Combined thoracic erector spinae plane and inter‐transverse plane blocks for awake breast surgery

Summary

Worldwide, breast cancer is the most commonly diagnosed cancer in women. Surgical procedures are typically performed using general anaesthesia, often complemented by regional anaesthesia to manage postoperative pain. However, avoidance of general anaesthesia for breast surgery may be desirable for clinical reasons or patient choice. It is theorised that the use of regional anaesthesia and the avoidance of volatile anaesthetics and opioid analgesia may have beneficial effects on oncological outcomes, and there is some evidence to support this. While many patients successfully undergo awake breast surgery, a limited number of anaesthetists possess direct experience of this approach, despite familiarity with regional anaesthesia techniques. Undertaking regional anaesthesia for awake breast surgery requires patient cooperation and excellent staff teamwork. Here, we present a case of a patient who underwent awake bilateral mastectomy with reconstruction. This was carried out under two ‘paravertebral‐by‐proxy’ blocks: the thoracic erector spinae plane and inter‐transverse plane blocks, with intravenous sedation.

Introduction

Breast cancer surgery is mostly performed under general anaesthesia (GA). However, for some patients, GA may carry a high risk of complications. These patients include those with a potentially difficult airway, frailty, pregnancy or specific comorbidities. Some patients may just prefer to avoid GA [1]. The use of regional anaesthesia techniques and the limitation of opioids or volatile anaesthetics may have a beneficial impact on cancer outcomes [2].

Awake breast surgery under regional anaesthesia presents a challenge for anaesthetists. Several regional blocks have been used for surgery under sedation. Of these, the thoracic paravertebral block is the most studied, and the evidence suggests that it might not be sufficient as the sole technique to produce effective anaesthesia for breast surgery [3].

The dermatomes relevant to breast surgery extend from C3 to T6. These are innervated by the supra‐clavicular nerve, the long thoracic nerve, the medial and lateral pectoral nerves, the thoracodorsal nerve and spinal thoracic nerves T2 to T6. The supra‐clavicular nerve originates from the C3 and C4 nerve roots of the superficial cervical plexus, the long thoracic nerve arises just above the clavicle from the proximal portion of the brachial plexus receiving contribution from roots of spinal nerves C5, C6 and C7, emerging from behind the middle scalene muscle and innervating the serratus anterior muscle. The medial and lateral pectoral nerves arise from the lateral and medial cord of the brachial plexus and innervate the pectoralis major and minor muscles, and the thoracodorsal nerve innervates the latissimus dorsi muscle, arising from the posterior cord of the brachial plexus. All nerves relevant for anaesthesia during breast surgery can be blocked adequately with around 40 ml of local anaesthetic solution distributed as follows: 20 ml at the interscalene level and paravertebral blocks at seven levels, from C8‐T6, using 3 ml per injection [4].

Several alternatives to the paravertebral block have recently been proposed, but none of these new techniques have yet been shown to provide anaesthesia equivalent to a paravertebral block. Multiple thoracic wall blocks are required for anaesthesia of the breast for surgery [5], and additional infiltration for skin incision may still be required in addition. A 2019 case report describes the successful use of the mid‐point transverse process to pleura block as the sole anaesthetic technique for a patient undergoing breast surgery [6]. The mid‐point transverse process to pleura block may be preferred as an alternative to the thoracic paravertebral block for breast surgery with a lower risk of complications [7]. When compared to the paravertebral block, the landmarks for the mid‐point transverse process to pleura block are more reliably identified, with the potential to reduce the chance of epidural or spinal injection, vascular puncture, pneumothorax or nerve damage associated with paravertebral block. The nomenclature of the mid‐point transverse process to pleura block has now been updated to inter‐transverse plane block.

In our case report, we describe anaesthesia for awake breast surgery. We used a combination of two ‘paravertebral‐by‐proxy’ blocks: the thoracic erector spinae plane block and inter‐transverse plane block. ‘Paravertebral by proxy’ implies an alternative approach to accessing the paravertebral space. The desired effect is achieved indirectly through injection of local anaesthetic solution outside the paravertebral space. The erector spinae plane block targets the plane deep to the erector spinae muscle, while the inter‐transverse plane block targets a deeper space than the erector spinae plane block which is superficial to the superior costo‐transverse ligament. This ligament constitutes the posterior portion of the paravertebral space (Fig. 1a). We used these blocks in combination with intravenous sedation to deliver effective anaesthesia for a patient requiring bilateral mastectomy with reconstruction.

Figure 1.

(a) A drawing of the erector spinae plane muscle compartment, thoracic paravertebral space and retro‐superior costo‐transverse ligament space (dotted arrows). (b) Ultrasound scan at T3 level showing the sonoanatomy of the posterior thoracic wall: trapezius, rhomboid and erector spinae muscles are shown. (c) Erector spinae plane block. Local anaesthetic solution is injected under the erector spinae plane muscle. Dotted arrow shows superior costo‐transverse ligament. (d) Inter‐transverse plane block. Through the same puncture, the needle tip is directed to the mid‐point between the transverse process and the pleura. Local anaesthetic solution is injected posterior to the superior costo‐transverse ligament in the retro‐superior costo‐transverse ligament space. (e) The pleural edge is displaced by the spread of injectate. ES: erector spinae; P: paravertebral space; SCTL: superior costo‐transverse ligament; TP: transverse process; R: rib; ESP: erector spinae plane; ITP inter‐transverse plane.

Case report

A 57‐year‐old woman (height 173 cm, weight 72 kg, BMI 24 kg.m⁻², ASA III), required bilateral mastectomy with reconstructive surgery. She expressed anxiety regarding GA because of a fear of difficult airway management. She had no history of a difficult airway although, at airway assessment ahead of surgery, she was found to be Mallampati class III with limited mouth opening. Following a discussion with the patient, the decision was made to undertake surgery under regional anaesthesia and sedation. Pre‐operatively, the patient was informed about regional techniques including the procedure for the regional blocks, the local anaesthetics and adjuvants to be used, and the use of intravenous sedation for optimal comfort. Risks of local anaesthetics such as allergy and toxicity were also discussed. Rescue options discussed included increased sedation, additional analgesia, local anaesthesia supplementation by the surgeon and the possibility of conversion to GA. Pre‐operatively we had a team brief to ensure the whole team were aware of the situation, especially the patient's choice to have awake surgery and their anxiety about GA, and our choice of regional anaesthesia techniques with the aim of avoiding GA. The whole team were aware of the potential challenges and requirements including the need for anaesthetic access to the airway. With the patient's written consent, we planned for ultrasound‐guided thoracic erector spinae plane block and inter‐transverse plane block to be performed bilaterally.

After we initiated routine monitoring, midazolam 2 mg was administered intravenously, prior to performing the blocks. Following skin asepsis, we performed ultrasound‐guided thoracic erector spinae plane blocks and inter‐transverse plane blocks via a single injection on each side at the T3 level using an echogenic 100 mm needle (Ultraplex™, B.Braun, Melsungen AG, Germany) with the ultrasound probe (EDGE II, FUJIFILM‐Sonosite™, Bothwell, WA) orientated in the paramedian sagittal plane, approximately 2 cm from the midline measured from the spinous process, while visualising the transverse process. We then inserted the needle in‐plane from a cranial to caudal direction until the needle tip reached the transverse process. Local anaesthetic was injected for the thoracic erector spinae plane block (Fig. 1b and c). Then, the inter‐transverse plane block was achieved by advancing the needle towards the mid‐point between the posterior border of the transverse process and the pleura (Fig. 1d and e). We used 20 ml ropivacaine 0.2%, dexamethasone 2 mg, and dexmedetomidine 6.25 μg for the erector spinae plane block, and 10 ml ropivacaine 0.2%, dexamethasone 2 mg, and dexmedetomidine 6.25 μg for the inter‐transverse plane block, both performed bilaterally. Forty minutes after the blocks were performed, adequate anaesthesia was confirmed in the dermatomes C7‐T6. We evaluated this using both pinprick and cold tests with a 70% isopropyl alcohol sponge. Oxygen was provided via nasal cannulae at a rate of 4 l.min⁻¹ and capnometry was used to monitor ventilation. Intra‐operative sedation was achieved with intravenous infusions of dexmedetomidine (range of infusion rate: 0.3 to 0.8 mcg.kg⁻¹.h⁻¹) and ketamine (range of infusion rate: 0.25 to 0.45 mg.kg⁻¹.h⁻¹). This was adjusted throughout the case to maintain a Bispectral index (BIS™, Medtronic, Minneapolis, MN, USA) between 70 and 90. The surgery lasted 210 min and was uneventful. No additional local anaesthetic infiltration was required. On leaving the theatre, the patient reported pain as a numeric rating scale score (NRS score, 0–10 where 0 = no pain and 10 = worst pain) of 0. At 2 h after surgery, an NRS score of 2 at rest and on movement was reported. At 6, 12, 24 and 36 h NRS scores at rest were 1, and on movement 2, without the need of additional analgesics. After three days, the patient was discharged home.

Discussion

In recent years, many regional anaesthesia techniques, specifically thoracic wall blocks, have been described for analgesia and anaesthesia in breast surgery.

Our report involves the provision of two posterior thoracic wall blocks for a patient who needed awake breast surgery. The erector spinae plane block was first described by Forero in 2016 and has been used effectively for analgesia after breast surgery. The mid‐point transverse process to pleura block was initially demonstrated in 2017 in three cadavers and replicated in two patients undergoing breast surgery as a new approach to the paravertebral space, by injecting local anaesthetic posterior to the superior costo‐transverse ligament [8]. The updated nomenclature for the mid‐point transverse process to pleura block is the inter‐transverse process block, as per ASRA‐ESRA consensus.

The erector spinae plane block's mechanism of action is thought to be due to the spread of injectate towards the paravertebral space. A recent review including 29 anatomical studies, reported that the erector spinae plane block consistently results in injectate spread within the erector spinae compartment in 100% of cases across up to nine spinal levels, with diffusion into the paravertebral space in 57% of cases with a mean spread 3.5 spinal levels [9]. Notably, a time‐related spread is observed, with 38% of observed paravertebral space diffusion having occurred at 30 min and 73% having occurred between 30 min and 2 h. The likely mechanism of action of the inter‐transverse plane block is the diffusion of local anaesthetic within the paravertebral space via the superior costo‐transverse ligament. Recent anatomical and three‐dimensional micro‐computed tomography (CT) scan studies of embalmed cadavers show that the superior costo‐transverse ligament only incompletely forms the posterior wall of the thoracic paravertebral space. The retro‐superior costo‐transverse ligament space, which contains the posterior branches of the spinal nerves, exhibits extensive connections to the costo‐transverse space, intervertebral foramen, erector spinae compartment and thoracic paravertebral space via slits in the superior costo‐transverse ligament [10]. The inter‐transverse plane block is delivered to the posterior aspect of the thoracic paravertebral space (retro‐superior costo‐transverse ligament space), and is described as a ‘paravertebral‐by‐proxy’ block [11]. A single‐injection inter‐transverse plane block has been demonstrated to be non‐inferior in comparison to multiple injections in terms of anaesthetised dermatomes [12]. Chen et al. observed that, compared to the erector spinae plane block, a single‐injection inter‐transverse plane block provides enhanced anterior and intercostal spread, albeit with reduced cephalocaudal spread [13]. The average segmental spread is reported as 10 for erector spinae plane blocks and 4.5 for inter‐transverse plane blocks [13].

As previously demonstrated in a patient undergoing a laparoscopic Nissen fundoplication [14], in our case we hypothesised that by combining the two techniques, the inter‐transverse plane block might enhance the erector spinae plane block. The inter‐transverse plane block, with the administration of the local anaesthetic deeper than the erector spinae plane provides anaesthesia in four to five levels from the T3‐T4 puncture. The erector spinae plane block with its potential spread to the paravertebral space (especially for high‐volume injectates after 30 min) and its consistent diffusion into the erector spinae compartment spanning up to 9–10 spinal levels extends coverage to more distant dermatomes. This compliments the inter‐transverse plane block, which has more limited spread and typically misses coverage of the nerves emerging from the brachial plexus.

The advantages of this approach are a single injection to each side, ensuring effective blockade of the nerves with sufficient duration to allow awake breast surgery. To achieve a similar coverage using alternative techniques, we would have had to perform many blocks (inter‐pectoral, pecto‐serratus and parasternal) together with multiple level paravertebral injections, resulting in longer procedural times, and the potential for patient discomfort and increased risks of vascular injection, local anaesthetic systemic toxicity and pneumothorax. The inter‐transverse plane block performed at the level of T3‐T4 ensures an anaesthetic coverage over four to five levels, where surgical incision is typically made. This may contribute to a reduced need for intravenous sedation. The surgical field remains dry, as no anaesthetic spread occurs through the fascial planes of the anterior thoracic wall.

We used a high‐volume of a diluted ropivacaine (0.2%) with two adjuncts, balancing the goals of reducing local anaesthetic‐related toxicity, enhancing the blocks' effects and prolonging analgesic duration. While regular paracetamol was prescribed for the first 48 h, opioids were not required. With regard to the perineural adjuncts, a meta‐analysis showed that combining dexmedetomidine with local anaesthetics for paravertebral block significantly improves postoperative pain, extends analgesia and reduces analgesic consumption. However, it could be associated with an increased risk of hypotension [15], with the dosage of dexmedetomidine in the included studies up to 1.2 mcg.kg⁻¹. In our case, despite using intravenous dexmedetomidine for sedation after its use as an adjunct, we did not observe hypotension or bradycardia.

It is important to note that our patient received substantial doses of sedative agents, and although we were able to accommodate the patient's wishes, it is worth considering whether there is a clinical benefit to this approach when compared with GA.

In procedural sedation, such as in the emergency department, dexmedetomidine is combined with a potent analgesic like ketamine to achieve conscious sedation. In our case, the dosage of ketamine administered throughout the entire surgery matches that of a bolus used for procedural sedation. For dexmedetomidine, we maintained a continuous infusion to ensure an optimal sedation plan throughout the surgery duration. This led to the administration of more substantial doses of dexmedetomidine.

Given that these are fascial blocks and they are large‐volume dependent, further research into optimal dosing regimens (local anaesthetic volume and concentration) will help in balancing the considerations of block onset, effectiveness and duration versus the risk of local anaesthetic systemic toxicity. Caution is mandatory to avoid exceeding the maximum safe dose of local anaesthetic, particularly in frail patients or when these techniques are combined. The role of local anaesthetic adjuncts or the continuous catheter techniques in prolonging the duration of analgesic benefits also require further study.

In conclusion, anaesthesia and analgesia for breast surgery remains a challenge. Our case report demonstrates that a combination of two ‘paravertebral‐by‐proxy’ blocks facilitated a bilateral mastectomy plus reconstruction with a relatively low dose of local anaesthetic, providing prolonged postoperative analgesia without side effects. We believe that awake breast surgery can be safely and effectively conducted with careful planning.