Anatomical landmarks for a proximal paravertebral thoracolumbar block in sheep: Cadaver and in-vivo study

Abstract

The objective of this study was to describe the anatomy of the spinal nerves, specifically the last thoracic nerve (T13) and the first to third lumbar nerves (L1 to L3), in order to safely carry out an accurate proximal paravertebral block (PPVB) in sheep. This study consisted of 2 phases. In Phase 1, 7 sheep cadavers were dissected to identify the path and relevant anatomical landmarks of spinal nerves T13 and L1 to L3. In Phase 2, 2 healthy sheep received bilateral injections of 0.35 mL/kg body weight (BW) for each hemithoracolumbar area (0.088 mL/kg BW per nerve) of a dye-lidocaine solution (50:50) using a PPVB approach and then assessed for 15 min for signs of systemic and local effects of lidocaine. After euthanasia, the infiltrated area was dissected to assess the spread of the dye. Successful nerve staining (> 2 cm in length), macroscopic evidence of intraneural/intravascular injection, and spread to the epidural space and the abdominal cavity were recorded. In Phase 1, each branch of all nerves was easily identified and located using the caudal aspect of the spinous apophysis and the lateral edge of the transverse process of the respective vertebrae. An overlap was observed between the costoabdominal (T13), the iliohypogastric (L1), and ilioinguinal (L2) nerves. In Phase 2, all nerves were stained at least 2 cm from the injection site. There was no diffusion of the dye into the epidural space or abdominal cavity. In conclusion, using the anatomical landmarks described specifically for sheep, the PPVB provided accurate perineural distribution of the injected dye-lidocaine solution, which could result in clinical analgesia of the flank.

Introduction

In ruminants, the sensory and motor innervation of the skin, muscles, and peritoneum of the abdominal flank are supplied by the dorsal and ventral branches of the last thoracic nerve (T13) and the first (L1), second (L2), and third (L3) lumbar nerves (1,2).

The anatomy of these nerves has been described extensively in cattle (3–6) and there is a tendency to assume that the anatomical structures in small ruminants are similar (7,8). However, significant anatomical differences have been described in the distribution and paths of cranial nerves and lumbar nerves in cattle, as well as between cattle and other ruminants. These differences could result in inaccuracies when such extrapolations are made (7,9,10).

Several anesthetic techniques have been used to desensitize the abdominal flank in ruminants (4–6) and among these, the paravertebral thoracolumbar nerve block is the most recommended technique (11,12). Two techniques (proximal and distal) have been described for this type of paravertebral block (3,6,13), both of which have been commonly used in sheep and goats by extrapolation from cattle (11,14–17).

Variations in the anatomy of sheep and cattle could result in incomplete local anesthetic blockade of the nerves, if the distance between the nerves and the site of injection is longer than the optimal diffusion distance through the tissues. In this case, a higher dose of local anesthetic drug would be required to produce a satisfactory block (10,18,19). Conversely, a high dose of local anesthetic drugs can result in a blockade of other neighboring nonselected nerves and toxicity from excessive systemic absorption (6,20,21).

There are no anatomical studies in sheep that could provide guidance for an accurate and efficient paravertebral thoracolumbar block, despite such studies in other small ruminants (14,16). The main objective of this study was to describe the relevant anatomy of the thoracolumbar paravertebral nerves in sheep and to define the relevant anatomical landmarks required to efficiently and safely complete a proximal paravertebral block (PPVB) in sheep.

Materials and methods

The project was approved by the Ethics Committee of the Faculty of Veterinary Medicine at the Universidad Autonoma de Yucatan (CB-CCBA-D-2019-003) and adheres to the Consensus Author Guidelines on Animal Ethics and Welfare for Veterinary Journals. It is reported according to the Animal Research: Reporting of In-vitro Experiments (ARRIVE) guidelines. This study consisted of 2 phases.

Phase 1: Anatomical dissection

Seven fresh sheep cadavers weighing 34 ± 5.6 kg were used after the sheep were euthanized for a different study with an intravenous overdose of barbiturate. Before euthanasia, sheep underwent a complete physical examination to rule out abnormalities of the spine that could interfere with the anatomical findings.

The cadaver was positioned in lateral recumbency for macroscopic examination by palpation of anatomical reference points (last rib and spinal processes of the first lumbar vertebrae). The spinal nerves were dissected according to Mansour et al (2) by making a dorsal longitudinal skin incision on the right side of the midline from the fifth thoracic vertebrae (T5) to the pelvic limb, followed by removal of the subcutaneous fat from the deep fascia of the trunk. The longissimus, semispinalis, and multifidus muscles were then dissected and removed to facilitate the visualization of the dorsal rami of the spinal nerves. The dorsal and ventral rami were then dissected along their path in the epaxial lumbar muscles and hypaxial lumbar and abdominal muscles, respectively.

The exact location, emergence site, thickness, and trajectory of the spinal nerves (T13, L1, L2, and L3) were determined following reference measurements adapted from Nev et al (16), using an electronic vernier caliper (Calibrador digital, Uline, Mexico) and plastic-headed pins. Specific reference measurements included depth, point where the nerve bifurcates into dorsal and ventral rami, and length of the transverse process. In addition, the structures adjacent to the nerves were described to facilitate determination of the nerve location. The dissection was recorded photographically.

Phase 2: Evaluation of dye-lidocaine spread

Two healthy, 1.8-year-old female sheep, weighing 36.3 and 40.5 kg and free of abnormalities of the spine, were used. The sheep were premedicated with acepromazine (Calmivet, 0.5%, Vetoquinol, Mexico City, Mexico), 0.02 mg/kg body weight (BW) administered into the jugular vein. After 30 min, the hair was clipped from the epaxial area of T12 to the iliac crest and the whole paralumbar fossa of both flanks and the area aseptically prepared.

To desensitize the dermis, subcutaneous tissue, and superficial muscles of the area, local anesthetic infiltration was carried out at the site of each injection for the PPVB with lidocaine 1% (Pisacaine 1%; PiSA Farmaceutica, Guadalajara, Jalisco, Mexico), 0.5 mL/kg BW, injected with a 22-gauge, 1.2-in (3.2-cm) hypodermic needle (BD Yale, Mexico City, Mexico).

The anatomical landmarks were the caudal edge of the spinous process of the vertebrae, which coincides with the point of bifurcation of the spinal nerve, as well as the lateral edge of the transverse process. The transverse process of each vertebra of interest (T13 and L1 to 3) was identified and a perpendicular line was traced from the transverse process towards the spinous process. The puncture site was located lateral to the spinous process and caudal to the transverse process, at 1/3 of the distance of the perpendicular line, which corresponded to approximately 2.4 cm from the lateral edge of the transverse process.

Ten minutes later, a 21-gauge, 1.5-in (3.8-cm) needle (BD Yale, Mexico) was inserted at a 90° angle relative to the skin, obliquely directing the needle in a lateral-to-medial direction with respect to the spinous process, until the tip of the needle reached the intertransverse ligament. An aspiration test was carried out and, if positive for blood or air, the needle was removed completely and repositioned 3 mm caudally and the approach repeated.

Once aspiration confirmed negative pressure, a dyelidocaine solution consisting of a mixture (50:50) of lidocaine (Pisacaine 2%; PiSA Farmaceutica) was already included and a tissue marker dye (Tissue Marker, #111 Noris Meat Marking Ink; Noris Color, Kulmbach, Germany) was injected slowly at a dose of 0.088 mL/kg BW at that location of the intertransverse ligament to block both rami of the nerve.

The procedure was repeated for each nerve block and hemiabdomen and the distance from the skin surface to the location of the tip of the needle at the intertransverse ligament was recorded for each nerve. The order of block for each hemiabdomen (left or right) and the order of nerve block (T13 to L3 or L3 to T13) was randomized using a computer-generated random number (GraphPad, https://www.graphpad.com/quickcalcs/randomN2/), so that 1 sheep was started on the left side and the other on the right side, and that 1 side was blocked in ascending order and the other in descending order for each sheep.

Sheep were observed for 15 min after the last injection of the solution for signs of toxicity from systemic absorption (depression, muscle tremors, or seizures), epidural diffusion of the anesthetic (loss of motor control of the hind limbs), and scoliosis of the lumbar region. Analgesia was assessed by cutaneous pinprick with a 21-gauge needle. A positive response to pain consisted of skin and muscle twitching, turning the head towards the stimulated area, or avoidance of the stimulus.

Sheep were then euthanized with an intravenous overdose of barbiturate (Pentobarbital injection 6.5%; Aranda Salud Animal, Querétaro de Arteaga, Mexico) and the anatomical structures of interest were dissected by an anatomical pathologist to evaluate the distribution of the dye, following the criteria described by Campoy et al (22). Staining of over 2 cm along the nerve indicated an effective block, whereas staining of more than 4 cm indicated that excessive volume had been injected (22). During the dissection, nerves and blood vessels were evaluated for macroscopic evidence of either intraneural or intravascular injection or epidural dispersion.

Statistical analysis

A Shapiro-Wilk test was used to test data for distribution patterns. A paired Student’s t-test was used to compare measurements on the left and right side of the spinal nerves. Fisher’s exact test was used to test for the presence or absence (yes/no) of nerve staining. Parametric data is presented as mean ± SD and nonparametric and ordinal variables are expressed as median and range. Differences were considered significant at a value of P < 0.05. Data was analyzed using GraphPad Prism Version 9.0 (GraphPad Prism Software, San Diego, California USA).

Sample calculations for both phases were based on an alpha value of 5% and a power of 80% and were calculated with an online calculator (Clin calc.com https://clincalc.com/stats/samplesize.aspx). For Phase I, the number of sheep was calculated based on similar cadaveric studies where anatomical landmarks and development of a local block technique were completed with a sample of 6 to 10 cadavers (16,22,23). For Phase II, a pilot study was completed using sheep cadavers to determine the volume of injectate that could spread from the lumbar injection site. Based on described total volumes of injection of 0.26 mL/kg BW to 1.1 mL/kg BW, which corresponded to total volumes of injection per nerve of 2 to 10 mL (15,24,25), a total volume of 0.35 mL/kg BW, divided in equal amounts among the 4 nerves, was appropriate when injected using the technique described for Phase I.

Results

The spinal nerves on both the left and right sides of the sheep were included.

Phase 1

Anatomical description

All 4 spinal nerves were easily recognized and the dorsal and ventral branch of each nerve were also identified. The relevant anatomical landmarks determined during anatomical dissection were: i) the caudal aspect of the spinous apophysis (it should be noted that the nerve passes ventrally after emerging from the intervertebral foramen); and ii) the lateral edge of the transverse process.

Anatomical dissection of the flank innervation revealed the dermatomes innervated by their respective spinal nerves. An overlap was observed with the costoabdominal (T13), iliohypogastric (L1), and ilioinguinal (L2) nerves (Figure 1), but this did not include the genitofemoral (L3) nerve.

Figure 1.

Lateral view of the dermatomes innervated by the thoracolumbar spinal nerves T13 (a), L1 (b), L2 (c), and L3 (d) in sheep.

Morphometric findings

There were no significant differences in the anatomical location of the 4 spinal nerves with regard to their vertical distance from the skin to the emergence from the intervertebral foramen, distance from the emergence point to bifurcation into dorsal and ventral rami, and the length of the transverse process (Table I).

Table I.

Anatomical location of spinal nerves T13, L1, L2, and L3 in sheep (n = 7). Data expressed as mean ± SD.

Nerve	Anatomical parameters (cm)	Side
		Left	Right
T13	DSE	4.0 ± 0.5	4.0 ± 0.5
	DEB	0.1 ± 0.09	0.1 ± 0.16
	LTP	3.5 ± 0.2	3.5 ± 0.2
L1	DSE	4.1 ± 0.5	4.1 ± 0.5
	DEB	0.1 ± 0.07	0.1 ± 0.03
	LTP	3.7 ± 0.2	3.7 ± 0.2
L2	DSE	4.2 ± 0.5	4.2 ± 0.5
	DEB	0.1 ± 0.06	0.1 ± 0.07
	LTP	3.7 ± 0.1	3.7 ± 0.1
L3	DSE	4.2 ± 0.5	4.2 ± 0.5
	DEB	0.2 ± 0.01	0.1 ± 0.04
	LTP	3.8 ± 0.3	3.8 ± 0.3
Mean values	DSE	4.1 ± 0.4
	DEB	0.1 ± 0.02
	LTP	3.6 ± 0.2

DSE — Distance from the skin to the emergence of the spinal nerve from the intervertebral foramen; DEB — Distance from the emergence point to bifurcation of the spinal nerve; LTP — Length of the transverse process.

Phase 2

Clinical evaluation

The volume of solution (0.35 mL/kg BW) used per nerve was 3.2 mL and 3.5 mL according to the weight of each sheep (36.3 kg and 40.5 kg, respectively). No clinical signs of epidural spread, toxicity, or scoliosis were observed within 15 min after injection.

Distribution of the dye

The anatomical landmarks identified in Phase I were easily located in all sheep. During the approach, the aspiration test was positive for blood in 2 out of 16 procedures. The mean distance from skin to the intertransverse ligament was 3.7 ± 0.1 cm, based on needle location during the PPVB and both dorsal and ventral rami of the spinal nerve could be dyed with the injection from this location.

The ventral and dorsal branches of the 16 spinal nerves analyzed were stained for over 2 cm and less than 4 cm (Figure 2). Although no spread of the dye was observed at the epidural level or in the abdominal cavity, slight staining of the psoas major muscle was observed on the right side of 1 sheep at the level of T13 (Figure 2).

Figure 2.

Distribution of dye after a proximal paravertebral block injection at the level of the intertransverse ligament caudal to the transverse process. A — Third lumbar vertebrae (L3) using 0.088 mL/kg BW of a dyelidocaine solution per nerve. B — Thirteenth thoracic vertebrae (T13). The distribution and spread of dye were predominantly dorsal and caudal to the transverse process. The red points in the transverse view of “A” show the dye distribution in the dorsal and ventral rami of the spinal nerve.

Discussion

In this study, a novel PPVB carried out according to sheep-specific anatomical landmarks is described. This technique resulted in accurate perineural deposition of the local anesthetic mixed with dye in the live tissue of sheep, which could result in clinical analgesia of the flank. No significant differences were noted between the measurements of the right and left sides of the spinal nerves. A similar symmetry has been previously reported in goats (16).

In sheep, the spinal nerves emerging from the intervertebral foramen show a strictly segmental dermatomal distribution (26), as they run obliquely on the lateral surface of the transverse abdominal muscle (2). In the present study, we determined that the nerve branches of the costoabdominal nerve (T13) run obliquely in the medial region of the abdomen, overlapping the segments corresponding to the iliohypogastric (L1) and ilioinguinal (L2) nerves, to reach the umbilical area (Figures 1 and 2).

This innervation pattern has been previously documented in sheep and consists of a segmental pattern of regular and continuous bands with a craniocaudal direction, overlapping across the dorsal and ventral midlines (27,28), where the caudal edge of each dermatome overlaps the cranial edge of the subsequent dermatome (28). This overlap of thoracolumbar spinal nerves is more common in sheep than in goats (27), but similar to cattle (29) and humans (30), although the pattern of overlap may vary between them.

From a clinical perspective, the overlapping of nerves on dermatomes can partially overcome the failure to properly block each nerve. Since more than a single nerve root innervates any dermatome, the blockade of adjacent nerves is recommended for better inclusion of the area to be blocked (31).

It is important to include the genitofemoral nerve (L3) in the PPVB if better sensory blocking of the caudal third of the flank is desired (6), since this nerve does not overlap with the other nerves of the PPVB. Kirk (28) described the communication of the sensitive area of the L2 and L3 nerve as overlaying the caudal border of the sensitive area of L2 nerve, with the cranial of the L3 nerve border extended onto all the area.

Other studies in goats and sheep did not include the L3 nerve in their methods and the volume of injection per nerve ranged from 2 to 10 mL injected near the intervertebral foramen (25) and was divided into equal volumes (5 mL each for the dorsal and ventral rami) (24) or 1 to 1.5 mL at the dorsal rami and 2 to 2.5 mL at the ventral rami (11,15).

In this study, the volume of injection per nerve was 3.2 to 3.5 mL, which also included L3. In addition, the injection to block both the dorsal and ventral rami was administered from a single point once the intertransverse ligament was reached, due to the thin nature of the ligament in small ruminants. The spread of dye indicated that this approach was effective.

In adult cattle, the injection is divided between the dorsal and ventral rami by injecting above and below the intertransverse ligament, respectively. The volume injected below the ligament is significantly larger (4-fold) since the ventral rami provides most of the innervation to the paralumbar fosa (6). In small ruminants, individual injection of the dorsal and ventral rami has also been described using the PPVB approach, with similar ratios and proportional volumes according to their weight (11,6) or larger but equal volumes for each ramus of each nerve (24).

The recommended depth of needle insertion for small ruminants ranges from 2.5 to 3.8 cm (6) to 5 to 7.5 cm (19). In this study, the mean depth from the skin to the emergence point of the spinal nerve from the intervertebral foramen was 4.1 ± 0.4 cm and the rami were reached using a 3.8-cm needle as they travel closer to the transverse process, since this distance was measured as 3.7 ± 0.1 cm from the point of needle insertion at the skin to the intertransverse ligament. Accurate estimation of the depth at which the needle should be advanced would allow the local anesthetic to be deposited with greater precision.

These distances will vary according to the size and breed of the sheep, as well as body condition, due to the arrangement of skin, muscle, and subcutaneous fat (16). In smaller goats, the distance was 3.0 ± 0.4 cm for West African dwarf goats that weighed 13.3 kg (16) and 3.3 ± 0.2 to 3.7 ± 0.1 cm for Black Bengal goats that weighed 15 to 20 kg (14).

In general, local anesthetics should be deposited as close to the nerve as possible, preferably in the tissue sheaths, e.g., brachial plexus or lumbar plexus, or in the epineural sheaths of nerves, e.g., femoral or sciatic (32). We observed such a pattern in the distribution of the dye in the dorsal and ventral rami of each spinal nerve involved in this study (T13 to L3), as well as on the point of emergence from the intervertebral foramen.

The dye distribution observed in the present study was optimal because a single anesthetic injection carried out using the PPVB, e.g., near the point of nerve emergence from the intervertebral foramen, had a similar effect to a distal paravertebral approach where 2 injections are required, 1 above (dorsal rami) and 1 below (ventral rami) the transverse process in a horizontal and lateromedial fashion from the lateral aspect of the respective transverse process.

Moreover, the staining pattern on the 16 nerves used in this study can be considered clinically effective, according to established criteria (22), in which a staining distance that is 5 times the minimal exposure length of 3 to 4 mm (3 nodes of Ranvier) is required to effectively block nerve transmission. Under these conditions, the conduction of nerve impulses should be completely blocked, thus desensitizing the whole flank wall and allowing muscular relaxation (33).

A limitation of this study was the short time for which analgesia was assessed after the PPVB. In this study, determining the spread of the dye was more relevant for the anatomical determinations. It is likely that the onset of lidocaine was not complete for blockade of the 4 nerves at the 15-minute post-administration period used in this study to assess the onset of the block and properly determine its characteristics, including speed of onset and extension of the block and duration, since the effective concentration of lidocaine andits degree of ionization is affected (32). In addition, lidocaine was diluted from 2% to 1% with the dye, which may limit the spread of the block, but also decrease the likelihood of adverse/toxic effects.

The use of proper localization techniques, e.g., anatomical surface markings, is one of the main determinants of successful nerve blocking. Localization techniques should be preferred over the use of larger volumes of anesthetic in order to increase the diffusion of the local anesthetic over the tissues, as there is a high risk of systemic toxicity when larger volumes are used.

Based on the staining pattern obtained, we propose that the total volume used (0.35 mL/kg BW) to carry out the PPVB was sufficient for successful individual spinal nerve block, when this volume is divided equally among the 4 nerves involved.

In conclusion, the sheep-specific anatomical landmarks described (the caudal aspect of the spinous apophysis and the lateral edge of the transverse process) are reliable for accurate perineural distribution of the local anesthetic in a proximal approach. A total volume of 0.35 mL/kg BW (0.088 mL/kg BW) of the dye-lidocaine solution was sufficient to stain ≥ 2 cm of the 4 target nerves.