Article Text

Download PDFPDF

Upper trunk block for shoulder analgesia with potential phrenic nerve sparing: a preliminary anatomical report
  1. José Cros Campoy1,
  2. Oscar Domingo Bosch2,
  3. Jaume Pomés3,
  4. Jing Lee2,
  5. Ben Fox4 and
  6. Xavier Sala-Blanch5,6
  1. 1 Anesthesia, Royal Wolverhampton Hospitals NHS Trust, Wolverhampton, UK
  2. 2 Anesthesia, Worcestershire Acute Hospitals NHS Trust, Worcester, UK
  3. 3 Radiology, Hospital Clinic de Barcelona, Barcelona, Spain
  4. 4 Anaesthesia, Queen Elizabeth Hospital King's Lynn NHS Foundation Trust, King's Lynn, UK
  5. 5 Anesthesia, Hospital Clinic de Barcelona, Barcelona, Spain
  6. 6 Human Anatomy and Embriology, Universitat de Barcelona, Barcelona, Spain
  1. Correspondence to Dr Xavier Sala-Blanch, Anesthesia, Hospital Clinic de Barcelona, Barcelona 08036, Spain; xavi.sala.blanch{at}


Background and objectives Ipsilateral phrenic nerve palsy (PNP) is an undesirable side of conventional approaches to interscalene brachial plexus blocks. The purpose of this study was to demonstrate whether or not the phrenic nerve can be spared by dye when injected at the division of the upper trunk of the brachial plexus.

Methods Under ultrasound guidance, 5 mL of radiolabeled dye was injected between the anterior and posterior division of the upper trunk in two fresh, cryopreserved cadavers. CT scan analysis, cadaveric dissection, and cryosectioning were performed to examine the spread of the injectate.

Results We found staining of the injectate over the entire upper trunk with its anterior and posterior divisions, the suprascapular nerve under the omohyoid muscle and the lateral pectoralis nerve, and the C5 and C6 roots. The middle trunk was partially stained. There was no evidence of dye staining of the lower trunk, anterior aspect of the anterior scalene muscle, or the phrenic nerve.

Conclusions Our study offers an anatomical basis for the possibility of providing shoulder analgesia and avoiding a PNP.

  • interscalene block
  • nerve block
  • shoulder surgery
  • postoperative analgesia
  • ultrasound
  • anatomy

Statistics from


Interscalene brachial plexus block remains the gold standard for regional anesthesia and analgesia of the shoulder. However, it is associated with a significant risk of incidental phrenic nerve palsy (PNP), which limits its use in respiratory-compromised patients.1 2 To reduce the risks of PNP, a number of modified, interscalene brachial plexus blocks have been described3——so-called ‘diaphragm-sparing’ techniques.

Ultrasound-guided peripheral nerve blocks have a pivotal role in the reduction of the incidence of PNP since they enable the use of lower volumes of local anesthetic (LA) and permit a more precise deposition near the relevant nerves.4 Brachial plexus blockade in the supraclavicular fossa produces equivalent analgesic efficacy after shoulder surgery, with a lower incidence of PNP.5–7 Aguirre et al 8 have described a new method of blockade at the divisions of the upper trunk in the supraclavicular fossa. They proposed that injection of LA in the subomohyoid plane, in close proximity to the suprascapular nerve, at the level of the divisions of the upper trunk, will not only provide analgesia to the regions supplied by this nerve, (the acromioclavicular and glenohumeral joints, the infraspinatus and supraspinatus muscles) but the LA might also spread proximally to the upper trunk to supply analgesia to the shoulder via other branches of C5 and C6 nerve roots. Such distribution of LA could be expected to provide superior analgesia compared with a suprascapular nerve block by the posterior approach.

We conducted an anatomical study, using the technique described by Aguirre et al 8 in cadavers with dye to determine its spread to the relevant nerve structures which innervate the shoulder, and its potential for phrenic nerve sparing.


Ultrasound guided dye injection

Ultrasound scanning and dye injections were carried out in the supraclavicular fossa of both cadavers bilaterally with a linear transducer (6–13 MHz) and a 50 mm, 22 gage (30° beveled) block needle. The needle was inserted using an in-plane technique from a posterolateral to anteromedial direction. The C5 and C6 nerve roots were identified and tracked distally to view their union into the upper trunk, and further, until the level of the divisions which were located posterolaterally to the subclavian artery. The target point was between the anterior and posterior divisions within the paraneural sheath (figure 1). Here, 5 mL of injectate (60% NaCl 0.9%, 20% methylene blue and 20% iodine contrast) was administered using a pump at a constant flow of 20 mL/min. With methylene blue, dissection and cryosectioning would reveal the in vitro distribution of the injected fluid.

Figure 1

Preinjection and postinjection sonographic view of the upper trunk with clear visualization of anterior and posterior divisions. Labeled view of the upper trunk identifying the posterior division (PD) and the anterior division (AD) (white contour). Point of injection (white asterisk) with spread of injectate around the divisions of upper trunk (yellow).

CT imaging

After block performance, CT imaging of the cadaver was performed. A scout view of 512 mm was obtained. The images were obtained with a rotation time of 0.5 s, slice collimation of 0.6 mm, 120 kV, and 90 mA effective current. Axial, coronal, and sagittal reconstructions were performed with 3 mm sections. The images were evaluated by a radiologist (JP, blinded in primary objective of the study) to confirm the spread of the injectate in relation to the anatomical locations of the divisions of the upper trunk, the suprascapular, lateral pectoralis, and phrenic nerves, and the anterior aspect of the anterior scalene muscle. Cross-sectional and three-dimensional(3D) reconstruction images were used to assess the spread of the injectate. Postacquisition accurate 3D assessment of spread was done with Digital Imaging and Communication On Medicine (DICOM) images obtained in CT scan by the anesthesiologists to confirm the distribution, using computer software.

Anatomic dissection and cryosectioning

On one of the cadavers, bilateral dissection was performed by experienced anatomists who were blinded to both the injections and the CT images. The injection involved plane-by-plane dissection until the upper trunk and associated structures were exposed. Dye spread was examined to determine the nerves that would have been potentially been blocked by a similar injection in vivo.

The other cadaver, following CT scanning, was frozen and cryosectioned for analysis. On the cadaver’s right side, 2 cm slices were taken in the sagittal plane; on the left side, 2 cm slices were obtained in the axial plane.


Ultrasound-guided upper trunk block was successfully performed in both cadavers bilaterally, with accurate placement of the needle at the point between the anterior and posterior divisions.

The point of injection between the anterior and posterior divisions, and the distribution of injectate are shown in figure 1. In figure 2, the CT scan and anatomical findings corroborate the distribution of injectate, with the distribution of dye following the brachial plexus in a craniomedial to caudolateral direction. In figure 3, the distal end of C5 and C6 roots (but not C7 root), as well as the upper trunk and its anterior and posterior divisions are stained. Staining is also seen along the proximal suprascapular nerve and at the proximal origin of the lateral pectoral nerve. Caudally, spread was limited to the middle trunk and the inferior trunk was not affected. Anteriorly, there was no spread of dye as the supraclavicular nerve (superficial cervical plexus) was not stained (figures 3 and 4). Medial spread of dye was not observed and the phrenic nerve was also spared, see figure 5.

Figure 2

3D CT scan reconstruction of the spread of the injectate in the right side of cadaver one. The volume of injectate is shown in yellow; the spread is mainly proximal reaching the level of the transverse process of C5 (the highest proximal spread of the four injections performed in the study). The distal spread stopped at a level between the first and second rib. 1, first rib; 2, second rib; 3, clavicle; 4, thyroid cartilage.

Figure 3

Anatomical dissection of cadaver 1, showing the staining of brachial plexus structures in the right and left side with of methylene blue. 1, anterior division of upper trunk; 2, posterior division of upper trunk; 3, suprascapular nerve; 4, lateral pectoral nerve; 5, subclavian artery; 6, subclavian vein; 7, anterior scalene muscle.

Figure 4

Cryosection slices of cadaver 2; right side in sagittal section (A 2 cm lateral to B sagittal section) and left side in axial section (A 2 cm cephalic to B). 1, sternocleidomastoid muscle; 2, anterior scalene muscle; 3, middle-posterior scalene muscle; 4, subclavian artery; 5, first rib; 6, vertebral vessels (artery and vein); 7, carotid artery; 8, longus colli muscle; 9, C7 vertebral body; 10, T1 vertebral body.

Figure 5

Spread of dye affecting completely the upper trunk and its divisions but was not observed in phrenic nerve. UT, ultrasound

All four injections performed in the two cadavers demonstrated a posterior spread, with involvement of the proximal portion of the suprascapular nerve but not extending as far as the suprascapular notch (table 1).

Table 1

Distribution of the injectate


This study demonstrates that when a 5 mL volume of fluid is injected at the level of, and between, the anterior and posterior divisions of the upper trunk, its distribution remains limited to an area relatively close to the point of injection.

In an alive human, this limited spread may spare the phrenic nerve by the absence of proximal diffusion to C4 root and antero-medial spread to the anterior aspect of the scalenus anterior muscle.

PNP is an unwanted consequence of the interscalene approach and limits its use in patients with compromised lung function.1 2 The introduction of ultrasonography and lower volume usage has reduced its incidence but not eradicated it,9 and the interscalene brachial plexus block continues to be relatively contraindicated in these patients. More selective blocks have been proposed where interscalene block is contraindicated.10–12 Our cadaveric findings of limited spread of dye using a novel block technique provides an anatomical basis to advocate for further clinical research.


This study was a project for the Master in Advanced Medical Competences in Regional Anaesthesia Based on Anatomy, School of Medicine, Universitat of Barcelona, Barcelona, Spain, for JCC). The authors thank Dr Laura Perry and Dr James Stimpson for their initial idea that inspired our study and Dr Barrie Fischer for his expert opinion and final supervision of the article.



  • Presented at This article has been presented as an e-poster and oral presentation during the ESRA meeting, Dublin, 2018.

  • Contributors JCC participated in study design, conduction, data collection, data analysis, and manuscript preparation. Attestation: JCC approved the final manuscript. ODB participated in study conduction, data collection, data analysis, and manuscript preparation. Attestation: ODB approved the final manuscript. JP participated in data collection and data analysis. Attestation: JP approved the final manuscript. JL participated in study conduction and data collection, and manuscript preparation. Attestation: JL approved the final manuscript. BF participated in study conduction and data collection, and manuscript preparation. Attestation: BF approved the final manuscript. XS-B designed the study, participated in study conduction, data collection, data analysis, and manuscript preparation. Attestation: XS-B approved the final manuscript.

  • Funding The authors received internal funding from the Faculty of Medicine, Universitat of Barcelona, Spain, for this article.

  • Competing interests All authors have participated in the design and conduct of the study, have read and approved the document, they meet the requirements of authorship, have no conflicts of interest and consider the manuscript presents an honest work and have respected all ethical principles of the World Medical Association and the Declaration of Helsinki.

  • Patient consent for publication Not required.

  • Ethics approval This study was approved by the Research Ethics Committee and conducted at the Medical School, University of Barcelona.

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

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.