Case report reveals cardiac arrest from presumed vertebral artery injection in a chronic pain patient, highlighting persistent risks despite real-time ultrasound guidance.
We report a case of cardiac arrest immediately following stellate ganglion block performed under ultrasound guidance. We believe this to be the first reported case of probable inadvertent vertebral artery injection during stellate ganglion block under ultrasound guidance. A 60-year-old, overweight male was listed for left stellate ganglion block for chronic facial pain following multiple maxillofacial operations refractory to conservative therapies. Several previous stellate ganglion blocks had been effective and performed without incident. The procedure was carried out by two experienced operators, one holding the probe and needle and the other injecting the local anaesthetic. A 5–13 MHz linear ultrasound probe (GE Healthcare, Hatfield, Herts, UK) was used to guide the needle via a lateral approach. The sixth cervical transverse process, left common carotid artery, left internal jugular vein, longus colli muscle and oesophagus were clearly identifiable at all times. A 50-mm 22-G pencil-point nerve stimulator needle (Polymedic, Carrieres-sur-Seine, France) was used and positioned below the prevertebral fascia under direct in-plane ultrasound guidance. A total of 10 ml bupivacaine 0.5% was injected in 2-ml increments over one minute and the local was observed spreading during injection. Before and after each 2-ml bolus, the syringe was aspirated and at no point was blood withdrawn. The tip of the needle was visible at all times. Immediately after completion of the procedure, the patient felt unwell. He began twitching around the mouth and eyes, then lost consciousness and became cyanosed. This was followed by a generalised tonic-clonic seizure. The patient’s airway became obstructed and no pulse was palpable, so ventilation with 100% oxygen and cardiac massage were commenced. There was return of spontaneous circulation after two minutes. Post-arrest observations were satisfactory. The patient was discharged home the next morning. Vasovagal syncope was excluded by the presence of tonic-clonic seizures and sinus tachycardia before cardiac arrest which showed asystole on the ECG [1]. We surmise that the seizures led to respiratory then cardiac arrest. The likely cause of the complication was injection of local anaesthetic into the vertebral artery because of: (i) the temporal course of events; (ii) subsequent features of local anaesthetic toxicity; (iii) the fact that the carotid artery was clear of the needle on ultrasound; and (iv) the excitatory nature of the convulsions. Although it is likely that a small volume (< 2 ml) would have entered the vertebral artery, very small doses of local anaesthetic have been known to precipitate generalised seizures [2]. The advent of ultrasound-guided nerve blockade has changed the practice of regional anaesthesia, and its use in stellate ganglion block has gained popularity [3]. First described by Kapral et al. in 1995 [4], there is no consensus regarding the optimal anatomical approach for performing ultrasound-guided stellate ganglion block. The lateral approach has been advocated and validated in cadavers [5]. Ultrasound-guided regional anaesthesia techniques offer superior accuracy and require lower doses of local anaesthetic, thus improving patient safety over landmark techniques. However, in the absence of large-scale prospective trials, this serious adverse event acts as a timely safety warning. Clinicians must remain mindful of potentially life-threatening risks such as intra-arterial injection, despite the benefit of ultrasound-guidance. When performing stellate ganglion block under ultrasound guidance, we suggest routine safety precautions still apply and resuscitation equipment must be readily available, including lipid emulsion for local anaesthetic toxicity. Operators must have thorough knowledge of the anatomical region and be competent using ultrasound technology. The potential for life-threatening complications must be included in the patient consent. No external funding and no competing interests declared. Published with the patient’s written consent.
No takes yet. Share an insight, caveat, or question.
Rastogi et al. (2010) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: