| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Departments of Anesthesiology and Orthopedic Surgery, University Hospital Balgrist, Zurich, Switzerland
To the Editor:
We have concerns about the Neal et al. study (1) regarding suprascapular nerve block after nonarthroscopic shoulder surgery. First, why did Neal et al. use two different local anesthetics for performing the interscalene block (short-acting) and then the suprascapular nerve block (long-acting)? This limitation has been briefly discussed by the authors, but doubts remain since the suprascapular nerve has probably already been blocked by the ISB. The interscalene and subsequent suprascapular nerve block could have caused a double block of the suprascapular nerve. Unfortunately, the authors did not check the absence of the suprascapular nerve block after the initial ISB. To clarify this point, it would be important to know which approach was chosen for the block, since an accurate description of neither the technique of the interscalene block nor the suprascapular nerve block technique had been given. It is also surprising that the authors elicited (or looked for) paresthesias when they used the nerve stimulator. One of the advantages of using the neurostimulator technique is to avoid paresthesias and to be sure not to elicit any when injecting the drug. It is also surprising to administer such a large and—unnecessary—dose of fentanyl (3.5 µg/kg) if they had a successful single (or double) block.
The exact description of the patient position during assessment of the supra- and infrascapular test (the external rotation angle) is lacking. One has to know that the teres minor muscle function (innervated by the axillary nerve) is responsible for up to 45% of the power of external rotation and this muscle should be tested (with 100% sensitivity and 93% specificity) with the hornblower’s sign (2). The infraspinatus muscle function (innervated by the suprascapular nerve) could be tested (with 100% sensitivity and specificity) with the dropping sign. The supraspinatus muscle function (innervated by the suprascapular nerve) could be tested in upright position with the Jobe test.
Finally, we believe the authors should not have waited until the occurrence of pain to give supplemental oral analgesic, since a pain window will interfere with the preemptive analgesia concept (3). Moreover, the wound was infiltrated with long-acting local anesthetic (bupivacaine) at the end of surgery, making the evaluation of pain still more complicated.
In conclusion, we believe that so many factors could have been implicated in this study that make the validity of their results questionable.
References
Virginia Mason Medical Center, Seattle, WA
In Response:
We appreciate Karovic et al.’s interest in our study of suprascapular nerve block (SSNB) in the setting of simultaneous interscalene block (ISB) for open anterior shoulder surgery (1). Their title and conclusion suggest that indeed we are in agreement regarding the expected results of this study, even if not with the methodology used. Karovic et al.’s comment about "double block" reflects concern that one may be unable to distinguish the effects of a SSNB from those of an ISB, because the latter typically deposits local anesthetic prior to the suprascapular nerve (SSN) take-off, and thus should in itself produce SSN blockade. Their concern is precisely why we performed this study. Despite subjective observation that open shoulder surgery patients had less pain when an ISB was supplemented with a SSNB, this seemed inconsistent with aforementioned brachial plexus anatomy, plus the areas affected by anterior shoulder surgery are not primarily innervated by the SSN. The fact that patients randomized to SSNB experienced a moderate, but widely variable, prolongation of analgesia suggests to us that perhaps the acromioclavicular joint, subacromial bursa, and coracoclavicular ligament, which all receive small contributions from the SSN, derived some degree of analgesic benefit from the SSNB, but not enough to significantly affect overall outcome.
Karovic et al. noted some of the challenges we faced in designing this study. We will first address methodological criticisms that could have potentially affected our results. An intermediate-acting local anesthetic (mepivacaine) was chosen for the ISB because we wanted any effects of this approach to have dissipated prior to assessing effects of the SSNB (bupivacaine). Otherwise, had we observed benefits from blocking the SSN, we could not have differentiated whether it was blocked by the interscalene approach or the suprascapular approach. Karovic et al. are also concerned with the potential effects of three interventions on time to first significant pain (our primary outcome measure). They rightly question the need for fentanyl 3.5 µg/kg in the setting of adequate local anesthetic block, but we chose that amount as a means of eliminating patients with evidence of failed block. Indeed, patients averaged only 1 µg/kg fentanyl, effects of which should have been dissipated well in advance of block resolution 6 to 10 h later. Second, we did not allow patients to develop significant pain prior to starting oral analgesics. Rather, patients were asked to take them at the first sign of pain (block resolution), but not before (which may have compromised recognition of block resolution). Third, incisional infiltration of subcutaneous bupivacaine would not be expected to affect deeper sources of pain or to exceed peripheral nerve block duration. Most importantly, because all three interventions were provided to both groups, we have no reason to suspect that they unduly biased our results.
Karovic et al. query three other issues that we believe have no effect on result interpretation. We regret any confusion interpreting our attempt to gather evidence of SSN block, as indeed this was a minor part of our study. To ascertain if patients randomized to receive SSNB developed infraspinatus muscle weakness, SSN function was evaluated by resistance to external rotation of the arm at 0° abduction with the forearm at 45° external rotation. This maneuver was chosen specifically to eliminate teres minor influence on external rotation, as Karovic et al. suggest would be proper. We did not test external rotation separately after ISB, as doing so would have delayed operating room turnover by 15 min (estimated onset time for ISB). Ultimately we were not entirely confident in our ability to distinguish ISB from SSNB effects, or local anesthetic block from preexisting shoulder pathology. We reported the results only because there did appear to be a greater range of strength in that 25th to 75th percentile of patients who received a sham SSNB. Karovic et al. apparently missed our references to full description of the Moore technique for SSNB and Winnie technique for ISB. We suspect they also misinterpreted our use of peripheral nerve stimulation (PNS). Although PNS was our preferred technique for ISB, if an unintentional paresthesia presented itself prior to a motor response, we accepted it as an endpoint. This decision is documented in our manuscript (1) as being consistent with reports of successful shoulder surgery following either paresthesia or motor response to the anterior shoulder or arm.
In summary, we appreciate and understand Karovic et al.’s concerns regarding the methodologies chosen for our study, as we too struggled with the best methods to evaluate our hypothesis. Their letter reminds us of an occasional author frustration—that inadequately describing the trees can make it difficult to see the forest.
Reference
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
