including severe impaired coagulation profile or infection); or allergy to local anesthetics were excluded from the study.

Randomization, blinding and study intervention

The randomization table was created with a 1:1 allocation ratio in a computer-generated block randomization (Research Randomizer V.4.0,, and subjects were assigned to either the QLB group (n=28) or the ITM group (n=28). All blocks were performed by either one of two investigators who had already performed multiple nerve block procedures (RAK and JSK).4–6 Outcome assessment was performed by an investigator who was blinded to group allocation. Standard ASA monitoring and supplemental oxygen were applied to all participants in the operating room. Study subjects were not blinded because ITM was performed before the induction of general anesthesia and QLB was conducted after the induction of general anesthesia. ITM was performed in awake subjects due to the potential for neurological complications when performing the ITM procedure in general anesthetized patients. Details of both procedures are provided in online supplemental appendix 1.

Supplemental material

Perioperative pain management

All surgeries were conducted by two surgeons (JMK or G-SC).1 Surgical pleth index (SPI) was used to monitor intraoperative nociception-antinociception balance but was not used to adjust the analgesic dose.13 Intravenous pethidine (0.5 mg/kg) was injected at the time of the skin closure according to the institutional protocol.

All donors were transferred to the postanesthetic care unit (PACU) after surgery and stayed until the PACU discharge criteria were satisfied.14 Intravenous PCA with fentanyl programmed to deliver a 15 µg bolus (1 mL) dose and a lock-out interval of 15 min was administered in the PACU. Additionally to intravenous PCA, if donors expressed breakthrough pain (Numeric Rating Scale (NRS) >4), intravenous pethidine (25 mg) was injected as first-line therapy. If this did not provide sufficient analgesia after 15 min, intravenous fentanyl (25–50 µg) was injected as second-line therapy.

At surgical ward, all participants received 400 mg of intravenous ibuprofen (Huons, Seoul, Korea) every 6 hours, and when oral administration became possible at postoperative day (POD) 2, intravenous ibuprofen was replaced by an oral Mypol capsule (codeine phosphate 10 mg, ibuprofen 200 mg, acetaminophen 250 mg; SungWon Adcock Pharm, Seoul, Korea) every 8 hours. If donors expressed breakthrough pain (NRS >4), intravenous hydromorphone (2 mg) was injected.

Total arterial plasma ropivacaine concentration analysis

Blood sampling for the analysis of total ropivacaine concentration was performed only in the QLB group through the right radial artery at 30, 45, 60 min, and 4 hours after the end of QLB.6 15 16 The study’s initial design included the measurement of ropivacaine concentrations at baseline (preblock). However, the protocol was revised to include blood samples at 45 min instead of at baseline to better elucidate the maximum concentration of ropivacaine. Blood samples were kept at 4°C and transported to the Sample Logistics Central laboratory, where they were centrifuged at 2500 RPM for 5 min at 4°C and then transferred to a −70°C freezer for storage before shipping to SCL Healthcare facilities (Seoul, Korea) for batch processing. Total ropivacaine concentrations were determined using a quantitative assay based on the principle of liquid chromatography-tandem mass spectrometry (API 4000 LC-MS/MS system; Applied Biosystems, Foster City, California, USA). The lower limit of detection was 5 ng/mL, and the internal standard was Ropivacaine-D7 (Santa Cruz Biotechnology, Dallas, Texas, USA).

Outcome measures

The outcome assessor visited the patients at a predetermined time to record the total intravenous PCA dose, the resting pain scores, and the presence of nausea or vomiting, and pruritus. The primary outcome was the resting pain score at 24 hours postsurgery. Secondary outcomes included: (1) Resting pain scores during PACU stay (at admission, highest recorded, and at discharge) and at 48 and 72 hours postsurgery; (2) Cumulative opioid consumption during PACU stay and at 24, 48, and 72 hours postsurgery (rescue opioids administered up to that point were combined with intravenous fentanyl PCA and converted to intravenous morphine equivalents); (3) Presence of postoperative nausea or vomiting, or pruritus during 24 hours postsurgery; (4) Patient satisfaction with sleep on the POD 0 night and pain relief at 24 hours postsurgery was determined via a Likert scale (1=very dissatisfied, 2=dissatisfied, 3=neutral, 4=satisfied, and 5=very satisfied); (5) Quality of Recovery (QoR)−15 questionnaire scores (Korean version) before surgery and 48 hours postsurgery17; (6) Presence of block-related complications including postoperative hypotension (mean blood pressure below 65 mm Hg), postdural puncture headache, and respiratory depression within the 24 hours postsurgery (defined as oxygen saturation below 90% or a respiratory rate below eight breaths per minute); (7) Time to first flatus; and (8) Total plasma ropivacaine concentration in the QLB group.

Sample size calculations and data analyses

The sample size was calculated based on the primary objective according to the non-inferiority criterion.18 The predetermined non-inferiority margin (δ) was set to one point on the eleven-point NRS.4 Based on a previous study,4 an SD of 1.2 was assumed for the pain score distribution. As a result of the calculation, 25 patients per group were estimated to be required to achieve a significance level (α) of 0.05, and a power of 90%. Assuming a drop-out rate of 10%, we decided to enroll 28 donors per group.

Standardized differences were used to make balance comparisons. If the absolute value of the standardized difference for a factor was greater than 0.52 Embedded Image, it was considered as evidence of imbalance for that factor.19 Continuous variables are presented as mean (SD) or median (IQR), and categorical variables are presented as number (%). The normality of the data distribution was determined using the Shapiro-Wilk test. Continuous variables were compared using t-test or Wilcoxon rank sum test. Categorical variables were compared using the χ2 test or Fisher’s exact test. The primary outcome was analyzed according to the non-inferiority approach.18 20 The non-inferiority hypothesis was tested using a one-sided t-test at a significance level of 2.5%. The two-sided 95% CI, the upper limit of which was equivalent to the upper limit of the one-sided 97.5% CI of the mean difference in pain scores, is presented in relation to the predefined non-inferiority limit.20 All secondary outcomes were analyzed using a two-sided test, and a p value below 0.05 was considered significant. Bonferroni correction was used for multiple comparisons for all outcomes with repeated measures (pain scores, cumulative opioid consumptions, and QoR-15 scores). The peak plasma concentration of ropivacaine (Cmax) and time to Cmax (Tmax) were obtained directly based on the observed concentration-time data. Data analysis was performed using SPSS software (V.27.0; SPSS) and using R V.4.1.3 (R Foundation for Statistical Computing, Vienna, Austria).


Study participants

Between September 2020 and May 2021, 58 donors were assessed for eligibility (figure 1). Of these, two donors were excluded prior to randomization because they declined to participate. Fifty-six donors were randomly assigned to one of two groups (n=28 in each group). Two donors (one in each group) underwent a second operation due to surgical bleeding within 24 hours postsurgery, and they were excluded from the outcome assessment. Fifty-four donors completed the study (n=27 in each group) and were included in the data analysis. Baseline donor characteristics and surgical environment were comparable between the two groups (table 1). Both QLB and ITM were performed successfully in all donors, and there were no immediate complications.

Table 1

Patient characteristics discriminated between the quadratus lumborum block (QLB) and the intrathecal morphine (ITM) groups

Figure 1

CONSORT flow diagram of participants in the study. CONSORT, Consolidated Standards of Reporting Trials; ITM, intrathecal morphine; QLB, Quadratus lumborum block.

Primary outcome and pain scores

The mean resting pain score at 24 hours postsurgery was 4.19±1.66 in the QLB group and 3.07±1.41 in the ITM group, respectively (p=0.04). The mean differences (QLB-ITM) in pain scores were 1.11 (95% CI 0.27 to 1.95), and the upper limit of the 95% CI was higher than the prespecified non-inferiority margin (δ=1) (figure 2), indicating that non-inferiority had not been established. The mean resting pain score in the PACU at discharge was significantly higher in the QLB group compared with the ITM group (p=0.03) (figure 3). Resting pain scores at 48 and 72 hours postsurgery were similar between the two groups.

Figure 2

Non-inferiority diagram of Numeric Rating Scale pain score differences in QLB group and ITM group at 24 hours postsurgery. The solid line indicates a non-inferiority margin (δ) of 1. The black square indicates mean pain score difference, and the error bars indicate 95% CIs of the difference between groups. ITM, intrathecal morphine; QLB, Quadratus lumborum block.

Figure 3

Violin plot of the NRS pain scores at rest during the 72 hours after surgery. Dashed lines indicate median, and dot lines indicate IQR. The individual p values result from a Bonferroni correction for multiple comparisons. A p<0.05 was considered statistically significant. PACU, postanesthetic care unit.

Cumulative opioid consumption and recovery outcomes

Opioid consumption was significantly lower in the ITM group than the QLB group at 24 and 48 hours postsurgery (p<0.001, p<0.01, respectively) (table 2). There were no significant differences between the two groups in intraoperative outcomes, sleep satisfaction on the first night, QoR-15 scores at 48 hours postsurgery, time to first flatus, and length of hospital stay (table 2). The number of patients satisfied with pain relief at 24 hours was significantly higher in the ITM group compared with the QLB group (88.9% vs 59.3%, p=0.03). Pruritus was more common in the ITM group than in the QLB group (88.9% vs 3.7%, p<0.001) (table 3). Other postoperative complications were comparable between the two groups.

Table 2

Perioperative clinical outcomes discriminated between the quadratus lumborum block (QLB) and the intrathecal morphine (ITM) groups

Table 3

Postoperative complications discriminated between the quadratus lumborum block (QLB) and the intrathecal morphine (ITM) groups

Total arterial plasma ropivacaine concentrations

A full panel of ropivacaine assays was not available for eight of the donors (29.6%) from the QLB group because the second panel of ropivacaine assays (at 45 min) was not collected owing to changes in the blood sampling protocol. Therefore, the sequential changes in total arterial plasma ropivacaine concentrations in the QLB group shown in figure 4 were obtained based on data from 19 donors (70.4%). The Tmax value was 47±10 min, the Cmax value was 1.2±0.4 µg/mL, and the area under the curve was 194±46 µg/min/mL. None of the recorded values were close to the previously demonstrated toxic concentration of 4.3 µg/mL.15 No clinical symptoms or adverse events related to systemic toxicity were recognized.

Figure 4

Total arterial plasma ropivacaine concentration versus time in the QLB group. T1, T2, T3, and T4 indicate 30 min, 45 min, 60 min and 4 hours after ropivacaine administration, respectively. Boxes represent the medians and IQR, and whiskers indicate the ranges. QLB, quadratus lumborum block.


In this prospective randomized non-inferiority trial, we found that QLB exhibited an inferior performance compared with ITM in terms of resting pain score at 24 hours postsurgery in living liver donors after laparoscopic right hepatectomy. Significantly higher opioid consumption was observed in the QLB group compared with the ITM group at 24 and 48 hours postsurgery. The number of patients satisfied with pain relief at 24 hours was significantly lower in the QLB group compared with the ITM group. Although the incidence of postoperative pruritus was significantly lower in the QLB group than in the ITM group, other recovery outcomes were similar between the two groups. Consequently, our study did not demonstrate the non-inferiority of the analgesic effect of QLB over the ITM.

These findings are in line with a recent procedure-specific postoperative pain management review that observed no evidence to support the use of QLB after open liver resection.21 Recent studies on cesarean section22 and open pancreatic surgery23 also found that QLB alone provided inferior pain control than ITM. These reports suggested that QLB might provide additional analgesic benefits when combined with ITM as part of a multimodal analgesic regimen, especially during the early postoperative period. However, some studies have supported the effectiveness of QLB in the provision of analgesic effects during liver surgery. Anterior QLB reduced opioid consumption and pain scores in laparoscopic hepatectomy compared with oxycodone intravenous PCA.11 A previous study conducted at our institution demonstrated that posterior QLB provided postoperative analgesia in laparoscopic hepatectomy compared with erector spinae plane block.6 These conflicting results regarding the analgesic effects of QLB might be ascribed to differences in the type of surgery and QLB approaches. Different QLB approaches can lead to differences in block effects10 and block duration.24 Duration of the block was significantly shorter in the posterior QLB approach (mean 12 hours) when compared with the anterior QLB approach (mean 20 hours).24 Moreover, the optimal QLB approach in hepatectomy has not been established. It is conceivable that the use of adjuncts such as dexamethasone25 or epinephrine26 to prolong analgesia duration, or a continuous catheter technique, will lead to different outcomes. Another reason might be the differences in the mechanisms of action between the central neuraxial and the fascial plane blocks. In fact, in previous studies, QLB demonstrated inferior postoperative analgesia than ITM,22 but showed relatively effective postoperative analgesia compared with no-block11 or fascial plane block (eg, erector spinae plane block).6

Morphine sulfate is directly injected into the cerebrospinal fluid and can cover a broader range of somatic dermatomes more reliably during liver surgery23 compared with sensory dermatomes after QLB (T4–T12 or L1).10 15 Additionally, ITM can also provide visceral analgesia by interacting with μ-and κ-opioid receptors in the spinal cord.23 As a result, ITM can effectively provide both somatic and visceral analgesia. On the other hand, the analgesic effect of QLB is due to the spread of local anesthetic injected into the thoracic paravertebral space between the transversalis fascia and the quadratus lumborum muscle.10 Recent cadaveric and contrast studies have not consistently confirmed this paravertebral spread.27 28 Therefore, the analgesic effect becomes difficult to predict because the distinction of the fascial layers is usually not accurate with current ultrasound technology, and it is also difficult to confirm the volume of local anesthetic spread to the desired location.10 22 29 These reasons could explain why QLB offered inferior postoperative pain control effect to ITM.

One of the most concerning complications with nerve block is systemic toxicity due to the absorption of local anesthetics into the systemic circulation. Additionally, more extensive hepatectomy can lead to further hemostasis and drug metabolism disruptions.30 In our study, all measurements of plasma ropivacaine concentration in the QLB group were lower than the toxic threshold value (4.3 µg/mL). Based on this result, 150 mg of ropivacaine seems to be an appropriate dose that can be safely used for laparoscopic donor hepatectomy. In addition, although we did not set out to systemically collect data on the impact of QLB on muscle strength, a post hoc review of medical records did not find patients with muscle weakness after blocks and falls.

This study has several limitations. First, there was no control group that received systemic analgesia alone or a placebo, which would have further revealed the benefit rendered by ITM or QLB. However, our institutional ethics board objected to the use of placebos or sham blocks. Second, only the outcome assessor was properly blinded. Since the ITM procedure was performed before and the QLB procedure after the induction of general anesthesia, the subjects might have become aware of the group they had been assigned to. Third, the dose of ITM in this study was higher than that in a previous study in which 0.1–0.2 mg ITM was used.21 However, our dosage followed institutional standards,4 8 9 and we wanted to compare the analgesic effects of QLB against them specifically. Fourth, the predetermined non-inferiority margin of 1 might be small. A larger delta may yield different results. Finally, we did not evaluate block success by checking for loss of skin sensation using a pinprick test or a cold alcohol swab, which could affect our results. However, since the mean duration of anesthesia in our study was 3.4 hours, the QLB may have contributed predominantly to intraoperative rather than postoperative analgesia, as reflected by comparable intraoperative remifentanil consumption and SPI score in both QLB and ITM groups.


Bilateral posterior QLB resulted in higher resting pain scores than ITM in the first 24 hours after the surgery and thus did not meet the criteria for non-inferiority. Further studies are warranted to illustrate the benefits rendered by QLB in this particular surgical procedure.

Data availability statement

Data are available on reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study was approved by the Samsung Medical Center Research Ethics Board (SMC 2020-04-222-003). Participants gave informed consent to participate in the study before taking part. The approval for change of blood sampling protocol was obtained by the Samsung Medical Center Research Ethics Board (SMC 2020-04-222-007).


We thank the statistics team of the Samsung Medical Center for their advice on the statistical analysis and reporting of the data.



  • SL and RAK are joint first authors.

  • SL and RAK contributed equally.

  • Contributors SL, RAK, and JSK were involved in the planning, conception, and design of the study, analyzed and interpreted the data, and wrote and revised the manuscript. MSG and GSK were involved in the planning and conducting the study, in the reporting and acquisition of data, analyzed the data, and provided critical comments. JMK and G-SC performed all surgeries and postoperative management, interpreted the data, and provided critical comments. JSK is responsible for the overall content as a guarantor.

  • 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.

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

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

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.