Veterinary Surgery, 19,2,93-101, 1990
Arthroscopic Approaches and lntraarticular Anatomy of the Equine Elbow ALAN J. NIXON, BVSc,
MS, DiplomateACVS
Periarticular anatomy and techniques for arthroscopic access to the equine elbow were studied in six joints from cadavers. Caudomedial and craniolateral approaches were evaluated subsequently in 11 anesthetized horses. The caudomedial approach was made between the flexor carpi radialis and flexor carpi ulnaris muscle bellies. Most of the caudal articular surfaces of the humeral condyles, the caudal perimeter of the radius, and the trochlear notch and portions of the anconeal process of the ulna could be identified. The voluminous caudal joint capsule cul-de-sac proximal to the anconeal process was readily entered. A 70' arthroscope allowed examination of more of the joint recesses and articular surfaces of the olecranon fossa than a 25' arthroscope. A second portal for intraarticular instrument manipulation was made caudal and slightly proximal to the arthroscope entry. Entry more proximal than the level of the radiohumeral articulation carried significant risk of damage to the ulnar nerve and collateral ulnar artery and vein. For examination of the cranial regions of the elbow, a craniolateral portal was established cranial to the lateral collateral ligament. An instrument portal was made through the muscle bellies of the extensor carpi radialis or common digital extensor muscles. The cranial articular surfaces of the humeral condyles were readily exposed by extension of the elbow. The weight-bearing articular surface of the radius could not be seen. Eight horses were euthanatized without recovery from anesthesia and the elbows were dissected for examination. Three horses were allowed to recover from anesthesia and were euthanatized on days 3, 30,and 60. The procedure resulted in mild lameness for 1 to 2 days. At necropsy, there was normal range of joint motion and none of the entry sites could be identified by days 30 and 60. No gross evidence of osteoarthritis had developed by day 60 and normal matrix glycosaminoglycan content was identified by histochemical staining at each time period.
are not common in horses but arise with sufficient frequency to warrant development of a technique for surgical exploration. Subchondral cystic lesions of the medial humeral condyle or medial portion of the proximal radius have been described in eight horses.' Although these horses were treated conservatively and most were eventually pasture sound, an increased number of athletes and a shorter time to recovery may have followed surgical debridement. Surgical debridement of subchondrdl cysts in the proximal aspect of the radius was described in three horses; some difficulty was encountered in precisely locating the subchondral cysts by an extraarticular transcortical approach.' An arthroscopic technique may allow examination ofthe joint surfaces and synovial lining tissues while providing the established benetits of arthroscopy, which include minimal entry wounds and rapid convalescence.
0
S T E O C H O N D R A L DISEASES OF T H E ELBOW J O I N T
The literature is replete with information on arthroscopy of the human knee but there is scant reference to arthroscopic elbow The periarticular neurovascular structures and inherent risks of elbow arthroscopy have been described.j In this article, similar neurovascular hazards in the horse and two arthroscopic approaches to the cranial and caudal portions ofthe equine elbow are described.
Materials and Methods Nineteen elbow joints were examined in 14 horses. The first six joints were examined in three equine cadavers to establish the proximity of important structures and accessible joint pouches. The remainder of the study was performed in anesthetized animals. The horses were 6 to 12 months old and weighed 280 to 327 kg. After the cadaver studies. three horses were anesthe-
From the Department of Clinical Sciences New York State College of Veterinary Meoicine. Cornell University. Ithaca, New York Reprint requests Alan J Nixon, BVSc, Department of Clinical Sciences New State College of Veterinary Medicine, Cornell University Ithaca, NY 14853
93
94
AH I HHUSLUP'Y U P I H t tLlUlNt tLBUW
Fig. 1. Caudornedial approach to the right elbow joint. The arthroscope entry point is between the flexor carpi radialis and flexor carpi ulnaris muscles, 2 cm distal to the radiohumeral articulation. The instrument entry incision is caudal and slightly proximal to the arthroscope entry and varies according to the hurneral condyle being approached.
tized and positioned in lateral recumbency with the elbow to be examined lowermost and the upper limb flexed and retracted caudally. This provided access to the medial surface of the elbow joint. After aseptic preparation and draping, a 5 mm skin incision was made 2 cm distal to the palpable level of the radiohumeral articulation and between the bellies of the flexor carpi radialis and flexor carpi ulnaris muscles. A 7.5 cm, 18 gauge spinal needle was inserted through the caudomedial aspect of the elbow joint capsule and 100 mL of lactated Ringer's solution was injected to distend the joint. After removing the needle, an arthroscope sleeve and sharp trocar were inserted and advanced toward the joint in an oblique proximal direction (Fig. 1). When cartilage or bone was encountered orjoint fluid returned through the egress outlets on the arthroscope sleeve, the sharp trocar was replaced by the blunt obturator and the sleeve was fully inserted into the joint. In many instances, the sleeve could be inserted to its limit as it penetrated into the caudal cul-de-sac of the elbow, proximal to the anconeal process of the ulna. The blunt obturator was removed and replaced by a 25" oblique forward-viewing, 4 mm arthroscope.* A fiberoptic light cable was attached to * Panoview Arthroscopc. Richard Wolf Medical Instruments, Roscm o n t 11.
supply light for diagnostic and photographic purposes from a dual light source and flash generator.? A video camera was attached for remote viewing. Lactated Ringer's solution was instilled under pressure throughout the procedure to provide distension of the joint capsule and to flush out blood and debris. Commencing at the caudal perimeter of the radiohumeral articulation, the caudal surfaces ofthe humeral condyles were examined by moving the tip of the arthroscope caudally, scanning the medial ulnarohumeral articulation, and penetrating the caudal cul-de-sac of the elbow joint. The 25" arthroscope was replaced by a 70" arthroscope and the caudal surface of the humerus was examined further. The site for a second entry for intraarticular manipulation was determined by placing a 7.5 cm spinal needle caudal and slightly proximal to the arthroscope. If the access angle and position were satisfactory, the needle was withdrawn and a 5 mm skin incision was made. Entry more proximal than the level ofthe radiohumeral articulation was avoided to reduce the risk of damage to the ulnar nerve and collateral ulnar vessels. The blunt obturator was used to establish a path into the joint and occasionally an 8.0 mm diameter Teflon sleeve$ was inserted to maintain patency for instruments such as probes, rongeurs, or motorized abraders. After the caudal region of the elbow joint was examined, excess lactated Ringer's solution was expressed from the joint and the instruments were removed. Each skin incision was sutured with a simple interrupted 2-0 monofilament nylon suture. The horse was rolled over to position the elbow being examined uppermost. The lateral aspect of the elbow was prepared aseptically and draped. An 18 gauge needle was inserted to redistend the joint with lactated Ringer's solution. The cranial perimeter of the radiohumeral articulation was palpated cranial to the lateral collateral ligament and a 5 mm stab incision was made 2 cm cranial to this border, avoiding the belly of the common digital extensor muscle cranially. The arthroscope was inserted and the cranial articular surfaces and cranial joint pouch were examined (Fig. 2). The arthroscope was inserted as deeply as possible to examine the craniomedial margins of the radius and humerus. A 70" arthroscope was used to examine the limits of the craniomedial region. A second 5 mm skin incision was made cranial or caudal to the arthroscope entry to provide access for hand instruments. Entry to the elbow cranially usually penetrated through the antebrachial extensor muscles. The mid-cranial region was avoided to minimize potential damage to the transverse cubital artery.
t Flash Generator S007. Richard Wolf Medical Instruments. Rosemont. IL. $ Univcrsal Cannula. Acufeex Microsurgical. Norwood. MA.
95
NIXON
photography, 5 mm sagittal slabs from the mid-medial and mid-lateral condyle regions of the humerus, radius, and ulna were cut and fixed in buffered formalin. The slabs were radiographed, decalcified in buffered formic acid, embedded in paraffin, and sectioned at 6 pm for histologic analysis. Sections were stained with hematoxylin-eosin, safranin-0 and fast green, and alcian bluenuclear fast red. Cellular derangement and matrix glycosaminoglycan depletion were evaluated.
Results Late
Fig 2 Craniolateral approach to the right elbow joint The arthroscope enters the joint caudal to the common digital extensor muscle belly The instrument portal can be made adjacent to the arthroscope or more cranial, depending on the target area Entry on the craniomedial surface IS prevented by the median artery, vein, and nerve
Upon completion of the examination the excess joint fluid was expressed, the instruments were removed, and the skin incisions were sutured. Each horse was euthanatized without recovering from anesthesia and the elbow was dissected to investigate the entry wounds, the proximity of important neurovascular structures, and the extent of iatrogenic damage to the joint capsule and articular surfaces. Eight horses were positioned in dorsal recumbency with one forelimb suspended vertically. The medial and lateral surfaces of the elbow were prepared aseptically and draped for surgery. The craniolateral and caudomedial approaches to the elbow were performed as before. In two horses, the procedure was performed bilaterally. Five horses operated on in dorsal recumbency were euthanatized at the completion of the procedure and the elbows were dissected for evaluation. Three horses, free of clinical forelimb lameness, were allowed to recover from anesthesia and were euthanatized and necropsied on day 3 , 30. or 60. Before euthanasia they were examined daily for periarticular swelling, pain on palpation, and lameness. Phenylbutazone (4.4 mg/kg orally) was administered for the first 2 days after surgery. The horses were confined to a stall for 30 days and the horse evaluated at 60 days was confined to a 0.5 acre pasture for the last 30 days. After euthanasia, gross examination. and
The elbow joint could not be examined from one entry point. The cranial and caudal joint pouches were examined through craniolateral and caudomedial entries, respectively. A thorough understanding of the surface and deep anatomy of the medial aspects of the elbow and axilla was required for safe entry by the caudomedial approach. Entry cranial to the flexor carpi radialis muscle entailed risk of damage to the median nerve, artery, and veins, especially if the entry point was more than 5 cm distal to the radiohumeral articulation. The short medial collateral ligament and its eminence for attachment on the medial epicondyle of the humerus were important palpable landmarks to define the radiohumeral articulation. The space between the flexor carpi radialis and flexor carpi ulnaris muscles was somewhat obscured by the overlying superficial pectoral muscles and antebrachial fascia, but could be reliably located by palpating the myotendinous junctions of these muscles and then moving proximad. Penetration of the joint proximal to the level of the radiohumeral articulation placed the ulnar neurovascular structures at risk, particularly when a second entry was made caudal to the flexor carpi ulnaris muscle. After entering the caudomedial aspect of the elbow, the medial humeral condyle was readily visible. The caudal perimeter of the radiohumeral articulation was a convenient landmark to commence examination of the joint (Fig. 3 ) . The articular surfaces of the humeral condyles. particularly the medial condyle, were evaluated. Flexion of the joint improved the exposure of the condyles. Portions of the weight-bearing articular surface of the radius could be seen but the arthroscope could not be advanced between the radius and humerus (Fig. 4). With the horse positioned in lateral rather than dorsal recumbency, it was easier to distract the medial aspect of the radiohumeral joint by abducting the limb. Manipulating the tip of the arthroscope caudad brought the trochlear notch of the ulna and apposing articulation ofthe humeral condyles into view (Fig. 5). Further caudad, the limit of the medial epicondyle of the humerus, the trochlear notch of the ulna, and the large tendon of origin of the humeral head of the deep digital flexor tendon were seen (Fig. 6). The space between the trochlear notch of the
96
ARTHROSCOPY OF THE EQUINE ELBOW
Fig 3 Caudornedial views of the right elbow A The intraarticular field of view (circle) B lntraarticular view with a 25' arthroscope The medial (M) and lateral (L) condyles of the humerus, the radius (R), and the ulna (U) are visible The caudal extent of the synovial fossa of the humeral head is evident (arrow)
ulna and this movable tendon provided access to the caudal joint cul-de-sac proximal to the anconeal process of the ulna (Fig. 7). The arthroscope readily entered this cul-desac to provide an assessment of the proximal aspects of the anconeal process and the extensive synovial membrane,
which had a cobblestone appearance (Fig. 8). A 70" arthroscope was necessary for a complete assessment of this culde-sac and the anconeal process. An instrument entry portal caudal to the arthroscope entry allowed surgical manipulation by triangulation.
Fig 4 Caudomedial view of the right elbow A The intraarticular field of view deeper and cranial to the view in Figure 3 B lntraarticular view Portions of the capitulum fovea of the radius (R) are visible
NIXON
97
Fig 5 Caudomedial view of the right elbow A The field of view in the caudal portion of the elbow with the arthroscope redirected more caudad B The medial (M) and lateral (L) hurneral condyles and their articulation with the trochlear notch of the ulna (U) are evident
The most suitable path for instrument entry was selected by inserting a 7.5 cm spinal needle. There was little difficulty in gaining access to the medial condyle of the humerus; however. the lateral condyle was less accessible and the central regions ofthe capitular fovea ofthe radius could not be reached. A separate caudolateral instru-
Fig 6
ment entry was required for manipulation within the caudal joint capsule cul-de-sac proximal to the anconeal process. When the horses were in lateral recumbency, it was necessary to reposition them for the craniolateral approach to the cranial portions of the elbow joint (Fig. 9).
Caudomedial view of the right elbow A The caudomedial field of view with the arthroscope tip directed more caudad than in Figure 5
B The medial condyle and epicondyle of the humerus (H) and the trochlear notch of the ulna (U) are visible
98
ARTHROSCOPY OF THE EQUINE ELBOW
Fig. 7. Caudomedial view of the right elbow. A. The intraarticular field of view (circle) when the arthroscope is inserted deeper and directed more caudad than in Figure 66. Arthroscopic view of the junction of the medial epicondyle of the humerus (H), trochlear notch of the ulna (U), and the humeral head of the deep digital flexor tendon (DF).
The cranial portion of the joint cavity was quite extensive and the arthroscope could be inserted deeply in a lateral-to-medial direction to examine the cranial articular surfaces of the humeral condyles and the cranial pe-
nmeter of the radius (Fig. 10). A synovial fossa was frequently present in the humeral cartilage between the lateral and medial condyles. Maximum exposure of the humeral condyles was obtained by extending the elbow.
Fig 8 Caudomedial view of the right elbow A The intraarticular field of view (circle) with the arthroscope at the entry to the caudoproximal cul-de-sac B The 25" arthroscope is advanced between the epicondyle of the humerus (H). ulna (U), and deep flexor tendon (DF) This allows entry and inspection of the voluminous caudal joint cul-de-sac proximal to the ulnarohumeral articulation Most of the anconeal process also can be examined by using a 70" arthroscope
NIXON
99
Fig. 9. Craniolateral view of the right elbow. A. The intraarticular field of view with a 25' arthroscope (circle). 6.The lateral (L) and medial (M) humeral condyles articulate with the radius (R). A synovial band extends from the radial fossa on the humerus to its attachment on the cranial aspect of the radius (arrow)
A 70" arthroscope provided visualization of the cranio-
medial surfaces of the joint as far caudal as the medial collateral ligament. The synovial membrane was relatively smooth throughout the cranial joint pouch: however, several synovial bands could occasionally be seen (Fig. 9). Withdrawing the arthroscope toward the lateral articular limit and redirecting it down the lateral surface of the joint exposed the large and inelastic lateral collateral ligament (Fig. 1 I ) . It was not possible to place the arthroscope beyond the lateral collateral ligament to gain access to the caudal regions ofthe elbow from this cranial entry. An instrument entry point was made either cranial or caudal to the arthroscope entry incision. An approach cranially penetrated through the common digital extensor or extensor carpi radialis muscle belly. Hemorrhage was minimal because the mid-cranial location of the transverse cubital artery was avoided. The radial and humeral articular surfaces could not be distracted sufficiently by traction on the lower limb or with forceps placed intraarticularly to allow visualization of the capitular fovea of the radius. Positioning horses in dorsal recumbency obviated the need to reposition and redrape for the craniolateral and eaudomedial approaches. However, dorsal recumbency led to increased difficulty in establishing the caudomedial entry. The craniolateral entry was relatively simple with either recumbency. Nevertheless, once in the joint the exposure and assessment of all regions of the elbow were similar regardless of the recumbency used.
Three designated horses recovered from anesthesia and surgery without complication. Penarticular fluid which had accumulated at surgery dissipated over the following 2 to 3 days. Mild lameness was evident at a walk for 12 to 24 hours but not at a walk or trot after day 2. The sutured skin wounds healed without complication and were difficult to detect by day 30. In the eight horses euthanatized immediately after the arthroscopic procedure, it was found that the arthroscope consistently entered the caudomedial joint capsule in the space between the caudal perimeter of the radius and the large tendon of origin of the humeral head of the deep digital flexor muscle. The entry wound in the joint capsule was typically 5 to 7 m m long. The instrument portal frequently penetrated between the flexor carpi ulnaris and deep digital flexor muscle bellies. The articular cartilage surfaces were rarely damaged except where a linear mark was made on the humeral condyles to later identify the limits of the exposed condyle at necropsy inspection. The craniolateral approach entered the cranial portion of the elbow joint through a cul-de-sac between the common digital extensor muscle and the craniolateral angle ofthe radiohumeral articulation. The entry wound made a 4 to 5 mm opening in the joint capsule and was safely proximal to branches of the radial nerve. There was no damage to the articular surfaces. Instrument portals penetrating through the lateral portions of the antebrachial extensor muscles caused a 3 to 4 mm zone of hemorrhage around the pathway of repeated instrument entry.
100
ARTHROSCOPY OF THE EQUINE ELBOW
Fig. 10. Craniolateral view of the right elbow. A. Field of view with the arthroscope advanced deeper into the joint. B. The synovial fossa (arrow) between the lateral (L) and medial (M) humeral condyles is visible. Further exposure can be obtained by extending the elbow.
No extensive muscle damage was seen and the transverse cubital artery and branches of the radial nerve were not encountered. Results of necropsy examination of the horses euthanatized on days 3 , 30, and 60 showed minimal iatrogenic damage. Although entry sites for the arthroscope and instruments could be identified in the horse examined on
day 3 , there was no evidence of entry wounds at days 30 or 60. The linear identification marks on the medial condyle of the humerus were still visible. Histologically, tissues from the midlateral and midmedial sections of the humeral condyle and radial capitular fovea slabs had normal cartilage and subchondral bone architecture. N o depletion of the matrix glycosami-
Fig. 11. Craniolateral view of the right elbow. A. Field of view with the arthroscope withdrawn and directed caudad. B. The lateral perimeter of the radius (R) and humerus (H) and the lateral collateral ligament (LC) are visible.
101
NIXON noglycan content was recognized by safranin-0 or alcian blue histochemical staining at any of the time periods studied. Discussion Two separate and independent arthroscopic approaches were needed to examine the cranial and caudal regions of the equine elbow. The surgical approach and positioning of the animal were varied according to the region being examined. The caudomedial approach was generally more versatile and provided adequate exposure of the medial humeral condyle. The subchondral cysts involving the humerus reported in the literature were all located in the medial condyle.' This approach would allow a lesion in the capitular fovea of the radius to be viewed remotely, but instrument access into the lesion would not be possible. The extraarticular approach to radial cysts may still be the only available surgical therapy.' Only the medial portion of the lateral humeral condyle could be viewed and the cartilage surfaces probed. Access for larger hand tools was limited. With practice and the knowledge gained by prior anatomic dissections, the caudomedial approach was consistently achieved. Nevertheless, it was common to make several attempts to locate the arthroscope correctly within the joint. Furthermore, the proximity of vital neurovascular structures added to the complexity of the procedure. The median neurovascular bundle was directly vulnerable to injury if the caudomedial entry portal was inadvertently made cranial to the flexor carpi radialis muscle. An instrument portal caudal and proximal to the flexor carpi ulnaris muscle endangered the ulnar neurovascular structures, particularly the ulnar nerve. These structures coursed immediately caudal to the normal instrument entry portal, located between the flexor carpi ulnans and deep digital flexor muscles or occasionally directly over the muscle belly of the flexor carpi ulnaris. Similar hazards were reported for the medial approach to the elbow in humans.' The craniolateral approach was comparatively simple but provided access to regions of less clinical significance. Terminal branches of the radial nerve and collateral vessels were at some risk during entry of instruments between the extensor carpi radialis and common digital extensor muscles. Second entry portals, providing access to cranial structures within the elbow, were usually selected
'
over the belly of the common digital extensor muscle. More medially, the bulk of the extensor carpi radialis muscle restricted instrument mobility and a direct medial approach was precluded by the median nerve. The inherent neurovascular risks of the craniolateral approach to the elbow in horses followed a pattern similar to that in humans with the radial nerve being vulnerable.' The surgical approaches and intraarticular manipulations were facilitated by placing the horses in lateral recumbency. Dorsal recumbency made the caudomedial approach more difficult. Therefore, the site of the lesion should dictate the positioning of the animal. If the cranial and caudal aspects of the elbow joint need to be assessed, dorsal recumbency would be the logical choice. For discrete lesions, lateral recumbency would facilitate either craniolateral or caudomedial entry. Lesions of the radial capital fovea were not well addressed by the arthroscopic approaches described here. Inasmuch as subchondral cysts in the radial head are reportedly as common as those in the medial humeral condyle,' this is disappointing and is an area in need of further study. Distraction of the elbow with an ankle distractors or a fracture distractorll attached with threaded Steinmann pins in the humerus and radius was inadequate to allow complete assessment of the capitular surface of the radius. References 1. Hopen LA, Turner TA. Colahan P. Nixon AJ. Elbow subchondral
-.7 3.
4.
5. 6.
7.
8.
bone cysts in the horse: Conservative treatmeni of seven cases. J Am Vet Med Assoc (accepted for publication) IY90. Bertone AL. McIIwraith CW. Powers BE. et al. Subchondral osseous cystic lesions of the elbow of horses: conservative versus surgical treatment. J Am Vet Med Assoc 1986: 189:540-546. Lynch GJ. Meyers JF. Whipple TL. Caspari RB. Neurovascular anatomy and elbow arthroscopy: inherent risks. Arthroscopy I986:z: 191-197. Clarke RP. Symptomatic. lateral synovial fringe (plica) of the elbowjoint. Arthroscopy 1988:4:1 12-1 16. Guhl JF. Arthroscopy and arthroscopic surgery of the elbow. Orthopedics 1985;8: 1290-1196. Boe S. Arthroscopy ofthe elbow: diagnosis and extraction of loose bodies. Acta Orthop Scand 1986:57:52-53. Andrews JR. Carson WG. Arthroscopy of the elbow. Arthroscopy 1985: 197-107. Thomas MA. Fast A. Shapiro D. Radial nerve damage as a complication of elbow arthroscopy. Clin Orthop l987:2 15: 130- I3 I .
4 Ankle distractor. Acukx Microsurgical, Norwood, MA. (I
Femoral fracture distractor. Synthes Ltd (USA), Wayne. PA.
Arthroscopic Approaches and lntraarticular Anatomy of the Equine Elbow ALAN J. NIXON, BVSc,
MS, DiplomateACVS
Periarticular anatomy and techniques for arthroscopic access to the equine elbow were studied in six joints from cadavers. Caudomedial and craniolateral approaches were evaluated subsequently in 11 anesthetized horses. The caudomedial approach was made between the flexor carpi radialis and flexor carpi ulnaris muscle bellies. Most of the caudal articular surfaces of the humeral condyles, the caudal perimeter of the radius, and the trochlear notch and portions of the anconeal process of the ulna could be identified. The voluminous caudal joint capsule cul-de-sac proximal to the anconeal process was readily entered. A 70' arthroscope allowed examination of more of the joint recesses and articular surfaces of the olecranon fossa than a 25' arthroscope. A second portal for intraarticular instrument manipulation was made caudal and slightly proximal to the arthroscope entry. Entry more proximal than the level of the radiohumeral articulation carried significant risk of damage to the ulnar nerve and collateral ulnar artery and vein. For examination of the cranial regions of the elbow, a craniolateral portal was established cranial to the lateral collateral ligament. An instrument portal was made through the muscle bellies of the extensor carpi radialis or common digital extensor muscles. The cranial articular surfaces of the humeral condyles were readily exposed by extension of the elbow. The weight-bearing articular surface of the radius could not be seen. Eight horses were euthanatized without recovery from anesthesia and the elbows were dissected for examination. Three horses were allowed to recover from anesthesia and were euthanatized on days 3, 30,and 60. The procedure resulted in mild lameness for 1 to 2 days. At necropsy, there was normal range of joint motion and none of the entry sites could be identified by days 30 and 60. No gross evidence of osteoarthritis had developed by day 60 and normal matrix glycosaminoglycan content was identified by histochemical staining at each time period.
are not common in horses but arise with sufficient frequency to warrant development of a technique for surgical exploration. Subchondral cystic lesions of the medial humeral condyle or medial portion of the proximal radius have been described in eight horses.' Although these horses were treated conservatively and most were eventually pasture sound, an increased number of athletes and a shorter time to recovery may have followed surgical debridement. Surgical debridement of subchondrdl cysts in the proximal aspect of the radius was described in three horses; some difficulty was encountered in precisely locating the subchondral cysts by an extraarticular transcortical approach.' An arthroscopic technique may allow examination ofthe joint surfaces and synovial lining tissues while providing the established benetits of arthroscopy, which include minimal entry wounds and rapid convalescence.
0
S T E O C H O N D R A L DISEASES OF T H E ELBOW J O I N T
The literature is replete with information on arthroscopy of the human knee but there is scant reference to arthroscopic elbow The periarticular neurovascular structures and inherent risks of elbow arthroscopy have been described.j In this article, similar neurovascular hazards in the horse and two arthroscopic approaches to the cranial and caudal portions ofthe equine elbow are described.
Materials and Methods Nineteen elbow joints were examined in 14 horses. The first six joints were examined in three equine cadavers to establish the proximity of important structures and accessible joint pouches. The remainder of the study was performed in anesthetized animals. The horses were 6 to 12 months old and weighed 280 to 327 kg. After the cadaver studies. three horses were anesthe-
From the Department of Clinical Sciences New York State College of Veterinary Meoicine. Cornell University. Ithaca, New York Reprint requests Alan J Nixon, BVSc, Department of Clinical Sciences New State College of Veterinary Medicine, Cornell University Ithaca, NY 14853
93
94
AH I HHUSLUP'Y U P I H t tLlUlNt tLBUW
Fig. 1. Caudornedial approach to the right elbow joint. The arthroscope entry point is between the flexor carpi radialis and flexor carpi ulnaris muscles, 2 cm distal to the radiohumeral articulation. The instrument entry incision is caudal and slightly proximal to the arthroscope entry and varies according to the hurneral condyle being approached.
tized and positioned in lateral recumbency with the elbow to be examined lowermost and the upper limb flexed and retracted caudally. This provided access to the medial surface of the elbow joint. After aseptic preparation and draping, a 5 mm skin incision was made 2 cm distal to the palpable level of the radiohumeral articulation and between the bellies of the flexor carpi radialis and flexor carpi ulnaris muscles. A 7.5 cm, 18 gauge spinal needle was inserted through the caudomedial aspect of the elbow joint capsule and 100 mL of lactated Ringer's solution was injected to distend the joint. After removing the needle, an arthroscope sleeve and sharp trocar were inserted and advanced toward the joint in an oblique proximal direction (Fig. 1). When cartilage or bone was encountered orjoint fluid returned through the egress outlets on the arthroscope sleeve, the sharp trocar was replaced by the blunt obturator and the sleeve was fully inserted into the joint. In many instances, the sleeve could be inserted to its limit as it penetrated into the caudal cul-de-sac of the elbow, proximal to the anconeal process of the ulna. The blunt obturator was removed and replaced by a 25" oblique forward-viewing, 4 mm arthroscope.* A fiberoptic light cable was attached to * Panoview Arthroscopc. Richard Wolf Medical Instruments, Roscm o n t 11.
supply light for diagnostic and photographic purposes from a dual light source and flash generator.? A video camera was attached for remote viewing. Lactated Ringer's solution was instilled under pressure throughout the procedure to provide distension of the joint capsule and to flush out blood and debris. Commencing at the caudal perimeter of the radiohumeral articulation, the caudal surfaces ofthe humeral condyles were examined by moving the tip of the arthroscope caudally, scanning the medial ulnarohumeral articulation, and penetrating the caudal cul-de-sac of the elbow joint. The 25" arthroscope was replaced by a 70" arthroscope and the caudal surface of the humerus was examined further. The site for a second entry for intraarticular manipulation was determined by placing a 7.5 cm spinal needle caudal and slightly proximal to the arthroscope. If the access angle and position were satisfactory, the needle was withdrawn and a 5 mm skin incision was made. Entry more proximal than the level ofthe radiohumeral articulation was avoided to reduce the risk of damage to the ulnar nerve and collateral ulnar vessels. The blunt obturator was used to establish a path into the joint and occasionally an 8.0 mm diameter Teflon sleeve$ was inserted to maintain patency for instruments such as probes, rongeurs, or motorized abraders. After the caudal region of the elbow joint was examined, excess lactated Ringer's solution was expressed from the joint and the instruments were removed. Each skin incision was sutured with a simple interrupted 2-0 monofilament nylon suture. The horse was rolled over to position the elbow being examined uppermost. The lateral aspect of the elbow was prepared aseptically and draped. An 18 gauge needle was inserted to redistend the joint with lactated Ringer's solution. The cranial perimeter of the radiohumeral articulation was palpated cranial to the lateral collateral ligament and a 5 mm stab incision was made 2 cm cranial to this border, avoiding the belly of the common digital extensor muscle cranially. The arthroscope was inserted and the cranial articular surfaces and cranial joint pouch were examined (Fig. 2). The arthroscope was inserted as deeply as possible to examine the craniomedial margins of the radius and humerus. A 70" arthroscope was used to examine the limits of the craniomedial region. A second 5 mm skin incision was made cranial or caudal to the arthroscope entry to provide access for hand instruments. Entry to the elbow cranially usually penetrated through the antebrachial extensor muscles. The mid-cranial region was avoided to minimize potential damage to the transverse cubital artery.
t Flash Generator S007. Richard Wolf Medical Instruments. Rosemont. IL. $ Univcrsal Cannula. Acufeex Microsurgical. Norwood. MA.
95
NIXON
photography, 5 mm sagittal slabs from the mid-medial and mid-lateral condyle regions of the humerus, radius, and ulna were cut and fixed in buffered formalin. The slabs were radiographed, decalcified in buffered formic acid, embedded in paraffin, and sectioned at 6 pm for histologic analysis. Sections were stained with hematoxylin-eosin, safranin-0 and fast green, and alcian bluenuclear fast red. Cellular derangement and matrix glycosaminoglycan depletion were evaluated.
Results Late
Fig 2 Craniolateral approach to the right elbow joint The arthroscope enters the joint caudal to the common digital extensor muscle belly The instrument portal can be made adjacent to the arthroscope or more cranial, depending on the target area Entry on the craniomedial surface IS prevented by the median artery, vein, and nerve
Upon completion of the examination the excess joint fluid was expressed, the instruments were removed, and the skin incisions were sutured. Each horse was euthanatized without recovering from anesthesia and the elbow was dissected to investigate the entry wounds, the proximity of important neurovascular structures, and the extent of iatrogenic damage to the joint capsule and articular surfaces. Eight horses were positioned in dorsal recumbency with one forelimb suspended vertically. The medial and lateral surfaces of the elbow were prepared aseptically and draped for surgery. The craniolateral and caudomedial approaches to the elbow were performed as before. In two horses, the procedure was performed bilaterally. Five horses operated on in dorsal recumbency were euthanatized at the completion of the procedure and the elbows were dissected for evaluation. Three horses, free of clinical forelimb lameness, were allowed to recover from anesthesia and were euthanatized and necropsied on day 3 , 30. or 60. Before euthanasia they were examined daily for periarticular swelling, pain on palpation, and lameness. Phenylbutazone (4.4 mg/kg orally) was administered for the first 2 days after surgery. The horses were confined to a stall for 30 days and the horse evaluated at 60 days was confined to a 0.5 acre pasture for the last 30 days. After euthanasia, gross examination. and
The elbow joint could not be examined from one entry point. The cranial and caudal joint pouches were examined through craniolateral and caudomedial entries, respectively. A thorough understanding of the surface and deep anatomy of the medial aspects of the elbow and axilla was required for safe entry by the caudomedial approach. Entry cranial to the flexor carpi radialis muscle entailed risk of damage to the median nerve, artery, and veins, especially if the entry point was more than 5 cm distal to the radiohumeral articulation. The short medial collateral ligament and its eminence for attachment on the medial epicondyle of the humerus were important palpable landmarks to define the radiohumeral articulation. The space between the flexor carpi radialis and flexor carpi ulnaris muscles was somewhat obscured by the overlying superficial pectoral muscles and antebrachial fascia, but could be reliably located by palpating the myotendinous junctions of these muscles and then moving proximad. Penetration of the joint proximal to the level of the radiohumeral articulation placed the ulnar neurovascular structures at risk, particularly when a second entry was made caudal to the flexor carpi ulnaris muscle. After entering the caudomedial aspect of the elbow, the medial humeral condyle was readily visible. The caudal perimeter of the radiohumeral articulation was a convenient landmark to commence examination of the joint (Fig. 3 ) . The articular surfaces of the humeral condyles. particularly the medial condyle, were evaluated. Flexion of the joint improved the exposure of the condyles. Portions of the weight-bearing articular surface of the radius could be seen but the arthroscope could not be advanced between the radius and humerus (Fig. 4). With the horse positioned in lateral rather than dorsal recumbency, it was easier to distract the medial aspect of the radiohumeral joint by abducting the limb. Manipulating the tip of the arthroscope caudad brought the trochlear notch of the ulna and apposing articulation ofthe humeral condyles into view (Fig. 5). Further caudad, the limit of the medial epicondyle of the humerus, the trochlear notch of the ulna, and the large tendon of origin of the humeral head of the deep digital flexor tendon were seen (Fig. 6). The space between the trochlear notch of the
96
ARTHROSCOPY OF THE EQUINE ELBOW
Fig 3 Caudornedial views of the right elbow A The intraarticular field of view (circle) B lntraarticular view with a 25' arthroscope The medial (M) and lateral (L) condyles of the humerus, the radius (R), and the ulna (U) are visible The caudal extent of the synovial fossa of the humeral head is evident (arrow)
ulna and this movable tendon provided access to the caudal joint cul-de-sac proximal to the anconeal process of the ulna (Fig. 7). The arthroscope readily entered this cul-desac to provide an assessment of the proximal aspects of the anconeal process and the extensive synovial membrane,
which had a cobblestone appearance (Fig. 8). A 70" arthroscope was necessary for a complete assessment of this culde-sac and the anconeal process. An instrument entry portal caudal to the arthroscope entry allowed surgical manipulation by triangulation.
Fig 4 Caudomedial view of the right elbow A The intraarticular field of view deeper and cranial to the view in Figure 3 B lntraarticular view Portions of the capitulum fovea of the radius (R) are visible
NIXON
97
Fig 5 Caudomedial view of the right elbow A The field of view in the caudal portion of the elbow with the arthroscope redirected more caudad B The medial (M) and lateral (L) hurneral condyles and their articulation with the trochlear notch of the ulna (U) are evident
The most suitable path for instrument entry was selected by inserting a 7.5 cm spinal needle. There was little difficulty in gaining access to the medial condyle of the humerus; however. the lateral condyle was less accessible and the central regions ofthe capitular fovea ofthe radius could not be reached. A separate caudolateral instru-
Fig 6
ment entry was required for manipulation within the caudal joint capsule cul-de-sac proximal to the anconeal process. When the horses were in lateral recumbency, it was necessary to reposition them for the craniolateral approach to the cranial portions of the elbow joint (Fig. 9).
Caudomedial view of the right elbow A The caudomedial field of view with the arthroscope tip directed more caudad than in Figure 5
B The medial condyle and epicondyle of the humerus (H) and the trochlear notch of the ulna (U) are visible
98
ARTHROSCOPY OF THE EQUINE ELBOW
Fig. 7. Caudomedial view of the right elbow. A. The intraarticular field of view (circle) when the arthroscope is inserted deeper and directed more caudad than in Figure 66. Arthroscopic view of the junction of the medial epicondyle of the humerus (H), trochlear notch of the ulna (U), and the humeral head of the deep digital flexor tendon (DF).
The cranial portion of the joint cavity was quite extensive and the arthroscope could be inserted deeply in a lateral-to-medial direction to examine the cranial articular surfaces of the humeral condyles and the cranial pe-
nmeter of the radius (Fig. 10). A synovial fossa was frequently present in the humeral cartilage between the lateral and medial condyles. Maximum exposure of the humeral condyles was obtained by extending the elbow.
Fig 8 Caudomedial view of the right elbow A The intraarticular field of view (circle) with the arthroscope at the entry to the caudoproximal cul-de-sac B The 25" arthroscope is advanced between the epicondyle of the humerus (H). ulna (U), and deep flexor tendon (DF) This allows entry and inspection of the voluminous caudal joint cul-de-sac proximal to the ulnarohumeral articulation Most of the anconeal process also can be examined by using a 70" arthroscope
NIXON
99
Fig. 9. Craniolateral view of the right elbow. A. The intraarticular field of view with a 25' arthroscope (circle). 6.The lateral (L) and medial (M) humeral condyles articulate with the radius (R). A synovial band extends from the radial fossa on the humerus to its attachment on the cranial aspect of the radius (arrow)
A 70" arthroscope provided visualization of the cranio-
medial surfaces of the joint as far caudal as the medial collateral ligament. The synovial membrane was relatively smooth throughout the cranial joint pouch: however, several synovial bands could occasionally be seen (Fig. 9). Withdrawing the arthroscope toward the lateral articular limit and redirecting it down the lateral surface of the joint exposed the large and inelastic lateral collateral ligament (Fig. 1 I ) . It was not possible to place the arthroscope beyond the lateral collateral ligament to gain access to the caudal regions ofthe elbow from this cranial entry. An instrument entry point was made either cranial or caudal to the arthroscope entry incision. An approach cranially penetrated through the common digital extensor or extensor carpi radialis muscle belly. Hemorrhage was minimal because the mid-cranial location of the transverse cubital artery was avoided. The radial and humeral articular surfaces could not be distracted sufficiently by traction on the lower limb or with forceps placed intraarticularly to allow visualization of the capitular fovea of the radius. Positioning horses in dorsal recumbency obviated the need to reposition and redrape for the craniolateral and eaudomedial approaches. However, dorsal recumbency led to increased difficulty in establishing the caudomedial entry. The craniolateral entry was relatively simple with either recumbency. Nevertheless, once in the joint the exposure and assessment of all regions of the elbow were similar regardless of the recumbency used.
Three designated horses recovered from anesthesia and surgery without complication. Penarticular fluid which had accumulated at surgery dissipated over the following 2 to 3 days. Mild lameness was evident at a walk for 12 to 24 hours but not at a walk or trot after day 2. The sutured skin wounds healed without complication and were difficult to detect by day 30. In the eight horses euthanatized immediately after the arthroscopic procedure, it was found that the arthroscope consistently entered the caudomedial joint capsule in the space between the caudal perimeter of the radius and the large tendon of origin of the humeral head of the deep digital flexor muscle. The entry wound in the joint capsule was typically 5 to 7 m m long. The instrument portal frequently penetrated between the flexor carpi ulnaris and deep digital flexor muscle bellies. The articular cartilage surfaces were rarely damaged except where a linear mark was made on the humeral condyles to later identify the limits of the exposed condyle at necropsy inspection. The craniolateral approach entered the cranial portion of the elbow joint through a cul-de-sac between the common digital extensor muscle and the craniolateral angle ofthe radiohumeral articulation. The entry wound made a 4 to 5 mm opening in the joint capsule and was safely proximal to branches of the radial nerve. There was no damage to the articular surfaces. Instrument portals penetrating through the lateral portions of the antebrachial extensor muscles caused a 3 to 4 mm zone of hemorrhage around the pathway of repeated instrument entry.
100
ARTHROSCOPY OF THE EQUINE ELBOW
Fig. 10. Craniolateral view of the right elbow. A. Field of view with the arthroscope advanced deeper into the joint. B. The synovial fossa (arrow) between the lateral (L) and medial (M) humeral condyles is visible. Further exposure can be obtained by extending the elbow.
No extensive muscle damage was seen and the transverse cubital artery and branches of the radial nerve were not encountered. Results of necropsy examination of the horses euthanatized on days 3 , 30, and 60 showed minimal iatrogenic damage. Although entry sites for the arthroscope and instruments could be identified in the horse examined on
day 3 , there was no evidence of entry wounds at days 30 or 60. The linear identification marks on the medial condyle of the humerus were still visible. Histologically, tissues from the midlateral and midmedial sections of the humeral condyle and radial capitular fovea slabs had normal cartilage and subchondral bone architecture. N o depletion of the matrix glycosami-
Fig. 11. Craniolateral view of the right elbow. A. Field of view with the arthroscope withdrawn and directed caudad. B. The lateral perimeter of the radius (R) and humerus (H) and the lateral collateral ligament (LC) are visible.
101
NIXON noglycan content was recognized by safranin-0 or alcian blue histochemical staining at any of the time periods studied. Discussion Two separate and independent arthroscopic approaches were needed to examine the cranial and caudal regions of the equine elbow. The surgical approach and positioning of the animal were varied according to the region being examined. The caudomedial approach was generally more versatile and provided adequate exposure of the medial humeral condyle. The subchondral cysts involving the humerus reported in the literature were all located in the medial condyle.' This approach would allow a lesion in the capitular fovea of the radius to be viewed remotely, but instrument access into the lesion would not be possible. The extraarticular approach to radial cysts may still be the only available surgical therapy.' Only the medial portion of the lateral humeral condyle could be viewed and the cartilage surfaces probed. Access for larger hand tools was limited. With practice and the knowledge gained by prior anatomic dissections, the caudomedial approach was consistently achieved. Nevertheless, it was common to make several attempts to locate the arthroscope correctly within the joint. Furthermore, the proximity of vital neurovascular structures added to the complexity of the procedure. The median neurovascular bundle was directly vulnerable to injury if the caudomedial entry portal was inadvertently made cranial to the flexor carpi radialis muscle. An instrument portal caudal and proximal to the flexor carpi ulnaris muscle endangered the ulnar neurovascular structures, particularly the ulnar nerve. These structures coursed immediately caudal to the normal instrument entry portal, located between the flexor carpi ulnans and deep digital flexor muscles or occasionally directly over the muscle belly of the flexor carpi ulnaris. Similar hazards were reported for the medial approach to the elbow in humans.' The craniolateral approach was comparatively simple but provided access to regions of less clinical significance. Terminal branches of the radial nerve and collateral vessels were at some risk during entry of instruments between the extensor carpi radialis and common digital extensor muscles. Second entry portals, providing access to cranial structures within the elbow, were usually selected
'
over the belly of the common digital extensor muscle. More medially, the bulk of the extensor carpi radialis muscle restricted instrument mobility and a direct medial approach was precluded by the median nerve. The inherent neurovascular risks of the craniolateral approach to the elbow in horses followed a pattern similar to that in humans with the radial nerve being vulnerable.' The surgical approaches and intraarticular manipulations were facilitated by placing the horses in lateral recumbency. Dorsal recumbency made the caudomedial approach more difficult. Therefore, the site of the lesion should dictate the positioning of the animal. If the cranial and caudal aspects of the elbow joint need to be assessed, dorsal recumbency would be the logical choice. For discrete lesions, lateral recumbency would facilitate either craniolateral or caudomedial entry. Lesions of the radial capital fovea were not well addressed by the arthroscopic approaches described here. Inasmuch as subchondral cysts in the radial head are reportedly as common as those in the medial humeral condyle,' this is disappointing and is an area in need of further study. Distraction of the elbow with an ankle distractors or a fracture distractorll attached with threaded Steinmann pins in the humerus and radius was inadequate to allow complete assessment of the capitular surface of the radius. References 1. Hopen LA, Turner TA. Colahan P. Nixon AJ. Elbow subchondral
-.7 3.
4.
5. 6.
7.
8.
bone cysts in the horse: Conservative treatmeni of seven cases. J Am Vet Med Assoc (accepted for publication) IY90. Bertone AL. McIIwraith CW. Powers BE. et al. Subchondral osseous cystic lesions of the elbow of horses: conservative versus surgical treatment. J Am Vet Med Assoc 1986: 189:540-546. Lynch GJ. Meyers JF. Whipple TL. Caspari RB. Neurovascular anatomy and elbow arthroscopy: inherent risks. Arthroscopy I986:z: 191-197. Clarke RP. Symptomatic. lateral synovial fringe (plica) of the elbowjoint. Arthroscopy 1988:4:1 12-1 16. Guhl JF. Arthroscopy and arthroscopic surgery of the elbow. Orthopedics 1985;8: 1290-1196. Boe S. Arthroscopy ofthe elbow: diagnosis and extraction of loose bodies. Acta Orthop Scand 1986:57:52-53. Andrews JR. Carson WG. Arthroscopy of the elbow. Arthroscopy 1985: 197-107. Thomas MA. Fast A. Shapiro D. Radial nerve damage as a complication of elbow arthroscopy. Clin Orthop l987:2 15: 130- I3 I .
4 Ankle distractor. Acukx Microsurgical, Norwood, MA. (I
Femoral fracture distractor. Synthes Ltd (USA), Wayne. PA.
