Pneumothorax following serratus anterior plane block

Summary

Serratus anterior plane blocks may provide analgesia for rib fractures, thoracic surgery and breast surgery. There remains uncertainty regarding the location of injection, be it superficial or deep to the serratus anterior muscle. We describe the case of a 63‐year‐old ASA physical status 3 woman undergoing a wire‐guided wide local excision of a right breast lump. A modified right serratus anterior plane block was performed under ultrasound guidance, injecting 20 ml of bupivacaine 0.25% deep to the serratus anterior muscle. No immediate complications were noted. Intra‐operatively the patient developed arterial desaturation associated with high airway pressures. A subsequent chest radiograph in the post‐anaesthetic care unit demonstrated a large right‐sided pneumothorax which was treated with immediate needle decompression and chest drain insertion. This is the first case of pneumothorax reported from this approach to serratus anterior plane blockade and serves as a reminder of a potential serious complication.

Introduction

Serratus anterior plane blockade has been increasingly utilised for analgesia for rib fractures, thoracic surgery and breast surgery. Since the first reported ultrasound‐guided serratus anterior plane block ¹, various approaches have been described. Analgesia is provided by blocking the lateral branches of the intercostal nerves (T2–T6), the long thoracic nerve and the thoracodorsal nerve. We report a case that serves as a reminder of a serious complication associated with serratus anterior planes blocks and reflect specifically on two approaches where local anaesthetic is deposited either superficial or deep to the muscle.

Report

A 63‐year‐old woman with a background of hypertension, type 2 diabetes mellitus, ex‐smoker (quit 20 years previously) and a body mass index of 39 kg.m⁻² was scheduled for a wire‐guided right breast wide local excision with sentinel lymph node biopsy for a grade 2 invasive ductal carcinoma. During the pre‐assessment visit on the day of surgery, she was given paracetamol, ibuprofen, omeprazole and metoclopramide. She then underwent a wire insertion into the breast lesion in the radiology department before presenting for general anaesthesia. After establishing standard monitoring, intravenous general anaesthesia was induced in the supine position, a size 4 iGel supraglottic airway device (Intersurgical Ltd, Wokingham, Berkshire, UK) was inserted and intermittent positive pressure ventilation commenced. No neuromuscular blockade was administered. After initiation of positive pressure ventilation, the patient started coughing, which resolved with a bolus dose of propofol.

Anaesthesia was maintained with sevoflurane in an oxygen and air mixture. A modified serratus anterior plane block was then performed using a 6–13 MHz linear ultrasound transducer (Sonosite Xporte, Fujifilm SonoSite, Amsterdam, the Netherlands) under aseptic conditions with the patient supine. The probe was positioned over the fifth rib in the mid‐axillary line, allowing identification of latissimus dorsi, serratus anterior muscle and underlying ribs in the coronal plane. A 22G, 80 mm Sonoplex needle (Pajunk^® GmbH Medizintechnologie, Geisingen, Germany) was used with an in‐plane technique, with the needle directed from caudad to cephalad in the coronal plane to contact the fifth rib. Ten millilitres of bupivacaine 0.25% was injected in the fascial plane deep to the serratus anterior muscle, the needle trajectory was flattened and advanced to extend the hydro‐dissection to the fourth rib and a further 10 ml of bupivacaine 0.25% was injected. No immediate complications were noted.

Intra‐operatively the patient developed a significant episode of oxygen desaturation with elevated peak airway pressures to 37 cmH₂O. This temporarily improved by increasing the fractional inspired concentration of oxygen, performing a positive pressure recruitment manoeuvre, applying positive end‐expiratory pressure and increasing the inspiratory to expiratory ratio. The peripheral arterial oxygen saturations increased from 80% to 92% on 60% inspired oxygen concentration. The patient remained haemodynamically stable throughout the surgical procedure, which lasted approximately 90 minutes. At the end of the procedure the patient was transferred to the bed, sat upright, re‐established on spontaneous respiration and was emerged from anaesthesia. The supraglottic airway device was removed, the patient was given 10 l.min⁻¹ oxygen through a facemask and transferred to the post‐anaesthesia care unit. Postoperatively, the patient's oxygen requirement remained high with peripheral arterial oxygen saturation readings between 85% and 88%, although the patient was completely asymptomatic. A subsequent chest radiograph showed a large right‐sided pneumothorax, which was treated with immediate needle decompression and insertion of an open surgical chest drain in the fifth intercostal space. The chest drain was connected to an underwater seal and low pressure suction at 2 kPa. The chest drain remained in‐situ for three days and a chest radiograph following removal showed a small residual apical pneumothorax. As the patient was asymptomatic, she was discharged home for follow‐up four weeks later under the respiratory physicians. The patient made a full recovery and a follow‐up chest radiograph demonstrated complete lung re‐expansion.

Discussion

Since their initial description, serratus anterior plane blocks have been increasingly discussed in the literature as a means of providing analgesia for thoracic and breast surgery as well as in the management of multiple rib fractures ^{2, 3}. The first ultrasound guided block was reported in 2013 ¹ and since then various approaches have been described, including a high or low approach with injection of local anaesthetic superficial or deep to the serratus anterior muscle, respectively ^{1, 4}. Analgesia results from blocking lateral branches of the intercostal nerves (T2–T6), the long thoracic nerve and the thoracodorsal nerve. For thoracic surgery, when compared with thoracic epidurals, serratus anterior plane blockade may provide a safe and equally effective form of analgesia whilst avoiding potential deleterious consequences such as autonomic disruption ².

Two distinct compartments have been identified by Blanco et al. ¹ and the question of whether to block superficial or deep to the serratus anterior muscle belly remains a matter of contention ⁵. Blanco et al. provide a descriptive study which suggests local anaesthetic injection superficially is more effective at blocking the lateral cutaneous branches of the intercostal nerves as well as providing a longer duration of sensory block, a desirable feature for day case procedures ¹.

Superficial deposition of local anaesthetic, whilst posing lower risk of pneumothorax, may also block the long thoracic nerve (the motor nerve to the serratus anterior muscle), resulting in winging of the scapula in the early postoperative period. This is particularly important to consider as the long thoracic nerve can be damaged during axillary dissection and damage to the nerve from either cause can have a disabling impact on the patient. In contrast, depositing local anaesthetic deep to the serratus anterior muscle could reduce the risks of this and may be practically more straightforward to perform, with rib contact providing a definitive end‐point for needle advancement.

This case demonstrates that despite purported advantages of local anaesthetic deposition deep to the serratus anterior muscle, there is also the potential for significant patient harm. As the first case of a pneumothorax reported from this approach, it serves as a reminder of a serious complication and the need for vigilance, as such complications can present in a subtle manner.