| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Department of Anesthesia and Critical Care, Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts; Department of Anesthesiology and Perioperative Medicine, Medical College of Georgia, Augusta, Georgia
Address correspondence and reprint requests to Carl Rosow, MD, PhD, Department of Anesthesia and Critical Care, Massachusetts General Hospital, 55 Fruit St., Boston, MA 02114. Address e-mail to crosow{at}partners.org.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Regardless of the specific approach used, IC block is frequently performed by using electrical stimulation and observing the activation of specific muscle groups (6). The cord first encountered (lateral versus posterior versus medial) depends on the position of the needle with respect to the neural bundle. Most published studies report that a single injection of local anesthetic produces an adequate block (5,7–10), although one group reported frequent failures using this technique (11).
The present study was based on the premise that localizing the posterior cord during a single-injection IC block would place the needle centrally within the infraclavicular portion of the brachial plexus and allow an even spread of local anesthetic. We therefore hypothesized that posterior cord stimulation would be associated with more frequent block success than stimulation of the lateral or medial cords. "Success" in this case was defined as rapid onset of motor block and sensory block adequate to perform surgery.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Our standard IC block technique was similar to that described by Wilson et al. (7). After placement of an IV catheter and premedication with 1–2 mg midazolam and/or 50–100 µg fentanyl, patients were placed supine with the arm abducted to approximately 90º. The coracoid process was palpated and a point 2 cm inferior and 2 cm medial to the process identified. The skin overlying this point was cleaned and infiltrated with 1% lidocaine. A 50-mm short-bevel insulated needle connected to a neural stimulator (Stimuplex, B. Braun, Bethlehem, PA) was then inserted perpendicular to the skin. The stimulator was set to deliver rectangular direct current impulses with a frequency of 2 Hz and pulse width of 100 ms. The initial stimulator current was set at 1.0 mA. Once proximity to a cord was identified by visible contraction of an appropriate muscle group, the current was incrementally reduced, and the needle slowly inserted until muscle activity resumed. The cord was identified by observation of the specific muscles responding:
Local anesthetics were injected when either (a) muscle activity was observed at a stimulator current of 0.3 mA or less or (b) the operator felt a loss of resistance as the needle entered the plexus sheath, and a motor response was visible at a stimulator current of 0.5 mA. At least one of the motor responses listed was required before injection. Other responses were not considered acceptable unless they occurred in combination with one of those listed.
All patients in this study received 30 mL 1.5% mepivacaine with 3 mL 8.4% sodium bicarbonate, injected over 1–2 min with intermittent aspiration. This was followed immediately by 10 mL 0.75% bupivacaine injected over 1–2 min. Clinicians were not required to seek or avoid a particular cord. They were not limited as to the number of times cord stimulation could be performed, but all of the local anesthetic was injected at the same site. Fifteen min after injection, evaluation of motor and sensory function was performed to assess block of the radial, median, ulnar, musculocutaneous, medial brachial, and antebrachial cutaneous nerves. An assessment time of 15 min was chosen because it allowed ample time to detect block onset but did not unduly delay surgery. After transfer to the operating room (usually 20–30 min after block placement), any patient with a block that was felt to be inadequate for surgery was treated with supplemental local anesthetic or offered general anesthesia.
The primary independent variable was the specific cord stimulated immediately before the injection of local anesthetic. We used two primary outcome measures:
Demographic information was collected, including age, gender, height, weight, and operative procedure. We recorded the block "operator," indicating the level of training of staff performing the block (attending, resident, or attending taking over from resident). We did not specify criteria for an attending taking over a block. Assessments were made about the technical aspects of block placement (minimum stimulator current, sensing loss of resistance, elicitation of paresthesias, blood on aspiration). Because the experience of those performing blocks varied widely, we did not collect data on the total block time or the number of needle passes. Because the operative site could also influence the probability of block success, we broadly categorized the location of surgery as above the wrist versus wrist and below.
A review of our clinical database indicated that clinical success rate after posterior cord stimulation was approximately 90%, and we assumed (incorrectly) that the 3 cords of the brachial plexus would be stimulated with similar frequency. To demonstrate that stimulation of another cord resulted in a 15% decrease in success rate, a study population of 339 (113 per group) was required (
= 0.05, two-tailed; 1-ß = 0.8).
Patients were grouped depending upon the cord(s) stimulated. Demographic variables were compared using analysis of variance for continuous measurements and 
2 test for categorical variables. In most of the analysis, the block extent (number of nerves blocked) was prospectively dichotomized into 2 categories: 0–2 and 3–4. 2 test was also used to compare the rate of successful block and the rate of high block extent (3 or 4 nerves blocked) among the 4 groups. Multiple ordinal regression models were used to evaluate the potential predictors of failure and to examine the difference between study groups adjusted for confounders. Adjusted odds ratios from these models are reported.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The demographic and clinical data for all groups are presented in Table 1. There were no statistically significant differences among the groups with respect to age, gender, height, weight, body mass index, site of surgery, or operator performing the block. There was a small, but highly significant intergroup difference in minimum stimulator current (posterior>multiple>lateral>medial). Because minimum current did not subsequently prove to be a predictor of outcome, it was not a confounder in our analysis.
|
Of the 364 blocks, 322 (88.5%) were judged as clinically successful, and 243 (66.8%) had evidence of motor block of all 4 nerves at 15 min (Table 2). The risk of block failure as well as incomplete motor block was strongly predicted by the cord stimulated. Both of these poorer outcomes were significantly more likely if the lateral or medial cord was stimulated rather than the posterior cord (all with P < 0.05). There was no significant difference in outcome between posterior versus multiple cord stimulation.
|
Multiple logistic regression models were used to identify other predictors for poor outcomes and to compare study groups adjusting for these predictors. Male gender was associated with a lower extent of motor block and more frequent clinical block failure. Greater body mass index was associated with more frequent clinical block failure whereas staff take-over from a resident was associated with a lower extent of motor block. When adjustment was made for these potential confounders, posterior cord stimulation was still a highly significant predictor of greater motor block and higher block success compared with the lateral and medial cords (Table 3).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Before initiating this protocol we were concerned that a preexisting bias (in favor of seeking the posterior cord) might influence our results. Only one attending physician stated that she specifically looked for that cord, and all participants agreed that we would accept other cords. Using this needle approach, however, the posterior cord was encountered much more frequently than expected by chance. It is entirely possible that our conclusions might not apply if the needle were inserted at a different angle.
This concept of placing a needle centrally to increase the success rate of IC block is not new. For example, Borgeat et al. (5) reported a 97% rate of IC block success when nerve stimulation elicited a distal response consistent with central placement. More recently, Porter et al. (13) described three cases using ultrasound for IC block placement and speculated that injection of local anesthetic posterior to the axillary artery (also a central placement) would predict a successful block for the same reason. Ultrasound may be a direct and reliable method of confirming central placement (14), but this technique is not yet as popular as nerve stimulation and requires additional equipment and training. Given the 88.5% rate of successful block we achieved overall, it is unclear how much improvement we could achieve with ultrasound.
In summary, we have shown that stimulating the posterior cord (or multiple cords) before local anesthetic injection is associated with a more frequent success rate for IC blockade than stimulation of either the medial or lateral cord. The results must be interpreted cautiously because the study was not randomized, and the occurrence of posterior nerve stimulation was much greater than expected by chance. The potential drawbacks of searching for a specific cord have not been investigated. We do not know whether deliberately seeking the posterior cord results in a significantly longer time for block placement, increased complications from needle reinsertion, or greater patient discomfort. These factors will need to be investigated prospectively before we can uniformly recommend the technique.
The authors wish to acknowledge the support and participation of the attending and resident anesthesia staff in the Massachusetts General Hospital Same Day Surgical Unit.
|
Footnotes |
|---|
Accepted for publication January 12, 2006.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Levesque, N. Dion, and M.-C. Desgagne Endpoint for successful, ultrasoundguided infraclavicular brachial plexus block Can J Anesth, May 1, 2008; 55(5): 308 - 308. [Full Text] [PDF]
|
|
|
|
|
|
M. Morimoto, J. Popovic, J. T. Kim, H. Kiamzon, and A. D. Rosenberg Case series: Septa can influence local anesthetic spread during infraclavicular brachial plexus blocks: [Les septa peuvent influencer la diffusion de l'anesthesique local durant les blocs infraclaviculaires du plexus brachial] Can J Anesth, December 1, 2007; 54(12): 1006 - 1010. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. H. De Tran, A. Clemente, J. Doan, and R. J. Finlayson Brachial plexus blocks: a review of approaches and techniques: [Les blocs du plexus brachial : compte-rendu des approches et techniques] Can J Anesth, August 1, 2007; 54(8): 662 - 674. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Minville, O. Fourcade, B. Bourdet, M. Doherty, C. Chassery, J.-C. Pourrut, C. Gris, B. Eychennes, A. Colombani, K. Samii, et al. The Optimal Motor Response for Infraclavicular Brachial Plexus Block Anesth. Analg., February 1, 2007; 104(2): 448 - 451. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
