Background Intrathecal morphine (ITM) provides effective postoperative analgesia in living donor hepatectomy but has significant adverse effects. Studies support the efficacy of erector spinae plane (ESP) blocks in laparoscopic abdominal surgery; we therefore hypothesized that they would provide non-inferior postoperative analgesia compared with ITM and reduce postoperative nausea/vomiting and pruritus. We conducted a randomized, controlled, non-inferiority trial to compare the analgesic efficacy of ITM and bilateral single-injection ESP blocks in laparoscopic donor hepatectomy.
Methods Fifty-four donors were randomized to receive bilateral ESP blocks with 20 mL 0.5% ropivacaine (n=27) or 400 µg ITM (n=27). Primary outcome was resting pain score 24 hours postoperatively measured on an 11-point numeric rating scale. The prespecified non-inferiority limit was 1. Incidences of postoperative nausea/vomiting and pruritus were assessed.
Results The mean treatment difference (ESP–ITM) in the primary outcome was 1.2 (95% CI 0.7 to 1.8). The 95% CI upper limit exceeded the non-inferiority limit. Opioid consumption and all other pain measurements were similar between groups up to 72 hours postoperatively. The ESP group had significantly lower incidences of postoperative vomiting (p=0.002) and pruritus (p<0.001).
Conclusions Bilateral single-injection ESP blocks resulted in higher resting pain scores 24 hours postoperatively compared with ITM and thus did not meet the study definition of non-inferiority. However, the pain intensity with ESP blocks was mild (mean pain scores <3/10) and associated with reduced incidence of postoperative vomiting and pruritus. It warrants further investigation as an analgesic option after laparoscopic living donor hepatectomy.
Trial registration number KCT0003191.
- regional anesthesia
- pain outcome measurement
- opioids, adverse effects
Statistics from Altmetric.com
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.
Living donor liver transplantation is an established treatment option for patients with end-stage liver disease.1 Although advances in surgical techniques and perioperative anesthetic management have rendered donor hepatectomy a relatively safe procedure,2 improving safety and quality of recovery for living donors remains an important concern. Liver donors are more vulnerable to postoperative pain than patients undergoing hepatectomy for tumor resection,3 and optimizing analgesia is therefore a priority.
At our institution, we have evaluated several strategies to manage postoperative pain in living liver donors, including epidural analgesia,4 intrathecal morphine (ITM) with intravenous patient-controlled analgesia (PCA),2 intravenous PCA alone,2 and continuous wound infusion of ropivacaine.5 A single 400 µg dose of ITM is currently used as the standard protocol because it is relatively easy to administer while providing effective analgesia for up to 30 hours after surgery.2 However, limitations include frequent postoperative nausea and vomiting, pruritus, and importantly, the potential for delayed respiratory depression.2 We therefore sought to identify an alternative to ITM that would be as effective but produce fewer adverse effects.
The erector spinae plane (ESP) block is a regional anesthetic technique that provides abdominal analgesia when performed at the lower thoracic vertebral level (T7 or T8).6–8 In addition, case report data suggest that it may be a viable analgesic modality for living donors undergoing open hepatectomy.9 However, the analgesic efficacy of ESP blocks in liver surgery has not been systematically investigated nor has it been compared with more established regional analgesic modalities such as ITM. We conducted a randomized, non-inferiority clinical trial and hypothesized that bilateral single-injection ESP blocks performed at the T8 level would provide non-inferior postoperative analgesia compared with ITM, while reducing opioid-related adverse effects in living donors undergoing laparoscopic right hepatectomy.
This trial was prospectively registered on the Clinical Trial Registry of Korea (principal investigator: JSK; registration date: 9 July 2018). To ensure donor safety, all donors were evaluated by our multidisciplinary transplant team and met standard tests and refined criteria for liver donation. We enrolled adult patients with American Society of Anesthesiologists (ASA) Physical Status classification I–II, scheduled for elective laparoscopic right hepatectomy between July 2018 and March 2019 at Samsung Medical Center, Seoul, Korea. We excluded patients who refused to participate in the study or had contraindications to peripheral nerve block or allergy to local anesthetics.
Randomization and blinding
A hospital staff member who was not otherwise involved in the study performed computer-generated block randomization in a 1:1 ratio to allocate enrolled patients to an ITM group (n=27) or an ESP group (n=27). The allocation of each patient was concealed in an opaque envelope, which was only opened on the day of surgery by the anesthesiologist performing the blocks. All blocks were performed by one of two investigators proficient in regional anesthesia (JSK or RAK). These investigators were not involved in outcome assessment or further patient management. All other personnel involved in intraoperative and postoperative care of the patients (anesthesiologists, surgeons, and nursing staff), as well as assessors of study outcomes, were blinded to group allocation.
ITM and ESP block administration
Patients received no premedication. Preoperative ITM or ESP blocks were performed in the operating room prior to induction of general anesthesia. After applying standard ASA monitoring and supplemental oxygen, intravenous midazolam (1–2 mg) was administered as an anxiolytic. The patient was turned into the right lateral position. In the ITM group, an intrathecal injection of morphine 400 µg (0.4 mL of morphine sulfate 1 mg/mL diluted in 1.6 mL cerebrospinal fluid aspirated at the time of dural puncture) was administered at the L3–L4 or L4–L5 level with a 27 G Whitacre spinal needle. In the ESP group, ultrasound-guided bilateral single-injection ESP blocks at the level of T8 transverse process were performed as described by Chin et al 6 using the following systematic three-step protocol. Dosing guidelines (volume and type of local anesthetic, and site of ESP block) were based on a combination of existing evidence and institutional experience.5 6
The T8 vertebral level was identified via ultrasound by counting down from the first rib, and marked on the skin. A high-frequency linear-array transducer was placed in a transverse orientation to identify the T8 transverse process and then rotated into a longitudinal parasagittal orientation over its tip (figure 1A).
The left ESP block was performed first. A 22 G, 100 mm echogenic needle was inserted in-plane in a cranial to caudal direction to contact the T8 transverse process. Correct needle tip position was confirmed by a linear pattern of injectate spread lifting the erector spinae muscle off the transverse process (figure 1B); 20 mL 0.5% ropivacaine with 5 µg/mL epinephrine was injected at this point.
The right ESP block was performed in an identical manner without repositioning the patient. The total amount of administered local anesthetics was 40 mL.
Immediately following completion of ITM administration or ESP blocks, patients were returned to the supine position and general anesthesia was induced with intravenous pentothal sodium (3–6 mg/kg), vecuronium (0.1 mg/kg), and remifentanil (0.05–0.1 µg/kg/min) followed by tracheal intubation. Anesthesia was maintained with isoflurane in a 1:1 oxygen:air mixture and intravenous remifentanil (0.01–0.05 µg/kg/min) infusion. Isoflurane concentration and remifentanil dose were adjusted to achieve a bispectral index of 40–60 and to maintain mean arterial blood pressure and heart rate within 20% of the preinduction value. All patients received intravenous meperidine 25 mg 30 min before the end of surgery for additional analgesia.
A narrative overview of the pure laparoscopic donor right hepatectomy surgical technique used at our institution has been previously published.10 Briefly, five trocar ports were placed as follows (figure 2): one 12 mm port receiving a 30° optical device at the umbilicus; two 12 mm operative trocars, one at the subcostal margin in the right midaxillary line and one in the midline between the umbilicus and xiphoid process; and two 5 mm trocars for instrumental assistance, placed along the subcostal margin in the left mid-clavicular line and in the subxiphoid region. Ultrasound observation was followed by parenchymal sectioning using an ultrasonic dissector (Sonicision; Medtronic, Minneapolis, MN, USA) and a cavitron ultrasonic surgical aspirator (CUSA Excel; Integra LifeSciences, Plainsboro, NJ, USA). The major hepatic veins were saved for reconstruction. After completion of parenchymal dissection, the remnant bile duct stump was sutured. The hepatic graft was removed through a 12–14 cm Pfannenstiel incision in the suprapubic area. Surgeons did not perform local anesthetic wound infiltration of any of the surgical incisions.
After surgery, donors were transferred to the postanesthesia care unit (PACU), where they remained until they met PACU discharge criteria. Postoperative supplemental analgesia was standardized. Pain severity was measured using an 11-point (0=no pain and 10=worst pain) numeric rating scale (NRS). Intravenous PCA with fentanyl was initiated in PACU at the first complaint of pain (NRS≥1/10) and programmed to deliver a background infusion of 15 µg/hour (1 mL/hour) with 15 µg bolus (1 mL) dose and 15 min lockout interval. If patients presented with breakthrough pain (NRS>4/10) despite intravenous PCA administration, intravenous morphine 5 mg was administered. If this was ineffective after 15 min, intravenous fentanyl 25–50 µg was administered. Postoperative nausea or vomiting was treated with intravenous metoclopramide 10 mg. A blinded nurse recorded all PACU data, including opioid consumption, presence or absence of nausea or vomiting, and highest, lowest, and average pain scores during PACU stay.
On arrival to the ward, all donors were managed with a multimodal analgesic regimen comprising scheduled intravenous ibuprofen 400 mg (Huons, Seoul, South Korea) every 6 hours on postoperative day (POD) 0–1 (maximum six doses), intravenous PCA fentanyl, and intravenous rescue opioids as needed. Intravenous ibuprofen was replaced by one tablet of oral cetamadol (acetaminophen 325 mg/tramadol HCl 37.5 mg; Il-dong Pharmacy, Seoul, South Korea) every 8 hours on POD 2. The IV PCA was continued until POD 5. In donors with breakthrough pain (NRS>4/10), intravenous hydromorphone 1 mg was administered every 4 hours as needed. All donors received 2 L/min of oxygen via nasal cannula and were continuously monitored with pulse oximetry until at least the morning of POD 1. Hospital discharge was determined by the surgical team according to institutional guidelines. Discharge criteria included the ability to void after urinary catheter removal, oral intake of solid food, and adequate pain control without need for intravenous opioids or antiemetic medication.
The primary outcome was the resting pain score at 24 hours postoperatively. Secondary outcomes included highest, lowest, and average pain scores in PACU; resting pain scores at 8, 48, and 72 hours postoperatively; cumulative opioid consumption in PACU, and at 8, 24, 48, and 72 hours postoperatively; the incidence of postoperative nausea, vomiting, or pruritus within 24 hours after surgery; donor satisfaction with pain relief at 24 hours postoperatively; and sleep disturbance on the first night. The blinded assessor visited each donor at 8, 24, 48, and 72 hours postoperatively and tabulated pain scores, postoperative opioid consumption (converted into intravenous morphine equivalents),11 overall satisfaction with analgesia at 24 hours, and quality of sleep on the first night using a Likert scale (1=very dissatisfied, 2=dissatisfied, 3=neutral, 4=satisfied, and 5=very satisfied). Block performance time was measured as follows: spinal anesthesia: from time of palpation to completion of intrathecal injection, ESP block: from time of transducer placement on skin to completion of the second injection. Procedure-related complications including hypotension (defined as mean arterial blood pressure <60 mm Hg requiring vasopressor therapy or intravenous fluid boluses ≥500 mL), postdural puncture headache, and respiratory depression within the first 24 hours postoperatively (oxygen saturation <90% or respiratory rate <8 breaths per minute) were assessed.
Sample size calculations
The sample size was calculated based on the primary endpoint according to the non-inferiority hypothesis. The predetermined non-inferiority limit (δ) was set to 1 point on the 11-point NRS scale. Based on preliminary analyses (unpublished), an SD of 1.2 was assumed for the NRS distribution. With α=0.05 and power of 90%, 25 patients were required in each group. Assuming a 10% dropout rate, we decided to enroll 27 patients per group. The primary outcome was analyzed according to the non-inferiority approach.12 The non-inferiority hypothesis for the primary outcome was tested using the one-sided t-test (null hypothesis that the difference in pain scores was ≥1 point vs the alternative hypothesis that the difference in pain scores was <1 point) under 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 by treatment, is presented in relation to the predefined non-inferiority limit and null effect.
After determining the normality of data distribution using the Shapiro-Wilk test, continuous variables were analyzed using the t-test or Mann-Whitney U test as appropriate. Parametric and non-parametric data were reported as mean±SD and median (IQRs), respectively. Categorical variables were analyzed using the χ2 test or Fisher’s exact test. Bonferroni correction was used for multiple comparisons. Data analyses were performed using SPSS software (V.25.0; SPSS). For all analyses, p<0.05 was taken to indicate significance. Two-sided tests were used, except for the one-sided t-test of non-inferiority.
From July 2018 to March 2019, fifty-six donors scheduled for pure laparoscopic right hepatectomy were assessed for eligibility and two donors who did not meet the inclusion criteria were excluded (figure 3). All enrolled donors (n=54) were randomly assigned to one of the two groups (n=27 each) and completed the study. Baseline patient and surgical characteristics of the two groups were comparable except for a higher proportion of male patients in the ITM group (p>0.05) (table 1). Administration of ITM and bilateral ESP blocks were performed successfully in all patients without any immediate procedural complications. Surgery was successful in all donors and there were no cases of conversion to open surgery.
The mean pain score at rest at 24 hours postoperatively was 1.3±1.1 for the ITM group (95% CI 0.8 to 1.7) and 2.5±1.0 for the ESP group (95% CI 2.1 to 2.9). The mean treatment difference (ESP–ITM) in the pain score at 24 hours between the two groups was 1.2 (95% CI 0.7 to 1.8). As the upper limit of the 95% CI (1.8) was higher than the prespecified non-inferiority margin (δ=1), non-inferiority was not established (figure 4).
The highest and average pain scores at PACU as well as the resting pain scores at 8 and 24 hours postoperatively were significantly higher in the ESP group than in the ITM group (table 2). There was no significant difference in resting pain scores between the two groups at 48 and 72 hours postoperatively. Intraoperative remifentanil consumption was similar between the groups (table 2). Cumulative opioid consumption at PACU was significantly lower in the ITM group than the ESP group, but there was no difference between the two groups at 24, 48, and 72 hours postoperatively. Both groups reported similar quality of sleep on the first postoperative night and satisfaction with analgesia at 24 hours postoperatively (table 2).
The incidence of nausea within 24 hours was similar between groups (ITM vs ESP; 59.3% vs 44.4%, p=0.414), whereas the incidences of vomiting and pruritus were significantly lower in the ESP group than the ITM group (vomiting: 2 (7%) vs 13 (48%) patients, p=0.002; pruritus: 2 (7%) vs 25 (93%) patients, p<0.001) (table 3). One donor in the ITM group requested discontinuation of the intravenous PCA on the first night of surgery due to severe nausea and vomiting. There was no significant difference in the incidence of perioperative hypotension between groups. There were no block-related complications, including postdural puncture headache or respiratory depression, or other major perioperative complications observed in either group (table 3). Median hospital stay was 7 (7–9) days in the ITM group and 8 (7–9) days in the ESP group (p=0.327).
In this study, ESP blocks provided inferior postoperative analgesia compared with ITM after laparoscopic donor right hepatectomy, as measured by resting pain scores at 24 hours postoperatively. In addition, the pain scores reported in PACU and at 8 hours postoperatively were significantly higher in patients who received ESP block. Nevertheless, pain was still well controlled in the ESP block group during the first 24 hours, as evidenced by mean pain scores <3/10, opioid consumption that was comparable to the ITM group, and similar degrees of reported patient satisfaction and sleep quality on the first postoperative night. Notably, the incidence of vomiting and pruritus was significantly lower during the first 24 hours in the ESP block group. There was no difference in pain scores or opioid consumption between 24 and 72 hours postoperatively.
Living liver donors are healthy individuals who undergo major surgery without any direct therapeutic benefit from the procedure.1 Severe postoperative pain can cause substantial physical and psychological distress in donors and impair functional recovery; effective postoperative pain control should therefore be an integral part of donor management. At our institution, we have sequentially evaluated several modalities of postoperative pain control in living liver donors.2 4 5 In 2007, we implemented an analgesic strategy of a single preoperative injection of 400 µg ITM combined with intravenous PCA, which proved to be an effective method of providing long-lasting analgesia (mean duration of analgesia at rest: up to 30 hours).2 We noted several advantages of ITM over thoracic epidural analgesia, including the fact that it could be performed at the lumbar level, and was more reliable with a higher success rate. In addition, it reduced the risk of epidural-related complications, including perioperative hypotension, postdural puncture headache and epidural hematoma. Consequently, ITM plus intravenous PCA was established as the routine postoperative analgesic regimen following donor hepatectomy in our institution. However, based on 12 years of institutional experience encompassing approximately 600 living liver donors, we have noted drawbacks associated with ITM analgesia; most notably, high incidences of postoperative nausea, vomiting, and pruritus. Furthermore, there is a risk for delayed respiratory depression which requires close monitoring during the first 24 hours postoperatively. We therefore sought to compare the analgesic efficacy and side effect profile of ITM with an alternative strategy incorporating bilateral single-injection ESP blocks.
The ESP block has emerged as a potentially valuable regional anesthesia technique in various types of abdominal surgery, including laparoscopic ventral hernia repair,7 bariatric surgery,6 open epigastric hernia repair,8 laparoscopic cholecystectomy,13 and open living donor hepatectomy.9 Its analgesic effect is ascribed to local anesthetic spread into the paravertebral space through intertransverse connective tissues,14 15 thereby blocking the ventral and dorsal rami of spinal nerves (somatic pain) and the rami communicantes that contain sympathetic nerve fibers (visceral pain).14–18 Although this mechanism has been questioned by cadaveric studies that argue instead for a predominant effect on lateral cutaneous branches of intercostal nerves,19–21 there is both radiological and clinical evidence that paravertebral or epidural spread can occur in actual patients and it is thus likely that this accounts for at least part of the clinical effect.6 15–18 It may however be a question of how consistently and to what extent this occurs. Furthermore, the volume of injection may influence physical spread, the mass of local anesthetic that reaches neural targets in the paravertebral space, and subsequent block intensity and duration.
In this study, we did not set out to systematically collect data on the qualitative nature of postoperative pain. However, a post hoc informal review of the pain assessment datasheet used routinely by PACU nursing staff indicated that while patients in the ESP group had little to no somatic incisional pain at emergence, they sometimes described deep abdominal pain or discomfort which led to demands for additional opioid analgesia. We postulate that this may reflect visceral pain, due in part to the prolonged (4–5 hour) pneumoperitoneum,22 and this in turn may explain the discrepancy in analgesic efficacy when compared with studies of ESP block in shorter laparoscopic procedures13 or surgery where the pain is primarily somatic in nature.8 It may be helpful for future studies to collect data on pain characteristics other than severity. Another consideration is the limited duration of analgesic effect obtained from single-injection ESP block. Currently available data suggest that this ranges from 5 to 12 hours for a 20 mL injection of 0.25%–0.375% bupivacaine.8 13 Given that the average duration of anesthesia in our study was 4 hours, the ESP block may have been responsible for mainly intraoperative rather than postoperative analgesia, which would account for the similar intraoperative remifentanil consumption in both ESP and ITM groups. Continuous ESP blockade with insertion of bilateral catheters, although technically more complex, may potentially address this issue.23
Although ITM provided adequate analgesia in the first 24 hours postoperatively, it did result in higher incidences of vomiting and pruritus. These are common adverse effects of intrathecal opioids that can be as distressing to patients as postoperative pain, and should factor into the choice of analgesic strategy. There is also the small but significant risk of delayed respiratory depression, especially at higher doses of ITM. One meta-analysis reported an incidence of 8.8% with ITM doses of 300 mcg or more,24 and this correlates with an incidence of 7.6% in our own patient database (unpublished data). The ESP block, on the other hand, has a highly favorable adverse effect profile. There is little to no risk of hemodynamic instability25 and more importantly, the absence of major blood vessels and neural structures in the immediate vicinity minimizes concerns regarding the development of a clinically significant hematoma in patients who are at risk of complications associated with postoperative coagulopathy. This includes living liver donors4 and liver transplant recipients as highlighted in a recent report.23 In this regard, the ESP block can also be used as a postoperative rescue block for patients experiencing significant pain despite other analgesic measures.26
Probably the most significant major risk associated with the ESP block is local anesthetic systemic toxicity due to systemic absorption. This is fortunately a rare occurrence (<0.3%) and symptoms are usually minor.25 Nevertheless, we routinely add epinephrine to the local anesthetic and observe maximum recommended local anesthetic dose limits to minimize the risk.27
Our study did have several limitations. First, the timing of surgery and the presence of postoperative wound dressings as well as an abdominal binder prevented us from effectively assessing sensory loss to confirm successful ESP blockade. All regional anesthesia techniques have an inherent failure rate (this was estimated at 6.5% for the ESP block in one series25) and this may have contributed to the observed results. In addition, ESP blocks at the T8 level may not have reliably covered the Pfannenstiel incision in the lower abdomen, and may be another reason why the non-inferiority limit was not achieved. We chose however to perform the ESP block at T8 transverse process as upper abdominal coverage was deemed most important with regard to sites of surgical pain, impact on patient comfort, and respiratory function. A possible strategy for future study is the utilization of bilevel ESP injections for greater abdominal coverage.28 Second, we did not have a control group of systemic analgesia alone, which might have better elucidated the true analgesic efficacy of ESP block. However, as the current standard of care in our institution was the use of ITM, the use of a control group was deemed inappropriate by our institutional ethics board. Third, there is a risk of bias from lack of patient blinding; however, logistical and safety considerations prevented us from using sham or placebo blocks to achieve this. Fourth, our results may not be generalizable to other techniques of donor hepatectomy (eg, laparoscopic assisted or open), or to other combinations of multimodal analgesics. We employed a relatively conservative regimen that comprised non-steroidal anti-inflammatory drugs for 24–36 hours, followed by low-dose oral acetaminophen and tramadol. The combination of acetaminophen and tramadol has been recommended as a safe and reasonable option in liver transplantation29 and is currently our standard of care for postoperative analgesia in this setting. Finally, our results may similarly only be applicable to the specific single-injection ESP block technique used. It is conceivable that the use of different volumes and doses of local anesthetics, the use of adjuncts such as dexamethasone or dexmedetomidine for prolonging the duration of analgesia,30 or a continuous catheter technique, will produce different outcomes.
Preoperative bilateral single-injection ESP blocks resulted in higher pain scores compared with ITM in the first 24 hours after surgery in donors undergoing pure laparoscopic donor right hepatectomy and thus did not meet criteria for non-inferiority. However, postoperative analgesia in donors who received ESP blocks appeared adequate and was associated with reduced risk of postoperative vomiting and pruritus. The ESP block may thus warrant further investigation as an analgesic option for living liver donors in the context of a multimodal analgesic regimen.
RK and KJC are joint first authors.
Contributors RAK, KJC, and JSK: planning, conception and design of the study, analyzed the data, interpretation of data, wrote the manuscript, and revised the manuscript. MSG, GSK, and SJC: planning and conducting the study, reporting and acquisition of data, analyzed the data, and gave critical comments. JMK and GSC: performed all surgeries, postoperative management, interpretation of data, and gave critical comments.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Not required.
Ethics approval This study received approval from the Samsung Medical Center Research Ethics Board (SMC 2018-05-117-001). Written informed consent was obtained from all participants.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available upon reasonable request.