Current teaching of regional anaesthesia recommends establishment of regional blockade in awake patients in order to minimize the risk of nerve damage. There is no firm data to confirm this point of view relating to interscalene block. The publication by Benumof  pointed out several factors possibly contributing to the increased risk of interscalene block. Even though similar complications were described previously [2,3] the authors tried to analyse the causative factors of such complications. They believe that general anaesthesia should be considered as a relative contraindication to interscalene block. This opinion contradicts routine practice adopted in our hospital, where interscalene blocks for shoulder surgery as a rule are performed on anaesthetized patients. The necessity to stay awake for the block may cause discomfort to some patients (especially in countries where patients traditionally expect to have all unpleasant procedures performed under general anaesthesia). Some of them may subsequently refuse to have regional block in spite of clear clinical benefits.
Materials and methods
We analysed retrospectively 548 cases of interscalene block performed in our hospital over the last 3 yr for arthroscopic shoulder surgery. Two anaesthetists, experienced in this particular technique, performed 83 % of these blocks. Trainee anaesthetists performed the remainder under supervision.
All patients were anaesthetized using propofol and fentanyl for induction followed by spontaneous respiration with nitrous oxide, oxygen and sevoflurane or isoflurane via a laryngeal mask.
Interscalene block was performed after induction of general anaesthesia, but before the start of surgery, using the modified Winnie approach, with a nerve stimulator and 22-G 50-mm stimulating needle (B. Braun Melsungen AG, Melsungen, Germany). A current of 1.5 mA at 2 Hz was used to identify the nerve trunk either by contraction of the biceps (musculocutaneous nerve) or deltoid muscle (axillary nerve). The needle tip was positioned to lose clinical signs of contractions at a threshold of 0.4-0.5 mA.
Bupivacaine 0.5% 20 mL or ropivacaine 0.75% 30 mL was injected after a negative aspiration test. The choice of local anaesthetic was at the discretion of the anaesthetist. Patients were positioned in the semi-recumbent position with the head and upper neck fixed for the duration of surgery.
Postoperatively patients were assessed for adequacy of pain relief and analgesics were administered as required, but there was no formal pain scoring system used. Patients remained in hospital overnight before re-assessment to establish that the block had worn off and that the patients fitted the criteria for discharge.
Detailed clinical assessment of patients was performed within the period from 4 to 8 weeks after surgery by the surgeon in order to establish the clinical outcome and also to reveal any signs of possible neurological damage related to the interscalene block.
Postoperative examination of patients conducted by the surgeon within the period of 4-8 weeks postoperation failed to reveal any permanent or long-lasting neurological complication attributable to interscalene block. Examination was carried out on a clinical basis without the use of specialized techniques.
124 patients received ropivacaine; 424 received bupivacaine. The procedures performed are shown in Table 1. Duration of the block was defined as the time from the moment of completion of the block to the moment of return of pain sensation and was 13.26 ± 7.86 and 11.87 ± 6.75 h in the ropivacaine and bupivacaine groups, respectively.
24% of patients required no additional pain relief, 63% required simple analgesics (paracetamol, diclofenac) and 13% required opiates for analgesia within 24 h postoperatively. It was noted that the success rate for anaesthetists involved with shoulder surgery on a regular basis was 96%, whereas for those doing this type of anaesthesia only occasionally it was as low as 80%. The block was considered to be a failure if there was a requirement for additional analgesia within 4 h of the procedure.
Five patients had to stay in hospital for more then 24 h due to medical problems unrelated to the surgery or anaesthesia. These were re-stabilization of insulin dependent diabetes in four cases and management of labile hypertension in one case.
It is believed that when an interscalene block is performed after induction of general anaesthesia, there may be an increased risk of epidural, intrathecal or intracord injection of local anaesthetic or a higher risk of damage to the brachial plexus. The first point was illustrated by Benumof , who presented a report describing four cases of permanent loss of cervical spinal cord function when interscalene block was performed on anaesthetized or heavily sedated patients. In all four cases the length of the stimulating needles used was 37.5-50 mm. A description of the anaesthetic technique was given in only one case where the interscalene block was performed by ‘walking’ the needle off the posterior aspect of the C6 transverse process. The direction of the stimulating needle was not mentioned in any of these cases.
Benumof believes that if the neurological damage was caused by intracord injection of local anaesthetic, then the following measures should be taken in order to minimize the risk of such a dramatic complication:
- General anaesthesia should be considered a relative contraindication for interscalene block.
- Stimulating needles longer than 37.5 mm should not be used.
- The stimulating needle should have definite caudal direction at C6 level.
- The physician should ensure that the patient does not unexpectedly move.
- In obese patients landmarks may be obscure.
While we agree with the points about the caudal direction of the needle (and indeed we believe that this is the key point for a safe interscalene block), distortion of anatomical landmarks in obese patients, the prevention of unexpected patient movements and the other points are debatable.
The technique of interscalene block as described by Winnie  used a medial, dorsal and slightly caudal direction of the needle. With the passage of time the caudal direction was largely ignored and it became routine practice to direct the needle nearly perpendicular to the skin of the neck.
In our hospital we use a modified approach, with the needle at the level of C6 directed caudally and laterally at an angle of approximately 30-40° to the spine. No attempt is made to identify the transverse process or to make contact with the bone. The needle is directed laterally towards the apex of the axilla beneath a finger identifying the interscalene grove (Fig. 1).
Using the above technique, the direction of the needle has primary importance over other factors. It is our opinion that general anaesthesia is not a relative contraindication to interscalene block because the direction of the needle prevents inadvertent penetration of the epidural or subarachnoid spaces as well as direct needling of the spinal cord. The length of the needle also becomes less important and 50-mm stimulating needles can safely be used. General anaesthesia also prevents sudden movements of the patient during the block, thus making it easier to perform and possibly safer. In our study Horner's syndrome was not considered to be a complication because it rarely reached clinical significance. Phrenic nerve block occurs in 100% of cases  during interscalene blocks and is believed to be a sign of a successful block, rather than a complication.
The second point, a potentially increased risk of damage to the brachial plexus has also not been clarified by relevant studies thus far and remains uncertain. Nevertheless several authors have expressed the opinion that an interscalene block is not the major factor in the development of neurological complications after shoulder surgery. McFarland and colleagues  in their review of complications of shoulder arthroscopy state that shoulder arthroscopy itself can cause neurological injury to the axillary, median, radial, ulnar and musculocutaneous nerves. The majority of injuries are a neuropraxis. McFarland quotes the rate of this complication as from 0% to 30% according to different authors. The paper lists direct injury to the nerves due to the placement of the endoscope portal, strain injury from traction and positioning errors as possible causes. Broadman and Cofield  believe that the majority of neurological complications after shoulder surgery are related to the surgical procedure itself (such as shoulder replacement), rather than to interscalene block.
The study of long-term complications related to interscalene block also revealed a low complication rate. Borgeat and colleagues  in their study of 521 cases (both single shot and catheter technique) state the rate of short- and long-term complications as 0.4%. In two cases they observed long-lasting neurological sequelae, though the authors did not specify what kind of surgery was performed in these cases.
Interscalene block is not, therefore, the only factor nor the most likely in nerve injury after shoulder surgery. Fisher in his editorial  points out that ‘the major factors in preventing neurological sequelae are the experience and dexterity of the anaesthetist, proper training and supervision of trainees, detailed assessment of pre-existing vascular and neurological diseases, avoidance of repeated attempts due to technical difficulties’. In the same article he quotes unpublished data from his department, where 85% of 9000 central and 7000 peripheral blocks were performed on anaesthetized patients without any serious neurological sequelae.
In our study of 548 cases of arthroscopic shoulder surgery with interscalene block we failed to reveal any short- or long-term neurological complications. We believe that our result may be explained by several factors:
- The modified Winnie approach was used as a routine technique.
- Only anaesthetists experienced in interscalene block were involved and trainees were supervised at all times.
- Arthroscopic operations (rather than open operative procedures) were included in our study.
The presence of a control group in our study certainly would have added more weight to the conclusion we reached. But this particular practice of anaesthesia for shoulder surgery has been adopted in our hospital for considerable period of time (more than 10 yr). We analysed 548 consecutive cases for only 3-yr period and failed to find any major complications related to interscalene block in this group of patients.
In conclusion, we believe that modification of Winnie's original technique allows the interscalene block to be performed safely on anaesthetized patient. Modification of the technique rather than the timing of regional block in relationship to general anaesthesia is the main factor in reducing the risk of serious complications. There is, however, a need in further studies on the subject.
1. Benumof JL. Permanent loss of cervical spinal cord function associated with interscalene
block performed under general anesthesia. Anesthesiology
2. Passannante AN. Spinal anesthesia and permanent neurological deficit after interscalene
block. Anesth Analg
3. Gologorsky E, Leanza RF. Contralateral anesthesia following interscalene
block. Anesth Analg
4. Winnie AP. Interscalene
brachial plexus block. Anesth Analg
5. Urmey WF, Talts KH, Sharrock NE. One hundred percent incidence of hemidiaphragmatic paresis associated with interscalene
brachial plexus anesthesia as diagnosed by ultrasonography. Anesth Analg
6. McFarland EG, O'Neil OR, Hsu CY. Complications
arthroscopy. J South Orthop Assoc
7. Broadman ND, Cofield RH. Neurologic complications
surgery. Clin Orthop
8. Borgeat A, Ecotodramis G, Kalberer F, Benz C. Acute and non-acute complications
associated with interscalene
block and shoulder
9. Fisher H. Regional anaesthesia
- before or after general? Anaesthesia