- Email: [email protected]
Arthroscopic portals of the wrist: An anatomic study
Arthroscopic Portals of the Wrist: An Anatomic study Reid A. Abrams, MD, Michael Petersen, MD, Michael J. Botte, MD, San Diego, CA Wrist arthroscopy useful therapeutic
has become
an accepted
diagnostic
are at risk of injury during this procedure; make wrist arthroscopy
however,
safer. Wrist arthroscopic
understanding
joint
portals were relatively
sensory nerve branches and tendons.
copy portals is emphasized.
(J Hand
_ From the University of California, San Diego, Department of Orthopedic Surgery, Hand and Microvascular Surgery Service,
and Veterans Administration Hospital, Department of Rehabilitation Medicine. San Diego, CA. Received for publication Sept. 16, 1993; accepted in revised form Feb. 16, 1994. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Reprint requests: Reid A. Abrams, MD, University of California, San Diego, 200 West Arbor, 8894, San Diego, CA 92103. 940
The Journal of Hand Surgery
while
anatomy
should
in 19 fresh cadaver
the midcarpal,
safe. Even “safe”
A safe technique
Surg 1994;
Wrist arthroscopy was initially reported in 1979, and as understanding of wrist mechanics has increased, so has the interest and utility of wrist arthroscopy.‘-9 It is anticipated that with further advancement in small joint endoscopic technology and increased understanding of the kinematics of the wrist, diagnostic and interventional arthroscopy will become more common. Several reports have discussed the anatomy and technique of wrist arthroscopy with mention of potential for injury to tendons, nerves, and vessels when establishing portals.5,8.9 Although attention has been given to methods of avoiding injury to these structures, we have not found published data delineating quantitative periportal anatomy. The purpose of this study was to quantitatively describe the anatomy of wrist arthroscopy portals, stressing soft tissue interrelationships, in the hope of minimizing risk of injury to dorsal wrist structures.
periportal
anatomy was described and quanti-
6R, and 6U portals were the most perilous,
and distal radioulnar adjacent
and it is starting to be a
portals were established
wrists, after which the limbs were dissected and periportal fied. The l-2,
technique,
tool. Extensor tendons, the radial artery, and dorsal sensory nerve branches
3-4,
4-5,
portals had occasional
of establishing
wrist arthros-
19A:940-944.)
Materials
and Methods
Portal Nomenclature Previously described wrist arthroscopy portals and their established nomenclature were used in the study5T9 (Fig. 1). The most commonly used portals are between the third and fourth dorsal wrist compartments (3-4 portal) and between the fifth and sixth dorsal wrist compartments (on the radial side of the sixth compartment, hence the nomenclature 6 R portal). Radial and ulnar midcarpal joint portals are often used and are located 1 cm distal to the 3-4 portal (between the extensor carpi radialis brevis and extensor digitorum comminus and the 4-5 portal (between the extensor digitorum communis and the extensor digiti quinti, respectively. Other described portals are the l-2 and 4-5 portals (between corresponding dorsal wrist compartments at the level of the radiocarpal joint), the 6U portal (just ulnar to the sixth dorsal wrist compartment at the level of the ulnar carpal joint), and the distal radioulnar joint portal (between the fifth and sixth dorsal wrist compartments over the distal radioulnar joint.‘T9 Study Method Twenty-three fresh unembalmed cadaver extremities were examined under 3.5 x loupe magnification, studying the periportal anatomy for all of the
The journal
of Hand Surgery
I Vol. 19A No. 6 November
1994
941
Figure 1. (A) Overview schematic of arthroscopic portal anatomy. SBRN, superficial branch of the radial nerve; DBUN, dorsal branch of the ulnar nerve; RA, radial artery; DRUJ, distal radio-ulnar joint: 1-2, 3-4,4-S: radiocarpal joint portals between first and second, third and fourth, and fourth and fifth dorsal wrist compartments, respectively; 6 R and 6 U: ulnocarpal joint portals radial and ulnar to the sixth dorsal compartment, respectively; MCR: midcarpal radial portal; MCU: midcarpal ulnar portal. The solid bold lines represent cutaneous nerves (SBRN, DBUN). Bold dashed lines represent variably present cutaneous branches from the DBUN. The most proximal transversely oriented bold dashed line represents the transverse branch of the dorsal branch of the ulnar nerve. (B) Enlarged schematic detailing the 3-4, 4-5,6R, and DRUJ portals. (C) Enlargement of the i-2 and MCR portals. (D) Enlargement of the 6U and MCU portals. Numbers represent distances in millimeters of portals (bold dots) to indicated structures. Numbers in parentheses represent the minimal distances found in our specimens.
942
Abrams et al. / Arthroscopic
Portals of the Wrist
above-mentioned portals. Three specimens were discarded due to preexisting disease distorting the anatomy. The first 10 specimens were evaluated as follows: 5-6 cc saline solution was instilled into the radiocarpal joint with an 18-gauge needle through the 3-4 portal. All portals were established by incising the skin longitudinally just through the dermis, taking care not to penetrate the subcutaneous layer. A fine hemostat was used to spread the subcutaneous tissue down to the capsule, after which the capsule was penetrated with a blunt trocar. The portal tracks were then labeled with India ink. The skin was removed, and the location of tendons, radial and ulnar cutaneous nerves, and radial arteries was measured to the nearest millimeter with respect to the blackened portal tracks. The tracks were followed into the joint to confirm proper portal placement. The data from one further specimen in this group was discarded due to poor portal placement. The next 10 specimens were similarly studied, except instead of marking portal tracks with India ink, a 2.7 mm arthroscope (Concepts, Berkely, CA) was introduced to confirm portal placement and to more closely simulate the actual procedure. For each portal, the arthroscope was replaced with a trochar, the surrounding skin was excised, and distances of neurovascular structures and tendons from the trochar were measured. It was noted if any of these structures had been injured while placing the instruments. For each specimen, a sketch of periportal anatomy was made. All portals in the 19 included specimens were measured except the l-2 portal, which was quantified in only 14 specimens. Normality of measurements was confirmed, and means and standard deviations were calculated with the Maclntosh Stat View II program (Abicus, Concepts Berkley, CA).
Results Measurements from 4 of the 23 specimens were discarded due to erroneous portal placement (1 specimen) or preexisting disease distorting anatomy (3 specimens). Thus, 19 cadaver wrists were included in this study (Fig. 1) l-2
mm; SD = 2.9) to the portal. The radial artery was a mean of 3 mm radial (range, l .-5 mm; SD = 1.3) to the portal (Fig. 1A. B). 3-4
Portal
The mean 3-4 portal tendon interval was 8 mm (range, 4-12 mm; SD = 2.5). In 16 of the specimens only one branch of the SBRN was in proximity and radial to this portal. In two specimens, there were two radially located branches, and in one there were branches radial and ulnar to the portal. The branches of the SBRN radial to the portal were located at a mean distance of 16 mm (range, 5-22 mm: SD = 5.8). The branch ulnar to the portal was 6 mm away. In seven cadavers, the distance of the 3-4 portal from the radial artery was measured, with a mean of 26.3 mm (range 20 to 30 mm: SD = 4.1). Due to the remoteness of the artery from this portal, we deemed it unnecessary to perform further measurements (Fig. lA, C). 4-5
Portal
The 4-5 portal tendon interval was a mean of 6.7 mm (range, 4-8 mm; SD = 1.5). Sensory nerves were remote to this portal, except in one specimen where an aberrant branch of the SBRN was found 4 mm radial to the portal (Fig. 1A, C). 6R Portal The interval between the fifth and sixth dorsal wrist compartments at the 6R portal was a mean of 4.5 mm (range, 1 to 7 mm; SD = 1.7). The mean distance of the dorsal sensory branch of the ulnar nerve (DBUN) to the 6R portal was 8.2 mm (range, 0 to 14 mm; SD = 3.6). The DBUN at the level of this portal usually consisted of one trunk residing ulnar to the portal. In one specimen, there were two branches, one of which lay directly over the portal. In 12 of 19 specimens the transverse radioulnar sensory branch of the DBUN (innervating the distal radioulnar joint)‘O (TRUDBUN) was found. In these specimens. the TRUDBUN was a mean of 2 mm proximal to the portal (range, 0 to 6 mm; SD = 1.5) (Fig. lA, C).
Portal
The l-2 portal was examined in 14 specimens. The mean interval between the tendons of the first and second dorsal wrist compartments at the site of the l-2 portal was 10 mm (range, 4-14 mm; SD = 2.6). Two branches of the sensory branch of the radial nerve (SBRN) were in proximity, with one a mean of 3 mm radial (range, l-6 mm; SD = 1.5) and the other a mean of 5 mm ulnar (range, 2-12
6U Portal The 6U portal is bounded radially by the extensor carpi ulnaris tendon and ulnarly by the DBUN. The interval between these structures was a mean of 8.3 mm (range, 5 to 12 mm; SD = 2.5). The mean distance of the DBUN from the 6U portal was 4.5 mm (range, 2 to 10 mm; SD = 2.7). In 11 of the 19 specimens, at the level of the portal, there were two
The Journal of Hand Surgery / Vol.
branches of the DBUN. In eight, one branch was a mean of 1.9 mm (range, 0 to 4 mm; SD = 1.5) radial, and the other branch was a mean of 4.8 mm ulnar to the portal. In two, both branches were ulnar to the portal by a mean 4.5 mm. In one specimen, one branch was 3 mm distal and the other was 8 mm ulnar. In 12 of the 19 specimens where a TRUDBUN was found, all of them but one were proximal to the portal by a mean of 2.5 mm (range, 1 to 4 mm; SD = 0.93). One TRUDBUN was 1 mm distal to the portal (Fig. IA, D). Distal Radioulnar
Joint Portal
With the wrist in neutral rotation or in pronation, the distal radioulnar joint portal is bounded radially by the extensor digitorum comminus, and ulnarly by the extensor carpi ulnaris. The tendon interval was a mean of 6.8 mm (range, 2 mm to 12 mm; SD = 2.4). The TRUDBUN was the only sensory nerve in proximity by a mean distance of 17.5 mm distally (range, 10 to 20 mm; SD = 3.6) (Fig. IA, C). Radial Midcarpal
Portal
The radial midcarpal portal is bounded radially by the extensor carpi radialis brevis and ulnarly by the tendons of the fourth dorsal wrist compartment. The tendon interval was a mean of 7.2 mm (range, 2 to 12 mm; SD = 2.7). In all 19 specimens, branches of the SBRN were found radial to the portal, with the closest being located at a mean distance of 15.8 mm (range, 5 to 26 mm; SD = 6.3). In two specimens, branches of the SBRN were also found ulnar to the portal, with one located at 2 mm and the other at 4 mm (Fig. IA, B). Ulnar Midcarpal
Portal
The ulnar midcarpal portal is bounded by the tendons of the fourth and fifth dorsal wrist compartments. The tendon interval was a mean of 8.1 mm (range, 6 to 12 mm; SD = 1.9). In all but one specimen. the SBRN branches were remote to this portal. In one specimen, however, a branch of the SBRN came within 1 mm radial to this portal. Branches of the DBUN were a mean of 15.1 mm ulnar to this portal (range, 4 to 25 mm; SD = 4.6). One DBUN aberrant branch ran directly over the portal. By using the described portal entry technique, no structures were injured in any specimen, despite frequent close proximity of nerves and tendons. However, while establishing several of the portals, the tendon sheaths of the extensor digitorum comminus, extensor digiti quinti, and extensor carpi ulnaris were occasionally entered at the level of the exten-
19A No. 6 November
1994
943
sor retinaculum. The sheath of the extensor digitorum comminus was entered once, four, and three times when establishing the 3-4, radial midcarpal, and ulnar midcarpal portals. respectively. The extensor digiti quinti sheath was violated once when establishing the 6R and twice when placing the DRUJ portals.
Discussion Wrist arthroscopy has become a useful diagnostic and therapeutic technique, as understanding of wrist kinematics, arthroscopic methods, and technology have improved. Extensor tendons, sensory nerve branches, and the radial artery are at risk of injury, due to their proximity to arthroscopic portals. Several authors have discussed wrist arthroscopic technique, but little emphasis has been given to the anatomy of dorsal wrist tendons and neurovascular structures and their relationship to the arthroscopic portals.5.8,9 We demonstrated that the l-2,6R, and 6U portals were the most perilous, due to close proximity of the radial artery and dorsal radial and ulnar sensory nerve branches. The midcarpal, 3-4.4-5, and distal radioulnar joint portals were relatively safe, since neurovascular structures were usually remote. However, even the “safe” portals occasionally had close aberrant sensory nerve branches. The extensor tendons that border each portal are also at risk. Due to the high variability of sensory nerve anatomy on the dorsum of the wrist and the proximity of extensor tendons to all portals, we feel that the technique of establishing the portal is especially important in safely performing wrist arthroscopy. Unlike establishing a portal for knee or shoulder arthroscopy (where a stab incision is frequently made through the skin, subcutaneous layer, and capsule into the joint), the incision used when establishing a wrist portal should be longitudinal and go no deeper than just through the dermis. The subcutaneous tissue should be entered and separated with a fine hemostat down to the capsule and. in turn, the capsule can be penetrated with a blunt trochar and sleeve. This technique helps protect nearby structures between the dermis and capsule from laceration. The methods employed in this study differ from standard arthroscopic techniques in that traction was not applied to our cadaver wrists. Although it would be difficult to substantiate without disrupting much of the stabilizing soft tissue (e.g., small cutaneous nerve branches, dorsal hand and wrist fascia and retinaculum), we anticipate that longitudinal
944
Abrams et al. / Arthroscopic
Portals of the Wrist
traction might affect distal/proximal values; but transverse distances should be minimally affected. The authors thank Richard Lieber for statistical assistance and William Collins. Marijo Sanson. and Rick Wilson for assistance
with anatomical preparations.
References 1. Kelley EP, Stanley JK. Arthroscopy of the wrist. J Hand Surg 1990;15B:236-42. 2. Takagi K. The classic arthroscope. Clin Orthop 1982; 167:6-8. 3. Altchek DW, Warren RF, Skyhar MJ. Shoulder arthroscopy. In: The shoulder. Rockwood CA, Matsen FA, eds. Philadelphia: WB Saunders 1990:258-77. 4. Yung-Cheng C. Arthroscopy of the wrist and finger joints. Orthop Clin North Am 1979;10:723-33.
5. Roth JH, Poehling GG, Whipple TL. Arthroscopic surgery of the wrist. AAOS Instructional Course Lectures 1988;37: 183-94. 6. Hanker GJ. Diagnostic and operative arthroscopy of the wrist. Clinical Orthop 1991;263: 165-74. 7. North ER, Thomas S. An anatomic guide for arthroscopic visualization of the wrist capsular ligaments. J Hand Surg 1988;13A:815-22. 8. Cooney WP, Dobyns JH, Linscheid RL. Arthroscopy of the wrist: anatomy and classification of carpal instabilty. Arthroscopy 1990;6: 133-40. 9. Botte MJ, Cooney WP, Linscheid RL. Arthroscopy of the wrist: anatomy and technique. J Hand Surg 1989;14A:313-16. 10. Lourie GM, King J, Kleinman WB. The transverse radio-ulnar sensory branch from the dorsal sensory ulnar nerve: its clinical and anatomical significance. J Hand Surg. 1994;19A:241-245.
has become
an accepted
diagnostic
are at risk of injury during this procedure; make wrist arthroscopy
however,
safer. Wrist arthroscopic
understanding
joint
portals were relatively
sensory nerve branches and tendons.
copy portals is emphasized.
(J Hand
_ From the University of California, San Diego, Department of Orthopedic Surgery, Hand and Microvascular Surgery Service,
and Veterans Administration Hospital, Department of Rehabilitation Medicine. San Diego, CA. Received for publication Sept. 16, 1993; accepted in revised form Feb. 16, 1994. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Reprint requests: Reid A. Abrams, MD, University of California, San Diego, 200 West Arbor, 8894, San Diego, CA 92103. 940
The Journal of Hand Surgery
while
anatomy
should
in 19 fresh cadaver
the midcarpal,
safe. Even “safe”
A safe technique
Surg 1994;
Wrist arthroscopy was initially reported in 1979, and as understanding of wrist mechanics has increased, so has the interest and utility of wrist arthroscopy.‘-9 It is anticipated that with further advancement in small joint endoscopic technology and increased understanding of the kinematics of the wrist, diagnostic and interventional arthroscopy will become more common. Several reports have discussed the anatomy and technique of wrist arthroscopy with mention of potential for injury to tendons, nerves, and vessels when establishing portals.5,8.9 Although attention has been given to methods of avoiding injury to these structures, we have not found published data delineating quantitative periportal anatomy. The purpose of this study was to quantitatively describe the anatomy of wrist arthroscopy portals, stressing soft tissue interrelationships, in the hope of minimizing risk of injury to dorsal wrist structures.
periportal
anatomy was described and quanti-
6R, and 6U portals were the most perilous,
and distal radioulnar adjacent
and it is starting to be a
portals were established
wrists, after which the limbs were dissected and periportal fied. The l-2,
technique,
tool. Extensor tendons, the radial artery, and dorsal sensory nerve branches
3-4,
4-5,
portals had occasional
of establishing
wrist arthros-
19A:940-944.)
Materials
and Methods
Portal Nomenclature Previously described wrist arthroscopy portals and their established nomenclature were used in the study5T9 (Fig. 1). The most commonly used portals are between the third and fourth dorsal wrist compartments (3-4 portal) and between the fifth and sixth dorsal wrist compartments (on the radial side of the sixth compartment, hence the nomenclature 6 R portal). Radial and ulnar midcarpal joint portals are often used and are located 1 cm distal to the 3-4 portal (between the extensor carpi radialis brevis and extensor digitorum comminus and the 4-5 portal (between the extensor digitorum communis and the extensor digiti quinti, respectively. Other described portals are the l-2 and 4-5 portals (between corresponding dorsal wrist compartments at the level of the radiocarpal joint), the 6U portal (just ulnar to the sixth dorsal wrist compartment at the level of the ulnar carpal joint), and the distal radioulnar joint portal (between the fifth and sixth dorsal wrist compartments over the distal radioulnar joint.‘T9 Study Method Twenty-three fresh unembalmed cadaver extremities were examined under 3.5 x loupe magnification, studying the periportal anatomy for all of the
The journal
of Hand Surgery
I Vol. 19A No. 6 November
1994
941
Figure 1. (A) Overview schematic of arthroscopic portal anatomy. SBRN, superficial branch of the radial nerve; DBUN, dorsal branch of the ulnar nerve; RA, radial artery; DRUJ, distal radio-ulnar joint: 1-2, 3-4,4-S: radiocarpal joint portals between first and second, third and fourth, and fourth and fifth dorsal wrist compartments, respectively; 6 R and 6 U: ulnocarpal joint portals radial and ulnar to the sixth dorsal compartment, respectively; MCR: midcarpal radial portal; MCU: midcarpal ulnar portal. The solid bold lines represent cutaneous nerves (SBRN, DBUN). Bold dashed lines represent variably present cutaneous branches from the DBUN. The most proximal transversely oriented bold dashed line represents the transverse branch of the dorsal branch of the ulnar nerve. (B) Enlarged schematic detailing the 3-4, 4-5,6R, and DRUJ portals. (C) Enlargement of the i-2 and MCR portals. (D) Enlargement of the 6U and MCU portals. Numbers represent distances in millimeters of portals (bold dots) to indicated structures. Numbers in parentheses represent the minimal distances found in our specimens.
942
Abrams et al. / Arthroscopic
Portals of the Wrist
above-mentioned portals. Three specimens were discarded due to preexisting disease distorting the anatomy. The first 10 specimens were evaluated as follows: 5-6 cc saline solution was instilled into the radiocarpal joint with an 18-gauge needle through the 3-4 portal. All portals were established by incising the skin longitudinally just through the dermis, taking care not to penetrate the subcutaneous layer. A fine hemostat was used to spread the subcutaneous tissue down to the capsule, after which the capsule was penetrated with a blunt trocar. The portal tracks were then labeled with India ink. The skin was removed, and the location of tendons, radial and ulnar cutaneous nerves, and radial arteries was measured to the nearest millimeter with respect to the blackened portal tracks. The tracks were followed into the joint to confirm proper portal placement. The data from one further specimen in this group was discarded due to poor portal placement. The next 10 specimens were similarly studied, except instead of marking portal tracks with India ink, a 2.7 mm arthroscope (Concepts, Berkely, CA) was introduced to confirm portal placement and to more closely simulate the actual procedure. For each portal, the arthroscope was replaced with a trochar, the surrounding skin was excised, and distances of neurovascular structures and tendons from the trochar were measured. It was noted if any of these structures had been injured while placing the instruments. For each specimen, a sketch of periportal anatomy was made. All portals in the 19 included specimens were measured except the l-2 portal, which was quantified in only 14 specimens. Normality of measurements was confirmed, and means and standard deviations were calculated with the Maclntosh Stat View II program (Abicus, Concepts Berkley, CA).
Results Measurements from 4 of the 23 specimens were discarded due to erroneous portal placement (1 specimen) or preexisting disease distorting anatomy (3 specimens). Thus, 19 cadaver wrists were included in this study (Fig. 1) l-2
mm; SD = 2.9) to the portal. The radial artery was a mean of 3 mm radial (range, l .-5 mm; SD = 1.3) to the portal (Fig. 1A. B). 3-4
Portal
The mean 3-4 portal tendon interval was 8 mm (range, 4-12 mm; SD = 2.5). In 16 of the specimens only one branch of the SBRN was in proximity and radial to this portal. In two specimens, there were two radially located branches, and in one there were branches radial and ulnar to the portal. The branches of the SBRN radial to the portal were located at a mean distance of 16 mm (range, 5-22 mm: SD = 5.8). The branch ulnar to the portal was 6 mm away. In seven cadavers, the distance of the 3-4 portal from the radial artery was measured, with a mean of 26.3 mm (range 20 to 30 mm: SD = 4.1). Due to the remoteness of the artery from this portal, we deemed it unnecessary to perform further measurements (Fig. lA, C). 4-5
Portal
The 4-5 portal tendon interval was a mean of 6.7 mm (range, 4-8 mm; SD = 1.5). Sensory nerves were remote to this portal, except in one specimen where an aberrant branch of the SBRN was found 4 mm radial to the portal (Fig. 1A, C). 6R Portal The interval between the fifth and sixth dorsal wrist compartments at the 6R portal was a mean of 4.5 mm (range, 1 to 7 mm; SD = 1.7). The mean distance of the dorsal sensory branch of the ulnar nerve (DBUN) to the 6R portal was 8.2 mm (range, 0 to 14 mm; SD = 3.6). The DBUN at the level of this portal usually consisted of one trunk residing ulnar to the portal. In one specimen, there were two branches, one of which lay directly over the portal. In 12 of 19 specimens the transverse radioulnar sensory branch of the DBUN (innervating the distal radioulnar joint)‘O (TRUDBUN) was found. In these specimens. the TRUDBUN was a mean of 2 mm proximal to the portal (range, 0 to 6 mm; SD = 1.5) (Fig. lA, C).
Portal
The l-2 portal was examined in 14 specimens. The mean interval between the tendons of the first and second dorsal wrist compartments at the site of the l-2 portal was 10 mm (range, 4-14 mm; SD = 2.6). Two branches of the sensory branch of the radial nerve (SBRN) were in proximity, with one a mean of 3 mm radial (range, l-6 mm; SD = 1.5) and the other a mean of 5 mm ulnar (range, 2-12
6U Portal The 6U portal is bounded radially by the extensor carpi ulnaris tendon and ulnarly by the DBUN. The interval between these structures was a mean of 8.3 mm (range, 5 to 12 mm; SD = 2.5). The mean distance of the DBUN from the 6U portal was 4.5 mm (range, 2 to 10 mm; SD = 2.7). In 11 of the 19 specimens, at the level of the portal, there were two
The Journal of Hand Surgery / Vol.
branches of the DBUN. In eight, one branch was a mean of 1.9 mm (range, 0 to 4 mm; SD = 1.5) radial, and the other branch was a mean of 4.8 mm ulnar to the portal. In two, both branches were ulnar to the portal by a mean 4.5 mm. In one specimen, one branch was 3 mm distal and the other was 8 mm ulnar. In 12 of the 19 specimens where a TRUDBUN was found, all of them but one were proximal to the portal by a mean of 2.5 mm (range, 1 to 4 mm; SD = 0.93). One TRUDBUN was 1 mm distal to the portal (Fig. IA, D). Distal Radioulnar
Joint Portal
With the wrist in neutral rotation or in pronation, the distal radioulnar joint portal is bounded radially by the extensor digitorum comminus, and ulnarly by the extensor carpi ulnaris. The tendon interval was a mean of 6.8 mm (range, 2 mm to 12 mm; SD = 2.4). The TRUDBUN was the only sensory nerve in proximity by a mean distance of 17.5 mm distally (range, 10 to 20 mm; SD = 3.6) (Fig. IA, C). Radial Midcarpal
Portal
The radial midcarpal portal is bounded radially by the extensor carpi radialis brevis and ulnarly by the tendons of the fourth dorsal wrist compartment. The tendon interval was a mean of 7.2 mm (range, 2 to 12 mm; SD = 2.7). In all 19 specimens, branches of the SBRN were found radial to the portal, with the closest being located at a mean distance of 15.8 mm (range, 5 to 26 mm; SD = 6.3). In two specimens, branches of the SBRN were also found ulnar to the portal, with one located at 2 mm and the other at 4 mm (Fig. IA, B). Ulnar Midcarpal
Portal
The ulnar midcarpal portal is bounded by the tendons of the fourth and fifth dorsal wrist compartments. The tendon interval was a mean of 8.1 mm (range, 6 to 12 mm; SD = 1.9). In all but one specimen. the SBRN branches were remote to this portal. In one specimen, however, a branch of the SBRN came within 1 mm radial to this portal. Branches of the DBUN were a mean of 15.1 mm ulnar to this portal (range, 4 to 25 mm; SD = 4.6). One DBUN aberrant branch ran directly over the portal. By using the described portal entry technique, no structures were injured in any specimen, despite frequent close proximity of nerves and tendons. However, while establishing several of the portals, the tendon sheaths of the extensor digitorum comminus, extensor digiti quinti, and extensor carpi ulnaris were occasionally entered at the level of the exten-
19A No. 6 November
1994
943
sor retinaculum. The sheath of the extensor digitorum comminus was entered once, four, and three times when establishing the 3-4, radial midcarpal, and ulnar midcarpal portals. respectively. The extensor digiti quinti sheath was violated once when establishing the 6R and twice when placing the DRUJ portals.
Discussion Wrist arthroscopy has become a useful diagnostic and therapeutic technique, as understanding of wrist kinematics, arthroscopic methods, and technology have improved. Extensor tendons, sensory nerve branches, and the radial artery are at risk of injury, due to their proximity to arthroscopic portals. Several authors have discussed wrist arthroscopic technique, but little emphasis has been given to the anatomy of dorsal wrist tendons and neurovascular structures and their relationship to the arthroscopic portals.5.8,9 We demonstrated that the l-2,6R, and 6U portals were the most perilous, due to close proximity of the radial artery and dorsal radial and ulnar sensory nerve branches. The midcarpal, 3-4.4-5, and distal radioulnar joint portals were relatively safe, since neurovascular structures were usually remote. However, even the “safe” portals occasionally had close aberrant sensory nerve branches. The extensor tendons that border each portal are also at risk. Due to the high variability of sensory nerve anatomy on the dorsum of the wrist and the proximity of extensor tendons to all portals, we feel that the technique of establishing the portal is especially important in safely performing wrist arthroscopy. Unlike establishing a portal for knee or shoulder arthroscopy (where a stab incision is frequently made through the skin, subcutaneous layer, and capsule into the joint), the incision used when establishing a wrist portal should be longitudinal and go no deeper than just through the dermis. The subcutaneous tissue should be entered and separated with a fine hemostat down to the capsule and. in turn, the capsule can be penetrated with a blunt trochar and sleeve. This technique helps protect nearby structures between the dermis and capsule from laceration. The methods employed in this study differ from standard arthroscopic techniques in that traction was not applied to our cadaver wrists. Although it would be difficult to substantiate without disrupting much of the stabilizing soft tissue (e.g., small cutaneous nerve branches, dorsal hand and wrist fascia and retinaculum), we anticipate that longitudinal
944
Abrams et al. / Arthroscopic
Portals of the Wrist
traction might affect distal/proximal values; but transverse distances should be minimally affected. The authors thank Richard Lieber for statistical assistance and William Collins. Marijo Sanson. and Rick Wilson for assistance
with anatomical preparations.
References 1. Kelley EP, Stanley JK. Arthroscopy of the wrist. J Hand Surg 1990;15B:236-42. 2. Takagi K. The classic arthroscope. Clin Orthop 1982; 167:6-8. 3. Altchek DW, Warren RF, Skyhar MJ. Shoulder arthroscopy. In: The shoulder. Rockwood CA, Matsen FA, eds. Philadelphia: WB Saunders 1990:258-77. 4. Yung-Cheng C. Arthroscopy of the wrist and finger joints. Orthop Clin North Am 1979;10:723-33.
5. Roth JH, Poehling GG, Whipple TL. Arthroscopic surgery of the wrist. AAOS Instructional Course Lectures 1988;37: 183-94. 6. Hanker GJ. Diagnostic and operative arthroscopy of the wrist. Clinical Orthop 1991;263: 165-74. 7. North ER, Thomas S. An anatomic guide for arthroscopic visualization of the wrist capsular ligaments. J Hand Surg 1988;13A:815-22. 8. Cooney WP, Dobyns JH, Linscheid RL. Arthroscopy of the wrist: anatomy and classification of carpal instabilty. Arthroscopy 1990;6: 133-40. 9. Botte MJ, Cooney WP, Linscheid RL. Arthroscopy of the wrist: anatomy and technique. J Hand Surg 1989;14A:313-16. 10. Lourie GM, King J, Kleinman WB. The transverse radio-ulnar sensory branch from the dorsal sensory ulnar nerve: its clinical and anatomical significance. J Hand Surg. 1994;19A:241-245.