Shoulder region of the ratAnatomy and fiber composition of some suprascapular nerve branches.код для вставкиСкачать
THE ANATOMICAL RECORD 239:332-342 (1994) Sho ilder Region of the Rat: Anatomy and Fiber Composition of - Some Suprascapular Nerve Branches R. NORLIN, C. HOE-HANSEN, G. OQUIST, AND C. HILDEBRAND Departments of Orthopedic Surgery (R.N., C.H.-H.) and Cell Biology (G.O., C.H.), Faculty of Health Sciences, University of Linkoping, Linkoping, Sweden ABSTRACT Background: The pathophysiology of chronic supraspinatus tendinitis is not fully understood. This may be due to the scarcity of experimental studies on this issue. Methods: In search for a system suitable for experimental analysis, the present study describes the relevant gross anatomy of the rat shoulder region (dissection), and examines the fiber composition of relevant suprascapular nerve branches (electron microscopy, selective denervations). Results: The rat shoulder region is similar to the human shoulder in terms of gross anatomy. The average suprascapular nerve (SSC) is derived mainly from the spinal cord segment C5 and contains 3,435 axons, 74% of which are unmyelinated. The supraspinatus branch (SSP)contains 627 fibers. Of the SSP fibers, 52% are myelinated, including 32% motor and 20% sensory axons. Of the C-fibersin the SSP 16%are sympathetic efferents and 32% are sensory. Many of the latter disappear after neonatal capsaicin treatment. The SSC emits a subacromial articular branch (ART),with some 260 axons, about 90% of which are unmyelinated. The myelinated ART fibers are sensory, and of the unmyelinated ones about 24% are sympathetic efferents and 66% are afferents. The latter resist neonatal capsaicin treatment. Conclusions: In view of the anatomy of the supraspinatus muscle, of the subacromial space, and of relevant nerves, the rat shoulder should be appropriate for experimental studies on inflammatory conditions in the subaCrOmia1 space. 0 1994 Wiley-Liss, Inc. Key words: Rat, Shoulder region, Gross anatomy, Subacromial space, Supraspinatus muscle, Suprascapular nerve, Fiber composition, Electron microscopy Many patients in primary health care, occupational medicine, rheumatology, and orthopedic surgery suffer from disorders of the shoulder region (Petersson, 1986; Moren Hybinette, 1987; McCormack et al., 1990; Dimberg, 1991). Traumatic andlor degenerative changes in the rotator cuff are common causes of pain and chronic disability in this area. The supraspinatus muscle tendon and the associated subacromial bursa (Birnbaum and Lierse, 1992) may be affected by chronic inflammatory conditions, which sometimes end up in tendon rupture. It has been suggested that impingement of the rotator cuff beneath the bony acromion and the coracoacromial ligament is a n important causative factor in this respect (Neer, 1972; Watson, 1989; Dines e t al., 1990; Birnbaum and Lierse, 1992). Accordingly, observations at dissection, during surgery, and by radiography show that the supraspinatus tendon can be impinged when the arm is abducted, and surgical measures against impingement, i.e., anterior acromioplasty, relieve the pain (Neer, 1972). The development of increasingly more sophisticated surgical techniques has improved the results of active treatment of this 0 1994 WILEY-LISS, INC type of disorder (Ellman, 1987; Gartsman et al., 1988; Norlin, 1989), but our understanding of the basic pathophysiological mechanisms remains very incomplete. This may be due to the fact that few experimental studies deal with this issue. Several studies suggest that neural elements may be involved in the development and maintenance of inflammatory conditions in a variety of tissues (see Discussion). It seems possible that the emergence of chronic inflammatory conditions in the shoulder might involve more complex mechanisms than the mechanical models discussed so far. In the present paper we examine the rat shoulder region, with emphasis on the supraspinatus muscle and the subacromial space and on the course and fiber composition of some relevant suprascapular nerve branches. The main purpose is to see if the rat shoulder may be Received November 30, 1993; accepted February 17, 1994. Address reprint requests to Claes Hildebrand, Department of Cell Biology, Faculty of Health Sciences, University of Linkoping, 581 85 Linkoping, Sweden. RAT SHOULDER REGION a n appropriate system for experimental studies on the pathophysiology of chronic supraspinatus tendinitis. MATERIAL AND METHODS Dissection Rats to be used for dissection (Sprague-Dawley, n = 5) were anesthetized with chloral hydrate (30 mgl 100 g b.w., ip), and exsanguinated. Dissection was performed under a n operation microscope. A binocular dissection microscope with a Nikon F-300 camera containing Kodak Tri-X film was used for photography. 333 fixed during 2 hours in the perfusate, and rinsed in buffer. Segments and ganglia were immersed overnight in Millonig’s buffer with 30% sucrose, and serially sectioned at 60 pm in a freezing microtome. The sections were mounted on slides and processed according to the TMB method for HRP visualization (Mesulam, 1982). Capsaicin treatment Five rat pups were anesthetized by hypothermia and injected under the dorsal skin with a solution of capsaicin (1g capsaicin [analytical grade, KEBO, Sweden] Radiography in a mixture of 10 ml Triton X-100,lO ml ethanol, and One rat was killed a s described above, and subjected 80 ml saline) on postnatal days 2, 3 (50 mg/kg, b.w.), to radiographic examination. Standard radiographs, and 4 (100 mg/kg, b.w.1. Three control rats received including anteroposterior, lateral, and outlet views vehicle only (Jancso et al., 1985; Fried et al., 1988). (special Y view) were taken, using a Siemens Orbix These rats were perfused after a survival time of 3 apparatus supplied with Kodak mammography OM 1 months. Nerve specimens were collected and processed film. for electron microscopy, as described above. Fiber counts were made on sections from SSP and ART specElectron Microscopy imens. In order to check that treatment had been effective, the dorsal roots L5 were collected from 3 capNormal material saicin-treated and 3 vehicle-injected animals. The Adult male and female Sprague-Dawley rats ( n = 5 , average occurrence of C-fibers in the surroundings of a t 300-500 g) were anesthetized with chloral hydrate and least 500 myelinated fibers was determined in each of perfused with Tyrode’s solution, followed by 5% glutar- the six cases. The analysis showed the presence of aldehyde in a 300 mOsm Millonig buffer with 0.1M 71.8% C-fibers in vehicle-injected cases and 26.1% in sucrose. After perfusion, suprascapular nerve speci- capsaicin-treated cases. Hence, the treatment worked mens were removed, postfixed in glutaraldehyde, well. rinsed in buffer, osmicated (2% OsO, in phosphate buffer), dehydrated in acetone, and embedded in Vesto- Deeff erentation pal W. Thin transverse sections from the suprascapular The proportion of myelinated sensory and motor finerve trunk (SSC), from the supraspinatus branch (SSP), and from a small subacromiaUarticular branch bers in the SSP and ART was determined through se(ART) were collected on one hole formvar-coated copper lective ventral rhizotomies in 3 adult rats anesthetized grids. After contrasting, the sections were examined in with chloral hydrate. After a cervical hemilamineca JEOL JEM 1200 EX electron microscope. Montages tomy, the ventral roots C4-C6 on the right side were of electron micrographs ( x 1,500) completely covering divided. Ten days later the animals were perfused, and each cross-cut nerve specimen, were prepared and used the deefferented right suprascapular nerve and the for myelinated axon countings. In order to determine normal contralateral nerve were collected and prethe number of C-fibers the sections were reexamined in pared for electron microscopy. The segmental levels afthe electron microscope ( x 10,000), using the montages fected by the rhizotomy were verified through dissecfor orientation. I n each of five nerves of each type, the tion. Analyses of the number and size distribution of cross-sectional areas of all myelinated fibers present intact myelinated fibers were made on sections from were measured on the montages, using a Kontron Vid- SSP and ART specimens, as described above. eoplan 2 equipment. Fiber diameters (D, including myelin sheaths) were calculated from the areas, assuming Sympathectomy circularity (Karnes et al., 1977). Three adult rats were anesthetized with chloral hydrate and used for surgical sympathectomy. The cerviHRP- labelling cal portion of the sympathetic trunk was exposed, moIn order to determine the segmental level of origin of bilized, and cut out, from the superior cervical ganglion the SSC, we performed retrograde labelling of “supras- to the level of the first rib. The animals were allowed to capular” ventral horn and dorsal root ganglion neurons survive for 10 days. The operated animals showed a n with horseradish peroxidase (WGA-HRP; Sigma). The obvious left-sided ptosis. For chemical sympathectomy animals (n = 5) were anesthetized with chloral hydrate. rat pups were given daily injections of guanethidine The SSC was exposed a t the neck of the scapula and cut (1.5%, 50 mgikg b.w.1 under the dorsal skin, during 3 with a pair of microscissors. The proximal stump was weeks, beginning day 3 after birth (Chad et al., 1983; immersed in 2% WGA-HRP in saline. After one hour Johnson and Manning, 1984). These animals were left the nerve stump was carefully cleaned and enclosed to survive until they were 3 months old. The injected within a sealed Parafilm pocket, and the wound was animals were sparsely furred and they had slightly closed. These animals were perfused 24 hours later, a s subnormal weights and loose stools, but no obvious ptodescribed above (perfusate: Ringer’s solution followed sis. The sympathectomized animals were perfused and by 1%paraformaldehyde and 1.5% glutaraldehyde in the suprascapular nerves were collected and processed Millonig’s phosphate buffer). The cervical spinal cord for electron microscopy, as described above. Counts of segments and dorsal root ganglia were collected, post- myelinated and unmyelinated fibers were made on SSP 334 R. NORLIN ET AL. Fig. 1. This dissected preparation shows the scapula, the clavicula, and the humerus held together by the humeroscapular and acromioclavicular joints. The preparation is viewed from the head end of the animal in a and from the lateral side in b. Note the distinct acromial process from the long spina scapulae (a,b). Thin arrow in b points a t tendon of long head of biceps muscle in intertubercular sulcus. Note the absence of an incisura scapulae along the cranial edge of the scapula (b). Arrowhead in b indicates the hook-like insertion site of the deltoid muscle. and ART specimens from 3 cases. In order to check that the guanethidine treatment had been effective, the thoracic sympathetic trunk was cut out and examined under a dissection microscope. The paravertebral ganglia were clearly seen in sympathetic trunks from normal rats, but not in trunks from treated cases. This impression was confirmed by light microscopic examination of sections from the superior cervical ganglion (not shown). Hence, treatment had been effective. All the experimental procedures described above were approved by the local ethical committee. RESULTS Gross Anatomy The thickened medial and lateral ends of the weakly curved rat clavicula were anchored by ligaments in the sternoclavicular and acromioclavicular joints. Two ligaments attached the lateral end of the clavicle to the RAT SHOULDER REGION 335 Fig. 2. Radiographs showing antero-posterior (a) and lateral (b) views of the shoulder region. SS = subacromial space, C =clavicle, CP = coracoid process, DM = insertion site of deltoid muscle. Course of the SSC and its Branches medially directed hook-like coracoid process. The rostral margin of the rat scapula lacked the incisure Emerging as a distinct trunk from the brachial present in man. Since the r a t scapula is longer in the plexus, the SSC traversed the supraclavicular fossa, dorso-ventral t h a n in the rostro-caudal direction, the accompanied by a branch of the subscapular nerve (Fig. spina scapulae is relatively longer than in man (Fig. 1). 3b). The SSC rounded the levator scapulae muscle and A distinct acromion covered the humeroscapular joint approached the scapula without branching. While (Figs. 1, 2). The oval articular surface of the glenoid passing collum scapulae, the SSC emitted the ART, formation, which was enlarged by a cartilagineous la- which was composed of 1-3 fascicles (Fig. 3c). This thin brum, was connected to the body of the scapula through branch could be followed into the subacromial space, a comparatively long and narrow neck (Figs. 1,2). The where it ended in relation to the humeroscapular joint. caput humeri had a half-spherical articular surface, a Having entered the fossa supraspinata, the SSC split tuberculum majus, a tuberculum minus, and a n inter- into two branches while on the deep side of the sutubercular sulcus. A prominent deltoid tuberosity praspinatus muscle. The smaller SSP branch coursed formed a broad-based hook-like process laterally on the dorsally on the deep side of its muscle, sending fiber proximal half of the humerus (Figs. 1,2). The humero- bundles into it. The larger infraspinatus branch scapular joint was enclosed by a distinct capsule, the rounded the ventral edge of the spina scapulae reachouter aspect of which was apposed by flattened ing the fossa infraspinata, where i t entered the inscapulo-humeral muscle tendons. The lateral end of the fraspinatus muscle from its deep side. clavicle, the acromion, and a coracoacromial ligament Fiber Composition of the SSC, SSP, and ART formed on osteofibrous arch delimiting the subacromial space. This space contained a synovial bursa, the ten- ssc The labelling experiments demonstrated that SSC don of the supraspinatus muscle, a portion of the joint capsule, and part of the tendon of the long head of the motor axons originate mainly from neurons in the segbiceps muscle, which traversed the joint capsule (Figs. ment C5, with additional components in the caudal half of the segment C4, or the cranial half of the seg2, 3a). 336 R. N O R I J N ET AL. Fig. 3. These photographs show different views on a dissected rat shoulder. a: View from cranial end. S M = supraspinatus muscle, SP = spina scapulae, C =clavicle, A = acromion. Black triangle is located in subacromial space on top of the supraspinatus tendon. b View from ventral side. This picture shows the cranial part of the brachial plexus (BP). Arrow indicates suprascapular nerve. c: Dorso- lateral view, showing the site where the suprascapular nerve (SSC) rounds the levator scapulae muscle (LSM), approaches the upper border of the scapula (SCAP), and splits into three branches (ART= articular nerve, SSP = supraspinatus nerve, INF = infraspinatus nerve). SM = supraspinatus muscle. ment C6. The sensory fibers come from corresponding dorsal root ganglia (data not shown). The average normal SSC was monofascicular (Fig. 4a), and contained 3,435 axons, of which 26% were myelinated and 74% unmyelinated (Table la). The myelinated fibers exhibited a distincly bimodal size distribution, ranging from 2 pm to about 17 pm. The two peak frequencies were located at about 3 pm and 13-14 pm, and there was a minimum a t 6-8 pm (Fig. 5a). lb). The myelinated fibers exhibited a bimodal size distribution, ranging from 2 pm to about 16 pm. The two peak frequencies were located at about 3 pm and 10-12 pm, with a minimum at 6-8 pm (Fig. 5b). After ventral rhizotomy the average SSP contained 116 myelinated axons, a reduction by 64% (Table lb). The deefferented SSP showed a reduced myelinated fiber size range and a single peak at 3-4 pm. The upper peak seen in the normal nerve was not obvious (Fig. 5c). In capsaicin-treated animals, the proportion of unmyelinated axons was clearly subnormal (Table lb). No reductions occurred in vehicle-treated animals (data not shown). Both after surgical and chemical sympathec- SSP The average normal SSP (Fig. 4b) contained 627 f i bers. Almost half of these were unmyelinated (Table RAT SHOULDER REGION Fig. 4. Electron micrographs showing general appearance of a)the suprascapular nerve ( x 450), b) the supraspinatus nerve ( x 525), and c ) the articularisubacromial nerve ( x 4,000). 337 338 R. NORLIN ET AL. TABLE lc. Fiber composition of the subacromial/articularnerve in the normal state and in various experimental conditions' TABLE la. Numbers of myelinated (M) and unmyelinated (U)fibers in the normal SuDrascaDular nerve' Normal Mean '% = 913 888 948 940 840 905 2,546 2,579 2,580 2,480 2,472 2,530 73.6 74.3 73.1 72.5 74.6 73.6 proportion of unmyelinated axons, Tot = 3,459 3,458 3,528 3,420 3,312 3,435 total number of axons. TABLE lb. Fiber composition of the supraspinatus nerve in the normal state and in various experimental conditions' Normal Mean Rhitzotomy Mean Capsaicin Mean Guanethidine Mean Surg. sympathectomy Mean M 344 285 371 312 309 324 U 285 286 313 310 319 303 % 45.3 50.0 45.8 49.8 50.8 48.3 Normal Mean Rhizotomy Mean Capsaicin Tot 629 571 684 622 628 627 100 124 124 116 Mean Guanethidine Mean Surg. sympathectomy 233 256 24 1 243 102 97 90 96 30.4 27.5 27.2 28.4 335 353 33 1 340 258 369 281 303 105 188 144 146 28.9 33.8 33.8 32.2 363 557 425 444 299 274 230 268 108 200 152 153 26.5 42.2 39.8 36.4 407 474 382 42 1 'M = number of myelinated axons, U = unmyelinated axons, % proportion of unmyelinated axons, Tot = total number of axons. = tomy, the proportion of C-fibers was subnormal (Table lb). ART The average normal ART (Fig. 4c) contained 259 axons. About 90% of all axons in the normal ART were unmyelinated (Table lc). Although some myelinated fibers reached up to 7-9 pm in diameter, most of them had diameters below 4-5 km, with a peak a t about 3 pm (Fig. 5d). In cases subjected to ventral rhizotomy, the proportions of myelinated and unmyelinated axons were normal. In capsaicin-treated animals, the proportion of unmyelinated ART axons was 84% (Table lc), i.e., slightly subnormal. No abnormality was observed in vehicle-injected cases (data not shown). In animals subjected to surgical sympathectomy, the ART contained 90% C-fibers, like in control cases. Following Mean M 26 29 31 24 26 27 U 255 245 222 193 245 232 90.7 89.4 86.0 88.9 92.7 89.5 Tot 281 274 253 217 271 259 19 13 31 21 175 169 234 193 90.2 92.9 88.3 90.2 194 182 265 214 37 36 51 41 293 250 155 233 88.8 87.4 75.2 83.8 330 286 206 274 45 62 41 49 58 118 134 103 56.3 65.6 76.6 66.2 103 180 175 153 34 59 37 43 274 615 317 402 89.0 91.2 89.5 89.9 308 674 354 445 % 'M = number of myelinated axons, U = number of unmyelinated axons, % = proportion of unmyelinated axons, Tot = total number of axons. chemical sympathectomy with guanethidine, the proportion of C-fibers was 66%, i.e., markedly subnormal (Table lc). DISCUSSION The present study was prompted by the deficient understanding of the pathophysiology of chronic inflammatory disorders affecting the subacromial space and the supraspinatus muscle tendon in man. The r a t appeared to be generally suitable for analysis since i t has a clavicle, unlike many other quadrupeds. Our analysis showed that the rat shoulder region has obvious anatomical similarities with the human shoulder. This might, conceivably, be related to the fact that the rat uses its forepaws for object manipulation. The scapula has a distinct spine with a well developed acromion that articulates with the clavicle. As in man, the acromion, the coracoid process, and bridging ligaments form the roof of a narrow subacromial space containing a synovial bursa, the tendon of the supraspinatus muscle, and the long head of the biceps muscle. The other parts of the human "rotator cuff" are also present in the rat. As in man, the deltoid muscle covers the shoulder joint, but in the rat the bony insertion of this muscle on the humerus is much more prominent than the human deltoid tuberosity (Fig. lb). We conclude that, in terms of gross anatomy, the rat shoulder is surprisingly similar to the human shoulder. In agreement with previous observations in man (Gardner, 1948; Kato, 1989) and rat (Greene, 1963; 339 RAT SHOULDER REGION :] % % 20 4 I 10 10 5 5 D 0 0 5 10 15 20 m) D 0 0 5 10 15 20 m) b a il& 46 15 10 - 0 5 5 10 d 20 D 6 4 Fig. 5. Representative histograms showing size distribution of myelinated fibres (D, myelin sheaths included) in the suprascapular nerve (a), the normal supraspinatus nerve (b), the deefferented supraspinatus nerve (c), and the articular/subacromial nerve (d). Hebel and Stromberg, 1986), our dissections showed that the SSC sends branches to the supra- and infraspinatus muscles. According to our labelling experiments, this nerve originates from neurons in the segment C5, with contributions from either of the adjacent segments. The same levels have been indicated by other workers as segments of origin for the SSC in the rat (Kitamura et al., 1981), and kangaroo (Kato and Hopwood, 1993), and in man (Kerr, 1918; Horiguchi, 1980; Kato, 1989). The rat shoulder joint is innervated from the segments C3-C5 (Yoshida et al., 1992). Our counts show that the SSC contains some 3,400 axons, about 314 of which are unmyelinated. This proportion of C-fibers is higher than in other muscle nerves, which suggests that the SSC may contain nonmuscular components. The size spectrum of the SSC showed a bimodal distribution, as in other muscle nerves (Romero and Skoglund, 1965), but the proportion of small myelinated fibers was relatively high. Since the cutaneous branch sometimes emerging from the human SSC (Horiguchi, 1980) was absent in the rat, the surplus of unmyelinated and small myelinated fibers in the rat SSC should not represent cutaneous axons. Instead, i t might reflect the presence of articular axons, some of which leave the SSC with the ART. Joint nerves typically contain high proportions of C-fibers and small myelinated axons (Langford and Schmidt, 1983a; Hildebrand et al., 1991). The muscular supraspinatus branch (SSP)was found to contain about 620 axons, half of which were unmyelinated. Nerves to various r a t hindlimb muscles contain 44-63% unmyelinated axons (Jenq and Cogges- hall, 1984, 1985; Peyronnard et al., 1986). In the rat sternomastoid nerve, 50% of all axons are unmyelinated and 40% of these are sympathetic efferents (Sandoz and Zenker, 1986). The rat phrenic nerve contains 43% unmyelinated axons (Langford and Schmidt, 1983b). About 113 of the myelinated SSP fibers should be sensory, since they persisted after ventral rhizotomy. This figure is within the range presented for other nerves (e.g., Sherrington, 1895; Gottschall et al., 1980). The myelinated axons exhibited a distinctly bimodal muscle-type size distribution. Our measurements on rhizotomized cases showed that most of the muscle afferents are relatively thin, although some reach relatively large sizes (Sherrington 1895; Lloyd and Chang, 1948), and that a large proportion of the thick myelinated axons are motor efferents. Following neonatal capsaicin treatment, the proportion of C-fibers had decreased from 48% to 28%. This suggests a substantial loss of C-fibers. The number of myelinated axons was also low, but we feel that this may be due to individual variations. According to other workers, both myelinated and unmyelinated axons are lost from peripheral nerves after neonatal capsaicin treatment, but the figures show a considerable variation (see Lawson, 1981; Jancso et al., 1985). We conclude that the SSP contains many unmyelinated capsaicin-sensitive axons. The functional role of these fibers remains to be found out. Following both surgical and chemical sympathectomy, the proportion of C-fibers in the SSP was reduced from the normal level of about 48% to some 32%. This indicates that 113 of the C-fibers in this nerve may by sympathetic. Chad et al. (1983) concluded that sympathetic 340 R. NORLIN ET AL The supraspinatus n e n e (SSP) ’ I h e subacromial n e n e (.\RT) 100 % 25 10 sens 6 Fig. 6. Diagrams showing estimated relative proportions of different fiber types in the supraspinatus nerve (SSP) and the subacromialiarticular nerve (ART). postganglionic unmyelinated axons account for 20 25%of all unmyelinated axons in cutaneous, muscular, and mixed nerves. According to Peyronnard et al. (1986), sympathetic axons account for 23-34% of all C-fibers in r a t hindlimb muscle nerves. Our analysis showed that the average ART is composed of 10% myelinated axons, all of which are sensory, and 90% C-fibers. The myelinated axons showed a unimodal size distribution and a limited size range. Similar characteristics have been observed in the posterior articular nerve of the r a t knee joint (Hildebrand et al., 1991). Following neonatal capsaicin treatment, the proportion of C-fibers in the ART was close to normal. Hence, this nerve contains few capsaicin-sensitive units. Usually, a large proportion of the sensory C-fibres, as well as some small myelinated axons, disappear from peripheral sensory nerves after treatment of neonatal rat pups with capsaicin. The content of substance P in the dorsal horn is also markedly reduced (Lawson, 1981; Nagy et al., 1983; Jancso et al., 1985; Holzer, 1988). Since the examined dorsal roots exhibited a dramatically decreased proportion of C-fibers, the very limited alteration in the proportion of C-fibers in the ART should be real. A relative insensitivity to neonatal treatment with capsaicin has also been noted in the posterior articular nerve of the knee joint (Hildebrand et al., 1991), in pulpal nerves (Holje et al., 1985; Fried et al., 1988), and in the spinal pia mater (Karlsson and Hildebrand, 1994) of the rat. Capsaicin primarily affects neurons with unmyelinated and small myelinated axons involved in polymodal or chemical nociception (see Nagy et al., 1983). The insensitivity of articular, pulpal, and pial nerves could reflect a basically mechanosensitive function (Dubner et al., 1978; Schaible and Smith, 1983a,b). In animals subjected to surgical sympathectomy, the proportion of C-fibers in the ART remained unchanged at 90%. Therefore, the surgery failed to remove ganglia responsible for sympathetic axons in the ART, although i t did affect the SSP. After chemical sympathectomy, however, the proportion of C-fibers in the ART had decreased to 66%. This suggests that 114 of all C-fibers in this nerve may be sympathetic, Some 50% of the numerous C-fibers present in cat articular nerves are sympathetic (Langford and Schmidt, 1983a), and in the posterior articular nerve of the rat knee joint about 213 of the C-fibers are sympathetic (Hildebrand et al., 1991). In man, the SSC provides sensory fibers to a major part of the shoulder joint. It emits articular branches which follow a course similar to that of the ART in the rat, and which ramify in the region where the supraspinatus tendon enters the capsule and around the acromioclavicular joint (Gardner, 1948; Kato, 1989; Gad0 and Emery, 1993). Whether the human ART projects to the subacromial bursa, which exhibits a n extensive neural network (Tomita et al., 19921, is unknown. On the basis of our anatomical observations, the typical “articular” fiber composition of the ART, and the discussion above, we suggest that this suprascapular nerve branch provides the rat shoulder joint (and possibly the subacromial bursa) with sensory and sympathetic axons. In humans, articular branches may also reach the shoulder joint through, e.g., the infraspinatus branch of the suprascapular nerve and the axillary nerve (Gardner, 1948). There is increasing evidence t h a t sensory and sympathetic C-fibers may play active roles in inflammatory conditions in the locomotor system (Levine et al., 1985; Fitzgerald, 1989; Kidd e t al., 1990; Weinstein, 1991) and in various other tissues (Saria et al., 1983; Foreman and Jordan, 1984; Szolcsanyi, 1988; Bienenstock e t al., 1991; Donnerer and Amann, 1993; Dray and Perkins, 1993). Afferents from arthritic joints have a n abnormal spontaneous activity and the articular receptors are markedly responsive to normally innocuous stimuli (Guilbaud et al., 1985; Grigg et al., 1986). Injection of substance P (SP)into arthritic r a t knee joints aggravates inflammation and destruction (Levine et al., 1984). The occurrence of SP is increased in joints and sciatic nerves of arthritic rats (Lembeck et al., 1981; Levine et al., 1984). It has also been suggested that activity in sympathetic axons can amplify neurogenic inflammation (Coderre e t al., 1989). These and other data indicate that nociceptive and autonomic nerve fibers may be actively involved in inflammatory conditions, pain, and tissue destruction forming a vicious circle (see Fitzgerald, 1989; Lam and Ferrell, 1991).To what extent the rat suprascapular nerve participates in experimental inflammatory conditions in the subacromial space remains to be found out. RAT SHOULD'ER REGION 34 1 1985 Morphological effects of capsaicin and its analogues in newborn and adult mammals. In: Tachykinin Antagonists. R.M. H i This report was supported by grants from Clas Grokansson and F. Sundler, eds. Elsevier, Amsterdam, pp. 35-44. schinsky's Memorial Foundation, The Swedish Medical Jenq, C.-B., and R.E. Coggeshall 1984 Effects of sciatic nerve regeneration on axonal populations in tributary nerves. Brain Res., Research Council (Proj. no. 3761), the County of Oster295t91-100. gotland, and from the Faculty of Health Sciences. Dr. Jenq, C.-B., and R.E. Coggeshall 1985 Numbers of regenerated axons Staffan Hammerby provided valuable help with the rain tributary nerves following neonatal sciatic nerve crush in rat. Neurosci. Lett., 61:43-48. diography. Johnson, E.M., and P.T. Manning 1984 Guanethidine-induced destruction of sympathetic neurons. Int. Rev. Neurobiol., 25t1-37. LITERATURE CITED Karlsson, M., and C. Hildebrand 1994 Sensory C-fibers in rat ventral Bienenstock, J., G. Macqueen, P. Sestini, J.S. Marshall, R.H. Stead, roots are capsaicin-insensitive and they do not mediate extravaand M.H. Perdue 1991 Inflammatory cell mechanisms. Mast cell/ sation from pial vessels. Brain Res., in press. nerve interactions in vitro and in vivo. Am. Rev. Respir. Dis., Karnes, J., R. Robb, P.C. OBrien, E.H. Lambert, and P.J. Dyck 1977 143tS55-S58. Computerized image recognition for morphometry of nerve atBirnbaum, K., and W. Lierse 1992 Anatomy and function of the bursa tribute of shape of sampled transverse sections of myelinated fisubacromialis. Acta Anat., 145t354-363. bres which best estimates their average diameter. J . Neurol. Sci., Chad, D.W., W.G. Bradley, C. Rasool, P. Good, S. Reichlin, and J. 34t43-51. Zivin 1983 Sympathetic postganglionic unmyelinated axons in Kato, K. 1989 Innervation of the scapular muscles and its morphothe rat peripheral nervous system. Neurology, 33t841-847. logical significance in man. Anat. Anz., 168t155-168. Coderre, T.J., A.I. Basbaum, and J.D. Levine 1989 Neural control of Kato, K., and H. Hopwood 1993 Innervation of the levator scapulae, vascular permeability: interactions between primary afferents, the serratus anterior, the rhomboideus and the scapular muscles mast cells and sympathetic efferents. J. Neurophysiol.,62t48-58. in the gray kangaroo (Macropus giganteus). Comparisons are Dimberg, L. 1991 Symptoms from the neck, shoulders and arms in an drawn to the crab-eating macaque and man. Ann. Anat. 175:21industrial population, and some related problems. Thesis, Uni28. versity of Gothenburg. Kerr, A.T. 1918 The bracnial plexus of nerves in man, the variations Dines, D.M., R.F. Warren, and A.E. Inglis 1990 The coracoid impingein its formation and branches. Amer. J. Anat., 23t285-395. ment syndrome. J. Bone Joint Surg., 72-Bt314-316. Kidd, B.L., P.I. Mapp, D.R. Blake, S.J. Gibson, and J.M. Polak 1990 Donnerer, J., and R. Amann 1993 The inhibition of neurogenic inNeurogenic influences in arthritis. Ann. Rheum. Dis., 49t649flammation. Gen. Pharmac., 24t519-529. 652. Dray, A., and M. Perkins 1993 Bradykinin and inflammatory pain. Kitamura, S., N. Fujita, T. Nishiguchi, and A. Sakai 1981 A study on Trends Neurosci., 16t99-104. the localization of the motoneurons supplying the scapular musDubner, R., B.J. Sessle, and A.T. Storey 1978 The Neural Basis of cles of the rat by means of the HRP method 1. Supraspinatus, Oral and Facial Function. Plenum Press, New York. infraspinatus and teres major muscles. J . Osaka Dent. Sch., 21: Ellman, H. 1987 Arthroscopic subacromial decompression: analysis of 121-129. one to three years results. Arthroscopy, 3t173-181. Lam, F.Y. and W.R. Ferrell 1991 Neurogenic component of different Fitzgerald, A.C. 1989 Arthritis and the nervous system. Trends Neumodels of acute inflammation in the rat knee joint. Ann. Rheum. rosci., 12t6-7. Dis., 5Ot747-751. Foreman, J.C., and C.C. Jordan 1984 Neurogenic inflammation. Langford, L.A., and R.F. Schmidt 1983a Afferent and efferent axons Trends Pharmacol. Sci., 5116-119. in the medial and posterior articular nerves ofthe cat. Anat. Rec., Fried, K., H. Aldskogius, and C. Hildebrand 1988 Proportion of un206t71-78. myelinated axons in rat molar and incisor tooth pulps following neonatal capsaicin treatment and/or sympathectomy. Brain Res., Langford, L.A., and R.F. Schmidt 198313 An electron microscopic analysis of the left phrenic nerve in the rat. Anat. Rec., 205t207-213. 463r118-123. Lawson, S.N. 1981 Dorsal root ganglion neurons and dorsal roots: Gado, K., and P. Emery 1993 Modified suprascapular nerve block with effect of neonatal capsaicin. In: Spinal Cord Sensations: Sensory Bupivacaine alone effectively controls chronic shoulder pain in Processing in the Dorsal Horn. A.G. Brown, ed. Scottish Press, patients with rheumatoid arthiritis. Ann. Rheum. Dis., 52t215Edinburgh, pp. 67-71. 218. Lembeck, F., J. Donnerer, and F.C. Coluaert 1981 Increase of subGardner, E. 1948 The innervation of the shoulder joint. Anat. Rec., stance P in primary afferent nerves during chronic pain. Neuro102tl-18. peptides, 1t175-180. Gartsman, G.M., E.B. Major, P.C. Noble, J.B. Bennett, and H.S. Tullos Levine, J.D., R. Clark, M. Devor, C. Helms, M.A. Moskowitz, and A. 1988 Arthroscopic subacromial decompression. An anatomical Basbaum 1984 Intraneuronal substance P contributes to the sestudy. Am. J. Sports Med., 16t48-50. verity of experimental arthritis. Science, 225t547-549. Gottschall, J., W. Zenker, W. Neuhuber, A. Mysicka, and M. MdnLevine, J.D., S.J. Dardick, A.I. Basbaum, and E. Scipio 1985 Reflex tener 1980 The sternomastoid muscle of the rat and its innervaneurogenic inflammation. i. Contribution of the peripheral nertion, muscle fiber composition, perikarya and axons of efferent vous system to spatially remote inflammatory responses that foland afferent neurons. Anat. Emhryol., 16Ot285-300. low injury. J. Neurosci., 5t1380-1386. Greene, E.C. 1963 Anatomy of the Rat, vol. 27. Trans. Amer. Phil. Lloyd, D.P.C., and H.-T. Chang 1948 Afferent fibers in muscle nerves. SOC.,Hafner Publ. Comp., London, pp. 1-370. J . Neurophysiol., 11t199-208. Grigg, P., H. Schaible, and R.F. Schmidt 1986 Mechanical sensitivity McCormack, R.R., R. Inman, A. Wells, C. Berntsen, and H. Imbus of group I11 and IV afferents from posterior articular nerve in 1990 Prevalence of tendinitis and related disorders of the upper normal and inflamed cat knee. J. Neurophysiol., 55t635-643. extremity in a manufacturing workforce. J. Rheumatol., I7t958Guilbaud, G., A. Iggo, and R. Tegner 1985 Sensory receptors in ankle 964. joint capsules of normal and arthritic rats. Exp. Brain Res., 61: Mesulam, M.M. 1982 Principles of horseradish peroxidase neurohis164-168. tochemistry and their applications for tracing neural pathwaysHebel, R., and M.W. Stromberg 1986 Anatomy and Embryology ofthe axonal transport, enzyme histochemistry and light microscopic Laboratory Rat, BioMed Verlag, Worthsee. analysis. In: Tracing Neural Connections with Horseradish PerHildebrand, C., G. Oquist, L. Brax, and F. Tuisku 1991 Anatomy of oxidase. M.M. Mesulam, ed. Wiley and Sons, Chichester, pp. the rat knee joint and fibre composition of a major articular nerve 3-135. Anat. Rec., 229545-555. Moren Hybinette, I. 1987 The painful diabetic shoulder. Relationships Holje, L., C. Hildebrand, and K. Fried, 1985 Proportion of unmyelibetween diabetes mellitus, rheumatic symptoms and disability nated axons in the rat inferior alveolar nerve and mandibular pensioning. Thesis, University of Lund, Sweden. molar pulps after neonatal administration of capsaicin. Brain Res., 266t133-136. Nagy, J.I., M.L. Iversen, M. Goedert, D. Chapman, and S.P. Hunt 1983 Dose-dependent effects of capsaicin on primary sensory neuHolzer, P. 1988 Local effector functions of capsaicin-sensitive sensory rons in the neonatal rat. J. Neurosci., 3t399-406. nerve endings: involvement of tachykinins, calcitonin gene-related peptide and other neuropeptides. Neurosci. 24t739-768. Neer, C.S. 1972 Anterior acromioplasty for the chronic impingement syndrome of the shoulder. J. Bone Joint Surg., 54-A:41-50. Horiguchi, M. 1980 The cutaneous branch of some human suprascapular nerves. J. Anat., 13Ot191-195. Norlin, R. 1989 Arthroscopic subacromial decompression versus open acrornioplasty. Arthroscopy, 5t321-323. Jancso, G., M. Ferensic, G. Such, E. Kiralyi, A. Nagy, and M. Bujdoso ACKNOWLEDGMENTS 342 R. NORLIN ET AL. Petersson, C.J. 1986 Painful shoulders in patients with rheumatoid arthritis. Scand. J . Rheumatol., 15r275-279. Peyronnard, J.M., L. Charron, L. Lavoie, and J.P. Messier 1986 Motor, sympathetic and sensory innervation of rat skeletal muscles. Brain Res., 373t288-302. Romero, C., and S. Skoglund 1965 Methodological studies of the technique in measuring nerve fibre diameters. Acta Morphol. N e e d Scand., 6t107-114. Sandoz, P.A., and W. Zenker 1986 Unmyelinated axons in a muscle nerve. Electron microscopic morphometry of the sternomastoid nerve in normal and sympathectomized rats. Anat. Embryol., I74:207-213. Saria, A., J.M. Lundberg, G. Skofitsch and F. Lembeck 1983 Vascular protein leakage in various tissues induced by substance , capsaicin, bradykinin, serotonin, histamine and by antigen challenge. Naunyn Schmiedebergs Arch. Pharmacol., 324r212-218. Schaible, H.-G., and R.F. Smith 1983a Activation of groups I11 and IV sensory units in medial articular nerve by local mechanical stimulation of knee joint. J . Neurophysiol., 49t35-44. Schaible, H.-G., and R.F. Smith 1983b Responses of fine medial articular nerve afferents to passive movements of knee joint. J. Neurophysiol., 49tll18-1126. Sherrington, C.S. 1895 On the anatomical constitution of nerves of skeletal muscles; with remarks on recurrent fibres in the ventral spinal nerve-root. J. Physiol. (Lond.), 17t211-258. Szolcsanyi, J . 1988 Antidromic vasodilatation and neurogenic inflammation. Agents Actions, 23t4-11. Tomita, Y., J. Ozaki, and S. Tamai 1992 Sensory nerve endings in the subacromial bursa. Acta Orthop. Sacnd. Suppl., 247r63t22-23. Watson, M. 1989 Rotator cuff function in the impingement syndrome. J. Bone Joint Surg., 71 -Br361-366. Weinstein, J. 1991 Neurogenic and nonneurogenic pain and inflammatory mediators. Orthop. Clin. N. Am., 22r235-246. Yoshida, N., S.-I. Kikuchi, and Y. Sugiura 1992 Sympathetic and sensory innervation of the shoulder joint in the rat and human materials. Abstract 5th Int. Conf. Surg. Should. Paris, 010.