- Email: [email protected]
Brachial plexus stretching injuries: Microcirculation of the brachial plexus
Brachial plexus stretching injuries: Microcirculation of the brachial plexus Toshio Kitamura, MD, Katsurnasa Takagi, MD, Makio Yamaga, MD, and Keizo Morisawa, MD, Kumamota, Japan This study was undertaken to investigate the pathogenesis of brachial plexus stretching injuries at an intensity level not severe enough to cause avulsion injury. While we performed traction on 64 forelegs of 32 rats, we evaluated changes in the blood flow in the extrinsic and intrinsic microvascular systems of the brachial plexus. While we laterally stretched the brachial plexus during 80 ~ shoulder abduction, we measured the blood flow at the bifurcation of the brachial plexus and at the median nerve with the hydrogen washout technique. During weak traction the blood flow decreased markedly in the extrinsic system, causing an imbalance in the two systems. In the median nerve, however, no such imbalance occurred. On histologic examination the axon and myelin in the brachial plexus and the median nerve showed no morphologic change. However, in parts of the brachial plexus we noted hypertrophic connective tissue or granulomatous inflammation in tissue surrounding the extrinsic system. The extrinsic system's apparent susceptibility to injury by acute traction may be a factor in the pathogenesis of the brachial plexus stretching injuries. (J SHOULDER ELBOWSURG 1995;4:118-23.)
Because the brachial plexus has many branches it is clinically regarded as being sensitive to stimuli. Owing to the high mobility of the shoulder girdle adjoining it, traction of the upper limb can cause avulsion or various injuries of lesser degree. Schwartzman ~1 reported that one factor responsible for traumatic thoracic outlet syndrome is stretching of the brachial plexus at an intensity level not severe enough to cause avulsion injury. Kataoka et al/" 5 reported a neurograph of the brachial plexus showing stretching in some cases of thoracic outlet syndrome. Swift and Nichols 12 and Clein 1 reported on the droopy shoulder syndrome, in which symptoms are caused by stretching of the brachial plexus. It is little understood, however, how traction not intense enough to cause avulsion injury is involved in the onset of pain or numbness. Clinically, relaxing the stretching stimuli by passive shoulder girdle elevation improves From the Department of OrthopaedJc Surgery, Kumamoto Universib, School of Medicine, Kumamoto, Japan. Reprint requests: Toshio Kitamura, Department of Orthopaedic Surgery, Kumarnoto Universily School of Medicine, 1-1-1 Honio , Kurnamoto 860 Japan. Copyright 9 1995 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/95/$3.00 + 0 32/1/61388
118
these symptoms. 1 Therefore it is possible that functional disturbances such as changes in the blood flow influence the onset of symptoms. Peripheral nerves are well-vascularized structures with separate but extensively interconnected microvascular systems in the epineurium, perineurium, and endoneurium. An "intrinsic" vascular system is present, ft consists of vascular plexa in the epineurium, perineurium, and endoneurium. An "extrinsic" system involving segmental regional vessels approaching the nerve trunk at various levels along its course is also present. These regional vessels run in the loose "adventitia" or along the paraneural tissue surrounding the nerve, allowing the nerve trunk considerable mobility in its bed. 7' 8 Numerous reports have suggested that when the peripheral nerves are stretched, the surrounding blood vessels are extended, resulting in decreased blood flow within the nerve bundie.7, ~o, ~3 However, no report has yet been published regarding changes in the blood flow at the bifurcations of the brachial plexus during traction of the upper limbs. In this study we examined such changes of the blood flow during traction at an intensity insufficient to cause avulsion injury. We also investigated the resistance of the extrinsic and intrinsic systems to tractive force.
J. Shoulder Elbow Surg. Volume 4, Number 2
Kitamura et al.
119
Figure 1 Neurograph of rat brachial plexus. Lucentarea is nerve (arrows). Brachial plexus is stretchedand straightduring 80 ~ shoulderabduction.
METHODS Experimental animals. The C8 ramus ventrails of the brachial plexus and the median nerve at the center of the brachium of 32 Wistar rats (64 limbs) were used. The body weight of each animal was 300 to 330 gm (average 320 gin). With the use of a steel wire inserted into the radius and ulna of each rat, we pulled the upper limb in the direction of abduction. Neurographic and macroscopic observation of the brachiol plexus allowed us to confirm that traction during 80 ~ shoulder abduction caused the brachial plexus and median nerve to be pulled straight (Figure 1), so we adopted this as our model. Method of traction. While the rats were under intraperitoneal pentobarbital sodium anesthesia (5 rag/100 gin), the wire insertion sites at the radius and ulna were pulled with a constantly increasing weight (0 to 800 gm at 100 gm/60 sec). We confirmed that neither rupture nor avulsion of the brachial plexus was caused even at a load of 800 gin. Elongation. The perineurium between the brachial plexus and the median nerve of four rats was marked with 6-0 nylon at intervals of 0.5 ram. With the use of these marks the relationship between the load and the percent of elongation was assessed according to the theory of Lundborg/ Blood flow measurement. After 20 rats were placed in a supine position, the skin was incised from the medial side of the elbow to the sternoclavicular joint. After the median nerve and
ulnar nerve were identified immediately below the incised area of the elbow, the center of the superficial and deep pectoral muscles was dissected to expose the brachial plexus in the axillary region. (The C8 ramus ventralis can be easily observed, because the bifurcation of the medial pectoral nerve is thick.) While we performed these steps, we took care to avoid touching the nerves. To measure the blood flow of the intrinsic system, the subepineurid space was incised under a stereoscopic microscope. This procedure was followed by the insertion of an iridium-coated hooklike platinum electrode (90 IJm) (Unique Medical, Tokyo) into the nerve bundle. Resistance was felt during passage through the perineurium. When resistance was no longer felt, indicating that the electrode had passed through the perineurium, we confirmed by microscope that the electrode was within the nerve bundles. Subsequently, to measure the blood flow of the extrinsic system another iridium-coated platinum electrode was placed within the subepineurial tissue immediately above the electrode in the nerve bundle. With other rats electrodes were similarly placed in the median nerve 1 cm distal to the C8 measuring point. With these electrodes the blood flow was measured when the depth of anesthesia was stable and the respiration rate varied little (i.e., from 60 to 90 minutes after the induction of anesthesia). The animals were allowed to spontaneously inhale a mixture of 50% hydrogen and 50% oxygen for 1 minute. Blood flow was measured by a hydrogen
120
J. Shoulder Elbow Surg. March/April 1995
Kitamura et al.
(%)
(~o) 100
80
80
,
60
40
zo
1 "1"'~
[
20
o ~;~oo~oo 3oo 4~o soo6oo 7oo 8'oo Lo~d<~) I
,
,
0 0.20.31 0.5
I o.~
1.0
1.35
load
(g)
1.875
;.5
/ body weight
65100
20'0 3 0 0
400
I ' ' 0 0.20.310.5 6.63 LO
(g)
500
600
700
8'00
load
(g)
I 1.25 load
(g)
1.875 / body weight
2.5 (g)
Figure 2 Blood flow of brachial plexus during each
Figure 3 Blood flow of median nerve during each load
load (n = 26). Blood flow in extrinsic system decreased more markedly than in intrinsic system (p < 0.005, MannWhitney's U test). Solid line, intrinsic system; broken line, extrinsic system.
(n = 14). Blood flow in extrinsic system began to decrease earlier than in intrinsic system (p < 0.01, Mann-Whitney's U test). Solid line, intrinsic system; broken line, extrinsic system.
Table I Blood flow recovery after load released Intrinsic system of brachial plexus (C8) (%)
Extrinsic system of brachial plexus (C8) (%)
Intrinsic system of median nerve (%)
Extrinsic system of median nerve (%)
104.8+21.7
82.9_+25.8
81.8_+21.0
95.9_+ 14.8
clearance tissue blood flow meter (MHG-D1, Unique Medical, Tokyo). Room temperature was kept at 25 ~ C. At each load level the electrode position was confirmed with a stereoscopic microscope before the blood flow was measured. After the release of each load the blood flow was measured again.
Histologic changes after repetitive mechanical traction. After repetitions of traction were done with a 320 gm load (equal to the body weight) for 2 weeks (90 min of traction/day) with eight rats, histologic changes were examined with hematoxylin-eosin and Luxol Fast blue stains.
RESULTS The percentage of elongation of the C8 ramus ventralis (Y = 1.300 x 10 3x + 1.987; r = 0.97) was higher than that of the median nerve at the
brachial center (Y = 2.113 x 10-4x + 1.979; r-0.99). In the extrinsic system of the brachial plexus the blood flow decreased sharply in the presence of relatively small loads (0 to 200 gm, 0 to 0.63 x body weight), and it showed a slow linear decrease when the load was greater than 200 gm (0.63 x body weight). In the intrinsic system the blood flow decreased markedly when the load exceeded 160 gm (0.5 x body weight). The response to tractive force between the two systems differed significantly (Figure 2). The blood flow in the intrinsic system of the median nerve at the center of the brachium decreased linearly at loads greater than 100 gm (0.31 x body weight) and that in the extrinsic system also decreased linearly at loads greater than 65 gm (0.2 x body weight). The decrease in
J. Shoulder Elbow Surg. Volume 4, Number 2
Kitamura et al.
121
Figure 4 A, Morphologic change in brachial plexus. Partsof brachial plexus showed hypertrophicconnectivetissuederangementof epineuriumand surrounding tissue, suggesting mild adhesions (hematoxylin-eosinstain, original magnification xlO0). B, Vascularization and monocytes (arrows)(hematoxylin-eosin stain, original magnification x400). blood flow occurred significantly earlier in the extrinsic system (Figure 3). After the load was released, the blood flow in the brachial plexus and the median nerve recovered by 81.8% to 104.8% without any significant difference between the two systems (Table I). After 2 weeks of repetitive traction the brachial plexus and the median nerve, the axon, and the myelin showed no morphologic change under light microscopic examination. However, parts of the brachial plexus showed hypertrophic connective
tissue and vascularization of the epineurium and the surrounding tissue, suggesting mild granulomatous inflammation (Figure 4). DISCUSSION The brachial plexus is anatomically characterized by numerous bifurcations, the presence of a highly mobile shoulder girdle in its vicinity, and loose connective tissue around it. Because of these features it is clinically more susceptible to injury by tractive forces than are other peripheral nerves.
122
Kitamura et al.
In this study the excursion of the brachial plexus was greater than that of the median nerve at the center of the brachium. This finding indicates that traction of the upper limb does not lead to uniform elongation of the nerve between the central nerve root and the periphery but that elongation is greater in some portions than in others depending on the nature of articular movement. Also, as Wilgis and Murphy ~5 reported with fresh cadaver arms, the excursion of the brachial plexus after movement of the shoulder girdle is greater than that of the median and ulnar nerves. This finding suggests that the excursion of the brachial plexus is a likely cause of nerve injury. Our quantitative analysis of microcirculation revealed that an imbalance occurred in the blood flow between the extrinsic and intrinsic systems of the brachial plexus. This occurrence seems to be attributable to a sharp decrease in the blood flow through the extrinsic system of the brachial plexus in the presence of a small tractive force. On the other hand, the blood flow through the intrinsic system showed a slow decrease similar to that of the median nerve at the center of the brachium. A possible explanation for this difference is that when a nerve is elongated nonuniformly, the blood flow is supplied from the less elongated portion to the intrinsic system to maintain balance as Lundborg suggested, z' 8 Similar to Ugaji's blood flow analysis of the sciatic nerve, 13 when we used the electrochemically generated hydrogen method, our analysis of blood flow through the median nerve at the brachial center revealed no imbalance between the extrinsic and intrinsic systems. It is unknown whether the imbalance observed in the brachial plexus is involved in the dysfunction of this nerve. However, the sharp decrease in the blood flow through the extrinsic system suggests that the brachial plexus is likely to show changes in blood flow, possibly leading to local elevation of vascular permeability and edema in the early phase. 7' 8 These decreases were reversible, because the blood flow recovered when the load was released. This finding supports the idea that relaxing the stretching stimuli by passive shoulder elevation improves these symptoms. Histologic examination under light microscopic examination revealed no abnormalities in the axon or myelin of the brachial plexus or in the median nerve after 2 weeks of repetitive traction. However, in parts of the brachial plexus we noted hypertrophic connective tissue or granulomatous
J. Shoulder Elbow Surg. March/April 1995 inflammation in the surrounding tissue of the extrinsic system. These local histologic changes seem to be attributable to local ischemia and injury to small vessels in the extrinsic system,9 because a marked decrease was seen in the blood flow in the extrinsic system, and because, as stereoscopic microscopic examination of the brachial plexus during traction revealed, the epineurium and surrounding tissue of the extrinsic system were subject to high tension even at small loads. In this connection, Liu et al. ~ reported that the blood vessels in the epineurium were less resistant to tractive stimuli than were the blood vessels in the endoneurium. DennyBrown and Doherty3 emphasized the pathologic importance of tractive stimuli based on the experimental finding that early formation of edema relates to damage to small blood vessels in the epineurium. Wall et al. TM reported that elongation by 6% or 12% was accompanied by epineurial microbleeding and edema, although the myelin and axon remained unchanged. According to published clinical reports fibrosis is often seen around the brachiat plexus during operations on patients with traumatic thoracic outlet syndrome7 These histologic changes are probably caused by local damage to the fulcrum in which the perineurium-perforating small vessels bind tightly to the perineurium. The resultant adhesion hampers smooth excursions of the nerve, which induces nerve irritation. In summary, when the brachid plexus was stretched at an intensity not severe enough to cause avulsion injury, the decrease in blood flow of the extrinsic system was greater than that of the intrinsic system. This imbalance in blood flow then probably induces nerve irritation and dysfunction. In some cases mild histologic changes of the tissue surrounding the extrinsic system can occur. These findings suggest that the extrinsic system of the brachial plexus is susceptible to injury by acute stretching, a possible factor precipitating the onset of symptoms. REFERENCES 1. CJein LJ. The droopy shoulder syndrome. CMAJ ] 976;114: 343-4. 2. Dellon AL. The results of supraclavicular brachial plexus neurolysis (without first rib resection) in management of posk traumatic "thoracic outlet syndrome." J Reconstr Microsurg 1993;9:11-6. 3. Denny-Brown D, Doherty MM. Effects of transient stretching of peripheral nerve. Arch Neurol Psychiatr 1945;54:11629.
J. Shoulder Elbow Surg. Volume 4, Number 2 4. Kataoka Y. Pathogenesis of thoracic outlet syndrome: diagnosis with neurography of the brachial plexus [In Japanese]. J Jpn Orthop Assoc 1994;68:357-66. 5. Kataoka Y, Takagi K, Morisawa K, Yamaga M. The relation bek,veen the classification and the results of surgical treatment in patients with thoracic outlet syndrome [In Japanese]. Shoulder Joint 1991 ;15:262-7. 6. Liu CT, Benda CE, Lewey FH. Tensile strength of human nerve. Arch Neurol Psychiatr 1948;59:323-36. 7. Lundborg G. Nerve injury and repair. New York: Churchill Livingstone, 1988:64-101. 8. Lundborg G. Structure and function af the intraneural microvessels as related to trauma, edema formation, and nerve function. J Bone Joint Surg [Am] 1975;57A:938-48. 9. Lundborg G, Rydevic BG. Effects of stretching the tibial nerve of the rabbit9 J Bone Joint Surg [Br] t973;55B:390401.
Kitamura et al.
123
10. Ogata K, Naitou M. Blood flow of peripheral nerve: effects of dissection, stretching and compression. J Hand Surg [Br] 1986;I 1B:I 0-4. 11. Schwartzman RJ. Brachial plexus traction injuries. Hand Clin 1991 ;7:547-56. 12. Swift TR, Nichols FT. The droopy shoulder syndrome. Neurology 1984;34:212-5. 13. Ugaji Y9 Experimental studies on the interneurd circulator y change of peripheral nerve after traction iniury-measurement of the local blood flow by electrochemically generated hydrogen9 J Juzen Med Soc 1987;96:599-612. 14. Wall EJ, MassieJB, Kwan MK, Rydevik BL, Myers RR, Garfin SR. Experimental stretch neuropathy: change in nerve conduction under tension. J Bone Joint Surg [Br] 1992;74B: 126-9. 15. Wilgis EFS, Murphy R. The significance of longitudinal excursion in peripheral nerves. Hand Clin 1986;2:761-6.
Because the brachial plexus has many branches it is clinically regarded as being sensitive to stimuli. Owing to the high mobility of the shoulder girdle adjoining it, traction of the upper limb can cause avulsion or various injuries of lesser degree. Schwartzman ~1 reported that one factor responsible for traumatic thoracic outlet syndrome is stretching of the brachial plexus at an intensity level not severe enough to cause avulsion injury. Kataoka et al/" 5 reported a neurograph of the brachial plexus showing stretching in some cases of thoracic outlet syndrome. Swift and Nichols 12 and Clein 1 reported on the droopy shoulder syndrome, in which symptoms are caused by stretching of the brachial plexus. It is little understood, however, how traction not intense enough to cause avulsion injury is involved in the onset of pain or numbness. Clinically, relaxing the stretching stimuli by passive shoulder girdle elevation improves From the Department of OrthopaedJc Surgery, Kumamoto Universib, School of Medicine, Kumamoto, Japan. Reprint requests: Toshio Kitamura, Department of Orthopaedic Surgery, Kumarnoto Universily School of Medicine, 1-1-1 Honio , Kurnamoto 860 Japan. Copyright 9 1995 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/95/$3.00 + 0 32/1/61388
118
these symptoms. 1 Therefore it is possible that functional disturbances such as changes in the blood flow influence the onset of symptoms. Peripheral nerves are well-vascularized structures with separate but extensively interconnected microvascular systems in the epineurium, perineurium, and endoneurium. An "intrinsic" vascular system is present, ft consists of vascular plexa in the epineurium, perineurium, and endoneurium. An "extrinsic" system involving segmental regional vessels approaching the nerve trunk at various levels along its course is also present. These regional vessels run in the loose "adventitia" or along the paraneural tissue surrounding the nerve, allowing the nerve trunk considerable mobility in its bed. 7' 8 Numerous reports have suggested that when the peripheral nerves are stretched, the surrounding blood vessels are extended, resulting in decreased blood flow within the nerve bundie.7, ~o, ~3 However, no report has yet been published regarding changes in the blood flow at the bifurcations of the brachial plexus during traction of the upper limbs. In this study we examined such changes of the blood flow during traction at an intensity insufficient to cause avulsion injury. We also investigated the resistance of the extrinsic and intrinsic systems to tractive force.
J. Shoulder Elbow Surg. Volume 4, Number 2
Kitamura et al.
119
Figure 1 Neurograph of rat brachial plexus. Lucentarea is nerve (arrows). Brachial plexus is stretchedand straightduring 80 ~ shoulderabduction.
METHODS Experimental animals. The C8 ramus ventrails of the brachial plexus and the median nerve at the center of the brachium of 32 Wistar rats (64 limbs) were used. The body weight of each animal was 300 to 330 gm (average 320 gin). With the use of a steel wire inserted into the radius and ulna of each rat, we pulled the upper limb in the direction of abduction. Neurographic and macroscopic observation of the brachiol plexus allowed us to confirm that traction during 80 ~ shoulder abduction caused the brachial plexus and median nerve to be pulled straight (Figure 1), so we adopted this as our model. Method of traction. While the rats were under intraperitoneal pentobarbital sodium anesthesia (5 rag/100 gin), the wire insertion sites at the radius and ulna were pulled with a constantly increasing weight (0 to 800 gm at 100 gm/60 sec). We confirmed that neither rupture nor avulsion of the brachial plexus was caused even at a load of 800 gin. Elongation. The perineurium between the brachial plexus and the median nerve of four rats was marked with 6-0 nylon at intervals of 0.5 ram. With the use of these marks the relationship between the load and the percent of elongation was assessed according to the theory of Lundborg/ Blood flow measurement. After 20 rats were placed in a supine position, the skin was incised from the medial side of the elbow to the sternoclavicular joint. After the median nerve and
ulnar nerve were identified immediately below the incised area of the elbow, the center of the superficial and deep pectoral muscles was dissected to expose the brachial plexus in the axillary region. (The C8 ramus ventralis can be easily observed, because the bifurcation of the medial pectoral nerve is thick.) While we performed these steps, we took care to avoid touching the nerves. To measure the blood flow of the intrinsic system, the subepineurid space was incised under a stereoscopic microscope. This procedure was followed by the insertion of an iridium-coated hooklike platinum electrode (90 IJm) (Unique Medical, Tokyo) into the nerve bundle. Resistance was felt during passage through the perineurium. When resistance was no longer felt, indicating that the electrode had passed through the perineurium, we confirmed by microscope that the electrode was within the nerve bundles. Subsequently, to measure the blood flow of the extrinsic system another iridium-coated platinum electrode was placed within the subepineurial tissue immediately above the electrode in the nerve bundle. With other rats electrodes were similarly placed in the median nerve 1 cm distal to the C8 measuring point. With these electrodes the blood flow was measured when the depth of anesthesia was stable and the respiration rate varied little (i.e., from 60 to 90 minutes after the induction of anesthesia). The animals were allowed to spontaneously inhale a mixture of 50% hydrogen and 50% oxygen for 1 minute. Blood flow was measured by a hydrogen
120
J. Shoulder Elbow Surg. March/April 1995
Kitamura et al.
(%)
(~o) 100
80
80
,
60
40
zo
1 "1"'~
[
20
o ~;~oo~oo 3oo 4~o soo6oo 7oo 8'oo Lo~d<~) I
,
,
0 0.20.31 0.5
I o.~
1.0
1.35
load
(g)
1.875
;.5
/ body weight
65100
20'0 3 0 0
400
I ' ' 0 0.20.310.5 6.63 LO
(g)
500
600
700
8'00
load
(g)
I 1.25 load
(g)
1.875 / body weight
2.5 (g)
Figure 2 Blood flow of brachial plexus during each
Figure 3 Blood flow of median nerve during each load
load (n = 26). Blood flow in extrinsic system decreased more markedly than in intrinsic system (p < 0.005, MannWhitney's U test). Solid line, intrinsic system; broken line, extrinsic system.
(n = 14). Blood flow in extrinsic system began to decrease earlier than in intrinsic system (p < 0.01, Mann-Whitney's U test). Solid line, intrinsic system; broken line, extrinsic system.
Table I Blood flow recovery after load released Intrinsic system of brachial plexus (C8) (%)
Extrinsic system of brachial plexus (C8) (%)
Intrinsic system of median nerve (%)
Extrinsic system of median nerve (%)
104.8+21.7
82.9_+25.8
81.8_+21.0
95.9_+ 14.8
clearance tissue blood flow meter (MHG-D1, Unique Medical, Tokyo). Room temperature was kept at 25 ~ C. At each load level the electrode position was confirmed with a stereoscopic microscope before the blood flow was measured. After the release of each load the blood flow was measured again.
Histologic changes after repetitive mechanical traction. After repetitions of traction were done with a 320 gm load (equal to the body weight) for 2 weeks (90 min of traction/day) with eight rats, histologic changes were examined with hematoxylin-eosin and Luxol Fast blue stains.
RESULTS The percentage of elongation of the C8 ramus ventralis (Y = 1.300 x 10 3x + 1.987; r = 0.97) was higher than that of the median nerve at the
brachial center (Y = 2.113 x 10-4x + 1.979; r-0.99). In the extrinsic system of the brachial plexus the blood flow decreased sharply in the presence of relatively small loads (0 to 200 gm, 0 to 0.63 x body weight), and it showed a slow linear decrease when the load was greater than 200 gm (0.63 x body weight). In the intrinsic system the blood flow decreased markedly when the load exceeded 160 gm (0.5 x body weight). The response to tractive force between the two systems differed significantly (Figure 2). The blood flow in the intrinsic system of the median nerve at the center of the brachium decreased linearly at loads greater than 100 gm (0.31 x body weight) and that in the extrinsic system also decreased linearly at loads greater than 65 gm (0.2 x body weight). The decrease in
J. Shoulder Elbow Surg. Volume 4, Number 2
Kitamura et al.
121
Figure 4 A, Morphologic change in brachial plexus. Partsof brachial plexus showed hypertrophicconnectivetissuederangementof epineuriumand surrounding tissue, suggesting mild adhesions (hematoxylin-eosinstain, original magnification xlO0). B, Vascularization and monocytes (arrows)(hematoxylin-eosin stain, original magnification x400). blood flow occurred significantly earlier in the extrinsic system (Figure 3). After the load was released, the blood flow in the brachial plexus and the median nerve recovered by 81.8% to 104.8% without any significant difference between the two systems (Table I). After 2 weeks of repetitive traction the brachial plexus and the median nerve, the axon, and the myelin showed no morphologic change under light microscopic examination. However, parts of the brachial plexus showed hypertrophic connective
tissue and vascularization of the epineurium and the surrounding tissue, suggesting mild granulomatous inflammation (Figure 4). DISCUSSION The brachial plexus is anatomically characterized by numerous bifurcations, the presence of a highly mobile shoulder girdle in its vicinity, and loose connective tissue around it. Because of these features it is clinically more susceptible to injury by tractive forces than are other peripheral nerves.
122
Kitamura et al.
In this study the excursion of the brachial plexus was greater than that of the median nerve at the center of the brachium. This finding indicates that traction of the upper limb does not lead to uniform elongation of the nerve between the central nerve root and the periphery but that elongation is greater in some portions than in others depending on the nature of articular movement. Also, as Wilgis and Murphy ~5 reported with fresh cadaver arms, the excursion of the brachial plexus after movement of the shoulder girdle is greater than that of the median and ulnar nerves. This finding suggests that the excursion of the brachial plexus is a likely cause of nerve injury. Our quantitative analysis of microcirculation revealed that an imbalance occurred in the blood flow between the extrinsic and intrinsic systems of the brachial plexus. This occurrence seems to be attributable to a sharp decrease in the blood flow through the extrinsic system of the brachial plexus in the presence of a small tractive force. On the other hand, the blood flow through the intrinsic system showed a slow decrease similar to that of the median nerve at the center of the brachium. A possible explanation for this difference is that when a nerve is elongated nonuniformly, the blood flow is supplied from the less elongated portion to the intrinsic system to maintain balance as Lundborg suggested, z' 8 Similar to Ugaji's blood flow analysis of the sciatic nerve, 13 when we used the electrochemically generated hydrogen method, our analysis of blood flow through the median nerve at the brachial center revealed no imbalance between the extrinsic and intrinsic systems. It is unknown whether the imbalance observed in the brachial plexus is involved in the dysfunction of this nerve. However, the sharp decrease in the blood flow through the extrinsic system suggests that the brachial plexus is likely to show changes in blood flow, possibly leading to local elevation of vascular permeability and edema in the early phase. 7' 8 These decreases were reversible, because the blood flow recovered when the load was released. This finding supports the idea that relaxing the stretching stimuli by passive shoulder elevation improves these symptoms. Histologic examination under light microscopic examination revealed no abnormalities in the axon or myelin of the brachial plexus or in the median nerve after 2 weeks of repetitive traction. However, in parts of the brachial plexus we noted hypertrophic connective tissue or granulomatous
J. Shoulder Elbow Surg. March/April 1995 inflammation in the surrounding tissue of the extrinsic system. These local histologic changes seem to be attributable to local ischemia and injury to small vessels in the extrinsic system,9 because a marked decrease was seen in the blood flow in the extrinsic system, and because, as stereoscopic microscopic examination of the brachial plexus during traction revealed, the epineurium and surrounding tissue of the extrinsic system were subject to high tension even at small loads. In this connection, Liu et al. ~ reported that the blood vessels in the epineurium were less resistant to tractive stimuli than were the blood vessels in the endoneurium. DennyBrown and Doherty3 emphasized the pathologic importance of tractive stimuli based on the experimental finding that early formation of edema relates to damage to small blood vessels in the epineurium. Wall et al. TM reported that elongation by 6% or 12% was accompanied by epineurial microbleeding and edema, although the myelin and axon remained unchanged. According to published clinical reports fibrosis is often seen around the brachiat plexus during operations on patients with traumatic thoracic outlet syndrome7 These histologic changes are probably caused by local damage to the fulcrum in which the perineurium-perforating small vessels bind tightly to the perineurium. The resultant adhesion hampers smooth excursions of the nerve, which induces nerve irritation. In summary, when the brachid plexus was stretched at an intensity not severe enough to cause avulsion injury, the decrease in blood flow of the extrinsic system was greater than that of the intrinsic system. This imbalance in blood flow then probably induces nerve irritation and dysfunction. In some cases mild histologic changes of the tissue surrounding the extrinsic system can occur. These findings suggest that the extrinsic system of the brachial plexus is susceptible to injury by acute stretching, a possible factor precipitating the onset of symptoms. REFERENCES 1. CJein LJ. The droopy shoulder syndrome. CMAJ ] 976;114: 343-4. 2. Dellon AL. The results of supraclavicular brachial plexus neurolysis (without first rib resection) in management of posk traumatic "thoracic outlet syndrome." J Reconstr Microsurg 1993;9:11-6. 3. Denny-Brown D, Doherty MM. Effects of transient stretching of peripheral nerve. Arch Neurol Psychiatr 1945;54:11629.
J. Shoulder Elbow Surg. Volume 4, Number 2 4. Kataoka Y. Pathogenesis of thoracic outlet syndrome: diagnosis with neurography of the brachial plexus [In Japanese]. J Jpn Orthop Assoc 1994;68:357-66. 5. Kataoka Y, Takagi K, Morisawa K, Yamaga M. The relation bek,veen the classification and the results of surgical treatment in patients with thoracic outlet syndrome [In Japanese]. Shoulder Joint 1991 ;15:262-7. 6. Liu CT, Benda CE, Lewey FH. Tensile strength of human nerve. Arch Neurol Psychiatr 1948;59:323-36. 7. Lundborg G. Nerve injury and repair. New York: Churchill Livingstone, 1988:64-101. 8. Lundborg G. Structure and function af the intraneural microvessels as related to trauma, edema formation, and nerve function. J Bone Joint Surg [Am] 1975;57A:938-48. 9. Lundborg G, Rydevic BG. Effects of stretching the tibial nerve of the rabbit9 J Bone Joint Surg [Br] t973;55B:390401.
Kitamura et al.
123
10. Ogata K, Naitou M. Blood flow of peripheral nerve: effects of dissection, stretching and compression. J Hand Surg [Br] 1986;I 1B:I 0-4. 11. Schwartzman RJ. Brachial plexus traction injuries. Hand Clin 1991 ;7:547-56. 12. Swift TR, Nichols FT. The droopy shoulder syndrome. Neurology 1984;34:212-5. 13. Ugaji Y9 Experimental studies on the interneurd circulator y change of peripheral nerve after traction iniury-measurement of the local blood flow by electrochemically generated hydrogen9 J Juzen Med Soc 1987;96:599-612. 14. Wall EJ, MassieJB, Kwan MK, Rydevik BL, Myers RR, Garfin SR. Experimental stretch neuropathy: change in nerve conduction under tension. J Bone Joint Surg [Br] 1992;74B: 126-9. 15. Wilgis EFS, Murphy R. The significance of longitudinal excursion in peripheral nerves. Hand Clin 1986;2:761-6.