I thank Dr Cornish for his interest and comments on our recent review [1]. He raises many points, so my responses will be brief.

A lack of artistic flair on the part of this author’s ability to predict which patients would experience moderate-to-severe pain after ‘minor’ arthroscopic shoulder surgery prompted us to perform a randomised trial. This trial demonstrated the benefit of the continuous technique in these patients [2]. Dr Cornish’s previous criticisms of that study have already been refuted; the issue does not warrant further comment here.

We believe Dr Cornish has misunderstood our discussion point regarding needle alignment for interscalene block. We stated that the out-of-plane needle-ultrasound probe alignment technique (‘anterolateral’ needle approach) places the needle along the long axis of the plexus, and together with a Tuohy tip, this confers significant safety because it ‘virtually eliminates the possibility of intraneural needle placement’ [1]. Nowhere in the article was it ever stated (or suggested) that this ‘makes needle trauma almost impossible’. The needle approach depicted in the supplied MRI figure bears no resemblance to the aforementioned needle orientation for the out-of-plane anterolateral approach. In the large prospective series to which Dr Cornish refers [3], the blocks were performed with a range of needle alignments in relation to the target nerves (15–90°); it did not, and we never claimed it did, test the hypothesis of safety with respect to needle-nerve alignment [3]. Neurological complications related to interscalene block continue to occur in 2010, with only a few ever reported [4]; the most serious are associated with unrecognised intraneural or intracordal injection using fine calibre (22-G or finer) needles. It remains my belief that regardless of the use (or not) of the commonly accepted and used techniques to prevent intraneural or intracordal injection (ultrasound, nerve stimulation, injection pressure monitoring or an awake patient), if these blocks were performed with 18-G Tuohy needles, a needle alignment along the direction of the brachial plexus and a needle direction away from the neuraxis (i.e. an ‘anterolateral’ needle direction), serious interscalene block related neurological complications would not occur.

The 25% failure rate and technical problems Dr Cornish cites from the first reported continuous interscalene analgesia series (more than 20 years ago) [5], which used an inappropriate tangential/medial (Winnie) needle approach, have been largely resolved with the abandonment of this approach.

The pump’s manufacturers did not recommend elastomeric device refilling because there were no data to support it. Our tightly controlled laboratory study confirmed the safety of this practice provided certain safety precautions are followed [6].

Dr Cornish suggests the suprascapular/axillary (circumflex) nerve block combination is not appropriate for the patient with severe respiratory disease because it is a single-injection technique. However, we specifically stated (and referenced) that supascapular catheter placement was possible [1]. The suprascapular block pneumothorax risk was also stated and referenced [1].

When directly questioned, 7% of healthy ambulatory patients receiving continuous interscalene analgesia reported ‘mild dyspnoea’ [1]. Many of these patients reported this condition as only an inability to take a deep breath and not as breathlessness as such. As that study was conducted, advances have been made to the dose regimen such that few patients now report this symptom.

Finally, our review focused primarily on postoperative analgesia– it was not intended to cover all peri-operative management aspects of the shoulder surgery patient.

MJF has received research support from the I-Flow Corporation. No other external funding or competing interests declared. Previously posted at the Anaesthesia Correspondence website:


  1. Top of page
  2. References