Article Text

Download PDFPDF

Quadratus lumborum block: an imaging study of three approaches
  1. Angela Lucia Balocco1,
  2. Ana M López1,
  3. Cedric Kesteloot2,
  4. Jean-Louis Horn3,
  5. Jean-François Brichant4,
  6. Catherine Vandepitte1,
  7. Admir Hadzic1 and
  8. Philippe Gautier2
  1. 1 Anesthesiology, Ziekenhuis Oost-Limburg, Genk, Limburg, Belgium
  2. 2 Anesthesiology, Clinique Sainte-Anne Saint-Remi, Brussels, Belgium
  3. 3 Anesthesiology, Stanford University, Stanford, California, USA
  4. 4 Anesthesiology and Intensive Care Medicine, Centre Hospitalier Universitaire de Liège, Liege, Belgium
  1. Correspondence to Dr Philippe Gautier, Anesthesiology, Clinique Sainte-Anne Saint-Remi, Brussels, 1070 Brussels, Belgium; p.gautier{at}skynet.be

Abstract

Background and objectives Different injection techniques for the quadratus lumborum (QL) block have been described. Data in human cadavers suggest that the transverse oblique paramedian (TOP) QL3 may reach the thoracic paravertebral space more consistently than the QL1 and QL2. However, the distribution of injectate in cadavers may differ from that in patients. Hence, we assessed the distribution of the injectate after the QL1, QL2, and TOP QL3 techniques in patients.

Materials and methods Thirty-four patients scheduled for abdominal surgery received QL blocks postoperatively; 26 patients received bilateral and 8 patients received unilateral blocks. Block injections were randomly allocated to QL1, QL2, or TOP QL3 techniques (20 blocks per each technique). The injections consisted of 18 mL of ropivacaine 0.375% with 2 mL of radiopaque contrast, injected lateral or posterior to the QL muscle for the QL1 and QL2 techniques, respectively. For the TOP QL3, the injection was into the plane between the QL and psoas muscles, proximal to the L2 transverse process. Two reviewers, blinded to the allocation, reviewed three-dimensional computed tomography (3D-CT) images to assess the distribution of injectate.

Results and discussion The QL1 block spread in the transversus abdominis plane (TAP), QL2 in the TAP, and posterior aspect of the QL muscle, whereas TOP QL3 spread consistently in the anterior aspect of the QL muscle with occasional spread to the lumbar and thoracic paravertebral areas.

Conclusions The spread of injectate after QL1, QL2, and QL3 blocks, resulted in different distribution patterns, primarily in the area of injection. The TOP QL3 did not result in consistent interfascial spread toward the thoracic paravertebral space.

  • regional anesthesia
  • pain
  • postoperative
  • ultrasonography

Statistics from Altmetric.com

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.

Introduction

The use of ultrasound in regional anesthesia allows identification of the interfascial planes through which the peripheral nerves innervating the abdominal wall travel.1 The quadratus lumborum (QL) block was developed from the transversus abdominis plane (TAP) block to accomplish more extensive postsurgical analgesia. Since the initial description of the QL block,2 several technique variations have been devised to try to extend the spread of local anesthetic either in a more medial and cranial direction to reach the thoracic paravertebral space, in a medial and caudal direction toward the lumbar plexus, or extend laterally to reach the TAP.

The QL1 technique was originally described as an injection at the lateral edge of the QL muscle, superficial to the transversalis fascia.3 Subsequently, the QL2 technique was introduced as a modification in which the injection was made at the posterior edge of the QL muscle, behind the middle layer of the thoracolumbar fascia.4 Finally, the anterior, transmuscular QL block (QL3 technique) was described, consisting of the injection into the fascial plane between the QL and psoas muscle at the L4 vertebral level.5 The QL3 technique is thought to result in local anesthetic spread cranially under the arcuate ligaments, potentially reaching the thoracolumbar paravertebral space and sympathetic chain. Additional modifications of the QL3 technique, the paramedian sagittal oblique, and transverse oblique paramedian (TOP) were introduced6 7 in an attempt to achieve a more cranial spread of local anesthetic. In TOP QL3 the transducer is positioned with a TOP orientation lateral to the L2 spinous process.

Defining the spread of the local anesthetic with different QL blocks is challenging because most studies were conducted in human cadavers, while very few have been performed in patients and or volunteers.8 The differences in the methodology in these studies (ie, QL techniques, volume, dye or local anesthetic, and methods to determine the dispersion of dye or contrast-containing injectate) make it difficult to deduce clinical relevance. In this study, we compared the spread of radiocontrast-containing local anesthetic among three QL block techniques in 34 patients using 3D-CT assessment.

Materials and methods

After Ethics Committee approval of the University Hospital of Liege, registration (EudraCT 2017-004221-32; https://www.clinicaltrialsregister.eu/ctr-search/trial/2017-004221-32/BE%23A, August 6, 2018), and written informed consent, 34 patients scheduled for abdominal surgery received QL blocks postoperatively. The patients received either bilateral or unilateral blocks, according to the clinical indications; 26 patients received bilateral and 8 patients received unilateral blocks (first patient enrollment August 21, 2018). The inclusion criteria were (1) an indication for QL blocks for postoperative pain management, (2) American Society of Anesthesiologists physical status I or II, (3) at least 18 years old, and able to understand the purpose and risks of the study, which included radiation risks from CT scan described in the informed consent. Exclusion criteria were (1) pregnancy, (2) body mass index above 35 kg/m2, (3) hepatic or renal insufficiency, and (4) history of allergic or adverse reactions to local anesthetics or contrast agents. The QL blocks were randomly allocated to the QL1, QL2 or TOP QL3 techniques, using a sealed envelope method. We aimed to investigate the injectate spread in 20 injections for each QL technique (total of 60 injections) to summarize the common distribution patterns in the three QL groups.

The injections were performed postoperatively (Canon US Applio 450 ultrasound machine) with a curvilinear transducer (2 to 5 MHz), and 21-gauge, 100 mm stimulating needle (SonoPlex Stim, Pajunk GmbH, Germany). An injection pressure monitor (B-Smart; B Braun Melsungen AG, Germany) standardized the injection force (<15 psi). After confirmation of the correct needle tip position with 2 mL of sodium chloride 0.9%, 20 mL of injectate containing 18 mL ropivacaine 0.375% (Fresenius Kabi Canada Ltd) with 2 mL of radiopaque contrast (Iomeron 300: 61.24% w/v of iomeprol equivalent to 30% iodine or 300 mg iodine/mL (Bracco UK Limited)) was injected. Thirty minutes after injection, a high-definition CT scan (Siemens) was obtained to determine the distribution patterns of the injectates. The 3D image reconstruction was assessed from the stored image data.

Block techniques

Two anesthesiologists, experienced with QL1, QL2, and TOP QL3 performed the blocks (PG and CK). Injections were done in a lateral decubitus position. All patients who received bilateral blocks were re-positioned in the lateral position for each block, with the side to be blocked facing up.

QL1 block

The transducer was placed in transverse orientation at the midaxillary line cranially to the iliac crest to visualize the posterior end of the TAP and QL muscle. The needle was inserted in-plane from anterior to posterior toward the lateral edge of the QL muscle, figure 1.

Figure 1

Cross-section (A) and frontal view (B) of the anatomy around the QL muscle showing the three different injection points for the QL blocks and the eight predefined areas of interest (blue lines). EO, external oblique muscle; ESP, erector spinae muscles; IO, internal oblique muscle; QL, quadratus lumborum; TA, transversus abdominis muscle; TAP, transversus abdominis plane; TOP, transverse oblique paramedian.

QL2 block

The transducer was placed in transverse orientation at the posterior axillary line to visualize the fascial plane between the posterior aspect of the QL muscle and the middle layer of the thoracolumbar fascia. The needle was inserted in-plane from anterior to posterior through the oblique muscles toward the lumbar iliofascial triangle at the posterior aspect of the QL muscle, figure 1.

TOP QL3 block

The transducer was placed in transverse, oblique, and paramedian orientation approximately 3 cm lateral to the L2 spinous process to visualize the acoustic shadow of the L2 transverse process, QL, psoas major, and erector spinae muscles. The needle was inserted in-plane from posterior to anterior through the QL until the tip reached the plane between the QL and psoas muscles, figure 1.

Assessment of the distribution patterns

Two radiologists, blinded to randomization, jointly reviewed the distribution of injectate in the high-definition CT and 3D image reconstructions to assess whether the injectate was present in eight predefined anatomical areas, table 1.

Table 1

Eight predefined anatomical areas of interest

Statistical analysis

Continuous and discrete variables were shown as mean±SD, and ordinal and nominal (categorical) variables as n (%). Descriptive statistics were used to present the findings.

Results

A total of 60 blocks in 34 patients (20 for each QL technique) were analyzed. Twenty-six patients received bilateral blocks, whereas eight patients received unilateral blocks. Table 2 presents the demographic data of the studied patients.

Table 2

Patient’s demographic characteristics

The typical CT axial and sagittal spread patterns after QL1, QL2, and TOP QL3 injections are shown in figure 2. The extent of spread to areas around the injection site was limited with QL1 and TOP injections, being confined mainly to the TAP plane after QL1 and to the anterior QL plane after TOP injections, table 3. The spread following a QL2 injection was more extensive, the majority reaching the TAP and posterior QL plane, and 45% reaching the anterior QL plane.

Figure 2

Axial (left) and sagittal (right) CT images showing the typical spread patterns after QL1 (A), QL2 (B), and TOP QL3 (C) injections. The red asterisk (*) indicates the position of the QL muscle. ESP, erector spinae muscles; K, kidney; P, psoas muscle; QL, quadratus lumborum; TAP, transversus abdominis plane.

Table 3

Distribution of the injectate in the proximity of the injection (TAP and QL muscle)

Spread in the TAP was found primarily after QL1 (90%) and after QL2 injections (75%). The sagittal extension of the spread in the TAP was less variable after QL1 injections and on average, it covered two vertebral levels. The maximum craniocaudal spread occurred between the L1 and L5 vertebrae, figure 3.

Figure 3

Sagittal spread measured in vertebral levels with each QL1 and QL2 injections that resulted in the spread to tap. QL, quadratus lumborum; TAP, transversus abdominis plane.

Spread in the posterior aspect of the QL muscle was found after all QL2 injections (100%), after three QL1 injections (15%), and after one TOP QL3 injection (5%). Sagittally, the maximum craniocaudal spread occurred between the L1 and L5 vertebrae, and on average, it covered two vertebral levels for the QL2 and three levels for the QL1, figure 4. The TOP QL3 injection that spread to the posterior aspect of the QL muscle extended from L1 to L5.

Figure 4

Sagittal spread measured in vertebral levels with each QL1 and QL2 injections that resulted in the spread to the posterior aspect of the QL muscle. QL, quadratus lumborum.

Spread in the anterior aspect of the QL muscle was found after all TOP QL3 injections (100%), after seven QL1 injections (35%), and after nine QL2 injections (45%). On average, the sagittal spread along the anterior aspect of the QL muscle after TOP QL3 injections extended for three vertebral levels. The maximum craniocaudal spread occurred between the T12 and L5 vertebrae after a TOP QL3 injection, figure 5.

Figure 5

Sagittal spread measured in vertebral levels with each QL1, QL2, and TOP QL3 injections that resulted in the spread to the anterior aspect of the QL muscle. QL, quadratus lumborum; TOP, transverse oblique paramedian.

Table 4 summarizes the distribution of injectate in the remote locations from the injection site after QL1, QL2, and TOP QL3 injections, and figures 6 and 7 show the typical spread patterns. Of the three studied techniques, only TOP QL3 injections reached the paravertebral space, lumbar nerve roots, and sympathetic chain, although not in a consistent pattern. The maximum craniocaudal extension of the paravertebral spread occurred at the L1 and L4 vertebrae as described in figure 8. Only three TOP QL3 injections (15%) reached the lateral side of the vertebral body (sympathetic chain) mainly at the L2 level. This spread did not demonstrate a wide craniocaudal extension.

Table 4

Injectate distribution with QL1, QL2, and TOP QL3 injections, remotely to the injection site

Figure 6

CT images showing injectate spread remotely from the injection site after TOP QL3 injections. Presence of injectate in the vicinity of the paravertebral space (A), presence of injectate on the lateral side of a vertebral body (B), and presence of injectate next to a nerve root (C). QL, quadratus lumborum; TOP, transverse oblique paramedian.

Figure 7

3D-CT scan showing injectate spread (white arrows) in the intercostal plane after QL1 injection (A), intercostal plane after QL2 injection (B), thoracic spread after QL2 injection (C), and thoracic spread after TOP QL3 injection (D). 3D-CT, three-dimensional computed tomography; QL, quadratus lumborum.

Figure 8

Sagittal spread measured in vertebral levels with each TOP QL3 injections that resulted in the spread to the lumbar paravertebral space and sympathetic chain. QL, quadratus lumborum; TOP, transverse oblique paramedian.

QL1 and QL2 injections reached the intercostal plane more frequently than the TOP QL3. In these cases, the spread reached the lateral side of the intercostal planes of the 10th and 11th ribs in more than half of the cases. Finally, the spread of the QL2 and TOP QL3 injections to the thoracic paravertebral space occurred only in one block each (5%). In these two cases, the most cranial extent of the spread reached the T12 transverse process after a QL2 injection and the T11 transverse process after a TOP QL3 injection.

Discussion

This is the first study that compares the distribution of injectate among three QL techniques in a group of patients. Previous studies deduced the spread of the local anesthetic of QL blocks, based mainly on data from studies on cadavers, with variable outcomes.

Multiple factors can influence the spread of local anesthetics when injected into a fascial plane.9 Among the anatomical studies with QL techniques, the great variability in the spread of injectate may be at least partially related to the differences in their methodology which makes the comparison of the various techniques difficult.8 In their study of QL1 blocks in volunteers, Carney et al showed a few isolated specks of dye in the thoracic paravertebral space using MRI imaging 1–4 hours after the injection.10 It is possible that the paravertebral spread found in Carney’s study may have enticed modifications of QL techniques by other investigators to increase the likelihood of a more consistent paravertebral spread with injections closer to the neuraxis.5 However, to document the actual spread, a continuous path of the dye should link the site of injection with the therapeutic site, which was not reported by Carney et al. The discontinuous spread in Carney’s study may have resulted through the lymphatic spread of gadolinium rather than interfascial spread, considering that this contrast agent has a propensity for lymphatic uptake.11 Indeed, the thoracic paravertebral spread in Carney’s study did not translate into the sensory and motor distribution expected with a paravertebral block. Furthermore, other investigators did not reproduce similar paravertebral spread with the QL1 technique.12 13

Our findings documented that the QL1, QL2, and TOP QL3 techniques result in different distribution patterns, dependent on the anatomical injection site, figure 9. Additionally, we evaluated whether the spread of the local anesthetic indeed extends to other areas of interest beyond the injection site (ie, the paravertebral space). As such, we believe that our data add incremental information on any potential benefit of one QL technique over its modifications. QL1 primarily distributes in the TAP fascial plane. The QL2 injections spread along the posterior fascia of the QL muscle, extended into the TAP, and nearly half of the injections also reached the anterior aspect of the QL muscle. Finally, in contrast to what is held as an advantage of TOP QL3 injections, the disposition of the injectate occurred mostly in the anterior aspect of the QL muscle. Although TOP QL3 injections do extend beyond the injection site, a substantial spread to the paravertebral space, nerve roots, sympathetic chain, and thoracic paravertebral space does not appear to occur. To date, distribution patterns of injectate with the TOP QL3 block had been only described in cadavers.14 15 These investigators reported a consistent spread of the injectate into the thoracic paravertebral space reaching the sympathetic chain in all specimens. These findings, however, have not been reproduced in our study, or in the study by Kumar et al, 16 where greater volumes of local anesthetic were used (40 mL). Of note, as two TOP QL3 injections (10%) reached the lumbar nerve roots, the value of this block for lower extremity pain management is questionable and its use for abdominal surgery analgesia may result in inadvertent lower extremity weakness.

Figure 9

Distribution patterns in the eight predefined anatomical areas of interest among the three QL techniques. QL, quadratus lumborum; TAP, transversus abdominis plane; TOP, transverse oblique paramedian.

Our study design and the variability of the injectate disposition does not allow meaningful statistical analysis of the distribution patterns. However, our data suggest that the spread after QL1 and QL2 blocks is predictable and more consistent than the more invasive and technically challenging TOP QL3. With QL1 and QL2 blocks, the injectate was commonly present in the TAP and intercostal planes of the 10th and 11th ribs, territories where the ilioinguinal, iliohypogastric, subcostal and low intercostal nerves travel. This finding correlates with results from cadaver studies13 17 18 and indicates the possible mechanism of action of these blocks in patients having abdominal surgery.

Limitations

The main focus of our study was to describe the distribution of the injectate after the performance of different QL block techniques, and not the sensory extension or clinical analgesic outcome. Therefore, preblock and postblock pain scores and the distribution of sensory motor loss was not assessed. Measuring pain and sensorimotor distribution as outcomes could cause confounding issues when different QL block injection techniques were performed on the same patient. In addition, we did not assess whether higher volumes of local anesthetic could have resulted in a greater spread or if the use of different contrast agents could have affected the results by changing the physical properties of the injectate.

Conclusion

Our findings are similar to those available in studies performed in cadavers in that the different QL techniques resulted in a spread of injectate predominantly in the planes where the subcostal, iliohypogastric, and ilioinguinal nerves traverse. The QL1, QL2, and TOP QL3 injection techniques resulted in different distribution patterns, and the preferential spread of injectate depends on the anatomical injection site. We did not document sufficient differences in injectate distribution that support the proposed benefits of TOP QL3 over QL1 or QL2. Contrary to the commonly held belief, the more medial, cranial, and deeper TOP QL3 technique does not appear to result in a more extensive and reproducible interfascial spread toward the paravertebral space, nerve roots, or sympathetic chain. The routine use of the TOP QL3 technique in protocols for the management of acute postoperative pain should be carefully evaluated until the exact mechanism of action is determined.

Acknowledgments

The authors would like to acknowledge Dr Yvan Demerlier, Dr Dahin Laurent, and the department of radiology of Clinique Sainte-Anne Saint-Remi for their invaluable help in reviewing the 3D-CT images, and the entire NYSORA team for their assistance in preparing this manuscript.

References

Footnotes

  • Contributors ALB and AML helped in analysis and interpretation of data, manuscript preparation, and revisions. CK helped in conducting the study and data collection. J-LH, J-FB, and CV helped in manuscript preparation and revisions. AH helped in analysis and interpretation of data, manuscript preparation, and revisions. PG helped in study design, conducting the study, data collection, data analysis, manuscript preparation, and is the archival author.

  • 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 AH has consulted and/or performed sponsored research for Philipps, Pacira, BBraun Medical, and Heron Therapeutics. He owns and manages NYSORA.com (education), MedXpress.Pro (IP), and VisionExpo (medical design). He has developed and licenses technologies to the medical device industry (BBraun, LifeTech). Other authors declare no competing interests.

  • Patient consent for publication Not required.

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

  • Data availability statement Data are available upon reasonable request from Dr Philippe Gautier.