Regional anaesthesia: awake or asleep?

Abstract

We read with interest Feely et al.’s survey and were surprised by their finding, that the vast majority of anaesthetists performed both upper, and particularly lower limb blocks after induction of general anaesthesia [1]. Furthermore, for certain blocks over half of those who believed it was safer to perform the block prior to general anaesthesia admitted that this was not their actual practice. It would have been intriguing to explore why this was the case. Presumably, one reason was to avoid patient discomfort. Whilst we consider it important to minimise patient discomfort during peripheral nerve blockade (PNB), this can be achieved without resorting to general anaesthesia. We perform over 2500 regional anaesthetics annually at the Toronto Western Hospital, none of them under general anaesthesia. Instead we sedate the patients as required with titrated doses of intravenous midazolam. Occasionally we will supplement this with fentanyl or propofol if the block is particularly painful or the patient is particularly anxious. We avoid over-sedation so as to be able to detect significant pain, paraesthesia or early signs of local anaesthetic toxicity during block performance. Unlike the majority of the respondents, we routinely use ultrasound to localise nerves. Compared to neurostimulation, ultrasound can reduce performance time, needle passes and procedure related pain [2, 3]. The principal risks in PNB are inadvertent intraneural and intravascular injection. None of the techniques in current practice, including ultrasound, can wholly prevent these complications [4–7]. With respect to nerve injury, although pain and paraesthesia are insensitive indicators of needle-nerve contact [8, 9], their positive predictive value is close to 100%. Hence we continue to regard them as useful warning symptoms in the awake patient, along with visual inspection on ultrasound for intraneural injection. The use of a peripheral nerve stimulator is no guarantee of safety; there is good laboratory and clinical evidence that current thresholds are poor predictors of needle-nerve proximity or even transfixion [8, 10, 11]. Regarding intravascular injection, there is evidence that early recognition and treatment of systemic toxicity with intravenous lipid emulsion is essential to good outcome [12, 13]. One of the early signs of significant local anaesthetic systemic toxicity is a decrease in conscious level; this will not be apparent if the patient is already anaesthetised. The authors of the survey conclude that central neuraxial blocks should not be performed on anaesthetised patients. In the absence of evidence to the contrary, we believe that this practice should be extended to include PNB for the reasons outlined above.