- Email: [email protected]
COMPARISON OF TWO METHODS OF AXILLARY BRACHIAL PLEXUS ANAESTHESIA
Br. J. Anaesth. (1988), 60, 841-844
COMPARISON OF TWO METHODS OF AXILLARY BRACHIAL PLEXUS ANAESTHESIA M. S. YOUSSEF AND D. A. DESGRAND
PATIENTS AND METHODS
Sixty patients older than 18 yr undergoing upper limb surgery were studied. Approval for the study was given by the local ethics committee and informed consent was obtained from each patient before entering the trial. The patients were allocated randomly to one of two groups. Brachial plexus blocks were performed using either the trans-arterial (group A) or the Winnie (group B) approaches. All patients were premedicated with temazepam 10-20 mg and metoclopramide 10 mg. On arrival of the patient in the operating theatre an indwelling cannula was placed in a vein in the arm not requiring surgery. The other arm was abducted to about 90°, with the hand resting on a pillow next to the patient's head. The axillary artery was identified as far proximally as possible. For both blocks, an extension set primed with local anaesthetic was connected to the needle. M . S. YoUSSEFfj M.B., CH.B., F.F.A.R.C.S.I., F.F.A.R.C.S.; D . A . DESGRAKD, M.B., CH.B., F.F.A.R.C.S. ; Queen Alexandra Hospi-
tal, Cosham, Portsmouth PO6 3LY. Accepted for Publication: November 11, 1987. t Present address, for correspondence: Anaesthetic Department, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF.
SUMMARY Comparison of the trans-arterial and Winnie techniques of axillary brachial plexus block was made in two groups of patients. This study did not find any statistical difference between the two techniques.
Group A The index and middle fingers of one hand were pressed over the artery as near to the anterior axillary fold as possible and a 23-S.W. gauge needle inserted between them into the artery. Aspiration was performed by an assistant continuously as the needle passed through the artery and once aspiration of blood no longer occurred, the solution was injected. After completion of the injection, the needle was quickly withdrawn and the arm was brought down to the patient's side while maintaining firm digital pressure over the puncture site for 5 min to prevent haematoma. Group B The index finger was placed over the artery and a 22-gauge B & D regional block needle was inserted immediately above the finger at an angle of 20° to the skin. It was directed parallel and just next to the artery. The needle was advanced slowly until a snap was felt as the needle penetrated the sheath or paraesthesiae were obtained. The syringe was aspirated and if no blood was obtained the solution was injected slowly. As soon as the injection was completed, the needle was withdrawn and the arm was brought down. Firm digital pressure was directly applied behind the needle during and immediately after the injection, to encourage cephalad spread of local anaesthetic. If blood was aspirated, the patient was excluded from the trial.
Downloaded from http://bja.oxfordjournals.org/ at University of Victoria on April 25, 2015
Axillary brachial plexus block is a simple method of anaesthetizing the arm. It carries no risk of pneumothorax. Commonly used by anaesthetists who perform brachial plexus blocks occasionally, its success depends on the placement of the tip of the needle inside the axillary sheath. However, this is not always easy to achieve. The transarterial approach offers a definitive end-point for the position of the needle. This technique is described in several textbooks [1,2] and widely practised, but its results have never been evaluated in the literature. It was proposed to compare the trans-arterial approach with the more established technique advocated by Winnie [3].
842
BRITISH JOURNAL OF ANAESTHESIA from group A as a result of difficulty in penetrating the artery and the other, from group B, because of accidental puncture of the artery. Student's t test was used to compare the mean age, weight and height of the patients and Fisher's exact probability test was used to compare both the failure rate and the frequency of blocked nerves in all patients. RESULTS
In both groups 1 % prilocaine solution 40 ml was injected slowly over 2 min with repeated aspiration to detect and avoid intravascular injection. All blocks were performed randomly by the authors and then assessed by one of two independent observers, who did not know which technique was used. The sensory changes were assessed 40 min after the injection. Sensory loss was assessed with a 25-S.W. gauge needle using a scale 0-2 (0 = no sensory loss; 1 = loss of pinprick; 2 = loss of touch). Sensory testing was carried out in the areas supplied by the following nerves: axillary, musculocutaneous, radial, median, ulnar, medial cutaneous nerve of forearm and medial cutaneous nerve of arm (fig. 1). For the sake of simplicity, sensory blockade grades 1 and 2 have been combined together. We used definitions described by Vester-Andersen and colleagues [4,5]: a failed block was one where the sensory blockade involved only a single or none of the following nerves: median, ulnar, radial or musculocutaneous. Successful blocks were classified as either complete (loss of sensation in areas supplied by all cutaneous nerves peripheral to the axillary nerve), or incomplete. Two patients were excluded from the trial: one
TABLE I. Characteristics of patients (mean±SEM)
Group A (n = 29) Group B (n = 29)
Age (yr)
Weight (kg)
Height (cm)
42.8±4.1
68.9 + 2.0
163.0+1.6
43.1+2.7
64.9 ±1.7
167.0±1.7
TABLE II. Results of axillary brachial plexus anaesthesia
Complete block Incomplete block Total failure rate
Group A (n = 29)
Group B (n = 29)
14(48%) 9 (31 %) 6 (21 %)
17(59%) 9 (31 %) 3(10%)
TABLE III. Frequency of blockade of each cutaneous nerve
Group A (« = 29) Nerve Axillary Musculocutaneous Radial Median Ulnar Medial cutaneous, forearm Medial cutaneous, arm
No.
0/
/o
Group B (n = 29) No.
0/
/o
23 17 23 17 21 22
79.3 58.6 79.3 58.6 72.4 75.9
16 21 21 24 23 25
55.2 72.4 72.4 82.7 79.3 86.2
20
70.0
25
86.2
Downloaded from http://bja.oxfordjournals.org/ at University of Victoria on April 25, 2015
FIG. 1. Cutaneous innervation of the upper limb. 1 = Axillary nerve; 2 = musculocutaneous nerve; 3 = radial nerve; 4 = median nerve; 5 = ulnar nerve; 6 = medial cutaneous nerve of forearm; 7 = medial cutaneous nerve of arm.
The mean ages, weights and heights of the 29 remaining patients in the two groups were comparable (table I). The results of axillary brachial plexus anaesthesia in each group are given in table II. The level of incomplete block (31 %) was comparable to that of other reported series [4]. Six patients in group A had a failed block, but only three in group B. The difference was not statistically significant.
AXILLARY BRACHIAL PLEXUS ANAESTHESIA TABLE IV. General or additional local anaesthesia required before surgery
Group A Group B (fi = 29) (n = 29) General anaesthesia Nerve block Local infiltration
8 2 4
6 3 0
DISCUSSION
In this study we did not demonstrate any obvious advantage of one technique over the other. The success rate of 90 % in group B was similar to the results (94 % success) for the Winnie approach to the axillary brachial plexus block described by several investigators [5-7]. Although there was no significant difference between the two techniques, the failure rate was twice as high in group A as in group B. The lower success rate in group A (79%) was surprising, as the end-point for injection is clearly defined. We can only assume that the needle was inserted too deeply once it had passed through the artery. A steady hand is needed to control accurately the movement of the needle as it passes through the artery. We have found this easiest to achieve seated, with the arm manipulating the needle resting on afirmsurface. Mastery of any particular technique of brachial plexus block is related to regular practice. Hudon and Jacques [8] reported an increase in their success rate in the later part of a series of 165 blocks, compared with the beginning. It is a common occurrence to find unblocked nerves with axillary brachial plexus anaesthesia and in this respect our results from both groups were not unusual. De Jong [9] and Lanz, Theiss
and Jankovic [10] found the musculocutaneous, axillary and radial nerves were most likely to remain unblocked. In this study, it was found that the axillary nerve was more frequently blocked in group A (79.3 %) than in group B (55.2 %) (P = 0.08). This is probably explained by the fact that the axillary nerve is situated deep to the artery, where local anaesthetic agents are deposited in the trans-arterial approach. A lower frequency of median nerve blockade was recorded in group A (58.6%) than in group B (82.7%). Although this was not statistically significant (P = 0.07) it may have caused clinical problems requiring conversion to general anaesthesia. To improve the chances of blocking the median nerve, one could withdraw the needle to lie just superficial to the axillary artery and deposit some local anaesthetic there. Whatever technique is used, unblocked nerves will require either additional local infiltration or peripheral nerve block. All brachial plexus blocks have potential complications. For the axillary approach, haematoma formation [11], vascular spasm [12] and inadvertent intravascular injection [13] have all been reported previously. Although none of these problems occurred in our study, it is obvious that the trans-arterial technique carries a higher theoretical risk of vascular complications. We attempted to minimize haematoma formation by maintaining firm pressure for at least 5 min after the withdrawal of the needle. We also used an immobile needle technique [14] in an attempt to reduce the risk of intravascular injection. This allowed the anaesthetist to immobilize the needle while the assistant manipulated the syringe. Injection was slow, with repeated aspiration to detect minor needle displacement. We conversed with the patient during the injection in order to detect early CNS toxicity. Winnie [14] advocate that, when his technique is used, correct positioning of the needle inside the axillary sheath should be determined by seeking paraesthesia or a snap as the needle penetrates the axillary sheath. A problem may arise when neither occur (one patient in group B). Searching for paraesthesia may increase the risk of nerve damage [15], although this has been contested in a recent review [16]. Paraesthesia is also unpleasant for the patient and difficult to elicit in heavily premedicated patients. A snap was felt in 72% of patients in group B. Of the remainder, paraesthesia was encountered accidently in two and searched for and found in five.
Downloaded from http://bja.oxfordjournals.org/ at University of Victoria on April 25, 2015
Table III shows frequencies of the blockade of individual nerves following the two approaches. The difference did not reach statistical significance. In group A (table IV), eight patients received a general anaesthetic: six for a failed block and two because of extreme anxiety. In addition, six required local infiltration for a peripheral nerve block. In group B six patients received a general anaesthetic; three for a failed block and three for incomplete block. In addition three patients required a peripheral nerve block.
843
844
BRITISH JOURNAL OF ANAESTHESIA
The use of a short bevel needle makes the snap more easy to feel and should decrease the risk of nerve damage [17].
Downloaded from http://bja.oxfordjournals.org/ at University of Victoria on April 25, 2015
5. Vester-Andersen T, Christiansen C, Sorensen M, Kaalund Jorgensen HO. Sauugbjerg P, Schultz-Moller K. Perivascular axillary block II: Influence of injected volume of local anaesthetic on neural blockade. Ada Anaesthesiologica Scandinavica 1983; 27: 95-98. In conclusion, the trans-arterial approach to 6. Hollmen A. Axillary brachial block. A double blind study of 59 cases using mepivacaine and Lac 43. Acta Anaesbrachial plexus anaesthesia with its definitive endthesiologica Scandinavica 1966; (Suppl.) 21: 53-65. point, is an easy and safe technique. It can also be 7. Selander D. Catheter technique in axillary plexus block. self taught. However, haematoma, vascular inActa Anaesthesiologica Scandinavica 1977; 21: 324-329. sufficiency and inadvertent intravascular injection 8. Hudon F, Jacques A. Block of the brachial plexus by the axillary route. Canadian Anaesthetists Society Journal are all potential complications. We feel that a 1959;6: 400-405. larger series is needed to evaluate its relative 9. De Jong RH. Modified axillary block with block of merits before an attempt is made to encourage its the lateral antebrachial cutaneous (terminal musculo wider use. cutaneous) nerve. Anesthesiology 1965; 26: 615-618. 10. Lanz E, Theiss D, Janovic DD. The extent of blockade following various techniques of brachial plexus block. Anesthesia and Analgesia 1983; 62: 55-58. ACKNOWLEDGEMENTS 11. Bosomworth PP, Egbert LD, Hamelberg W. Block of the brachial plexus in the axilla: its value and complications. We thank Mr I. T. A. Jeffrey for permission to perform this Annals of Surgery 1961; 154: 911-914. study on his patients and Dr D. Saunders for his advice on the statistical analysis, and Drs P. Spreadbury and G. Smith for 12. Merrill DG, Brodsky JB, Hentz RV. Vascular insufficiency following axillary block of the brachial plexus. their advice on the preparation of the manuscript. Anesthesia and Analgesia 1981; 60: 162-164. 13. Moore DC, Bridenbaugh LD, Eather KF. Block of the upper extremity: Supraclavicular approach versus axillary approach. Archives of Surgery 1965; 90: 68-72. REFERENCES 14. Winnie AP. An "Immobile needle" for nerve blocks. 1. Moore DC. Regional Block. 4th edn. Springfield, I. L.: Anesthesiology 1969; 31: 577-578. Charles C. Thomas, 1976; 243-256. 15. Selander D, Edshage S, Wolff T. Paresthesia or no 2. Cousins MJ, Bridenbaugh PO. Neural Blockade. Philaparesthesia? Acta Anaesthesiologica Scandinavica 1979; 3: delphia: J. B. Lippincott Company, 1980; 296-319. 27-33. 16. Shanahan PT, Keinert HE. Anesthesia management of 3. Winnie AP. Perivascular techniques of brachial plexus upper extremity replantation surgery. Anesthesiology block. In: Plexus Anesthesia, Vol. 1. Edinburgh: Churchill Reviews 1983; 10: 10-22. Livingstone, 1984; 121-143. 17. Selander D, Dhuner KG, Lundborg G. Peripheral nerve 4. Vester-Andersen T, Christiansen C, Sorense M, Eriksen injury due to injection needles used for regional C. Perivascular axillary block 1: blockade following 40 ml anaesthesia. Acta Anaesthesiologica Scandinavica 1977; 1 % mepivacaine with adrenaline. Acta Anaesthesiologica 221: 182-188. Scandinavica 1982; 26: 519-523.
COMPARISON OF TWO METHODS OF AXILLARY BRACHIAL PLEXUS ANAESTHESIA M. S. YOUSSEF AND D. A. DESGRAND
PATIENTS AND METHODS
Sixty patients older than 18 yr undergoing upper limb surgery were studied. Approval for the study was given by the local ethics committee and informed consent was obtained from each patient before entering the trial. The patients were allocated randomly to one of two groups. Brachial plexus blocks were performed using either the trans-arterial (group A) or the Winnie (group B) approaches. All patients were premedicated with temazepam 10-20 mg and metoclopramide 10 mg. On arrival of the patient in the operating theatre an indwelling cannula was placed in a vein in the arm not requiring surgery. The other arm was abducted to about 90°, with the hand resting on a pillow next to the patient's head. The axillary artery was identified as far proximally as possible. For both blocks, an extension set primed with local anaesthetic was connected to the needle. M . S. YoUSSEFfj M.B., CH.B., F.F.A.R.C.S.I., F.F.A.R.C.S.; D . A . DESGRAKD, M.B., CH.B., F.F.A.R.C.S. ; Queen Alexandra Hospi-
tal, Cosham, Portsmouth PO6 3LY. Accepted for Publication: November 11, 1987. t Present address, for correspondence: Anaesthetic Department, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF.
SUMMARY Comparison of the trans-arterial and Winnie techniques of axillary brachial plexus block was made in two groups of patients. This study did not find any statistical difference between the two techniques.
Group A The index and middle fingers of one hand were pressed over the artery as near to the anterior axillary fold as possible and a 23-S.W. gauge needle inserted between them into the artery. Aspiration was performed by an assistant continuously as the needle passed through the artery and once aspiration of blood no longer occurred, the solution was injected. After completion of the injection, the needle was quickly withdrawn and the arm was brought down to the patient's side while maintaining firm digital pressure over the puncture site for 5 min to prevent haematoma. Group B The index finger was placed over the artery and a 22-gauge B & D regional block needle was inserted immediately above the finger at an angle of 20° to the skin. It was directed parallel and just next to the artery. The needle was advanced slowly until a snap was felt as the needle penetrated the sheath or paraesthesiae were obtained. The syringe was aspirated and if no blood was obtained the solution was injected slowly. As soon as the injection was completed, the needle was withdrawn and the arm was brought down. Firm digital pressure was directly applied behind the needle during and immediately after the injection, to encourage cephalad spread of local anaesthetic. If blood was aspirated, the patient was excluded from the trial.
Downloaded from http://bja.oxfordjournals.org/ at University of Victoria on April 25, 2015
Axillary brachial plexus block is a simple method of anaesthetizing the arm. It carries no risk of pneumothorax. Commonly used by anaesthetists who perform brachial plexus blocks occasionally, its success depends on the placement of the tip of the needle inside the axillary sheath. However, this is not always easy to achieve. The transarterial approach offers a definitive end-point for the position of the needle. This technique is described in several textbooks [1,2] and widely practised, but its results have never been evaluated in the literature. It was proposed to compare the trans-arterial approach with the more established technique advocated by Winnie [3].
842
BRITISH JOURNAL OF ANAESTHESIA from group A as a result of difficulty in penetrating the artery and the other, from group B, because of accidental puncture of the artery. Student's t test was used to compare the mean age, weight and height of the patients and Fisher's exact probability test was used to compare both the failure rate and the frequency of blocked nerves in all patients. RESULTS
In both groups 1 % prilocaine solution 40 ml was injected slowly over 2 min with repeated aspiration to detect and avoid intravascular injection. All blocks were performed randomly by the authors and then assessed by one of two independent observers, who did not know which technique was used. The sensory changes were assessed 40 min after the injection. Sensory loss was assessed with a 25-S.W. gauge needle using a scale 0-2 (0 = no sensory loss; 1 = loss of pinprick; 2 = loss of touch). Sensory testing was carried out in the areas supplied by the following nerves: axillary, musculocutaneous, radial, median, ulnar, medial cutaneous nerve of forearm and medial cutaneous nerve of arm (fig. 1). For the sake of simplicity, sensory blockade grades 1 and 2 have been combined together. We used definitions described by Vester-Andersen and colleagues [4,5]: a failed block was one where the sensory blockade involved only a single or none of the following nerves: median, ulnar, radial or musculocutaneous. Successful blocks were classified as either complete (loss of sensation in areas supplied by all cutaneous nerves peripheral to the axillary nerve), or incomplete. Two patients were excluded from the trial: one
TABLE I. Characteristics of patients (mean±SEM)
Group A (n = 29) Group B (n = 29)
Age (yr)
Weight (kg)
Height (cm)
42.8±4.1
68.9 + 2.0
163.0+1.6
43.1+2.7
64.9 ±1.7
167.0±1.7
TABLE II. Results of axillary brachial plexus anaesthesia
Complete block Incomplete block Total failure rate
Group A (n = 29)
Group B (n = 29)
14(48%) 9 (31 %) 6 (21 %)
17(59%) 9 (31 %) 3(10%)
TABLE III. Frequency of blockade of each cutaneous nerve
Group A (« = 29) Nerve Axillary Musculocutaneous Radial Median Ulnar Medial cutaneous, forearm Medial cutaneous, arm
No.
0/
/o
Group B (n = 29) No.
0/
/o
23 17 23 17 21 22
79.3 58.6 79.3 58.6 72.4 75.9
16 21 21 24 23 25
55.2 72.4 72.4 82.7 79.3 86.2
20
70.0
25
86.2
Downloaded from http://bja.oxfordjournals.org/ at University of Victoria on April 25, 2015
FIG. 1. Cutaneous innervation of the upper limb. 1 = Axillary nerve; 2 = musculocutaneous nerve; 3 = radial nerve; 4 = median nerve; 5 = ulnar nerve; 6 = medial cutaneous nerve of forearm; 7 = medial cutaneous nerve of arm.
The mean ages, weights and heights of the 29 remaining patients in the two groups were comparable (table I). The results of axillary brachial plexus anaesthesia in each group are given in table II. The level of incomplete block (31 %) was comparable to that of other reported series [4]. Six patients in group A had a failed block, but only three in group B. The difference was not statistically significant.
AXILLARY BRACHIAL PLEXUS ANAESTHESIA TABLE IV. General or additional local anaesthesia required before surgery
Group A Group B (fi = 29) (n = 29) General anaesthesia Nerve block Local infiltration
8 2 4
6 3 0
DISCUSSION
In this study we did not demonstrate any obvious advantage of one technique over the other. The success rate of 90 % in group B was similar to the results (94 % success) for the Winnie approach to the axillary brachial plexus block described by several investigators [5-7]. Although there was no significant difference between the two techniques, the failure rate was twice as high in group A as in group B. The lower success rate in group A (79%) was surprising, as the end-point for injection is clearly defined. We can only assume that the needle was inserted too deeply once it had passed through the artery. A steady hand is needed to control accurately the movement of the needle as it passes through the artery. We have found this easiest to achieve seated, with the arm manipulating the needle resting on afirmsurface. Mastery of any particular technique of brachial plexus block is related to regular practice. Hudon and Jacques [8] reported an increase in their success rate in the later part of a series of 165 blocks, compared with the beginning. It is a common occurrence to find unblocked nerves with axillary brachial plexus anaesthesia and in this respect our results from both groups were not unusual. De Jong [9] and Lanz, Theiss
and Jankovic [10] found the musculocutaneous, axillary and radial nerves were most likely to remain unblocked. In this study, it was found that the axillary nerve was more frequently blocked in group A (79.3 %) than in group B (55.2 %) (P = 0.08). This is probably explained by the fact that the axillary nerve is situated deep to the artery, where local anaesthetic agents are deposited in the trans-arterial approach. A lower frequency of median nerve blockade was recorded in group A (58.6%) than in group B (82.7%). Although this was not statistically significant (P = 0.07) it may have caused clinical problems requiring conversion to general anaesthesia. To improve the chances of blocking the median nerve, one could withdraw the needle to lie just superficial to the axillary artery and deposit some local anaesthetic there. Whatever technique is used, unblocked nerves will require either additional local infiltration or peripheral nerve block. All brachial plexus blocks have potential complications. For the axillary approach, haematoma formation [11], vascular spasm [12] and inadvertent intravascular injection [13] have all been reported previously. Although none of these problems occurred in our study, it is obvious that the trans-arterial technique carries a higher theoretical risk of vascular complications. We attempted to minimize haematoma formation by maintaining firm pressure for at least 5 min after the withdrawal of the needle. We also used an immobile needle technique [14] in an attempt to reduce the risk of intravascular injection. This allowed the anaesthetist to immobilize the needle while the assistant manipulated the syringe. Injection was slow, with repeated aspiration to detect minor needle displacement. We conversed with the patient during the injection in order to detect early CNS toxicity. Winnie [14] advocate that, when his technique is used, correct positioning of the needle inside the axillary sheath should be determined by seeking paraesthesia or a snap as the needle penetrates the axillary sheath. A problem may arise when neither occur (one patient in group B). Searching for paraesthesia may increase the risk of nerve damage [15], although this has been contested in a recent review [16]. Paraesthesia is also unpleasant for the patient and difficult to elicit in heavily premedicated patients. A snap was felt in 72% of patients in group B. Of the remainder, paraesthesia was encountered accidently in two and searched for and found in five.
Downloaded from http://bja.oxfordjournals.org/ at University of Victoria on April 25, 2015
Table III shows frequencies of the blockade of individual nerves following the two approaches. The difference did not reach statistical significance. In group A (table IV), eight patients received a general anaesthetic: six for a failed block and two because of extreme anxiety. In addition, six required local infiltration for a peripheral nerve block. In group B six patients received a general anaesthetic; three for a failed block and three for incomplete block. In addition three patients required a peripheral nerve block.
843
844
BRITISH JOURNAL OF ANAESTHESIA
The use of a short bevel needle makes the snap more easy to feel and should decrease the risk of nerve damage [17].
Downloaded from http://bja.oxfordjournals.org/ at University of Victoria on April 25, 2015
5. Vester-Andersen T, Christiansen C, Sorensen M, Kaalund Jorgensen HO. Sauugbjerg P, Schultz-Moller K. Perivascular axillary block II: Influence of injected volume of local anaesthetic on neural blockade. Ada Anaesthesiologica Scandinavica 1983; 27: 95-98. In conclusion, the trans-arterial approach to 6. Hollmen A. Axillary brachial block. A double blind study of 59 cases using mepivacaine and Lac 43. Acta Anaesbrachial plexus anaesthesia with its definitive endthesiologica Scandinavica 1966; (Suppl.) 21: 53-65. point, is an easy and safe technique. It can also be 7. Selander D. Catheter technique in axillary plexus block. self taught. However, haematoma, vascular inActa Anaesthesiologica Scandinavica 1977; 21: 324-329. sufficiency and inadvertent intravascular injection 8. Hudon F, Jacques A. Block of the brachial plexus by the axillary route. Canadian Anaesthetists Society Journal are all potential complications. We feel that a 1959;6: 400-405. larger series is needed to evaluate its relative 9. De Jong RH. Modified axillary block with block of merits before an attempt is made to encourage its the lateral antebrachial cutaneous (terminal musculo wider use. cutaneous) nerve. Anesthesiology 1965; 26: 615-618. 10. Lanz E, Theiss D, Janovic DD. The extent of blockade following various techniques of brachial plexus block. Anesthesia and Analgesia 1983; 62: 55-58. ACKNOWLEDGEMENTS 11. Bosomworth PP, Egbert LD, Hamelberg W. Block of the brachial plexus in the axilla: its value and complications. We thank Mr I. T. A. Jeffrey for permission to perform this Annals of Surgery 1961; 154: 911-914. study on his patients and Dr D. Saunders for his advice on the statistical analysis, and Drs P. Spreadbury and G. Smith for 12. Merrill DG, Brodsky JB, Hentz RV. Vascular insufficiency following axillary block of the brachial plexus. their advice on the preparation of the manuscript. Anesthesia and Analgesia 1981; 60: 162-164. 13. Moore DC, Bridenbaugh LD, Eather KF. Block of the upper extremity: Supraclavicular approach versus axillary approach. Archives of Surgery 1965; 90: 68-72. REFERENCES 14. Winnie AP. An "Immobile needle" for nerve blocks. 1. Moore DC. Regional Block. 4th edn. Springfield, I. L.: Anesthesiology 1969; 31: 577-578. Charles C. Thomas, 1976; 243-256. 15. Selander D, Edshage S, Wolff T. Paresthesia or no 2. Cousins MJ, Bridenbaugh PO. Neural Blockade. Philaparesthesia? Acta Anaesthesiologica Scandinavica 1979; 3: delphia: J. B. Lippincott Company, 1980; 296-319. 27-33. 16. Shanahan PT, Keinert HE. Anesthesia management of 3. Winnie AP. Perivascular techniques of brachial plexus upper extremity replantation surgery. Anesthesiology block. In: Plexus Anesthesia, Vol. 1. Edinburgh: Churchill Reviews 1983; 10: 10-22. Livingstone, 1984; 121-143. 17. Selander D, Dhuner KG, Lundborg G. Peripheral nerve 4. Vester-Andersen T, Christiansen C, Sorense M, Eriksen injury due to injection needles used for regional C. Perivascular axillary block 1: blockade following 40 ml anaesthesia. Acta Anaesthesiologica Scandinavica 1977; 1 % mepivacaine with adrenaline. Acta Anaesthesiologica 221: 182-188. Scandinavica 1982; 26: 519-523.