The utilisation of regional anaesthesia in children has significantly improved pain management.¹ The erector spinae plane block (ESPB) was a serendipitous finding first described by Forero et al.² for managing neuropathic pain in the thoracic region in adults.

When performing an ESPB, local anaesthetic is injected into the fascial plane situated deep to the erector spinae muscle (ESM), superficial and lateral to the tips of the transverse process (Figure 1).³ The use of ESPB for perioperative analgesia,⁴ and as a possible safe alternative to traditional paravertebral and neuraxial blocks⁵ which carry a potential risk of complications, including dural puncture and epidural haematoma has been previously described in the adult population.⁶

While the cervical ESPB has been shown to anaesthetise the brachial plexus in adults, its precise anatomic mechanism of action remains unclear.⁷ This block has potential applications for pain management in cervical spine and shoulder surgeries,⁷ including posterior atlanto-axial fusion surgery⁸ and shoulder arthroscopy⁹ as shown in adults.

The anatomy of the cervical ESPB in neonates and children has not been described. Thus, it is important to define the anatomy of young children, because extrapolating adult anatomy to children of different ages carries the risk of inaccuracies.^10,11

Regional anaesthesia can be performed safely with limited risk of neurological damage.⁵ Ultrasound guidance is essential to ensure accurate needle placement for ESPBs, particularly in the cervical region. The study by Elsharkawy et al.⁷ demonstrated that the block can be simulated in cadavers. Ultrasound provides a clear view of the relevant anatomical structures.^7,12

The purpose of the study was not to test the effectiveness of the block for specific procedures or injuries, but rather to detail the methodology and spread of the cervical ESPB in neonates and infants.

Because a description of the anatomy of the cervical erector spinae plane in the neonates is lacking, this study aimed to investigate the anatomical characteristics of the cervical ESPB in a neonatal cadaver model. The intent is to clarify the anatomical findings in a manner that is clinically relevant to paediatric anaesthesia providers. Our premise was that the contrast medium, injected according to current procedure guidelines, would spread to the cervical spinal nerves when injected at both the C6 and C7 vertebral levels.

The primary objective was to determine the spread of injectate across dermatomes following the injection of contrast medium into the fascial plane at C6 and C7 vertebral levels in fresh-frozen neonate cadavers using computed tomography (CT) scanning. A bilateral ESPB in the cervical region was simulated at vertebral levels C6 on the right and C7 on the left using a Samsung PT60A system (Samsung Medison Co., Ltd., Seoul, Korea) equipped with a 12.3 MHz high-frequency linear transducer, which was covered with a protective plastic sheath in all procedures.

The cone beam CT scans were acquired using the Planmeca Viso G7 CBCT unit (Planmeca OY, Helsinki, Finland), operated with Romexis software (version 6.4.5.136; Planmeca, Helsinki, Finland) to expose all the scans.

All dissections and scans were performed in the Department of Anatomy at the University of Pretoria. The data in this article were obtained as part of a Doctor of Philosophy (PhD) (Anatomy) study that started in 2023, and the degree is expected to be completed by the end of 2025.

Nine fresh-frozen low-normal birth weight neonatal cadavers with an average weight of 1.37 kg, consisting of four males and five females, obtained through the National Tissue Bank from the University of Pretoria and Sefako Makgato Health Science University, were defrosted over a period of 48 h before study. Bilateral ultrasound-guided ESPBs in the cervical region were simulated at vertebral level C6 on the right and C7 on the left.

Only cadavers without developmental abnormalities in the neck and thoracic regions, and without prior dissection in these regions, were included in the study.

With the cadaver in a prone, fixed position, the spinous processes were assessed with ultrasound, whereby the C7 vertebra was palpated, and the probe slid cranially to identify the C6 vertebra, whose transverse foramen transmits both the vertebral artery and vein. At the level of the C5 vertebra, the anterior and posterior tubercles are more widely spread, and the presence of the carotid bulb serves as a distinguishing feature between the C5 and C6. In contrast, the transverse foramen of C7 transmits only the vertebral vein.

To simulate the ESPB at the C6 vertebral level, the probe was placed parallel and perpendicular to the vertebral column over the transverse processes of C6 and C7, approximately 1 cm lateral to the spinous process. The technique of using ultrasound on fresh cadavers was similar, though not identical, to that described by Govender et al.¹³ Relevant anatomical landmarks include the transverse processes of the C6 (anterior – Chassaignac’s – and posterior tubercles) and C7 vertebrae,¹⁴ the erector spinae tissue plane, the anterior, middle, and posterior scalene muscles, the vertebral artery and vein, as well as C5, C6, and C7 nerve roots. The anterior tubercle of the transverse process of C6 and the transverse process of C7 serve as reliable reference points for needle placement, with the needle directed toward the tip of the transverse process (Figure 2).¹² Using the in-plane technique, with a linear array ultrasound probe, the needle was advanced through the surrounding muscle layers assuring the precise desired location at the tip of the transverse process as demonstrated in adults.¹²

Utilising the in-plane approach, a 5 cm 22G Tuohy needle was directed in a cephalo-caudal direction towards the tip of the transverse process until contact was made. Contrast medium (0.1 mL/kg) was injected, and its spread was observed within the deep fascial plane.

A similar approach was followed to simulate the ESPB at the C7 vertebral level; however, the probe was placed parallel to the superficial lateral tip of the transverse process.

Cone beam CT scans were acquired for imaging, with the cadavers positioned upright in a plastic tube. Standard imaging parameters were applied (100 kV, 9.0 mA, exposure time 12.2 s; field of view Ø 20.0 cm × 17.0 cm; voxel size 0.200 mm). Scans were obtained within 30 min following contrast injection.

All scans were analysed using three-dimensional (3D) reconstructions to assess the distribution of the contrast medium (Figure 3). The vertebral levels reached by the contrast medium and the corresponding spinal nerve roots were documented based on the spread, enabling the authors to calculate the dose per kg per spinal nerve root (mL) for each cadaver. Additionally, the percentage of cadavers exhibiting contrast spread to specific spinal nerve roots was recorded for both the C6 and C7 vertebral levels.

The cervical ESPB was simulated bilaterally in nine fresh-frozen neonatal cadavers, resulting in 18 block simulations. A total of 12 simulated blocks were successful (n = 12), while six were classified as failures (n = 6). Failed blocks were defined as blocks where the contrast medium spread into the vertebral canal, laterally at the injection site or at an incorrect vertebral level. These blocks were excluded from the analysis.

The spread of the contrast medium at the C6 and C7 vertebral levels was observed in varying patterns, with the most widespread at the C6 level. At C6, the contrast medium spread across the C5–C7 levels (100% of samples). The contrast medium also reached the C8 and T1 levels in 85.71% and 71.43% of cases, respectively. The least common spread was observed at C2 (14.29%) and T4 (14.29%) (Table 1).

For the C7 injection, the spread was most frequently observed between C6 and T1 (100% of samples), with spread to C3 and C4 seen less frequently (20% and 40%, respectively). The spread to the C5 (80%) and T2 (80%) levels was observed in most cases (Table 2).

The spread to each vertebral level, as a percentage, is presented in Table 3.

The spread of the contrast medium ranged from C2 to T4, with the most common coverage occurring between C5 and C7, consistent with previous research in adults.^1,6 At the C6 injection level, there was frequent spread to the C8 and T1 levels (85.71% and 71.43%, respectively), which is consistent with the anticipated effect on the brachial plexus. It was also observed that the C6 injection produced the most consistent and broadest spread, highlighting the anatomical significance of the C6 vertebral level for achieving optimal spread. The spread observed at C7 was slightly less consistent, with coverage typically extending from C6 to T1 in most cases. These patterns suggest that the C6 injection may be preferable for achieving wider dermatomal coverage, particularly in clinical settings where more extensive brachial plexus involvement is desired.

The findings of this study highlight the potential of ultrasound-guided cervical ESPB as a viable regional anaesthetic technique in neonates. This study builds on the previous anatomical work describing the ultrasound appearance of the erector spinae plane in neonatal cadavers, extending this by demonstrating the pattern of injectate spread within the fascial plane.¹³ Given the smaller anatomical structures, thinner muscle layers and incomplete ossification in neonates and infants, precise needle placement is essential to avoid complications and ensure effective anaesthetic spread.³ The use of ultrasound guidance enhances safety and accuracy,¹⁵ particularly in the cervical region, where critical neurovascular structures are in proximity. The ultrasound-based identification of anatomical landmarks in this study is consistent with previously described techniques in neonatal cadaveric models.¹³

The observed spread across multiple spinal nerve roots, especially from C5 to T1, aligns with prior adult studies and highlights the ESPB’s capacity to provide effective multi-dermatomal analgesia. This broad sensory coverage may reduce the reliance on systemic opioids,^1,6 which are associated with adverse effects such as respiratory depression and excessive sedation in neonates.¹ Furthermore, the anterior spread of injectate into the paravertebral space suggests that the ESPB may also affect sympathetic fibres, offering a more comprehensive block than initially anticipated.¹⁶ The mediolateral spread of the local anaesthetic solution is typically bound by the borders of the ESM when performing thoracic ESPBs¹⁷ but has not been described for cervical ESPB. Previous studies have shown that the spread of tissue plane blocks is volume dependent.⁶ It has been suggested that a volume of 0.1 mL/kg would suffice to block each spinal nerve root in infants and children.¹⁸ This study followed these guidelines to determine the extent of the spread in a simulated ESPB. The median spread of the dye was six spinal nerve roots at the C6 vertebral level and seven spinal nerve roots at the C7 vertebral level, including spinal nerve roots in the cervical and thoracic regions. In one instance, the contrast medium spread from the C2 to T4 spinal nerve roots following injection at the C6 vertebral level, which was not observed for any simulation of the C7 level injections. This may have been because of several factors, including the density, composition and thickness of the individual’s fascia.¹⁹ Because of the aforementioned anatomical factors that may potentially affect the spread of the ESPB block, this study based the evaluations on the consistently affected spinal nerve roots. These findings were similar to those observed in adult cadavers, where the contrast medium spread to the C5–C7 roots and occasionally to the T1 root.⁷

Limitations of this study are that the variability in the spread of the contrast medium, which could be attributed to several factors, including the unique anatomical properties of fresh-frozen neonatal cadavers, the volume and rate of injection, and the use of contrast medium as opposed to local anaesthetics. Local anaesthetic agents may behave differently from contrast medium injected into live patients because of differences in physicochemical properties. Further studies, particularly in live patients, are necessary to confirm our observations and assess the clinical efficacy of cervical ESPBs in neonates and infants.

This study also supports the claim that the ESPB technique spreads along a fascial plane, away from critical structures such as the dura mater and spinal cord.

The sensitive nature of procuring fresh neonatal cadavers was another major limitation. As a result, the small sample size makes it difficult to draw absolute conclusions. Several confounding variables may be introduced when using data obtained from cadavers to simulate the spread in live patients. The weight of the cadavers used in the study ranged from 0.7 kg to 2.4 kg, which is less than the average weight for newborns, thus perhaps contributing to small cadaver size as a variable.

Also, the use of previously frozen-fresh cadavers defrosted prior to simulated injection of the contrast medium may result in anatomic distortion because of the freezing and thawing process. Furthermore, the lack of demographic information, such as gestational age of the cadavers, and the state of preparation, causing difficulty in evaluating anatomical factors in full, and the lack of in vivo factors, for example, muscle tension and pressure or volume dynamics in live tissue, are all factors that may have affected the spread of the contrast medium. Additionally, the physicochemical properties of contrast medium differ from those of local anaesthetics, which may impact the spread observed in live patients. The contrast medium was injected to dissect the cadavers following the CT scans and to document the spread of methylene blue; however, upon dissection, the investigators found that methylene blue did not stain the surrounding tissue or nerves. Despite the limitations mentioned, this study is the first of its kind and provides invaluable information that has significant clinical value.

There are anatomical and physiological differences between neonates, infants, children and adults. Fascia is typically looser and has greater elasticity in neonates, which may affect the spread of anaesthesia when performing an ESPB.²⁰ This study offers valuable anatomical insights into the cervical ESPB in a neonatal cadaveric sample, demonstrating the consistent spread of the contrast medium across the C5–C7 spinal nerve roots, with additional spread observed to C8 and T1, as well as to parts of the cervical plexus (C2–C5). The additional spread observed may be attributed to the anatomical features of the fascial plane in this age category. These results suggest that the cervical ESPB may be a promising technique for neonates and infants; however, extrapolation of the data to infants should be done cautiously. The findings of the study should be interpreted as preliminary anatomical evidence rather than definitive clinical evidence.

Given the limitations of using cadaveric specimens and the differences in contrast medium versus local anaesthetics, further research is necessary to assess the clinical efficacy, safety and potential outcomes of cervical ESPB in live patients. Future studies should focus on clinical trials to evaluate efficacy, safety and the long-term impact of this technique in neonates and infants.