Secondary Logo

Journal Logo

Obstetric Anesthesiology: Original Clinical Research Report

The Analgesic Effect of Ultrasound-Guided Quadratus Lumborum Block After Cesarean Delivery: A Randomized Clinical Trial

Krohg, Anders MD*; Ullensvang, Kyrre MD*; Rosseland, Leiv Arne PhD, MD†,‡; Langesæter, Eldrid PhD, MD*; Sauter, Axel R. PhD, MD†,§

Author Information
doi: 10.1213/ANE.0000000000002648


  • Question: Does quadratus lumborum block improve pain treatment after cesarean delivery?
  • Findings: Opioid consumption in 24 hours was reduced, and the effective analgesic scores were significantly better in an active treatment group compared with the control group.
  • Meaning: Quadratus lumborum block reduces the postoperative opioid consumption and pain intensity as a part of a multimodal analgesic regimen.

Pain treatment after cesarean delivery is important for early mobilization and to enable the mother to care for the newborn child.

Ultrasound (US)-guided transversus abdominis plane (TAP) blocks are now widely used for postoperative analgesia after surgery with incision in the inferior part of the abdominal wall.1 However, standard TAP blocks provide inferior analgesia to neuraxial morphine and little benefit when added to a multimodal regimen that includes neuraxial morphine.2 The quadratus lumborum (QL) block, an US-guided abdominal wall block, was presented by Raphael Blanco as an abstract at the annual European Society of Regional Anaesthesia (ESRA) congress in 2007. A similar technique was later published as transversalis fascia plane block by Hebbard.3 In 2015, Blanco et al2 introduced a modified QL block technique with an injection site at the posterior border of the QL muscle. Theoretically, QL blocks might give better and longer lasting analgesia compared to the US-guided anterior TAP block due to a spread to the thoracic paravertebral space and sympathetic nerves in the thoracolumbar fascia.

The aim of our study was to evaluate the postcesarean analgesic effect of the US-guided lateral QL block.4 We tested the primary hypothesis that bilateral QL block with ropivacaine reduces cumulative ketobemidone consumption during the first 24 hours after QL block, when compared with bilateral placebo injection with saline. Secondary and exploratory analyses compared additional measures of analgesic quality and side effects between groups.


Ethical approval for this study was provided by the Committee for Medical Research Ethics, Region South East, Oslo, Norway, on June 25, 2013 (Ethical Committee Number 2013/1293-1; Document-id: 392062). The study was registered with a clinical trial registry ( identifier: NCT02036749). Good Clinical Practice guidelines were followed. The trial was conducted and reported according to the Consolidating Standards of Reporting Trials (CONSORT) 2010 statement.5 Study design was randomized, placebo-controlled, and double blind, with parallel-group comparison.

The study was conducted at the Department of Anesthesiology, Division of Emergency and Critical Care, Oslo University Hospital, Rikshospitalet, between March 2014 and April 2015. The study was funded by the Department of Anesthesiology. The Birth Clinic at Rikshospitalet is a tertiary care center, but the majority of the laboring women are healthy and reflect the general population in Southeast Norway. Forty healthy parturients scheduled for cesarean delivery via a Pfannenstiel incision gave written informed consent to participate in the study. Patients in American Society of Anesthesiologists classification II or stable American Society of Anesthesiologists III were included. Exclusion criteria were as follows: body mass index >32, chronic pain, neuropathy, age <18 or >45 years old, allergy to local anesthetic (LA), inadequate Norwegian language skills, or inability to operate a patient-controlled analgesia (PCA) pump.

All patients had spinal anesthesia with 10 mg of isobaric bupivacaine with 4 μg of sufentanil. Postoperatively both groups received a basic analgesic regimen of oral paracetamol 1 g and ibuprofen 400 mg, dosed together 4 times daily. A CADD-Legacy (Smith Medical MD, Inc, St Paul, MN) PCA pump with ketobemidone 1 mg/mL was programmed with a 1-mg demand dose, a lockout of 8 minutes, and a maximum dose of 7 mg/h. Patients were instructed to administer PCA boluses to achieve acceptable pain control not exceeding a pain level at rest of 3 on a numeric rating scale (NRS). Ketobemidone is regarded as equianalgesic to morphine.6,7

The QL blocks were performed in the postoperative care unit within the first hour after cesarean delivery, before the patients experienced any postoperative pain or pain during the QL block procedure. Routine monitoring included electrocardiogram, pulse oximetry, and noninvasive arterial blood pressure. Three consultant anesthesiologists experienced in US-guided regional anesthesia (A.K., K.U., and A.R.S.) performed the blocks.

QL Block Performance

A modified technique combining the techniques described by Blanco et al2 and Hebbard3 were used for the blocks. A SonoSite Edge ultrasound unit (SonoSite, Bothell, WA) with a HFL50, 6–15-MHz linear transducer was used. The patients were in the supine position. The side to be blocked was slightly elevated by pillows underneath the hip and shoulder. The transducer was placed in the transverse plane on the flank of the patient cranially to the iliac crest, at the level of the umbilicus (Supplemental Digital Content 1, Figure 1, The muscle layers of the abdominal wall were identified. The transducer was then moved posteriorly to visualize the aponeurosis of the transversus abdominis muscle. The pararenal fat and the QL muscle were imaged medial to the aponeurosis (Figure 1). A Vygon Echoplex 21G 100-mm needle (Vygon SA, Ecouen, France) was advanced in-plane under US guidance in anteroposterior direction through the muscle layers of the abdominal wall. The needle tip was advanced to the transversus aponeurosis and positioned superficial to the pararenal fat and lateral to the border of the QL muscle. Two milliliters of the study medicine (ropivacaine 2 mg/mL or saline 9 mg/mL) were injected to verify the needle position. If necessary, the needle was repositioned. On each side, volume of 0.4 mL/kg study solution with a maximum of 30 mL was then injected under repeated aspiration for every fifth milliliter injected. Hence, for the bilateral procedures in the active group, a total amount of 1.6 mg/kg ropivacaine with a maximum of 120 mg was used.

Figure 1.
Figure 1.:
QL block technique: an injection needle is advanced in-plane under US guidance in anteroposterior direction through the muscle layers of the abdominal wall to the transversus aponeurosis (white arrows). The needle tip is positioned superficial to the pararenal fat and lateral to the border of the QL muscle. EO indicates external oblique muscle; IO, internal oblique muscle; trans. abd., transversus abdominis muscle; LA, local anesthetic; QL, quadratus lumborum.

The exact time of each ketobemidone bolus was recorded in the internal memory of the PCA pumps and extracted to a computer with CADD-Sentry Medication Delivery Manager software package (Smith Medical MD, Inc, St Paul, MN) The patients registered pain severity at rest and when coughing 2, 4, 6, 12, 24, and 48 hours after the QL block using NRS scales (0 = no pain, 10 = worst pain imaginable). Nausea and fatigue were also registered by the patients using NRS scales at the same time intervals. In addition, the patients registered the first time standing on their feet and when they walked 5 m.

Outcome Variables

The primary outcome was ketobemidone consumption during the first 24 hours after QL block. Secondary outcomes were ketobemidone consumption during 12, 36, and 48 hours, pain intensity (PI) (NRS) scores, the integrated scores for pain, nausea, and fatigue, and time to mobilization. All other comparisons were exploratory.

Randomization and Blinding

Patients were randomly assigned to either QL block with ropivacaine 2 mg/mL (n = 20) or saline 9 mg/mL (placebo) (n = 20). A person not involved in the data collection or in patient care randomly assigned the patients in blocks of 8 or 6, into 2 groups of equal size using a list of random numbers,8 according to the Moses–Oakford algorithm.9 The block size (8 or 6) was also randomized using a list of random numbers. Block size and randomization codes were not revealed to the investigators until all measurements and calculations had been entered into the database for all patients. Each patient, the investigators, and all medical caregivers were blinded to group allocation.

One hour before the QL block procedure of an enrolled patient, a nurse, not otherwise participating in the study, opened a sealed opaque envelope containing group allocation. The nurse then filled 4 syringes (2 × 20 mL and 2 × 10 mL) labeled “study medicine” with the allocated solution, either ropivacaine 2 mg/mL or saline.

All data were entered in the database before entering the randomization codes. The principles for intention-to-treat analysis were followed.


All the statistical analyses were performed using Statistical Package for the Social Sciences (SPSS) version 22 (IBM Corp, Armonk, NY).

Histograms, box plots, Q-Q plots, and the Kolmogorov–Smirnov test were used to assess whether the variables and the residuals were normally distributed.

The primary hypothesis was that the main cumulative ketobemidone consumption after 24 hours (primary outcome variable) was less in the active treatment group compared with the control group. Data for this primary outcome variable ketobemidone consumption was estimated as log-normal distributed. Therefore t tests for independent samples were used with log-transformed data. The data are given as ratio of means and confidence intervals (CIs).

PI (NRS) and ketobemidone consumption were integrated in an effective analgesic score (EAS), both secondary outcomes of the study. EAS was calculated using the formula: (NRS + 1) × (1 – K/10), where K indicates doses (mg) of ketobemidone consumption 2 hours before registration time point of NRS.10 The original formula has been modified as NRS + 1 replaces PI to avoid a zero product in case of NRS = 0. The modified EAS score represents an integration of both pain and analgesic consumption at every time point, even if pain score is zero.

Oxycodone was converted into ketobemidone equivalents and added to cumulative ketobemidone consumption (5 mg of short-acting oxycodone was considered equivalent to 3.3 mg of intravenous ketobemidone distributed over 4 consecutive hours; 10 mg of long-acting oxycodone was considered equivalent to 6.7 mg of intravenous ketobemidone distributed over 12 consecutive hours).

Cumulative consumption of ketobemidone at 24 hours was the primary end point of our study. Given statistically significant group difference in the primary end point (level of significance P < .05), analyses of secondary end points were performed. Differences in EAS at rest (EAS rest) and during coughing (EAS evoked) at 2, 4, 6, 12, 24, and 48 hours were analyzed in the 2 treatment groups. We used the linear mixed model in SPSS to test if the development of EAS over time differed between the 2 treatment groups; that is, an interaction effect (type III test of fixed effects, F test). Baseline (time point 2 hours after study drug injection) was used as reference point. Level of significance in the secondary outcome analyses was P < .01 to account for multiple comparisons. The residuals were checked for normality, and the assumption was found satisfied.

For nonparametrical data including NRS scores for fatigue and nausea (secondary outcomes), 2-sample Wilcoxon rank sum tests were used.

Sample Size

Sample size calculation was performed using SPSS SamplePower (IBM Corp, Armonk, NY). Our primary hypothesis was that opioid consumption after 24 hours was less in the active treatment group. We considered a 40% reduction in cumulative opioid consumption as clinically relevant. Based on previous studies on opioid consumption, a standard deviation (SD) of 40% was estimated. A sample size of 17 patients in each group would give 80% power to detect a 40% reduction in opioid consumption, using t tests with α = .05. Estimating a mean cumulative ketobemidone consumption of 50 mg (SD, 20 mg), a reduction of 20 mg could be detected. Forty patients were included in the study to allow missing data or dropouts.


Table 1.
Table 1.:
Baseline Characteristics
Figure 2.
Figure 2.:
Study flowchart.

Sixty-one patients were considered eligible; of these 40 patients were randomly assigned and included in analysis (Figure 2). Baseline characteristics are shown in Table 1. The mean age showed a statistically significant difference of 2 years. Other baseline characteristics did not show significant differences. Three patients (1 in the active group and 2 in the control group) received oral oxycodone at 30, 32, and 33 hours, respectively, after the QL block.

Primary Outcome

Figure 3.
Figure 3.:
Cumulative ketobemidone consumption (mg) per hour for the patients treated with QL block (thick green line) with 95% percentiles (thin green lines) and for the patients in the control group (dashed blue thick line) with 95% percentiles (dashed blue thin lines). QL indicates quadratus lumborum.

Patients receiving the active QL block had lower cumulative ketobemidone consumption at 24 hours, compared with the control group (P = .04; ratio of means = 0.60; 95% CI, 0.37–0.97). Figure 3 shows the cumulative ketobemidone consumption per hour in both study groups.

Secondary Outcomes

At 12 hours, the cumulative ketobemidone consumption was significantly lower in the active group compared with the control group (P < .01; ratio of means = 0.52; 95% CI, 0.35–0.79). No statistically significant differences were found at 36 hours (P = .13; ratio of means = 0.71; 95% CI, 0.45–1.12) or at 48 hours (P = .20; ratio of means = 0.74; 95% CI, 0.47–1.18).

The analyses of the interactions between time (0–48 hours) and treatment showed statistically significant group differences in PI at rest (P < .01) and when the patients were coughing (P < .01) (Figure 4A, B). The interactions between time (0–48 hours) and treatment analyses showed statistically significant differences in EAS at rest (P < .01) and when the patients were coughing (P < .01) (Figure 4C, D). The PI at rest and during coughing and the estimated analgesic score at rest and during coughing are presented in Table 2 (mean difference and 95% CI).

Table 2.
Table 2.:
Pain Intensity and Estimated Analgesic Scores
Figure 4.
Figure 4.:
Pain intensity (NRS, 0–10) at rest (A) and during coughing (B) presented as mean (± 2 SEM), and EAS (= [NRS + 1] × [1 + K/10], where K = mg of ketobemidone the 2 h before registration) at rest (C) and during coughing (D) presented as mean (± 2 SEM). EAS indicates estimated analgesic score; NRS, numeric rating scale.

The patients’ experience of fatigue was similar in both groups. Median NRS scores for nausea were 0 at all measurement time points in both groups, and no significant difference could be found (Supplemental Digital Content 2, Table 1,

Patients in the active group were able to stand 14.5 hours (SD, 1.6) after QL block, while patients in the control group were able to stand after 13.5 hours (SD, 1.5) (P = .66; difference of mean = −1.00; 95% CI, −5.48 to 3.49). In the active group, patients were walking at least 5 m after 19.1 hours (SD, 1.8), while patients in the control group were walking 5 m or more after 17.2 hours (SD, 1.8) (P = .44; difference of mean = −1.97; 95% CI, −7.13 to 3.18).

There were no complications or adverse events.


Our randomized controlled trial demonstrated a 41% opioid-sparing effect of the QL block during the first 24 hours postoperatively after cesarean delivery, when administered with multimodal analgesia in the absence of neuraxial morphine.

In 2008, McDonnell et al11 described the TAP block to anesthetize the lower anterior abdominal wall after cesarean delivery. Patients receiving active TAP blocks had a morphine-sparing effect of 70% compared to the control group. The LA doses used by McDonnell et al11 were 150 mg of ropivacaine, compared to the maximum dose of 60 mg each side in our study. A higher dose of LA may increase the efficacy and/or the duration of the QL block and may also produce high serum concentrations of local anesthetic, leading to both a systemic analgesic effect12 and risk for systemic side effects and toxicity.13

In systematic review published by Abdallah et al1 in 2012, only 3 of 6 randomized controlled studies showed a reduction of morphine consumption when therapeutic TAP block is performed after cesarean delivery. In 2016, Champaneria et al14 published a meta-analysis of 20 studies that confirmed TAP blocks can be effective for acute pain relive after cesarean delivery. However, TAP blocks did not confer additional analgesia when intrathecal hydrophilic opioids were used.

Blanco et al2 published the first study investigating the analgesic effect of QL block after cesarean delivery, in which 0.2 mL/kg 0.125% bupivacaine was injected on the posterolateral border of the QL muscle. A significant reduction in the morphine consumption and visual analogue scores scores was found during 48 hours after QL block administration.2 In a second trial, this group compared QL block with TAP block and found a significantly superior effect of the QL block lasting from 6 to 48 hours.15

Two theories explain why injection at the posterolateral border of the QL muscle may achieve superior pain relief when compared with standard anterior TAP blocks. First, the QL block may facilitate spread of LA into the paravertebral space, theoretically prolonging the block and achieving visceral pain relief. Alternatively, Blanco et al2 posit that LA spread to a network of sympathetic nerves in the thoracolumbar fascia can explain the long-lasting analgesic effect.15

During pilot studies, our group of anesthetists was unable to consistently establish sensory blockades when injecting LA at the posterior border of the QL muscle. The QL block technique that was used in our study (Figure 1) is similar to the methods originally described by Raphael Blanco as an abstract at the annual ESRA congress in 2007 and the “fascia transversalis sheath block” published by Hebbard.3 When injecting between the transversus aponeurosis and the pararenal fat lateral to the QL muscle, a slow reabsorption of LA may lead to prolonged analgesic effect of a single injection.

In contrast to Hebbard,3 we did not try to identify the very thin fascia transversalis medially to the transversus aponeurosis by US. We therefore do not know if the LA was injected medial or lateral to the fascia transversalis. When dissecting the abdominal wall in cadavers, the anterior abdominal branches of L1 could be found within the fascia transversalis on the medial side of the transversus aponeurosis (unpublished data). When the aponeurosis turns into the transversus abdominis muscle, the nerves regularly enter the TAP plane (unpublished results). We expect that LA injected medial to the aponeurosis can reach the nerves from either side of the thin fascia transversalis layer.

We did not find a clinically relevant opioid-sparing effect of the QL block in the 24–48-hour period. Catheter-based QL blocks might prolong the analgesic effect.

Compared with the control group, the patients with QL block had a nonsignificant trend toward a delay in ambulation. Theoretically, a spread to the lumbar plexus can cause weakness of the psoas, iliacus, and quadriceps muscles as described in a case report by Wikner.16 However, no obvious effects of a lumbar plexus block have been observed in our patients.


US-guided needle placement is an operator-dependent technique. In our study, we wanted to ensure correct placements by having a second investigator confirm the correct needle placement as imaged by US. We did not assess block success by sensory testing in our study. The lack of a sensory test was chosen to preserve blinding of group allocation. The optimal dose of LA for QL blocks is not known, and our study cannot provide information on the adequate dose. A higher dose of ropivacaine may have improved and prolonged the analgesic effect. We excluded patients with body mass index >32 to ensure homogeneous patient groups. We therefore do not know if QL blocks are as effective in obese patients.

Postoperative ibuprofen and paracetamol were administered in addition to PCA ketobemidone. Despite this multimodal analgesic regime, which may have obscured some of the effects of the QL blockade, the study was adequately powered to identify an opioid-sparing effect in the first 24 hours. Patients were instructed to use their PCA pumps to achieve a NRS at 3 or less. This instruction may limit the value of examining the difference in PI between the groups. However, patient-reported PI was lower in the first 12 hours in the active group, and the analyses of EAS—integrating PI and PCA opioid consumption—supported the observation of clinically and statistically significant group differences.

The assessment of PI and analgesic consumption in clinical pain trials is challenging. The introduction of an estimated analgesic score enables the researchers to present a combined outcome measure of improved validity. Presenting PI, analgesic consumption, and the integration of both is informative, but the analyses of potential differences in treatment effects are still challenging and represent a limiting factor.

The study included pregnant women scheduled for planned cesarean delivery under standard anesthesia, and the postcesarean pain treatment was only slightly modified compared to the standard practice in the department. The sample is representative for the population at our birth clinic, and a majority of the screened patients were eligible for participation. However, generalizability is limited to those institutions and clinical contexts in which neuraxial morphine is not available.


Ropivacaine US-guided QL block reduced postoperative ketobemidone consumption and pain after cesarean delivery. The patients received small doses of ropivacaine, and further trials must be conducted to evaluate the ideal dose, volume, and injection site. Future studies should compare the analgesic efficacy with a multimodal regimen that includes neuraxial morphine; analgesic outcomes of QL block and more traditional US-guided TAP block should be directly compared.


The authors thank Audun Stubhaug (MD, PhD, Professor, Division of Emergencies and Critical Care, Department of Anesthesiology, Oslo University Hospital) for providing randomization and giving advice for the study.


Name: Anders Krohg, MD.

Contribution: This author helped design and conduct the study; collect, analyze, and interpret the data; and write the manuscript.

Conflicts of Interest: None.

Name: Kyrre Ullensvang, MD.

Contribution: This author helped design and conduct the study; interpret the data; and write the manuscript.

Conflicts of Interest: None.

Name: Leiv Arne Rosseland, PhD, MD.

Contribution: This author helped design the study; analyze and interpret the data; and write the manuscript.

Conflicts of Interest: None.

Name: Eldrid Langesæter, PhD, MD.

Contribution: This author helped design the study, interpret the data, and write the manuscript.

Conflicts of Interest: None.

Name: Axel R. Sauter, PhD, MD.

Contribution: This author helped design and conduct the study; collect, analyze, and interpret the data; and write the manuscript.

Conflicts of Interest: A. R. Sauter has advised B. Braun concerning regional anesthesia. He served as a consultant for Philips and received compensation for these services. He has received speaker fees by Siemens.

This manuscript was handled by: Jill M. Mhyre, MD.


1. Abdallah FW, Laffey JG, Halpern SH, Brull R. Duration of analgesic effectiveness after the posterior and lateral transversus abdominis plane block techniques for transverse lower abdominal incisions: a meta-analysis. Br J Anaesth. 2013;111:721–735.
2. Blanco R, Ansari T, Girgis E. Quadratus lumborum block for postoperative pain after caesarean section: a randomised controlled trial. Eur J Anaesthesiol. 2015;32:812–818.
3. Hebbard PD. Transversalis fascia plane block, a novel ultrasound-guided abdominal wall nerve block. Can J Anesth. 2009;56:618–620.
4. El-Boghdadly K, Elsharkawy H, Short A, Chin KJ. Quadratus lumborum block nomenclature and anatomical considerations. Reg Anesth Pain Med. 2016;41:548–549.
5. Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Ann Intern Med. 2010;152:726–732.
6. Ohqvist G, Hallin R, Gelinder S, Lang H, Samuelson S. A comparison between morphine, meperidine and ketobemidone in continuous intravenous infusion for postoperative relief. Acta Anaesthesiol Scand. 1991;35:44–48.
7. Jylli L, Lundeberg S, Langius-Eklöf A, Olsson GL. Comparison of the analgesic efficacy of ketobemidone and morphine for management of postoperative pain in children: a randomized, controlled study. Acta Anaesthesiol Scand. 2004;48:1256–1259.
8. Armitage P, Berry G, Matthews JNS. Statistical Methods in Medical Research2008.Malden, MA: John Wiley & Sons
9. Meinert CL. ClinicalTrials: Design, Conduct and Analysis1986.New York, NY: Oxford University Press
10. Dai F, Silverman DG, Chelly JE, Li J, Belfer I, Qin L. Integration of pain score and morphine consumption in analgesic clinical studies. J Pain. 2013;14:767–777.e8.
11. McDonnell JG, Curley G, Carney J. The analgesic efficacy of transversus abdominis plane block after cesarean delivery: a randomized controlled trial. Anesth Analg. 2008;106:186–91.
12. Kato N, Fujiwara Y, Harato M. Serum concentration of lidocaine after transversus abdominis plane block. J Anesth. 2009;23:298–300.
13. Griffiths JD, Le NV, Grant S, Bjorksten A, Hebbard P, Royse C. Symptomatic local anaesthetic toxicity and plasma ropivacaine concentrations after transversus abdominis plane block for caesarean section. Br J Anaesth. 2013;110:996–1000.
14. Champaneria R, Shah L, Wilson MJ, Daniels JP. Clinical effectiveness of transversus abdominis plane (TAP) blocks for pain relief after caesarean section: a meta-analysis. Int J Obstet Anesth. 2016;28:45–60.
15. Blanco R, Ansari T, Riad W, Shetty N. Quadratus lumborum block versus transversus abdominis plane block for postoperative pain after cesarean delivery: a randomized controlled trial. Reg Anesth Pain Med. 2016;41:757–762.
16. Wikner M. Unexpected motor weakness following quadratus lumborum block for gynaecological laparoscopy. Anaesthesia. 2017;72:230–232.

Supplemental Digital Content

Copyright © 2017 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the International Anesthesia Research Society.