Incidence of peri‐operative peripheral nerve injuries: a reply

We would like to thank O'Sullivan and Sarai 1 for their comments on our article 2. We acknowledge the use of administrative datasets makes it difficult to draw clinically meaningful conclusions. However, due to the infrequency of these events, without the use of administrative datasets, we would be limited to self-reported registry data or smaller single centre cohorts at risk of their own inherent confounders and biases 3. The dataset used in our study allows us to estimate risk with a complete population denominator without the need for any extrapolation. In this sense, they are correct that the analysis can only give an impression of overall peri-operative risk. It does not allow attribution of causality to any individual modality, given the multifactorial nature of nerve injuries and our inability to determine timing or mechanisms of injury. Similarly, it cannot determine the timing of regional anaesthesia as occurring before or after induction of general anaesthesia. However, despite this being accepted practice in paediatric regional anaesthesia 4, there is a paucity of data in adults to suggest conscious feedback of paraesthesia and pain is protective against nerve injury 5. In our dataset, 88% of regional anaesthesia techniques performed over the period were combined with some degree of procedural sedation or general anaesthesia. The depth of hypnosis or drugs used could not be quantified from the dataset. This further confounds the capacity to make interpretations around the risk associated with regional anaesthesia 1. Risk in anaesthetic practice is modular, as is consenting for this risk. While a procedure performed under block only may be associated with lower risk of nerve injury, this must be weighed against the other risks of the procedure in addition to the risk of being awake for the entirety of the procedure. The material nature of all these risks must also be individualised for each patient. Overall, given the rarity of nerve injuries, the difference in absolute risk with and without sedation is likely difficult to appreciate for most patients and is likely immaterial 6. While the reduction in nerve injury rates appears to be reducing over time, the reduction in our subset of ‘specified nerve injuries’ over the same period was more modest, possibly due to further dilution of these infrequent events. While the suggestion of tracking the trend of injury rates over time across the different anaesthetic groups would be interesting, further dilution of these infrequent events would not provide for meaningful comparison. O'Sullivan and Sarai raise very valid points to consider; while prospective data would be preferable, the infrequent incidence of peripheral nerve injuries makes this often impractical. We await eagerly the findings from the 8th National Audit Project of the Royal College of Anaesthetists.