➢ Ulnar collateral ligament injuries are best classified according to chronicity of injury and grade of tear, with partial tears and nondisplaced tears being amenable to nonoperative treatment.
➢ In general, acute injuries may be primarily repaired with suture-anchor fixation, most often to the insertion of the ligament on the proximal phalanx, which allows for early joint range of motion.
➢ Symptomatic chronic injuries may be managed surgically with primary repair, reconstruction, or arthrodesis.
“Gamekeeper’s thumb” was first described by Campbell in 1955 as a mechanism for chronic ulnar thumb metacarpophalangeal (MCP) ligament instability among gamekeepers involved in the killing of rabbits1. According to Campbell, the repeated use of a pincer maneuver involving the ulnar side of the thumb and the forefinger in order to hyperextend the necks of the rabbits caused the ulnar collateral ligament (UCL) to become attenuated over time. Nearly 20 years later, Shultz and Fox described an acute injury to the UCL in skiers who had sustained forced thumb abduction and hyperextension, leading to rupture of the UCL or an avulsion of its phalangeal attachment2.
UCL injuries of the thumb are common, with an estimated incidence of 50 per 100,000 emergency room visitors3. Although the eponymous term “skier’s thumb” implies an acute injury with a specific mechanism, the majority of these acute injuries occur as a result of a fall on an outstretched hand, assault, or other sporting injury. The injury has been described in association with almost every sport, but it is more common in racquet and ball-handling sports4. A stable, pain-free thumb is necessary not only for sports and hobbies but also for activities of daily living. The injury usually is the result of a valgus or radially directed load to the thumb MCP joint. Because two-thirds of these injuries occur in working adults, the burden that these injuries place on productivity is substantial, with an average 88 days of work lost per injury5.
In the present report, the anatomy and diagnostic approach to thumb UCL injuries will be reviewed and treatment options will be stratified. Specifically, we will examine which procedure is most effective for particular injuries on the basis of ligamentous injury and chronicity.
Anatomy and Biomechanics of the Thumb MCP Joint
The thumb carpometacarpal joint is a diarthrodial ginglymoid joint that contains the articulation of the first distal metacarpal and proximal phalanx. Dynamic and static stabilizers provide stability to the joint while allowing mobility of the phalanx on the metacarpal in flexion-extension and adduction-abduction motion arcs as well as rotation. Uniquely, the thumb MCP joint must be stable in both flexion and extension to enable pinch and grasp6. Dynamic stabilizers include extrinsic and intrinsic muscles of the thumb, including the extensor pollicis longus, extensor pollicis brevis, flexor pollicis longus, flexor pollicis brevis, and adductor pollicis. The combined force of these dynamic stabilizers creates a net flexion moment and volar pull on the proximal phalanx at the MCP joint. The UCL, radial collateral ligament, volar plate, dorsal capsule, and articular congruity are all static stabilizers that provide mediolateral stability as well as dorsal support, preventing volar subluxation7.
The UCL is a thick band measuring 4 to 8 mm wide and 12 to 14 mm long8. Its origin is centered an average of 4.2 mm from the volar surface and an average of 5.3 mm from the articular surface of the thumb metacarpal. Its insertion is centered an average of 2.8 mm from the volar surface and an average of 3.4 mm from the articular surface of the proximal phalanx9 (Fig. 1).The UCL complex consists of a proper UCL and an accessory UCL, providing support against valgus stress and volar subluxation of the MCP joint7. However, these ligaments contribute differently to stability, depending on joint position. The proper UCL courses from the dorsal third of the metacarpal head to the volar aspect of the proximal phalanx. The accessory UCL originates volar to the proper UCL and inserts on the ulnar border of the volar plate. Thus, the accessory UCL becomes taut in extension, whereas the proper UCL is taut in flexion10 (Fig. 2). The normal valgus laxity of the MCP joint ranges from 6° in extension to 15° in flexion11,12. UCL insufficiency can lead to key-pinch and tip-pinch weakness13.
The adductor pollicis inserts on the proximal phalanx, volar plate, ulnar sesamoid, and adductor aponeurosis lying volar to the MCP axis of rotation14. The adductor aponeurosis normally lies superficial to the MCP capsule and UCL but may prevent apposition of the UCL back to the proximal phalanx in patients with severe injuries. In such patients, both the proper UCL and the accessory UCL are displaced superficial to the adductor pollicis aponeurosis as described by Stener15. Stener lesions have been reported to occur in as many 87% of patients with complete ruptures of the UCL. The implication of the Stener lesion is the inability for the UCL to heal as there is an interposed adductor pollicis aponeurosis16 (Fig. 3).
The treatment of UCL injuries requires accurate diagnosis of the severity of the injury17. Patients may recall an abduction or valgus-directed force to the thumb and may present with pain, swelling, ecchymosis, decreased range of motion, and tenderness at the MCP joint. A palpable mass may be indicative of a Stener lesion, but the absence of a mass does not necessarily rule out such a lesion15,16. Patients with chronic injuries may complain of weakness, deformity, or the feeling of instability with grasp or pinch.
Physical examination must include bilateral assessment of the UCL by providing a radially directed force to the proximal phalanx as the thumb metacarpal is stabilized12,18. The position of the MCP joint determines whether the accessory UCL or the proper UCL is being evaluated. At 0° of flexion the accessory UCL is stressed, whereas at 30° to 35° of flexion the proper UCL is better isolated. Mayer et al. demonstrated in a cadaver study that pronosupination may influence perceived UCL stability, with supination allowing for more laxity of the UCL. Therefore, it is recommended to test the thumb MCP joint in neutral rotation19.
Increased laxity of at least 35° without a firm end point or at least 10° to 15° more laxity on the involved side than on the unaffected side is often considered diagnostic of a complete tear16. However, in a study of 100 asymptomatic patients, 34% of subjects had a difference of ≥10° and 12% of patients had a difference of ≥15%20. This finding led the authors to recommend lack of a definite end point rather than comparison with the uninjured, contralateral side on stress examination to identify complete rupture of the UCL20. It is also important to understand that the physical examination findings can be subtle. During the examination, the patient needs to be relaxed and comfortable and the surgeon must be gentle and methodical. Gentle examination and palpation, with avoidance of sudden valgus stress, can often confirm the diagnosis. The presence of a large lump may be consistent with a Stener lesion. A local anesthetic may be injected into the MCP joint to allow for better clinical examination21.
Radiographs are routinely made to assess for any fractures or volar subluxation. Occasionally, a Stener lesion may be identified on radiographs22. Stress radiographs or arthrography may aid in the diagnosis23. However, stress radiographs can be uncomfortable for the patient and guarding may increase the rate of false-negative results. Similarly, care must be taken when relying on radiographs as both fracture and ligamentous tears may occur simultaneously, reinforcing the need for careful clinical examination24,25.
McKeon et al. used fluoroscopy and defined radial translation of the proximal phalanx as the shortest distance between a line drawn parallel to the longitudinal axis of the metacarpal as it passes through the radial border of the metacarpal head articular surface and the radial border of the proximal phalanx articular surface26. They found that radial translation of >2 mm was 100% specific for complete disruption of the UCL complex.
The use of ultrasound and magnetic resonance imaging (MRI) to improve diagnostic accuracy in patients with thumb UCL injuries has been extensively studied. Two findings that are both sensitive and specific for full-thickness UCL tears on ultrasound are the absence of UCL fibers and the presence of a heterogeneous mass-like abnormality proximal to the first MCP joint27. Jones et al. demonstrated that ultrasound had a positive predictive value of 94% for the diagnosis of UCL rupture3. A review of the literature demonstrated a sensitivity of 76%, a specificity of 81%, an accuracy of 81%, a positive predictive value of 74%, and a negative predictive value of 87%28.
The sensitivity and specificity of MRI for UCL injury detection were reported to be as high as 100% in an evaluation of 17 patients with UCL tears and 21 normal subjects29. In another study involving 34 asymptomatic volunteers, the UCL had high signal intensity in one-third of the subjects, which the authors suggested could lead to overdiagnosis of UCL injury30. No studies of the cost-effectiveness of routine MRI or ultrasound examination for patients with suspected UCL injuries have been performed, to our knowledge. Currently, advanced imaging is recommended if there is uncertainty with regard to the diagnosis or the extent of injury on the basis of clinical examination31.
Current Treatment for Acute UCL Injuries
Grading acute UCL injuries on an ordinal scale is extremely difficult when the injury pattern is on a continuous scale. However, grading is based on ligamentous laxity on a 3-grade scale31. Grade-I injuries present with tenderness along the UCL but with symmetrical laxity compared with the contralateral side, indicating a sprain. Grade-II injuries demonstrate an increase in laxity relative to the contralateral side, but with a firm end point, signifying a partial tear. Grade-III injuries are complete ruptures with an increase in laxity without a firm end point. While we are not aware of any consensus statements in the literature on the treatment of UCL injuries of the thumb, several authors have proposed operative indications, including gross instability, Stener lesions, and displaced avulsion fractures5 (Fig. 4).
Romano et al. divided UCL injuries into 5 categories on the basis of MRI imaging: Stener lesions, moderately displaced (≥3 mm) complete tears, minimally displaced (<3 mm) complete tears, nondisplaced complete tears, and partial tears32. This classification system was further modified by Milner et al., who combined nondisplaced and partial tears to create a 4-type treatment-oriented classification system for thumb UCL injuries33. They found Type-1 or Type-2 injuries (partial/nondisplaced tears or minimally displaced tears) could be successfully treated with immobilization. All 10 patients in their study with Type-3 injuries (displacement of ≥3 mm) were initially managed conservatively (with 9 subsequently undergoing operative treatment), whereas all patients with Type-4 injuries (Stener lesions) were managed operatively. Therefore, the authors concluded that patients with Type-1 or Type-2 injuries may be successfully managed with immobilization, whereas those with Type-3 or Type-4 injuries are likely to require operative repair (Table I).
Nonoperative treatment of UCL injuries is generally limited to partial tears or nondisplaced complete tears17. Immobilization can be performed with use of a variety of devices, from functional hinged splints to thumb spica casts, for up to 6 weeks. The goal of immobilization is to protect the MCP joint from stress while allowing interphalangeal joint range of motion. Sollerman et al. randomized 40 patients to nonoperative treatment with either a plaster thumb spica cast or a functional hinged splint34. They found no difference between the 2 groups in terms of range of motion, pinch strength, or return to work. However, the functional splint was better tolerated by patients. Athletic participation is often acceptable but may lead to the choice of 1 type of immobilization device over another4. For partial ligament tears or nondisplaced complete tears, nonoperative treatment yields generally favorable results with uneventful healing and little residual disability33,34.
Kuz et al. reported the results of nonoperative treatment in a study of 30 patients with UCL avulsion fractures who were treated with either cast immobilization or splinting for 4 to 6 weeks35. At an average of 3.1 years, 5 of 20 patients had a nonunion of the fracture and 3 had instability on stress testing. All 30 patients reported satisfaction with the results of treatment. Similarly, Sorene and Goodwin reported on 28 UCL avulsion injuries that were treated nonoperatively36. Despite a 60% rate of fibrous union, they reported no instability at 2.5 years and a 100% rate of satisfaction with nonoperative treatment.
However, in a smaller study of patients with minimally displaced (<2 mm) UCL avulsion fractures that were treated with immobilization, Dinowitz et al. reported that all patients had persistent pain and instability at 4 months, requiring operative repair37. Bowers and Hurst reported that 8 of 9 patients who had nonoperative treatment of avulsion fractures had similarly poor results23. Therefore, nonoperative treatment of UCL injuries should be limited to nondisplaced or partial ligament ruptures without displaced avulsion fractures.
Samora et al. recently performed a systematic review to compare nonoperative and operative treatment of UCL injuries as well as to compare surgical techniques38. While all but 2 of the 14 articles represented Level-IV evidence, the authors concluded that the outcomes of surgical treatment of both acute and chronic UCL injury are generally excellent regardless of operative technique. Common surgical options include direct repair with use of bone tunnels and pull-out sutures, direct repair with use of bone anchors, and tension-band fixation for osseous avulsions. Various other operative techniques have been described, including arthroscopic-assisted reduction39, interference screw docking40, and condylar shaving41. The goal of surgical treatment is to perform an anatomical reduction of the ligament to bone so as not to alter the normal mechanics of the MCP joint.
Suture-anchor fixation of the ruptured ligament to the base of the proximal phalanx has long been advocated as a way to initiate early active mobilization (Fig. 5). Kozin, in a study of 7 patients with acute UCL ruptures that were treated with suture anchors, reported nearly symmetrical range of motion and pinch strength, with no residual instability42. Weiland et al. and Zeman et al. studied larger cohorts of patients who were managed with suture anchors and reported no residual instability and high levels of patient satisfaction43,44.
In the study by Crowley et al., patients who were managed with suture anchors were randomized to either thumb spica immobilization or early mobilization45. The authors found that patients who were managed with early mobilization had earlier return to full hand function and earlier return to work, with no difference between the groups in terms of the final range of motion. Similarly, Katolik et al. demonstrated better motion, greater pinch strength, and fewer complications in patients who were managed with suture-anchor fixation and early mobilization than in those who were managed with a pull-out suture tied over a button and cast immobilization46.
Werner et al., in a study of college football players with complete UCL tears who were managed with suture-anchor repair, found that the average return to play was 4 weeks for nonskill-position players and 7 weeks for skill-position players47. All players returned to at least the same level of play. The average QuickDASH48 (an abbreviated version of the Disabilities of the Arm, Shoulder and Hand [DASH] questionnaire) score for the entire cohort was 1 of 100 after an average of 6 years of follow-up.
For fracture avulsions of the thumb UCL, fixation with a small screw or pin may be utilized. If the fragment is too small, it may be excised and the ligament may be repaired with suture anchors. Shin et al., in a cadaveric study, found that the hook-plate construct provided stronger fixation than did suture-anchor repair49.
Historically, some surgeons have augmented the repair of acute UCL injuries with temporary Kirschner-wire fixation39,50-52. When this procedure is performed, the MCP joint is pinned in slight flexion for 4 weeks and is protected with a cast or splint6,52. With an increased emphasis on early protected range of motion, the use of Kirschner wires to pin the MCP joint has been less commonly reported. However, MCP joint immobilization with use of a Kirschner wire may be beneficial for athletes who wish to return early to protected play4,52.
The systematic review by Samora et al. highlighted the paucity of literature on the acute repair of UCL injuries38. Of the 6 studies that met the inclusion criteria, 5 represented Level-IV evidence and 1 was a Level-III retrospective comparative study. Patients who were managed with suture anchors and early mobilization had significantly better motion (p < 0.05) and pinch strength (p < 0.05), with fewer complications (p = 0.08), than those who were managed with suture-button fixation and cast immobilization as reported by Katolik et al.46. However, clinical outcomes, including pain, motion, strength, and stability, were all improved regardless of operative technique. Furthermore, the longest average duration of follow-up among the 6 studies was 47 months38.
Current Techniques for Chronic UCL Injuries
The definition of the timing of chronic UCL injuries is debatable (usually described as being >3 to 6 weeks removed from the injury)53, but the determining feature in the treatment of these injuries is whether the ligament can be repaired primarily or needs reconstruction. Both dynamic and static stabilization procedures have been described. Arthrodesis may be used as a salvage procedure for patients with MCP joint arthrosis.
Primary repair generally has been avoided in cases of chronic UCL injuries in favor of reconstruction54. However, Christensen et al. reported on the outcome of repair of chronic UCL injuries in a study of 12 patients with a minimum duration of follow-up of 15 years53. The average delay between the injury and operative treatment was 2.6 years. At the time of the latest follow-up, 88% of patients had some degree of thumb MCP joint osteoarthritis and 67% of patients had some degree of clinical instability. Nevertheless, similar to the findings in other short-term outcome studies on UCL injuries38, patients were universally satisfied with the results of treatment of the affected thumb.
The use of tendon transfer for the treatment of a chronic UCL injury was first described in 1961 by Kaplan, who reported that the extensor indicis proprius was successfully transferred to the extensor pollicis longus with capsular repair55.
Dynamic reconstruction, as described by Neviaser et al.56, involves the use of adductor aponeurosis advancement as well as MCP joint capsulorrhaphy to stabilize the joint. In that study, 6 of 8 patients reported that the thumb was painless, stable, and mobile after 1 year of follow-up. McCue et al. similarly performed adductor advancement in athletes who had sustained injuries >4 weeks prior to operative treatment57. All of the athletes returned to play.
Sakellarides and DeWeese reported on 16 patients who were managed with transfer of a split extensor pollicis brevis tendon to the proximal phalanx58. All patients reported satisfactory results and had a stable MCP joint without loss of function, recurrence of instability, or disabling pain.
Static reconstruction with local tendon graft initially was described by Frykman and Johansson, who used a slip of the abductor pollicis longus13. Both Strandell59 and Fairhurst and Hansen60 used the extensor pollicis brevis to reconstruct the UCL. Strandell used a split extensor pollicis brevis in 2 patients, 1 of whom developed decreased motion, decreased pinch strength, and laxity compared with the uninjured side59. Fairhurst and Hansen reported favorable outcomes in their 2 patients, with a stable MCP joint and no loss of function60. Other autografts, including bone-retinaculum-bone61 and bone-periosteum-bone62, have been described.
Smith reported the results for 38 patients who had reconstruction with a free palmaris longus autograft7. All but 2 of the patients returned to full activities within 3 months after surgery. Breek et al. similarly used a free tendon graft for reconstruction of the UCL in 37 patients63. The authors reported equal pinch strength in 94% of patients and subjective satisfactory results in 92% of patients.
Glickel et al. described a reconstruction technique involving the passage of a free tendon graft through 2 holes in the proximal phalanx and 1 hole in the metacarpal neck in a triangular configuration to simulate the normal anatomy of the UCL64,65. Lee et al. performed a biomechanical study to compare the results of free tendon graft reconstructions in cadaveric specimens using 4 different configurations: (1) triangular configuration with apex proximal, (2) triangular configuration with apex distal, (3) cruciate configuration, and (4) parallel configuration66. They noted that all 4 configurations stabilized the MCP joint. However, only the triangular configuration with apex proximal stabilized the MCP joint while also maintaining flexion-extension range of motion (Fig. 6).
Mitsionis et al. reported on a series of 20 patients who were managed with bone suture anchors and free tendon graft for the reconstruction of chronic UCL injuries67. All patients had good or excellent results, albeit with loss of pinch strength and motion at the MCP joint compared with the unaffected side. Baskies et al. found no significant difference between reconstruction with a free tendon graft placed in a figure-of-8 fashion through drill-holes and reconstruction with the use of a Bio-Tenodesis Screw (Arthrex) in terms of peak load to failure68.
When reconstructing the UCL in patients with chronic injuries, restoring the anatomical footprint of the UCL is paramount. Carlson et al., in a cadaveric study, reported that accurate anatomical reconstruction of the UCL with palmaris longus graft and interference screw fixation showed no differences in terms of radial deviation in neutral or 30° of flexion in comparison with the values for intact controls69. However, Bean et al., in another cadaveric study, reported that volar displacement of the origin or dorsal displacement of the insertion increased radial deviation10. Similarly, distal or volar displacement of the UCL insertion on the phalanx was found to decrease flexion. While no static ligament reconstruction restores the normal stability characteristics of the UCL, Hogan et al. reported that a modification of the Glickel procedure demonstrated greater strength and stiffness than other static methods70. Martínez-Villén et al., in a study of 10 patients who were managed with apex proximal reconstruction as described by Glickel, reported an average 10.5° loss of flexion and 8° loss of extension71. All patients returned to work and had minimal loss of grip and key-pinch strength.
For patients with a chronically injured UCL and painful arthrosis, arthrodesis may be the preferred option. Careful examination of the range of motion of the uninjured thumb may reveal that there is limited motion of the MCP joint. Several groups have reported their results of arthrodesis for chronic UCL injuries. Although arthrodesis results in loss of motion, function is excellent and there is some restoration of pinch strength54,56,72.
Arnold et al. performed a retrospective review of patients with chronic UCL injuries who were managed with either primary repair, UCL advancement, reconstruction, or arthrodesis54. Although the numbers were small, the type of repair did not have an impact on outcomes. The authors concluded that arthrodesis is a reliable salvage procedure in appropriately selected patients.
The systematic review by Samora et al. included 6 studies that evaluated clinical outcomes after autograft UCL reconstruction38. Five of the studies represented Level-IV evidence, and 1 was a Level-II prospective cohort study. None of the studies compared different graft types or fixation techniques, and all techniques improved clinical outcomes in terms of pain, motion, strength, and stability at up to 75 months of follow-up. Thus, the treatment of chronic UCL injuries is based on a variety of factors, including patient symptoms, soft-tissue quality, and surgeon preference.
Rocchi et al., in a randomized trial of patients who were managed with postoperative immobilization with either a thumb spica or a modified spica that allowed for flexion-extension at the MCP joint, demonstrated that immediate postoperative motion led to better function at 1, 2, and 6 months, with no difference at 1 year73. Range of motion was greater in the experimental group at all time points, with quicker return to work and fewer physiotherapy sessions.
Earlier motion following acute UCL repair repeatedly has been shown to be safe. Firoozbakhsh et al., in a cadaveric biomechanical study, found that UCL repair with suture anchors may be able to tolerate a moderate range of active motion74. However, they cautioned against pinch activity during the early postoperative period. Similarly, a subsequent cadaveric study demonstrated that maximum load to failure following the repair of UCL injuries was 3 times greater than the load expected with active motion, suggesting that a controlled active-motion therapy protocol is safe after suture-anchor repair of a ruptured UCL75.
Little consensus exists in terms of the return to protected and unprotected play following the operative treatment of UCL injuries in elite athletes. Dy et al., in a survey of surgeons who treated professional athletes, found that most respondents recommended waiting 3 months before athletes returned to unprotected play76. However, surgeons who treated football players were more likely to recommend earlier return to protected play than non-football surgeons, whereas surgeons who treated basketball players were less likely to recommend earlier return to protected play than non-basketball surgeons.
In reviewing the numerous options for primary repair and reconstruction of these injuries, we have stratified UCL injuries according to ligamentous grade and chronicity (Table II). Few studies have directly compared treatment methodologies, and the majority of reports in the literature have been small case series or expert opinion. One systematic review highlighted the difficulties in performing a randomized controlled trial for UCL injuries38. Specifically, the results of operative treatment for these injuries are generally favorable regardless of the operative technique.
Investigation performed at the Mayo Clinic, Rochester, Minnesota
Disclosure: No external funds were received in support of this work. The Disclosure of Potential Conflicts of Interest forms are provided with the online version of this article.
- Copyright © 2017 by The Journal of Bone and Joint Surgery, Incorporated