FOUR-CORNER ARTHRODESIS BY ALEXANDER Y. SHIN, MD

Although nearly every combination of intercarpal arthrodeses has been described, excision of the scaphoid and fusion of the remaining carpal bones in neutral alignment was a unique concept when it was introduced nearly 20 years ago. This time-tested and established procedure that is commonly known as the 4-corner or 4-bone arthrodesis is based on the principle that the radiolunate articulation is often spared from degenerative changes from conditions that result in rotatory subluxations of the scaphoid. The 4-corner arthrodesis is a motion-sparing, limited arthrodesis that reliably results in pain relief, improved grip strength, and overall high patient satisfaction with low associated nonunion and complication rates. Copyright © 2001 by the American Society for Surgery of the Hand ongenital carpal fusions or coalitions are often discovered serendipitously on radiographs taken for other purposes.1-4 The wrists of these individuals are often normal, without symptoms or conditions that can be recognized by clinical examination. Because wrists with carpal coalitions often have normal appearance and function without pain, disability, weakness, or instability, it was postulated that intercarpal arthrodesis could restore acceptable function to an injured wrist and obviate the need for total wrist arthrodesis, which results in complete loss of motion.5 One of the earliest descriptions of an intercarpal arthrodesis was in 1924 by Thornton,6 who excised

C

From the Division of Hand & Microsurgery, Department of Orthopaedic Surgery, Naval Medical Center San Diego, San Diego, CA. The views expressed in this article are those of the authors and do not reflect the official policy or position of the Department of the Navy, Department of Defense, or the United States Government. Address reprint requests to Commander Alexander Y. Shin, MD, USN, c/o Clinical Investigations Department, Naval Medical Center San Diego, 34800 Bob Wilson Dr, San Diego, CA 92134-5000. Copyright © 2001 by the American Society for Surgery of the Hand 1531-0914/01/0102-0005$35.00/0 doi:10.1053/jssh.2001.23905

the base of the capitate and fused the hamate, trapezoid, scaphoid, lunate, and triquetrum. In 1946, Sutro7 reported 4 cases of scaphoid nonunions treated by intercarpal arthrodesis of the scaphoid fragments to the capitate. Similarly, in 1952 Helfet8 reported 7 cases of scaphocapitate arthrodesis for the treatment of scaphoid nonunions. By the 1960s the concepts and indications of intercarpal arthrodesis were further refined. Graner and associates9 described intercarpal arthrodesis with excision of the lunate or proximal scaphoid fragment in patients with Kienbo¨ck disease, scaphoid nonunions, or old carpal fracture-dislocations. Soon after Graner’s report, Peterson and Lipscomb5 of the Mayo Clinic reported the successful outcome of 8 patients treated with intercarpal arthrodesis performed for degenerative arthritis secondary to scaphoid nonunion, posttraumatic scaphoid subluxations, scaphoid nonunion, and Kienbo¨ck disease. One of the intercarpal arthrodeses they performed was the scaphoid-trapezium-trapezoid arthrodesis, subsequently popularized as the triscaphe arthrodesis by Watson.10-12 Early results of treatment of scaphoid nonunion by excision of part or all of the scaphoid often resulted in

JOURNAL OF THE AMERICAN SOCIETY FOR SURGERY OF THE HAND 䡠 VOL. 1, NO. 2, MAY 2001

93

94

FOUR-CORNER ARTHRODESIS 䡠 SHIN

weakness, pain, and loss of function of the wrist.7,13-15 As such, scaphoidectomy was generally not recommended for the treatment of scaphoid nonunions.7 However, when the scaphoid fragments were arthrodesed to the capitate, pain diminished and strength improved.5,7,8 Intercarpal arthrodesis of the scaphoid and capitate provided pain relief, stability, and the preservation of some wrist motion. As intercarpal arthrodeses evolved, nearly every conceivable combination of carpal bone fusion was described. Despite the historical recommendations against scaphoidectomy, Watson and associates16 described the unique concept of combining the scaphoidectomy, neutral alignment of the remaining carpal bones, and arthrodesis of the capitate, hamate, lunate, and triquetrum to maintain the neutral alignment. Arthrodesis of the capitate and lunate was difficult to achieve, with nonunion rates reported as high as 30%.17 Thus, to increase the union rates, the hamate and triquetrum were added to the capitolunate arthrodesis.16 The resultant arthrodesis of the capitate, hamate, lunate, and triquetrum was termed a 4-corner arthrodesis. In 1981, Watson and associates16 described the results of 2 patients (3 wrists) who underwent a 4-corner arthrodesis, excision of the scaphoid, and replacement with a silicone implant spacer for degenerative arthrosis. Three years later, Watson and Ballet18 described the scapholunate advanced collapse (SLAC) pattern of degenerative arthritis and reported on the results of 16 patients who underwent scaphoidectomy, neutral alignment, and 4-corner arthrodesis.18 A decade later, Ashmead and associates19 reported 100 cases in which scaphoidectomy and 4-corner arthrodesis were performed for SLAC wrist salvage. Today, nearly 20 years after its formal introduction, scaphoidectomy and 4-corner arthrodesis with neutral alignment of the carpus has become an accepted and time-tested procedure for the treatment of a variety of carpal maladies.

INDICATION

FOR

4-CORNER ARTHRODESIS

In 1984, Watson and Ballet18 reviewed 4,000 radiographs of the hand and wrist to determine the patterns of sequential change in degenerative arthritis of the wrist. The most common pattern of degenerative arthritis of the wrist was described as

SLAC and represented a final common pathway for a variety of carpal conditions, of which the most common were rotatory subluxations of the scaphoid and scaphoid nonunion.20-25 Most cases of SLAC wrist represent the late sequelae of scapholunate dissociation, either traumatic or secondary to attenuation of the scapholunate ligament.19 Scapholunate ligament attenuation by calcium pyrophosphate dihydrate deposition disease or rheumatoid arthritis can also result in SLAC wrist.21,26 Chronic scaphoid nonunions can also lead to a SLAC pattern of arthritis and have been more correctly termed scaphoid nonunion advanced collapse (SNAC) (Fig 1).27 Despite the etiology, the pattern of SLAC wrist shows a consistent progression of degenerative changes that first occurs between the tip of the radiostyloid and the scaphoid (stage I) (Fig 2).18 The arthritic changes then progress along the radioscaphoid articulation (stage II). The radiolunate joint is spared, and the degenerative changes progress to the capitolunate joint (stage III). Despite the etiology, the SLAC pathway has consistently spared the radiolunate articulation, even when there is severe or chronic dorsal intercalated segmental instability (DISI) stance of the lunate. This preservation of the radiolunate joint is the foundation of the 4-corner arthrodesis. Another group of patients who may benefit from the 4-corner arthrodesis includes those with chronic dynamic carpal instability, chronic perilunar instability28 or nondissociative carpal instability.29,30 When soft tissue reconstruction or repair has failed or when soft tissue reconstruction would result in a greater limitation of motion than limited arthrodesis, 4-corner arthrodesis with scaphoidectomy can decrease pain, improve strength, and preserve motion. The 4-corner arthrodesis is indicated in patients with nondissociative carpal instability that has failed soft tissue reconstructions and those with arthritic involvement of the radioscaphoid joint with or without capitolunate involvement, chronic dynamic carpal instability, or chronic perilunar instability not amenable to soft tissue procedures. Absolute contraindications to the 4-corner arthrodesis include ulnar translocation of the carpus and radiolunate degenerative changes, observed intraoperatively or in preoperative radiographs.19

FOUR-CORNER ARTHRODESIS 䡠 SHIN

95

FIGURE 1. An anteroposterior (A) and lateral (B) radiograph of a wrist with a SNAC. The radiolunate articulation and the radioscaphoid articulation proximal to the nonunion site are preserved, and the distal radioscaphoid and capitolunate articulation show significant degenerative arthritis.

BIOMECHANICAL CONSIDERATIONS Although a complete discussion of biomechanics of the wrist is beyond the scope of this article, basic understanding of the normal mechanics and the effects of the 4-corner arthrodesis on overall wrist motion and force transmission is essential in communicating the expected outcome to patients. In the normal wrist, there is a balanced synchrony between the proximal and distal carpal rows.31,32 When motion occurs in the flexion-extension plane, there is coordinated motion between the distal and proximal carpal rows. As the distal carpal row flexes, the proximal carpal row also flexes. Similarly, when the distal carpal row extends, the proximal carpal row extends. The relative contributions of flexion and extension of each of the carpal rows depends on the frame of reference evaluated. If the central portion of

the carpus (capitate-lunate-radius linkage) is considered, radiocarpal and midcarpal motion are equally divided in a third of wrists, with the remaining two thirds having approximately 60% of flexion at the midcarpal joint and 66% of extension at the radiocarpal joint.33-35 However, when the frame of reference is changed to the lateral portion of the carpus (radiusscaphoid-trapezium linkage), nearly two thirds of the global arc of motion occurs at the radioscaphoid joint.33,36,37 With radial and ulnar deviation, there is a complex reciprocating motion of the proximal and distal carpal rows. With radial deviation, the distal carpal row inclines radialward, extends, and supinates. The proximal carpal bones principally flex and translate ulnarward. With ulnar deviation the opposite occurs: The distal carpal row inclines ulnarward, flexes, and pronates, while the proximal carpal bones extend and translate radialward.37-39 Such complex

96

FOUR-CORNER ARTHRODESIS 䡠 SHIN

FIGURE 2. (A) SLAC follows a predictable pattern of degenerative arthrosis that starts at the radial styloid (stage I), advances to the radioscaphoid articulation (stage II), and ends with midcarpal (capitolunate) arthrosis (stage III). An anteroposterior radiograph illustrating (B) SLAC stage II and (C) SLAC stage III changes.

motions are required to maintain the carpal congruency and spatial consistency in all wrist positions.33,40,41 Because the proximal and distal carpal rows function separately, procedures linking the rows will result in profound effects on wrist range of motion and force

transmission across the radiocarpal joints. Simulation of intercarpal arthrodeses have shown that capitolunate arthrodesis resulted in the greatest reduction of dorsiflexion, palmar flexion, and ulnar deviation compared to scaphoid-trapezium-trapezoid, scaphocapitate, and triquetrohamate arthrodesis.42 Specifically,

FOUR-CORNER ARTHRODESIS 䡠 SHIN

97

TECHNICAL ASPECTS

FIGURE 3. A T-shaped incision, with the distal transverse limb centered on the carpometacarpal joints and the longitudinal incision in line with the third metacarpal, is an alternative incision that enables wide exposure of the carpus.

there was a reduction of 9° of ulnar deviation, 34° of dorsiflexion, and 25° of palmar flexion.42 Gellman and associates43 showed that intercarpal arthrodesis within a carpal row had minimal effects on wrist motion in all planes. However, when intercarpal arthrodesis occurred between rows, profound losses of motions were observed. Capitolunate arthrodesis resulted in the greatest loss of motion in the flexion-extension plane (loss of 30% flexion and 41% extension) and less in the radio-ulnar planes (loss of 11% ulnar deviation and 21% radial deviation).43 Elegant studies of load mechanics and force transmission across the radiocarpal by Viegas and associates44-48 have allowed the determination of normal and abnormal radiocarpal joint loading mechanics. By using pressure sensitive film, normal cadaver wrists were compared with the same wrists with simulated intercarpal arthrodeses. In the radiocarpal joint of the normal wrist, the scaphoid was found to transmit 60% of the load, whereas the lunate accounted for 40%. Overall, the amount of contact in the radiocarpal joint accounted for only 20% of the surface area but increased to 40% with increased load.49 In capitolunate simulated arthrodesis, in which the lunate was brought back into neutral alignment, there was improved loading across the radiocarpal joint that more closely approximated the normal wrist model.47

Surgical Exposure and Technique The exposure of the carpus should be planned to afford maximal exposure of the carpal elements and to allow for future salvage procedures should the 4-corner arthrodesis fail. A dorsal midline longitudinal incision centered over the third metacarpal-capitatelunate-radius axis is the traditional approach that allows for extensile exposure of the carpus. An alternative to this is the T incision, in which the transverse limb is centered over the carpometacarpal joints and the longitudinal limb is midline and extends proximally (Fig 3).50 Excellent carpal exposure with this incision can be obtained with no skin complications when used in elective wrist surgery. Additionally, when future salvage procedures were required (ie, total wrist arthrodesis), there have been no adverse wound problems. When a total wrist arthrodesis is performed after use of a T incision, a fusion plate that does not require exposure of the third metacarpal, such as the Cobra wrist fusion plate (Kinetikos Medical Inc, San Diego, CA), is preferred. The skin flaps are raised, and the superficial branch of the radial nerve is identified and protected. The extensor retinaculum over the extensor pollicis longus is divided in line with the tendon, and an ulnarly based flap of extensor retinaculum is created by divid-

FIGURE 4. After the T incision is made, the skin flaps are elevated, the extensor retinaculum over the extensor pollicis longus is divided, and a ulnar-based flap of retinaculum is created by dividing the septations between the third and fourth and fourth and fifth extensor compartments.

98

FOUR-CORNER ARTHRODESIS 䡠 SHIN

FIGURE 5. (A, B) A ligament-sparing capsulotomy is made in the lines of the dorsal intercarpal ligament, the dorsal radiocarpal ligament, and the radioscaphoid capsule.

ing the septations between the third, fourth, and fifth extensor compartments (Fig 4). The extensor tendons are retracted, and the dorsal wrist capsule is exposed. A longitudinal incision in the capsule can be made on the ulnar aspect of the fourth compartment to expose the carpal bones, or alternatively, a ligament-sparing capsulotomy can be made along the dorsal intercarpal, the dorsal radiotriquetral ligaments, and radioscaphoid joint (Fig 5).51 The ligament-sparing capsulotomy, when repaired, may prevent dorsal capsular contracture, and thus improve ultimate range of motion. A radial based flap of capsule is then raised, exposing the carpal bones (Fig 6). The radiolunate articulation is inspected for degenerative wear and if normal, the scaphoid is excised either piecemeal with a rongeur or in its entirety sharply. Care is taken not to injure the volar radiocarpal ligaments (radioscaphocapitate and long radiolunate ligaments), because injury to these structures may result in ulnar translocation of the remaining carpals. The articular surfaces of the capitolunate, capitohamate, triquetrohamate and lunotriquetral joints are denuded of articular cartilage, and further prepared by placement of 1-mm burr holes evenly spaced throughout the surfaces to be arthrodesed (Fig 7). Neutral Alignment of the Capitolunate Axis Reduction of the capitolunate axis in neutral position is critical to the success of the 4-corner arthrodesis. Without reduction of the capitolunate axis,

dorsal impingement on the radius can occur, often resulting in pain and limitation of extension. Correction of carpal alignment improves motion and prevents dorsal impingement. When significant DISI deformity is associated with the carpal pathology, neutral alignment can be difficult to obtain. Use of a K wire placed into the lunate to act as a joystick can be helpful; however, the K wire can often be in the way of future fixation methods. An alternative technique, which we prefer, was described by Linscheid.52 Under fluoroscopy, a lateral view of the flexed wrist is obtained. The wrist is flexed and slightly ulnarly

FIGURE 6. A radial-based flap of capsule is elevated after the incisions are made in the dorsal scaphotriquetral ligament, the dorsal radiotriquetral ligament, and the radioscaphoid capsule.

FOUR-CORNER ARTHRODESIS 䡠 SHIN

99

more fat.54,55 Autogenous iliac crest bone graft, on the other hand, is more dense and has greater numbers of osteogenic cells, but it requires an additional surgery site that is often more painful than the wrist surgery.54,55 Despite the histomorphogenic differences in iliac crest and distal radius bone graft, clinical studies have not borne out these differences and have shown both to be effective in affording arthrodeses.55 Allografts become attractive, because their use avoids additional surgery. However, the lack of osteoinductivity and osteogenic cells, in addition to the potential for disease transmission, tempers the decision to use allograft materials.53 The choice of bone graft and the surgeon’s preference should be discussed with the patient before surgery.

METHOD

OF

FIXATION

After bone graft is packed into the interstices of the denuded articular cartilage, the capitolunate axis is reduced, and a 0.045-inch K wire placed across the capitate and triquetrum maintains the alignment. Several methods of securing the 4-corner arthrodesis have been described and include fixation with K wires, staples, screws, and specially designed plates.

deviated until neutral alignment of the radius and lunate are seen. A 0.0625-inch K wire is then drilled from the dorsal distal radius into the lunate, with the lunate held in the reduced position. With the K wire in place, the wrist is extended, while the neutral alignment of the radiolunate joint is preserved.

K-Wire Fixation In Watson’s original description of the 4-corner arthrodesis, K wires were the method of fixation.16 After the carpal bones are temporarily held in neutral alignment and bone graft is placed, 0.045-inch K wires are placed from the capitate to the lunate, hamate to lunate, triquetrum to capitate, and triquetrum to lunate (Fig 8). The placement of wires is confirmed on fluoroscopy. The wires are cut and buried under the skin or left protruding through the skin. Additional bone graft is placed to fill the gaps.

Choice of Bone Graft Allograft or autologous bone graft obtained from the distal dorsal radius or the anterior iliac crest is packed into the interstices of the denuded articular surfaces. Autologous cancellous grafts are the most effective grafting material, because they are osteocondutive and osteoinductive and contain living osteogenic cells.53 Among the choices of autologous grafts, distal radius graft allows for a single incision for procurement and carpal exposure; however, the graft is less dense, has fewer osteogenic cells, and contains

Staple Fixation Since the advent of power staple fixation, the placement of staples has increased in accuracy and biomechanical properties.56 The 3M Staplizer (3M, Minneapolis, MN) is an air-powered stapler/impactor that uses staples in cartridges ranging from 7 mm to 16 mm in width and depth. Power-driven staples showed increased pullout strength compared with manually driven staples. When used in 4-corner arthrodesis, staples can be used in conjunction with K wires (Fig 9) or by themselves. When used by themselves, one of

FIGURE 7. The articular surfaces of the 4-corner arthrodesis are denuded of cartilage down to subchondral bone. A highspeed burr is used to create evenly spaced burr holes throughout the surfaces to be arthrodesed.

100

FOUR-CORNER ARTHRODESIS 䡠 SHIN

FIGURE 8. K-wire fixation of the 4-corner arthrodesis has been a very common method of securing the carpal bones. Typically, 0.045-inch K wires are placed percutaneously, engaging the capitate-lunate, capitate-triquetrum, hamate-lunate, and hamate-triquetrum bones (A, B). Additional K wires are placed as needed. After K-wire fixation, bone graft is packed between the bones to be arthrodesed (C).

FOUR-CORNER ARTHRODESIS 䡠 SHIN

101

FIGURE 9. (A, B) A combination of 3M Staples and K wires can be used for fixation of the 4-corner arthrodesis.

two configurations can be used. One configuration uses 4 staples in a “box” pattern (Fig 10). The staples span the capitolunate, capitohamate, triquetrohamate, and lunotriquetral articulations. The lunotriquetral staple is difficult to place and may impinge on the dorsal lip of the radius if improperly positioned. The second configuration, which avoids impingement on the radius, is placement of 3 staples that span the capitohamate, capitolunate, and triquetrohamate joints (Fig 11). Typically, a 13-mm– wide by 10mm– deep staple is used. Once the staples are placed, temporary K wires, which maintained carpal alignment, are removed and wrist motion is evaluated;

careful fluoroscopic evaluation of the depth of the staple is necessary in order to prevent irritation of the pisotriquetral joint or other volar structures.57 Additional bone graft is placed in the 4-corner area. Screw Fixation Screw fixation of the 4-corner arthrodesis with headless compression screws is a method of fixation that results in immediate stable fixation that would allow for early range of motion. Use of the cannulated Acutrak variable-pitched headless screw (Acumed, Beaverton, OR) or cannulated Herbert-Whipple screw (Zimmer, Warsaw, IN) is ideal. The cannulated

102

FOUR-CORNER ARTHRODESIS 䡠 SHIN

FIGURE 10. (A, B) 3M Staples used in a box configuration secure the capitolunate, lunotriquetral, capitohamate, and triquetrohamate articulations. Care must be used in placing the lunotriquetral staple, because it can impinge on the dorsal distal radius.

screw allows for accurate placement of the screw and allows the placement of multiple screws in a relatively confined space. Screws should be placed at a minimum across the capitolunate, triquetrohamate, and capitohamate joints when using screw fixation. A variety of combinations of screw placement can result in successful outcomes. Plate Fixation A specially designed plate for limited wrist arthrodesis called the Spider Limited Wrist Fusion Plate (Kinetikos Medical Inc, San Diego, CA) was introduced in 1999. The Spider plate is a lowprofile, conically shaped plate with 8 recessed screw holes that accept 2.4-mm screws (Fig 12). After correction of the DISI deformity with the radiolunate K wire, bone graft is packed into the interstices of the denuded carpal bones. A K wire is

placed from capitate to the triquetrum, provisionally fixing the carpal bones (Fig 13). A special conical rasp is used to create a recessed, accepting bed for the Spider plate, which is centered at the junction of the 4 corners (Fig 14). In the case of a type II lunate, the rasp is centered at the articulation of the lunate and hamate. The Spider plate is placed in the accepting bed and rotated to allow the placement of 2 screws into each of the carpal bones. A 1.5-mm drill is used to place the index screw hole, typically in the capitate. The screw length is measured, and a self-tapping 2.4-mm screw is placed into the capitate. Subsequently, a screw is placed in each of the remaining carpal bones, followed by placement of the second screw into each of the carpal bones (Fig 15). The provisional K wire fixation is removed; motion is evaluated, ensuring that no plate impingement occurs; and screw

FOUR-CORNER ARTHRODESIS 䡠 SHIN

103

FIGURE 11. (A, B) An alternative to the box configuration for placement of staples is the use of 3 staples placed across the capitohamate, capitolunate, and triquetrohamate joints. (C) Bone graft is placed between the bones before staple application and is further impacted between the bones after the staples are placed.

104

FOUR-CORNER ARTHRODESIS 䡠 SHIN

FIGURE 12. The Spider limited wrist fusion plate is a specially designed plate for 4-corner arthrodesis. It is a lowprofile, conical plate with 8 recessed screw holes (A), which accept 2.4-mm screws (B). (Courtesy of Kinetikos Medical, Inc)

lengths are confirmed with radiographs (Fig 16). Additional bone graft is packed into the center of the 4-corner region through the plate. The rigid fixation provided by the Spider plate allows for early active range-of-motion exercises. Closure After the 4-corner arthrodesis is secured and the temporary K wires removed, the radially based flap of

the ligament-sparing capsulotomy is replaced and repaired with nonabsorbable sutures (Fig 17). The extensor tendons are replaced, and the extensor retinaculum is repaired with the extensor pollicis longus dorsally transposed (Fig 18). Tourniquet is deflated, hemostasis is obtained, and the skin edges are reapproximated (Fig 19). A bulky hand dressing with a volar plaster splint immobilizes the wrist in a neutral position and the elbow at 90°.

POSTOPERATIVE CARE/REHABILITATION

FIGURE 13. K wires are used to maintain the reduction of the capitolunate axis when applying the Spider plate, and they must be placed volarly to avoid impinging on the rasp. As such, in addition to the radiolunate-reduction K wire used to reduce the DISI deformity, a K wire is used to secure the capitate to the triquetrum.

Postoperative care and rehabilitation are contingent on the method of fixation used; however, all patients are encouraged to commence digital range of motion, tendon-gliding exercises, and edema-control measures starting on the first postoperative day. Approximately 1 week after surgery, the bulky hand dressing and sutures are removed. If K wires or staples are used, a long arm cast with the elbow at 90° is applied for 3 weeks, at which time a short arm cast is applied until arthrodesis has occurred. When screw or Spider plate fixation is used, a short arm cast can be placed for 3 to 4 weeks, or alternatively, a removable custom-made orthoplast short arm splint can be applied to allow early active range-of-motion exercises. Serial interval radiographs should be taken to ensure the arthrodesis has united before return to full normal activities. Based on our experience, the average time to arthrodesis for K wires is approximately 8 to 10 weeks, for

FOUR-CORNER ARTHRODESIS 䡠 SHIN

105

FIGURE 14. A conical rasp is used to burr out an accepting bed for the Spider plate to lie in. The rasp is placed deep enough to allow the plate to sit flush with the dorsal cortex of the carpal bones.

staple fixation is approximately 6 to 8 weeks, and for Spider plate fixation is approximately 4 to 5 weeks. A formalized hand therapy program emphasizing range of motion, strengthening and endurance is often beneficial following prolonged immobilization.

CLINICAL OUTCOMES The earliest reported outcomes for 4-corner arthrodesis was by Watson and associates.16 They reported

FIGURE 15. The plate is situated such that each carpal bone will accept 2 screws. The initial screw is placed into the capitate and then into each of the remaining carpal bones. The remainder of the screws are then placed. The screws are self-tapping and vary in length from 8 mm to 14 mm in 2 mm increments.

on 2 patients with 3 wrists that underwent the 4-corner arthrodesis, of which only the results of 1 patient (1 wrist) were reported. This patient had a follow-up of 2 years and 6 months and had 10° of extension (40° opposite wrist), 50° flexion (55° opposite), 20° radial deviation (35° opposite), and 20° ulnar deviation (20° opposite) and reported that he was without pain. Three years later, Watson and Ballet18 reported on the treatment of 19 patients treated for SLAC wrist. Of these, only 9 were treated with silicone scaphoid replacement combined with the 4-corner arthrodesis and 1 with 4-corner arthrodesis. The average age of this subset of patients was 55 years, and the average follow-up was 17.5 months. Overall this cohort reported less pain, with the exception of 1 patient who reported more pain. The average grip strength of the operative wrist of this cohort was 73% of the unaffected wrist. The average flexion of the operative wrist was 30° (45% of the opposite), average extension was 33° (54% of opposite), average radial deviation was 11° (40% of opposite), and average ulnar deviation was 26° (66% of opposite). In 1993, Voche and Merle58 reported on 12 patients who underwent 4-corner arthrodesis that were followed up for an average of 18.8 months. They reported that 75% had pain relief and an average grip strength of 57% of the opposite, unaffected wrist. The average flexion of the operated wrist was 27°, extension was 18°, radial deviation was 4°, and ulnar deviation was 13°.

106

FOUR-CORNER ARTHRODESIS 䡠 SHIN

FIGURE 16. (A, B) After the plate is applied and provisional fixation is removed, radiographs of the wrist are obtained. The radiographs of the Spider plate for which the intraoperative photographs are illustrated in Figures 13, 14, and 15 are shown here.

The first large series to evaluate the outcome of scaphoidectomy and 4-corner was done by Ashmead and associates in 1994.19 These authors reported on 100 cases followed up for an average of 44 months. 85% of patients reported excellent or good pain relief. The average grip strength was 81% of the opposite unaffected wrist. The average flexion of the operated wrist was 32°, extension was 42°, radial deviation was 15°, and ulnar deviation was 22°. The average total flexion-extension arc of motion was 53% of that of the opposite wrist. Watson and associates12 reported their results of 252 patients who underwent the 4-corner arthrodesis for SLAC wrist. This report appears to include the patients in Ashmead’s series.19 Despite this, the average flexion-extension arc was 88° (53% of opposite

wrist), with 91% of patients reporting pain relief. The average grip was 80% of the opposite grip.

COMPLICATIONS Of the 431 cases reported in 8 series of 4-corner arthrodesis, the overall complication rate was 13.5%.12,16,18,19,27,57-60 Dorsal impingement of the capitate and radius occurred in 4.4% (Fig 20), reflex sympathetic dystrophy occurred in 3%, superficial infections occurred in 3%, nonunions occurred in 2%, deep infections occurred in 0.5%, and deQuervain’s tenosynovitis occurred in 0.5%. Only 7 failures of 4-corner arthrodesis that required conversion to total wrist arthrodesis were reported, representing a 2% incidence.

FOUR-CORNER ARTHRODESIS 䡠 SHIN

FIGURE 17. After fixation of the 4-corner arthrodesis with K wires, staples, screws or plates, the radially based flap of the ligament-sparing capsulotomy is closed with nonabsorbable sutures, with care taken not to overtighten the capsule.

ALTERNATIVE PROCEDURES Capitolunate Arthrodesis The 4-corner arthrodesis was innovated secondary to the difficulties in obtaining union in capitolunate arthrodesis.16 Several authors have described scaphoidectomy and capitolunate arthrodesis for the treatment of radioscaphoid arthritis.11,16,17,61 The average range of motion obtained with capitolunate arthrodesis was 48% of normal extension, 59% of normal flexion, 59% of normal radial deviation, and 59% of normal ulnar deviation.17 Grip strength averaged 67% of the

FIGURE 18. The extensor retinaculum is repaired, transposing the extensor pollicus longus dorsally, with nonabsorbable suture.

107

normal grip strength.17 A comparison of clinical results shows that the addition of the hamate and triquetrum to the capitolunate arthrodesis does not diminish motion.17,27,57,59 Despite the advantages of requiring a less extensive exposure, less bone graft and less surgical time compared with 4-corner arthrodesis, this procedure is not without complications. Kirshenbaum and associates reported a 33% nonunion rate and 33% pin tract infection rate.17 Another variant of a capitolunate arthrodesis was described by Calandruccio and associates,62 who described a capitolunate arthrodesis combined with excision of the scaphoid and triquetrum. In 14 wrists treated with this technique, they showed range of motions and grip strengths similar to those of other motion-sparing procedures (53° flexion-extension arc; 71% normal grip strength). The lower nonunion rate of 14% most likely represents the results of screw fixation compared with the K-wire fixation used by Kirshenbaum et al.17,62 Proximal Row Carpectomy An alternative to 4-corner arthrodesis in the treatment of stage I or II SLAC is the proximal row carpectomy (PRC), which involves the removal of the proximal carpal row and allows the capitate to articulate in the lunate fossa. Although concerns regarding the differential radius of curvature between the lunate fossa and capitate and the subsequent development of progressive degenerative arthritis have been discussed, long-term studies have not borne this out.63-66 Several studies have compared 4-corner arthrodesis to PRC in the treatment of SLAC wrist (Ta-

FIGURE 19. After tourniquet is released and hemostasis is obtained, the wound edges are reapproximated. The T incision has led to very few skin/wound complications when used for elective surgical procedures.

108

FOUR-CORNER ARTHRODESIS 䡠 SHIN

FIGURE 20. (A) Dorsal impingement of the capitate and radius can occur if the capitolunate axis is not arthrodesed in neutral alignment. (B) Radiograph of a patient with dorsal impingement secondary to failure to completely reduce the DISI deformity of the lunate in addition to excessive bone graft being placed.

ble 1).27,57,59 Tomaino and associates59 reported a retrospective comparison of 15 patients with SLAC wrist treated by PRC and 7 treated with 4-corner arthrodesis with an average follow-up of 5.5 years. All patients but 3 with PRC had satisfactory pain relief, grip strength, and functional performance. There were no statistical differences in subjective or objective

results with the exception of motion, where PRC had improved motion over 4-corner arthrodesis. Tomaino and associates concluded that for stage III SLAC, 4-corner arthrodesis is the motion-preserving treatment of choice but that for stage I and II SLAC, PRC is more appealing, because it avoids complications of nonunion, pin tract infections, and lengthy immobilization.

FOUR-CORNER ARTHRODESIS 䡠 SHIN

109

TABLE 1 A Comparison of Proximal Row Carpectomy and 4-Corner Arthrodesis in the Treatment of SLAC Wrist Tomaino et al59

Total flexion-extension arc of motion (degrees) Total flexion-extension arc of motion (percent opposite wrist) Grip strength (percent opposite) Failure (percent)

Wyrick et al57

Krakauer et al27

PRC (n ⫽ 15)

4-Corner (n ⫽ 7)

PRC (n ⫽ 11)

4-Corner (n ⫽ 17)

PRC (n ⫽ 12)

4-Corner (n ⫽ 23)

77

52

85

67

71

54

64 77 20

41 76 0

64 94 0

47 74 30

— — 17

— — 9

Wyrick and associates57 showed statistically significant differences with respect to total arc of wrist motion and grip strength. Specifically, in their retrospective comparison, they noted a total arc of motion of 64% in PRC compared with 47% in 4-corner arthrodesis and improved grip in PRC (94% of opposite wrist) compared with 4-corner arthrodesis (74% of opposite). These researchers recommended PRC for SLAC stage I or II. In a retrospective evaluation of treatments for SLAC wrists, Krakauer and associates27 compared 23 patients treated with 4-corner arthrodesis to 12 patients with PRC. Similar to the findings of the previous studies, they concluded that PRC best preserved wrist mobility and that 4-corner arthrodesis was reliable in diminishing pain in SLAC stage III wrist while maintaining a functional arc of motion.

CONCLUSIONS Four-corner arthrodesis with scaphoidectomy is a time-tested, motion-sparing wrist procedure that has evolved over the past 20 years. The indications

for 4-corner arthrodesis include radioscaphoid arthritis with or without midcarpal arthritis, failed intercarpal ligament reconstruction, or perilunar instabilities not amenable to soft tissue reconstruction. The absolute contraindication for 4-corner arthrodesis is radiolunate arthritis or ulnar translocation of the carpus. Four-corner arthrodesis is a biomechanically sound intercarpal fusion that results in near-normal load transmission through the radiolunate articulation. Patient satisfaction is high, and the procedure offers good to excellent pain relief. The range of motion after 4-corner arthrodesis ranges from 41% to 53% of the normal opposite wrist, and up to 76% of normal grip strength can be expected. Advances in surgical exposures, fixation techniques, and implants have allowed for rigid fixation that enables for rapid union and the commencement of early range of motion. Failure rates and complication rates are relatively low, and long-term outcomes have been promising.

REFERENCES 1. Campbell AC, Strick MJ. Coalition of the proximal row of the carpus. Ulster Med J 1994;63:119-120. 2. Delaney TJ, Eswar S. Carpal coalitions. J Hand Surg [Am] 1992;17A:28-31. 3. Resnik CS, Grizzard JD, Simmons BP, Yaghmai I. Incomplete carpal coalition. AJR Am J Roentgenol 1986;147:301-304. 4. Cope JR. Carpal coalition. Clin Radiol 1974;25:261-266. 5. Peterson HA, Lipscomb PR. Intercarpal arthrodesis. Arch Surg 1967;95:127-134. 6. Thornton L. Old dislocation of os magnum: open reduction and stabilization. South Med J 1924;17:430-431.

7. Sutro CJ. Treatment of nonunion of the carpal navicular bone. Surgery 1946;20:536-540. 8. Helfet AJ. A new operation for ununited fractures of the scaphoid. Proceedings of the South African Orthopaedic Association. J Bone Joint Surg Br 1952;34B:329. 9. Graner O, Lopes EI, Carvalho BC, Atlas S. Arthrodesis of the carpal bones in the treatment of Kienbock’s disease, painful ununited fractures of the navicular and lunate bones with avascular necrosis and old fracture dislocations of carpal bones. J Bone Joint Surg Am 1966;48A:767774.

110

FOUR-CORNER ARTHRODESIS 䡠 SHIN

10. Watson HK, Hempton RF. Limited wrist arthrodeses. I. The triscaphoid joint. J Hand Surg [Am] 1980;5A:320-327. 11. Watson HK. Limited wrist arthrodesis. Clin Orthop 1980: 126-136. 12. Watson HK, Weinzweig J, Guidera PM, Zeppieri J, Ashmead D. One thousand intercarpal arthrodeses. J Hand Surg [Br] 1999;24B:307-315. 13. Waugh RL, Reuling L. Ununited fractures of the carpal scaphoid. Am J Surg 1945;67:184-200. 14. Murray G. Bone graft for nonunion of the carpal scaphoid. Surg Gyn Obst 1935;60:540-541. 15. Davidson AJ, Horwitz MT. An evaluation of excision in the treatment of ununited fracture of the carpal scaphoid (navicular) bone. Ann Surg 1938;108:291-295. 16. Watson HK, Goodman ML, Johnson TR. Limited wrist arthrodesis. Part II: intercarpal and radiocarpal combinations. J Hand Surg [Am] 1981;6A:223-233. 17. Kirschenbaum D, Schneider LH, Kirkpatrick WH, Adams DC, Cody RP. Scaphoid excision and capitolunate arthrodesis for radioscaphoid arthritis. J Hand Surg [Am]1993;18A:780785. 18. Watson HK, Ballet FL. The SLAC wrist: scapholunate advanced collapse pattern of degenerative arthritis. J Hand Surg [Am] 1984;9A:358-365. 19. Ashmead D, Watson HK, Damon C, Herbert S, Paly W. Scapholunate advanced collapse wrist salvage. J Hand Surg [Am] 1994;19A:741-750. 20. Allende BT. Osteoarthritis of the wrist secondary to nonunion of the scaphoid. Int Orthop 1988;12:201-211. 21. Chen C, Chandnani VP, Kang HS, Resnick D, Sartoris DJ, Haller J. Scapholunate advanced collapse: a common wrist abnormality in calcium pyrophosphate dihydrate crystal deposition disease. Radiology 1990;177:459-461. 22. Harrington RH, Lichtman DM, Brockmole DM. Common pathways of degenerative arthritis of the wrist. Hand Clin 1987;3:507-527. 23. Ruby LK, Leslie BM. Wrist arthritis associated with scaphoid nonunion. Hand Clin 1987;3:529-539. 24. Vender MI, Watson HK, Wiener BD, Black DM. Degenerative change in symptomatic scaphoid nonunion. J Hand Surg [Am] 1987;12A:514-519. 25. Watson HK, Ryu J. Evolution of arthritis of the wrist. Clin Orthop 1986;202:57-67. 26. Resnick D, Niwayama G. Carpal instability in rheumatoid arthritis and calcium pyrophosphate deposition disease. Pathogenesis and roentgen appearance. Ann Rheum Dis 1977;36:311-318. 27. Krakauer JD, Bishop AT, Cooney WP. Surgical treatment of scapholunate advanced collapse. J Hand Surg [Am] 1994; 19A:751-759. 28. Herzberg G, Cooney WP. Perilunate fracture dislocations. In: Cooney WP, Linscheid RL, Dobyns JH, eds. The wrist: diagnosis and operative treatment. Vol 1. St. Louis: Mosby, 1998:651-683. 29. Wright TW, Dobyns JH, Linscheid RL, Macksoud W, Siegert J. Carpal instability non-dissociative. J Hand Surg [Br] 1994;19B:763-773. 30. Lichtman DM, Bruckner JD, Culp RW, Alexander CE. Palmar midcarpal instability: results of surgical reconstruction. J Hand Surg [Am] 1993;18A:307-315. 31. Amadio PC. Carpal kinematics and instability: A clinical and anatomic primer. Clin Anat 1991;4:1-12.

32. Berger RA. The anatomy and basic biomechanics of the wrist joint. J Hand Ther 1996;9:84-93. 33. Garcia-Elias M, Cooney WP, An KN, Linscheid RL, Chao EY. Wrist kinematics after limited intercarpal arthrodesis. J Hand Surg [Am] 1989;14A:791-799. 34. Sarrafian SK, Melamed JL, Goshgarian GM. Study of wrist motion in flexion and extension. Clin Orthop 1977:153-159. 35. Savelberg HH, Kooloos JG, De Lange A, Huiskes R, Kauer JM. Human carpal ligament recruitment and three-dimensional carpal motion. J Orthop Res 1991;9:693-704. 36. Smith DK, Cooney WP, An KN, Linscheid RL, Chao EYS. Effects of simulated unstable scaphoid fractures on carpal motion. J Hand Surg [Am] 1989;14A:283-291. 37. Ruby LK, Cooney WP, An KN, Linscheid RL, Chao EY. Relative motion of selected carpal bones: a kinematic analysis of the normal wrist. J Hand Surg [Am] 1988;13A:1-10. 38. Horii E, Garcia Elias M, An KN, et al. A kinematic study of luno-triquetral dissociations. J Hand Surg [Am] 1991;16A: 355-362. 39. Kauer JM. The interdependence of carpal articulation chains. Acta Anat (Basel) 1974;88:481-501. 40. Allieu Y, Brahin B, Asencio G. Carpal instabilities. Radiological and clinico-pathological classification. Ann Radiol (Paris) 1982;25:275-287. 41. Brahin B, Allieu Y. Compensatory carpal malalignments. Ann Chir Main 1984;3:357-363. 42. Douglas DP, Peimer CA, Koniuch MP. Motion of the wrist after simulated limited intercarpal arthrodeses. An experimental study. J Bone Joint Surg Am 1987;69A:1413-1418. 43. Gellman H, Kauffman D, Lenihan M, Botte MJ, Sarmiento A. An in vitro analysis of wrist motion: the effect of limited intercarpal arthrodesis and the contributions of the radiocarpal and midcarpal joints. J Hand Surg [Am] 1988;13A:378383. 44. Tencer AF, Viegas SF, Cantrell J, et al. Pressure distribution in the wrist joint. J Orthop Res 1988;6:509-517. 45. Viegas SF, Tencer AF, Cantrell J, et al. Load transfer characteristics of the wrist. Part II. Perilunate instability. J Hand Surg [Am] 1987;12A:978-985. 46. Viegas SF, Tencer AF, Cantrell J, et al. Load transfer characteristics of the wrist. Part I. The normal joint. J Hand Surg [Am] 1987;12A:971-978. 47. Viegas SF, Patterson RM, Peterson PD, et al. Evaluation of the biomechanical efficacy of limited intercarpal fusions for the treatment of scapho-lunate dissociation. J Hand Surg [Am] 1990;15A:120-128. 48. Viegas SF. Limited arthrodesis for scaphoid nonunion. J Hand Surg [Am] 1994;19A:127-133. 49. Viegas SF, Patterson R, Peterson P, Roefs J, Tencer A, Choi S. The effects of various load paths and different loads on the load transfer characteristics of the wrist. J Hand Surg [Am] 1989;14A:458-465. 50. Dao KD, Shin AY, Berger RA. T Incision for Exposure of the Distal Radius and Wrist. J Hand Surg [Br] 2000;25B:544547. 51. Berger RA, Bishop AT. A fiber-spitting capsulotomy technique for dorsal exposure of the wrist. Tech Hand Upper Extrem Surg 1997;1:2-10. 52. Linscheid RL, Rettig ME. The treatment of displaced scaphoid nonunion with trapezoidal bone graft. In: Gelberman RH, ed. Masters techniques in orthopaedic surgery. New York: Raven Press, 1994:119-131. 53. Gazdag AR, Lane JM, Glaser D, Forster RA. Alternatives to

FOUR-CORNER ARTHRODESIS 䡠 SHIN

54.

55. 56. 57.

58. 59.

autogenous bone graft: efficacy and indications. J Am Acad Orthop Surg 1995;3:1-8. Schnitzler CM, Biddulph SL, Mesquita JM, Gear KA. Bone structure and turnover in the distal radius and iliac crest: a histomorphometric study. J Bone Miner Res 1996;11:17611768. Biddulph SL. Bone donor site. Iliac crest or distal radius? J Hand Surg [Br] 1999;24B:645-646. Shapiro JS. Power staple fixation in hand and wrist surgery: new applications of an old fixation device. J Hand Surg [Am] 1987;12A:218-227. Wyrick JD, Stern PJ, Kiefhaber TR. Motion-preserving procedures in the treatment of scapholunate advanced collapse wrist: proximal row carpectomy versus four-corner arthrodesis. J Hand Surg [Am] 1995;20A:965-970. Voche P, Merle M. Arthrodesis of 4 bones of the wrist. Study of 12 follow-up cases. Rev Chir Orthop Reparatrice Appar Mot 1993;79:456-463. Tomaino MM, Miller RJ, Cole I, Burton RI. Scapholunate advanced collapse wrist: proximal row carpectomy or limited wrist arthrodesis with scaphoid excision? J Hand Surg [Am] 1994;19A:134-142.

111

60. Trumble T, Bour CJ, Smith RJ, Edwards GS. Intercarpal arthrodesis for static and dynamic volar intercalated segment instability. J Hand Surg [Am] 1988;13A:384-390. 61. Osterman AL, Mikulics M. Scaphoid nonunion. Hand Clin 1988;4:437-455. 62. Calandruccio JH, Gelberman RH, Duncan SF, Goldfarb CA, Pae R, Gramig W. Capitolunate arthrodesis with scaphoid and triquetrum excision. J Hand Surg [Am] 2000;25A:824832. 63. Tomaino MM, Delsignore J, Burton RI. Long-term results following proximal row carpectomy. J Hand Surg [Am] 1994;19:694-703. 64. Culp RW, McGuigan FX, Turner MA, Lichtman DM, Osterman AL, McCarroll HR. Proximal row carpectomy: a multicenter study. J Hand Surg [Am] 1993;18A:19-25. 65. Inglis AE, Jones EC. Proximal-row carpectomy for diseases of the proximal row. J Bone Joint Surg Am 1977;59A: 460-463. 66. Imbriglia JE, Broudy AS, Hagberg WC, McKernan D. Proximal row carpectomy: clinical evaluation. J Hand Surg [Am] 1990;15A:426-430.