- Email: [email protected]
Collateral ligament avulsion fractures from the heads of the metacarpals of the fingers
ARTICLE IN PRESS COLLATERAL LIGAMENT AVULSION FRACTURES FROM THE HEADS OF THE METACARPALS OF THE FINGERS D. J. SHEWRING and R. H. THOMAS From the Department of Orthopaedic Surgery, University Hospital of Wales, Cardiff, UK
Nineteen patients with collateral ligament avulsion fractures from the metacarpal heads of the fingers were treated during a 6 year period. Seven undisplaced fractures were initially treated conservatively. Four united with full movement. Three required internal fixation for symptomatic non-union and healed without problems. Eleven patients with displaced fractures were treated by primary internal fixation using a single lag screw through a dorsal approach. Seven of these achieved a full range of movement of the injured digit by 3 months. Four patients failed to regain full flexion of the metacarpophalangeal joint. One patient with a displaced and comminuted fracture was treated with internal fixation at 8 weeks when the fragment had consolidated. As with similar fractures from the proximal phalangeal bases, these fractures are prone to non-union when treated conservatively, even when undisplaced. If fixation becomes necessary, the delay of a trial of conservative treatment does not appear to affect the outcome. Internal fixation of displaced fractures allows gentle mobilisation and facilitates union. Journal of Hand Surgery (British and European Volume, 2006) 31B: 5: 537–541 Keywords: avulsion fracture, metacarpophalangeal joint, fingers
Avulsion fractures affecting the metacarpophalangeal joints occur through the pull of the collateral ligaments and may result in avulsion of a fragment from either the proximal phalangeal base or from the metacarpal head. Avulsion fractures from the base of the proximal phalanges of the fingers have previously been found to have a high rate of non-union when treated conservatively but excellent results when treated by internal fixation through a palmar approach (Shewring and Thomas, 1993). Avulsion fractures from the metacarpal heads (Fig 1) are rarer and there is scant reference to them in the literature. The only previous studies we have found are a case report (Gee and Pho, 1982) in which an avulsion fracture from the radial side of the fifth metacarpal head was treated with internal fixation and a study by Bischoff et al. (1994) in which two of these fractures were grouped along with 100 avulsion fractures of various types. The purpose of this prospective study was to investigate the outcomes of conservative and operative treatment of collateral ligament avulsion fractures from the metacarpal heads of the fingers and to describe the technique of surgical fixation.
soccer and one whilst playing ice hockey. The remainder occurred as the result of a simple fall onto the outstretched hand; five of these ten patients having tripped when running upstairs. The ulnar side of the index finger was most commonly affected, accounting for more than half of the injuries (Table 1). Three patients had concomitant fractures. There was an undisplaced spiral fracture of the shaft of the proximal phalanx in the same digit in one patient and a fracture of the metacarpal neck of an adjacent digit in another. One patient had an avulsion from the base of the proximal phalanx in an adjacent digit, which was also internally fixed through a separate palmar approach (Fig 2). Eleven patients with displaced fractures were treated within 5 days of injury by primary internal fixation using a single lag screw inserted through a dorsal approach. One patient with a displaced, but comminuted, fragment was initially treated conservatively. When the fragment itself had consolidated, internal fixation was carried out at 6 weeks (Fig 3). Operative technique (Figs 4a–c) A dorsal incision is made over the metacarpophalangeal joint. The extensor tendon is exposed and split in the line of its fibres. The capsule is opened and the fragment, with the collateral ligament attached to it, identified. The fragment and fracture site are gently curetted and the fracture reduced. With the fragment held in place with a skin hook, a 1.2 or 1.7 self-tapping titanium screw is inserted. When possible, a lag screw is used. In some patients, a single hole of a plate was used as a washer to spread the load of the head of the screw (Fig 5). The capsule and extensor tendon are each repaired with a
PATIENTS AND METHODS Over a 6 year period, a total of 19 patients with collateral ligament avulsion fractures from the metacarpal heads of the fingers presented to the hand trauma service in Cardiff. There were 12 men and seven women, with a mean age of 24.6 (range, 13–50) years. Nine fractures occurred whilst participating in contact sport. Of these, six occurred during rugby football, two during 537
ARTICLE IN PRESS 538
THE JOURNAL OF HAND SURGERY VOL. 31B No. 5 OCTOBER
Fig 1 Pre- and postoperative radiographs showing a displaced avulsion fracture from the ulnar side of the head of the third metacarpal with subsequent internal fixation. There is a concomitant fracture of the index metacarpal neck.
Table 1—Fracture distribution Injured finger
Index Middle Ring Small
Side of metacarpal head Radial
Ulnar
1 3 2 1
10 2 0 0
Fig 2 Pre- and postoperative radiographs showing avulsion fractures from both the base of the proximal phalanx and from the head of the adjacent metacarpal head with subsequent internal fixation.
running 5/0 Vicryl stitch prior to closure of the skin and the hand is immobilised in a palmar plaster of Paris slab in the position of comfort. Patients are seen on the fourth postoperative day and referred to the Hand Therapists. A resting thermoplastic splint is fashioned to protect the metacarpophalangeal
2006
Fig 3 A displaced, comminuted avulsion fracture, initially treated conservatively, then by delayed fixation after consolidation of the fragment.
joint, which rests in 70 degrees of flexion. The patients are instructed to regularly remove the splint when in a safe environment and to gently mobilise the joint within levels of comfort. The splint is discarded after 1 month. The interphalangeal joints are not protected and vigorous mobilisation of these is commenced immediately. Radiographs are taken at the first postoperative visit. A radiograph is taken at 3 months to confirm union of the fracture and the range of movement of the metacarpophalangeal and interphalangeal joints noted. Patients were discharged at 3 months if union was adjudged to have occurred and if a full range of movement had been achieved. Patients with ongoing stiffness continued with therapy and were observed for longer (range 5–12 months). Seven patients with undisplaced fractures were initially treated conservatively by ‘‘neighbour strapping’’ to the adjacent digit. The interphalangeal joints were vigorously mobilised, but the metacarpophalangeal joints were protected with a thermoplastic splint which allowed extension but prevented flexion to more than 70 degrees. This continued for 4 weeks after which the entire hand was mobilised vigorously. The fractures were monitored with radiographs taken at 2 week intervals for 6 weeks. The patients were then assessed clinically and radiographically at 3 months after the initial injury. If union had not occurred by then, internal fixation was carried out. The range of movement of the metacarpophalangeal and interphalangeal joints was noted at each review.
RESULTS Of the 11 patients treated with primary internal fixation within 5 days of injury, seven patients achieved a full range of movement of the injured digit with full
ARTICLE IN PRESS COLLATERAL LIGAMENT AVULSION FRACTURES
539
Fig 5 Fixation of particularly fragile fragments can be facilitated by using a single hole of a plate as a washer.
Fig 4 Technique of fracture fixation: dorsal approach splitting the extensor tendon and the capsule longitudinally. The fracture is exposed (a) and gently curetted with a Mitchell’s trimmer (b). The fracture is reduced with a skin hook (c) prior to fixation.
extension and flexion of the finger by 3 months after surgery. Although all patients regained a full range of movement of the interphalangeal joints, four patients had persisting mean loss of flexion of the injured metacarpophalangeal joint of 34 (range 20–40) degrees. Three of these had concomitant fractures and two of them developed signs of mild CRPS Type 1 (Reflex Sympathetic Dystrophy, Algodystrophy), which subse-
quently resolved. Another patient, whose initial fracture was markedly displaced, had a manipulation under general anaesthetic of the metacarpophalangeal joint at 10 months. This improved the range of flexion from 50 to 90 degrees, when measured 3 weeks later. He defaulted from subsequent review. All the fractures treated by immediate internal fixation had united at 3 months. One patient had a displaced fragment that was thought to be too comminuted for safe internal fixation. She was, therefore, initially treated conservatively and fixation carried out at 8 weeks when the fragment had consolidated (Fig 3). Review at 6 weeks after surgery found a loss of metacarpophalangeal joint flexion of 20 degrees but no functional problems. She defaulted from subsequent review. Of the seven patients treated conservatively, four achieved union of the fracture with a full range of movement by 6 weeks. Three remained symptomatic at 3 months with pain on abduction of the fully flexed metacarpophalangeal joint and no radiological signs of union. These patients underwent delayed internal fixation using the technique described previously. Bone graft was not used. At operation, the fracture fragments were all found to be mobile, with no sign of union, bony or fibrous. The collateral ligaments were all completely avulsed and attached to the fragments. Secondary surgery was not appreciably more difficult than primary fixation. On final review at 3 months after surgery, all three had obtained a full range of movement of the metacarpophalangeal and interphalangeal joints with radiological union (Fig 6).
DISCUSSION These injuries occur as the result of the application of an adduction or abduction force to a digit, usually during a
ARTICLE IN PRESS 540
Fig 6 Non-union of an avulsion fracture at 3 months with subsequent fixation and union.
fall onto the outstretched hand. The metacarpophalangeal joint is rendered unstable by detachment of the collateral ligament. If union does not occur, this instability persists. These injuries are less common than similar injuries whereby the collateral ligament is avulsed with a fragment of bone from the base of the proximal phalanx. A study of proximal phalangeal avulsions done over a similar period of time of 6 years in the same population yielded 33 patients (Shewring and Thomas, 1993). This study, also performed over 6 years, yielded 19 patients. In both series, the collateral ligament in every case was found at operation to have been completely avulsed and was attached to the fragment. Although both injuries, presumably, occur by a similar mechanism, it is unclear as to why some patients should sustain avulsion fractures from the base of the phalanx and others from the metacarpal head. Consideration of the role of the palmar plate in the aetiology of these injuries may provide an explanation. The palmar plate shares an attachment with the collateral ligament to the base of the proximal phalanx at the point where the avulsion occurs. If the injury occurs with the joint in extension, the collateral ligament would be slack but the palmar plate is taut and would contribute to the avulsion force, thereby predisposing to avulsion from the base of the proximal phalanx. With the metacarpophalangeal joint in flexion, the palmar plate would not be under tension but the collateral ligament would, so avulsion of the collateral ligament from the weaker, cancellous non-articular bone of the metacarpal head would occur. Since the fingers are more likely to be extended during a fall onto the outstretched hand, this may also explain the relative frequencies of the two injuries. In this series, avulsion of the collateral ligament from the metacarpal head on the ulnar side of the index finger accounted for more than half of the injuries (Table 1). This is, presumably, because the index is a border finger and more prone to abduction injury. This finding is
THE JOURNAL OF HAND SURGERY VOL. 31B No. 5 OCTOBER
2006
similar to that in our previous study (Shewring and Thomas, 1993) in which all of the avulsion fractures affected the ulnar side of the index or the radial side of the little finger. Previous literature on these fractures is sparse. Schubiner and Mass (1989) described ten cases of collateral ligament rupture of various types affecting the metacarpophalangeal joints of the fingers treated by surgical repair. Six cases had rupture of the ligament at the distal insertion. Two had avulsion fractures from the metacarpal head. All but two had flecks of bone attached to the torn ligament. The site of rupture in two patients was not described. The ligaments were reattached with a ‘‘pull-out’’ suture but the surgical approaches were not described. Pain-free stable joints were obtained and all but two patients regained full movement. Gee and Pho (1982) described a single case of an avulsion fracture of the proximal attachment of the radial collateral ligament of the fifth metacarpophalangeal joint. This led to gross instability, requiring internal fixation through a dorsal approach with a wire loop at 1 week. The patient was asymptomatic and had a full range of movement at 3 months. This paucity of information about these injuries prompted us to carry out this study, in the hope of providing a satisfactory means of management of the problem. Eleven patients with displaced fractures were treated by primary internal fixation. The rationale for this was that the displacement made these fractures less likely to unite and that they were less stable. At 3 months after surgery, all of these fractures had united. Seven obtained a full range of movement of the injured digit, although this took up to 3 months. This was longer than in patients with avulsion fractures from the proximal phalangeal base (Shewring and Thomas, 1993), in whom the time to achieve a full range of movement was 3 weeks. Although internal fixation facilitated early mobilisation, this was progressed cautiously because of the fragility of the fracture fragment and some protection was continued for 1 month. Three of those that did not achieve full movement of the metacarpophalangeal joint had concomitant fractures within the hand and one had a markedly displaced fracture. The poorer results in these four patients may reflect a more severe initial injury to the hand. With regard to the method of fixation, some authors have preferred tension band wiring to internal fixation for the treatment of various types of avulsion fractures in the hand, on the grounds that internal fixation may result in comminution of the fragment and excessive handling of the soft tissues (Bischoff et al., 1994; Jupiter and Sheppard, 1987) . Both of these series included a wide variety of avulsion fractures including bony mallet fractures, bony gamekeeper’s injuries, lateral phalangeal base fractures and fractures around the proximal interphalangeal joint. Any difference in outcome between the various types of fracture was not described.
ARTICLE IN PRESS COLLATERAL LIGAMENT AVULSION FRACTURES
Gee and Pho (1982) used a wire loop fixation. Technology has advanced, so that more refined screws, which are smaller, easier to insert and self-tapping, are now available. Insertion of a simple lag screw is quicker and less technically demanding than a tension band wire fixation. Furthermore, we feel that forcing a K-wire through an undrilled small fragment is more likely to cause comminution. We found that using a single hole of a plate as a washer improved the hold of the screwhead on more fragile fragments. Many of the fragments were very fragile and, in one of our cases, the avulsion fragment from the metacarpal head was comminuted. This is in contrast to collateral ligament avulsion fractures from the base of the proximal phalanges (Shewring and Thomas, 1993), none of which were comminuted. This may be because fractures avulsed from the phalangeal base have a portion of articular surface on the fragment. This is less likely to comminute than the fragments avulsed from the metacarpal head, which are entirely cancellous and non-articular. In our series, seven undisplaced collateral ligament avulsion fractures from the metacarpal heads of the fingers were initially treated conservatively and four of these united. Surgery was necessary for the remaining three and, although this was performed at 3 months, the delay did not appear to affect the outcome. It would seem reasonable, therefore, that undisplaced fractures be given a trial period of conservative treatment, which may avoid the need for surgery. Although the non-union rate was not as high as for the avulsion fractures from the base of the proximal phalanges in our previous study (Shewring and Thomas, 1993), this was still a problem for some patients after conservative management, even with undisplaced fractures. As with avulsion fractures from the base of the proximal phalanges, the high rate of non-union may be explained by the repeated pull of the collateral ligament on the fragment during daily movements of the joint. Flexion of the metacarpophalangeal
541
joint tightens the collateral ligament and may lead to displacement of the fragment, compromising union. This precludes early vigorous mobilisation of the metacarpophalangeal joint if conservative treatment is used. When the fractures initially treated conservatively were later secured with a lag screw, union occurred without the need for bone graft. It is our view that this indicates that these fractures will unite readily if adequate stability is provided. If patients are warned of the possibility of non-union and the consequent need for surgery, then some may prefer immediate surgery in the hope of avoiding a more extended period of incapacity. References Bischoff R, Buechler U, De Roche R, Jupiter J (1994). Clinical results of tension band fixation of avulsion fractures in the hand. Journal of Hand Surgery, 19A: 1019–1026. Gee TC, Pho RWH (1982). Avulsion-fracture at the proximal attachment of the radial collateral ligament of the fifth metacarpophalangeal joint: a case report. Journal of Hand Surgery, 7A: 526–527. Jupiter JB, Sheppard JE (1987). Tension wire fixation of avulsion fractures in the hand. Clinical Orthopaedics and Related Research, 214: 113–120. Schubiner JM, Mass DP (1989). Operation for collateral ligament ruptures of the metacarpophalangeal joints of the fingers. Journal of Bone and Joint Surgery, 71B: 388–389. Shewring DJ, Thomas RH (1993). Avulsion fractures from the base of the proximal phalanges of the fingers. Journal of Hand Surgery, 28B: 10–14.
Received: 16 August 2005 Accepted after revision: 8 May 2006 Mr D.J. Shewring, Consultant Hand Surgeon, Department of Trauma and Orthopaedic Surgery, University Hospital of Wales, Heath Park, Cardiff CF14 4XW, UK E-mail: [email protected]
r 2006 The British Society for Surgery of the Hand. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhsb.2006.05.009 available online at http://www.sciencedirect.com
Nineteen patients with collateral ligament avulsion fractures from the metacarpal heads of the fingers were treated during a 6 year period. Seven undisplaced fractures were initially treated conservatively. Four united with full movement. Three required internal fixation for symptomatic non-union and healed without problems. Eleven patients with displaced fractures were treated by primary internal fixation using a single lag screw through a dorsal approach. Seven of these achieved a full range of movement of the injured digit by 3 months. Four patients failed to regain full flexion of the metacarpophalangeal joint. One patient with a displaced and comminuted fracture was treated with internal fixation at 8 weeks when the fragment had consolidated. As with similar fractures from the proximal phalangeal bases, these fractures are prone to non-union when treated conservatively, even when undisplaced. If fixation becomes necessary, the delay of a trial of conservative treatment does not appear to affect the outcome. Internal fixation of displaced fractures allows gentle mobilisation and facilitates union. Journal of Hand Surgery (British and European Volume, 2006) 31B: 5: 537–541 Keywords: avulsion fracture, metacarpophalangeal joint, fingers
Avulsion fractures affecting the metacarpophalangeal joints occur through the pull of the collateral ligaments and may result in avulsion of a fragment from either the proximal phalangeal base or from the metacarpal head. Avulsion fractures from the base of the proximal phalanges of the fingers have previously been found to have a high rate of non-union when treated conservatively but excellent results when treated by internal fixation through a palmar approach (Shewring and Thomas, 1993). Avulsion fractures from the metacarpal heads (Fig 1) are rarer and there is scant reference to them in the literature. The only previous studies we have found are a case report (Gee and Pho, 1982) in which an avulsion fracture from the radial side of the fifth metacarpal head was treated with internal fixation and a study by Bischoff et al. (1994) in which two of these fractures were grouped along with 100 avulsion fractures of various types. The purpose of this prospective study was to investigate the outcomes of conservative and operative treatment of collateral ligament avulsion fractures from the metacarpal heads of the fingers and to describe the technique of surgical fixation.
soccer and one whilst playing ice hockey. The remainder occurred as the result of a simple fall onto the outstretched hand; five of these ten patients having tripped when running upstairs. The ulnar side of the index finger was most commonly affected, accounting for more than half of the injuries (Table 1). Three patients had concomitant fractures. There was an undisplaced spiral fracture of the shaft of the proximal phalanx in the same digit in one patient and a fracture of the metacarpal neck of an adjacent digit in another. One patient had an avulsion from the base of the proximal phalanx in an adjacent digit, which was also internally fixed through a separate palmar approach (Fig 2). Eleven patients with displaced fractures were treated within 5 days of injury by primary internal fixation using a single lag screw inserted through a dorsal approach. One patient with a displaced, but comminuted, fragment was initially treated conservatively. When the fragment itself had consolidated, internal fixation was carried out at 6 weeks (Fig 3). Operative technique (Figs 4a–c) A dorsal incision is made over the metacarpophalangeal joint. The extensor tendon is exposed and split in the line of its fibres. The capsule is opened and the fragment, with the collateral ligament attached to it, identified. The fragment and fracture site are gently curetted and the fracture reduced. With the fragment held in place with a skin hook, a 1.2 or 1.7 self-tapping titanium screw is inserted. When possible, a lag screw is used. In some patients, a single hole of a plate was used as a washer to spread the load of the head of the screw (Fig 5). The capsule and extensor tendon are each repaired with a
PATIENTS AND METHODS Over a 6 year period, a total of 19 patients with collateral ligament avulsion fractures from the metacarpal heads of the fingers presented to the hand trauma service in Cardiff. There were 12 men and seven women, with a mean age of 24.6 (range, 13–50) years. Nine fractures occurred whilst participating in contact sport. Of these, six occurred during rugby football, two during 537
ARTICLE IN PRESS 538
THE JOURNAL OF HAND SURGERY VOL. 31B No. 5 OCTOBER
Fig 1 Pre- and postoperative radiographs showing a displaced avulsion fracture from the ulnar side of the head of the third metacarpal with subsequent internal fixation. There is a concomitant fracture of the index metacarpal neck.
Table 1—Fracture distribution Injured finger
Index Middle Ring Small
Side of metacarpal head Radial
Ulnar
1 3 2 1
10 2 0 0
Fig 2 Pre- and postoperative radiographs showing avulsion fractures from both the base of the proximal phalanx and from the head of the adjacent metacarpal head with subsequent internal fixation.
running 5/0 Vicryl stitch prior to closure of the skin and the hand is immobilised in a palmar plaster of Paris slab in the position of comfort. Patients are seen on the fourth postoperative day and referred to the Hand Therapists. A resting thermoplastic splint is fashioned to protect the metacarpophalangeal
2006
Fig 3 A displaced, comminuted avulsion fracture, initially treated conservatively, then by delayed fixation after consolidation of the fragment.
joint, which rests in 70 degrees of flexion. The patients are instructed to regularly remove the splint when in a safe environment and to gently mobilise the joint within levels of comfort. The splint is discarded after 1 month. The interphalangeal joints are not protected and vigorous mobilisation of these is commenced immediately. Radiographs are taken at the first postoperative visit. A radiograph is taken at 3 months to confirm union of the fracture and the range of movement of the metacarpophalangeal and interphalangeal joints noted. Patients were discharged at 3 months if union was adjudged to have occurred and if a full range of movement had been achieved. Patients with ongoing stiffness continued with therapy and were observed for longer (range 5–12 months). Seven patients with undisplaced fractures were initially treated conservatively by ‘‘neighbour strapping’’ to the adjacent digit. The interphalangeal joints were vigorously mobilised, but the metacarpophalangeal joints were protected with a thermoplastic splint which allowed extension but prevented flexion to more than 70 degrees. This continued for 4 weeks after which the entire hand was mobilised vigorously. The fractures were monitored with radiographs taken at 2 week intervals for 6 weeks. The patients were then assessed clinically and radiographically at 3 months after the initial injury. If union had not occurred by then, internal fixation was carried out. The range of movement of the metacarpophalangeal and interphalangeal joints was noted at each review.
RESULTS Of the 11 patients treated with primary internal fixation within 5 days of injury, seven patients achieved a full range of movement of the injured digit with full
ARTICLE IN PRESS COLLATERAL LIGAMENT AVULSION FRACTURES
539
Fig 5 Fixation of particularly fragile fragments can be facilitated by using a single hole of a plate as a washer.
Fig 4 Technique of fracture fixation: dorsal approach splitting the extensor tendon and the capsule longitudinally. The fracture is exposed (a) and gently curetted with a Mitchell’s trimmer (b). The fracture is reduced with a skin hook (c) prior to fixation.
extension and flexion of the finger by 3 months after surgery. Although all patients regained a full range of movement of the interphalangeal joints, four patients had persisting mean loss of flexion of the injured metacarpophalangeal joint of 34 (range 20–40) degrees. Three of these had concomitant fractures and two of them developed signs of mild CRPS Type 1 (Reflex Sympathetic Dystrophy, Algodystrophy), which subse-
quently resolved. Another patient, whose initial fracture was markedly displaced, had a manipulation under general anaesthetic of the metacarpophalangeal joint at 10 months. This improved the range of flexion from 50 to 90 degrees, when measured 3 weeks later. He defaulted from subsequent review. All the fractures treated by immediate internal fixation had united at 3 months. One patient had a displaced fragment that was thought to be too comminuted for safe internal fixation. She was, therefore, initially treated conservatively and fixation carried out at 8 weeks when the fragment had consolidated (Fig 3). Review at 6 weeks after surgery found a loss of metacarpophalangeal joint flexion of 20 degrees but no functional problems. She defaulted from subsequent review. Of the seven patients treated conservatively, four achieved union of the fracture with a full range of movement by 6 weeks. Three remained symptomatic at 3 months with pain on abduction of the fully flexed metacarpophalangeal joint and no radiological signs of union. These patients underwent delayed internal fixation using the technique described previously. Bone graft was not used. At operation, the fracture fragments were all found to be mobile, with no sign of union, bony or fibrous. The collateral ligaments were all completely avulsed and attached to the fragments. Secondary surgery was not appreciably more difficult than primary fixation. On final review at 3 months after surgery, all three had obtained a full range of movement of the metacarpophalangeal and interphalangeal joints with radiological union (Fig 6).
DISCUSSION These injuries occur as the result of the application of an adduction or abduction force to a digit, usually during a
ARTICLE IN PRESS 540
Fig 6 Non-union of an avulsion fracture at 3 months with subsequent fixation and union.
fall onto the outstretched hand. The metacarpophalangeal joint is rendered unstable by detachment of the collateral ligament. If union does not occur, this instability persists. These injuries are less common than similar injuries whereby the collateral ligament is avulsed with a fragment of bone from the base of the proximal phalanx. A study of proximal phalangeal avulsions done over a similar period of time of 6 years in the same population yielded 33 patients (Shewring and Thomas, 1993). This study, also performed over 6 years, yielded 19 patients. In both series, the collateral ligament in every case was found at operation to have been completely avulsed and was attached to the fragment. Although both injuries, presumably, occur by a similar mechanism, it is unclear as to why some patients should sustain avulsion fractures from the base of the phalanx and others from the metacarpal head. Consideration of the role of the palmar plate in the aetiology of these injuries may provide an explanation. The palmar plate shares an attachment with the collateral ligament to the base of the proximal phalanx at the point where the avulsion occurs. If the injury occurs with the joint in extension, the collateral ligament would be slack but the palmar plate is taut and would contribute to the avulsion force, thereby predisposing to avulsion from the base of the proximal phalanx. With the metacarpophalangeal joint in flexion, the palmar plate would not be under tension but the collateral ligament would, so avulsion of the collateral ligament from the weaker, cancellous non-articular bone of the metacarpal head would occur. Since the fingers are more likely to be extended during a fall onto the outstretched hand, this may also explain the relative frequencies of the two injuries. In this series, avulsion of the collateral ligament from the metacarpal head on the ulnar side of the index finger accounted for more than half of the injuries (Table 1). This is, presumably, because the index is a border finger and more prone to abduction injury. This finding is
THE JOURNAL OF HAND SURGERY VOL. 31B No. 5 OCTOBER
2006
similar to that in our previous study (Shewring and Thomas, 1993) in which all of the avulsion fractures affected the ulnar side of the index or the radial side of the little finger. Previous literature on these fractures is sparse. Schubiner and Mass (1989) described ten cases of collateral ligament rupture of various types affecting the metacarpophalangeal joints of the fingers treated by surgical repair. Six cases had rupture of the ligament at the distal insertion. Two had avulsion fractures from the metacarpal head. All but two had flecks of bone attached to the torn ligament. The site of rupture in two patients was not described. The ligaments were reattached with a ‘‘pull-out’’ suture but the surgical approaches were not described. Pain-free stable joints were obtained and all but two patients regained full movement. Gee and Pho (1982) described a single case of an avulsion fracture of the proximal attachment of the radial collateral ligament of the fifth metacarpophalangeal joint. This led to gross instability, requiring internal fixation through a dorsal approach with a wire loop at 1 week. The patient was asymptomatic and had a full range of movement at 3 months. This paucity of information about these injuries prompted us to carry out this study, in the hope of providing a satisfactory means of management of the problem. Eleven patients with displaced fractures were treated by primary internal fixation. The rationale for this was that the displacement made these fractures less likely to unite and that they were less stable. At 3 months after surgery, all of these fractures had united. Seven obtained a full range of movement of the injured digit, although this took up to 3 months. This was longer than in patients with avulsion fractures from the proximal phalangeal base (Shewring and Thomas, 1993), in whom the time to achieve a full range of movement was 3 weeks. Although internal fixation facilitated early mobilisation, this was progressed cautiously because of the fragility of the fracture fragment and some protection was continued for 1 month. Three of those that did not achieve full movement of the metacarpophalangeal joint had concomitant fractures within the hand and one had a markedly displaced fracture. The poorer results in these four patients may reflect a more severe initial injury to the hand. With regard to the method of fixation, some authors have preferred tension band wiring to internal fixation for the treatment of various types of avulsion fractures in the hand, on the grounds that internal fixation may result in comminution of the fragment and excessive handling of the soft tissues (Bischoff et al., 1994; Jupiter and Sheppard, 1987) . Both of these series included a wide variety of avulsion fractures including bony mallet fractures, bony gamekeeper’s injuries, lateral phalangeal base fractures and fractures around the proximal interphalangeal joint. Any difference in outcome between the various types of fracture was not described.
ARTICLE IN PRESS COLLATERAL LIGAMENT AVULSION FRACTURES
Gee and Pho (1982) used a wire loop fixation. Technology has advanced, so that more refined screws, which are smaller, easier to insert and self-tapping, are now available. Insertion of a simple lag screw is quicker and less technically demanding than a tension band wire fixation. Furthermore, we feel that forcing a K-wire through an undrilled small fragment is more likely to cause comminution. We found that using a single hole of a plate as a washer improved the hold of the screwhead on more fragile fragments. Many of the fragments were very fragile and, in one of our cases, the avulsion fragment from the metacarpal head was comminuted. This is in contrast to collateral ligament avulsion fractures from the base of the proximal phalanges (Shewring and Thomas, 1993), none of which were comminuted. This may be because fractures avulsed from the phalangeal base have a portion of articular surface on the fragment. This is less likely to comminute than the fragments avulsed from the metacarpal head, which are entirely cancellous and non-articular. In our series, seven undisplaced collateral ligament avulsion fractures from the metacarpal heads of the fingers were initially treated conservatively and four of these united. Surgery was necessary for the remaining three and, although this was performed at 3 months, the delay did not appear to affect the outcome. It would seem reasonable, therefore, that undisplaced fractures be given a trial period of conservative treatment, which may avoid the need for surgery. Although the non-union rate was not as high as for the avulsion fractures from the base of the proximal phalanges in our previous study (Shewring and Thomas, 1993), this was still a problem for some patients after conservative management, even with undisplaced fractures. As with avulsion fractures from the base of the proximal phalanges, the high rate of non-union may be explained by the repeated pull of the collateral ligament on the fragment during daily movements of the joint. Flexion of the metacarpophalangeal
541
joint tightens the collateral ligament and may lead to displacement of the fragment, compromising union. This precludes early vigorous mobilisation of the metacarpophalangeal joint if conservative treatment is used. When the fractures initially treated conservatively were later secured with a lag screw, union occurred without the need for bone graft. It is our view that this indicates that these fractures will unite readily if adequate stability is provided. If patients are warned of the possibility of non-union and the consequent need for surgery, then some may prefer immediate surgery in the hope of avoiding a more extended period of incapacity. References Bischoff R, Buechler U, De Roche R, Jupiter J (1994). Clinical results of tension band fixation of avulsion fractures in the hand. Journal of Hand Surgery, 19A: 1019–1026. Gee TC, Pho RWH (1982). Avulsion-fracture at the proximal attachment of the radial collateral ligament of the fifth metacarpophalangeal joint: a case report. Journal of Hand Surgery, 7A: 526–527. Jupiter JB, Sheppard JE (1987). Tension wire fixation of avulsion fractures in the hand. Clinical Orthopaedics and Related Research, 214: 113–120. Schubiner JM, Mass DP (1989). Operation for collateral ligament ruptures of the metacarpophalangeal joints of the fingers. Journal of Bone and Joint Surgery, 71B: 388–389. Shewring DJ, Thomas RH (1993). Avulsion fractures from the base of the proximal phalanges of the fingers. Journal of Hand Surgery, 28B: 10–14.
Received: 16 August 2005 Accepted after revision: 8 May 2006 Mr D.J. Shewring, Consultant Hand Surgeon, Department of Trauma and Orthopaedic Surgery, University Hospital of Wales, Heath Park, Cardiff CF14 4XW, UK E-mail: [email protected]
r 2006 The British Society for Surgery of the Hand. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhsb.2006.05.009 available online at http://www.sciencedirect.com