Fractures of the metacarpals account for 18% and the phalanges for 23% of all below-elbow fractures in the United States, peaking in the third decade for men and second decade for women.1 The two most typical mechanisms of injury are accidental fall and direct blow; directblow injuries often result from one person against another person or object.1 Phalangeal fractures are nearly twice as common as metacarpal fractures. Most phalangeal fractures occur in the proximal phalanx, followed by the distal phalanx, then the middle phalanx.2 The small finger ray accounts for 38% of all hand fractures, with a relatively even distribution across the remaining four rays.2
Despite the high number of patients undergoing treatment of hand fractures, the literature remains unclear with respect to providing guidance to the clinician for specific phalangeal and metacarpal fractures. There are no well-performed, published studies providing level I or II evidence. Randomized controlled trials are prevented from being performed because of the broad spectrum of variation in fracture patterns and the many associated variables that are thought to affect treatment and outcome.
The forms of treatment discussed below for common fracture patterns in the hand are based on experience gained in a high-volume hand trauma specialty practice. This does not imply that all fractures of a certain pattern must be fixed according to that method of treatment. Details related to more comprehensive treatment of all phalangeal and metacarpal fractures are discussed by the author elsewhere.3,4
Review of the Literature
In a series of 924 hand fractures managed over a period of 10 years, overall results were excellent or good in >90% of thumbs but in only 59% to 76% of fingers. Reasons cited for poor results were comminution and open or multiple fractures.5 High-energy fractures with comminution predispose to stiffness of the proximal interphalangeal (PIP) joint, in particular; Gonzalez et al6 reported an average postoperative range of motion (ROM) of 66° in 27 patients (28 PIP fractures). Intraarticular extension was another variable that conferred a worse prognosis, with total active motion (TAM) of 169° compared with TAM of 213° in those without intra-articular extension.6 In another series of intraarticular fractures in patients following open reduction and internal fixation in the fingers, there were 9 excellent, 8 good, and 6 poor results. This study identified comminution and initial open fracture as the variables leading to the poorest results.7
Generally, the severity of the injury, in terms of energy imparted into the tissues and the associated softtissue lesions, serves as the primary determinant of outcome. Nevertheless, when a fracture is too unstable for nonsurgical management, a decision must be made regarding the method of fracture stabilization. Of the many combinations of internal fixation possible, Kirschner wires (K-wires) and screw-and-plate fixation predominate. Because of uncertainty regarding which method produces a superior outcome, decisions are often made with a goal of minimizing complications. The general concerns with smooth-sided, stainless steel K-wires are loss of reduction; tethering of soft tissues (primarily the extensor mechanism); and pin tract infection, potentially leading to osteomyelitis. Botte et al8 reported a smooth-pin fixation complication rate of 18%, including infection (7%), pin loosening (5%), and nonunion (4%).
Plate fixation has been noted to cause substantial loss of motion because of tendon adherence at the plate's surface as well as the added surgical dissection needed to reduce the fracture and apply the plate. Following plate fixation of phalangeal fractures, Kurzen et al9 reported a complication rate of 52%. These authors included TAM of <180° as a complication (24/64 fractures). The negative impact of tendon adhesions was not as great at the metacarpal level as at the proximal phalangeal level. Page and Stern10 reported that plate fixation yielded TAM of >220° in 76% of metacarpal fractures but in only 11% of phalangeal fractures. In another study, plate fixation produced a complication rate of 31% at the metacarpal level and of 82% at the phalangeal level.11
A specific concern with plate fixation in metacarpal fractures is delayed or nonunion that was found to be higher (30%) in transverse fracture patterns than in those with a broader interface (7%).12 Fixation by lag screw only, without a plate, represents something of a middle ground between K-wires and plates. It requires the added surgical dissection of the open approach but left a far lower hardware profile and, thus, less tendon irritation.
To date, there has been one attempt at a prospective, randomized, controlled study of long oblique and spiral isolated proximal phalanx (P1) fractures comparing percutaneous pinning and lag screws. However, the numbers are too small to provide meaningful conclusions.13 Notably, in 3 of 17 patients, fractures treated by K-wire required formal extensor tenolysis for tethering; this was not the case in any of the 15 patients whose fractures were treated by lag screw. There were, however, 8 malunions of the 15 resulting from lag screw fixation, a high number considering the opportunity for precise reduction and rigid fixation possible with this method.13
Fractures that are inherently stable (ie, limited initial displacement, lowenergy mechanism of injury) do not need surgical treatment; all other fractures should be considered for some form of additional stabilization. Ebinger et al14 reported that the intrinsic plus splinting position, with active PIP motion, achieved full active motion and complete fracture healing by 6 weeks in 44 of 48 displaced proximal phalanx fractures. The standard intrinsic plus splint is dorsally based from the level of the PIP joint to the base of the metacarpals, with the metacarpophalangeal (MCP) joints in full flexion. Immediate PIP joint motion exerts a dynamic tension-band effect with the extrinsic extensor that actually aids in the reduction of P1 fractures.
The general principles of hand fracture rehabilitation include early active joint ROM and tendon gliding using synergistic wrist positions and blocking techniques, including blocking splints.15 Hard splints may be discontinued in favor of side strapping (ie, buddy taping) protection by 3 weeks in P1 and metaphyseal metacarpal fractures and by 4 weeks in diaphyseal metacarpal fractures. At this time, the fracture line is routinely still visible on plain radiographs without evidence yet of periosteal callus. Even at approximately 6 weeks, when the fracture is considered clinically to be fully healed (ie, nontender to pressure on examination), the radiograph usually still reveals the original fracture line.
Preload-and-release tendon acceleration exercises have been described as a method to take advantage of the viscoelastic properties of both tendon and scar tissue in order to achieve differential gliding between the two.16 Motion is restrained by the ipsilateral thumb, while the extensor tendon is brought to maximum isometric tension followed by a rapid release of the joint block. This causes a sudden excursion of that segment of tendon directly overlying the fracture.
Less Frequently Used Treatment Methods
Virtually any method of providing additional stability to a fracture is a potentially valid strategy. Bone grafting alone into the site of impacted P1 base fractures provided “rigid internal fixation,” with reported good results and early return to work for 10 patients (13 fractures).17 Twenty-five patients with avulsion fractures of the base of the proximal phalanx were treated with a single lag screw via a palmar approach; excellent results were reported in all cases, with full ROM achieved by 3 weeks.18 Multiple intramedullary K-wires can be used at both the metacarpal and proximal phalangeal levels.19 Tensionband wiring was reported to achieve an average TAM of 188° in 16 P1 and 10 P2 fractures, with no complications.20 Of 36 metacarpal shaft fractures fixed with interosseous loop wire, 34 achieved full motion with no complications.21 External fixation of closed fractures has been reported to produce TAM of 230° or better in all metacarpal fractures and in 76% of phalangeal fractures.22
Unicondylar Proximal Phalanx Fractures
Unicondylar fracture of the head of the proximal phalanx typically occurs as the result of axial loading with angulation directed toward the side of the small fragment, producing an oblique fracture line through the metaphysis that originates in the intercondylar groove (Figure 1, A). Some fractures involve less than the whole condyle; others take part of the second condyle with the smaller fragment. A variable degree of extension into the diaphysis occurs proximal to the origin of the collateral ligament. The primary treatment goal is achievement of a congruent articular surface that restores coronal plane alignment of the digit and which should mitigate posttraumatic arthritis, necessitating stable fixation to resist axial shear forces.
Early Presentation and Easy Reduction
If the patient presents early, closed reduction often may achieve the goal of a congruent articular surface. If fixation is to be with smoothsided K-wires only, two wires that diverge are required to prevent slippage of the smaller fragment along the surface of one wire alone (Figure 1, C). The standard configuration is with the first wire placed transverse and subchondral, in order to level the articular surface; the second wire is placed obliquely from the condylar fragment into the opposite diaphyseal cortex to buttress the fragment against the opposite condyle.
Late Presentation or Difficult Reduction
With late presentation or a difficult reduction, open treatment may be required. Caution is required to avoid excessive soft-tissue dissection of the smaller fragment, especially the collateral ligament origin, which covers most of the fragment's cortex, to avoid osteonecrosis. Excessive dissection risks devascularization of the fragment, whose blood supply is contained in the collateral ligament. Fixation may be with either K-wires, as described, or with 1.3-mm lag screws. With smaller fragments, the collateral ligament origin interferes with the normal process of countersinking a lag screw. With an extended diaphyseal spike on the smaller fragment, lag screw fixation becomes an increasingly attractive option (Figure 1, B). An alternative method involves cannulated, percutaneous, variable pitch, headless screw fixation. Geissler23 reported that such fixation achieved PIP joint ROM of 5° to 85° in 18 patients.
Bicondylar Proximal Phalanx Fractures
Although a series of K-wires may be used in treating a bicondylar proximal phalanx fracture (usually one transverse between the two condyles and an oblique wire from each condyle to the opposite proximal cortex), more commonly, plate fixation serves to permit immediate mobilization of the PIP joint. When there are three large fragments (ie, two condyles and the shaft) without comminution, a 1.5-mm straight plate applied laterally will function as a load-sharing device, along with inherent fracture geometry.
With substantial comminution, the plate must be a load-bearing device; here the 1.5-mm condylar blade plate is preferred (Figure 2). Meticulous precontouring of the plate is required before implanting the blade because the surface of the shaft does not naturally match the plate as it comes from the manufacturer. However, repeated removal and reinsertion of the blade into the condylar fragments can create further comminution of the fracture to the point of rendering the fracture irreducible or not acceptable to plate fixation. Dorsal placement of a plate at this distal location under the extensor zone IV to zone III junction (Figure 3), right at the margin of the PIP joint, is even more functionally damaging than is a dorsal plate on the midshaft. This placement should be avoided if at all possible.
A rare bicondylar fracture pattern is the triplane fracture, which is difficult to recognize and technically challenging to treat. However, by virtue of the pattern, the triplane fracture can be stabilized with 1.5-mm lag screws only rather than a plate.24
Transverse Proximal Phalanx Neck Fractures
In treating a transverse proximal phalanx neck fracture, retrograde pinning through the small fragment and down the shaft can provide appropriate early fixation and fracture stability. Long-term PIP motion of 90° or better has been reported.25 Passage of a 0.045-inch K-wire retrograde through the PIP has risks, including pin tract infection that can lead directly to septic arthritis. Such pin positioning demands either more meticulous pin care than normal or cutting of the pin below the surface of the skin (Figure 4). PIP joint motion rehabilitation while the pin is in place creates a small longitudinal rent in the extensor mechanism, but this does not pose any short-term or long-term functional detriment. Far more detrimental would be failure to initiate early rehabilitation. In most patients, the pin should be removed by 3 weeks, and not later than 4 weeks, unless there are extenuating circumstances.
Transverse (Short Oblique) Proximal Phalanx Shaft Fractures
A transverse shaft fracture displaces into an apex palmar angulated position under the pull of the central slip on the distal fragment and the interossei insertions at the base of P1 (Figure 5, A). If allowed to heal in apex palmar malunion, the predicted corresponding extensor lags are for a 10° lag at 16° of angular deformity, a 24° lag at 27° of deformity, and a 66° lag at 46° of deformity.26 When the PIP joint is flexed, the extensor mechanism as a whole functions as a tension band to help reduce the fracture if the hand is positioned in an intrinsic plus splint. Collateral ligament, capsule, and intrinsic muscle attachments render transverse fractures in the proximal 6 to 9 mm of the P1 base more stable than fractures located distally.27 These fractures often can be well maintained in an intrinsic plus splint alone. At the least, these fractures reduce the pin requirements from two K-wires for rotational control to one K-wire.
Most transverse or short oblique P1 fractures without comminution are best stabilized by two 0.045-inch K-wires placed longitudinally through the fully flexed MCP joint (Figure 5, B and C). A single wire alone risks rotational malunion, but some fracture patterns may provide inherent rotational stability that would allow use of one wire for angular control. In a series of 12 transarticular intramedullary pinnings of extra-articular P1 fractures, the average TAM was 265°, with only one flexion contracture and one rotational deformity.28 Even with the minimal tissue trauma of percutaneous pins removed at 3 weeks and immediate PIP rehabilitation, avoiding PIP extensor lag can be difficult. In a series of 35 transmetacarpal longitudinal pinnings of unstable proximal phalangeal fractures, 32% developed a PIP joint flexion contracture averaging 18°, and 4 resulted in rotational deformity.29
Spiral (Long Oblique) Proximal Phalanx Shaft Fractures
Extensor lag at the PIP joint is the result of zone IV tendon adherence at the fracture site and may be worsened by apex palmar malunion as well as by the effects of shortening at the fracture site, which creates slack in the extensor mechanism. Extensor lag is predicted to be 12° for each millimeter of shortening, with the concept of an extensor tendon reserve of 2 to 6 mm.26 Spiral and long oblique P1 fractures tend to angulate less than do transverse and short oblique patterns. Also, spiral and long oblique P1 fractures more often shorten along the slope of the fracture, with varying degrees of axial rotational deformity. As long as the angular deformity is corrected and maintained in a transverse fracture, there is considerable inherent stability provided by the fracture fragment contact to resist the forces of rehabilitation.
This is not the case with long sloping fractures that offer very little help from fracture fragment contact unless significant interfragmentary compression can be achieved. Lag screws (1.3 mm is preferred to 1.5 mm) can achieve stability through interfragmentary compression in a two-part spiral or long oblique fracture that rivals the stability of an intact bone (Figure 6). This capability, combined with the need to achieve precise correction of rotation in spiral fractures, makes open reduction and lag screw fixation the treatment of choice for the spiral P1 shaft fracture. Although rotation is not as much of an issue in long oblique fractures, the benefits of open lag screw fixation still hold true for this pattern. An alternative approach is to attempt to place the lag screws percutaneously; however, there is a low likelihood of achieving excellent interfragmentary compression, which then causes the screws to function in a load-bearing rather than a load-sharing capacity.
The surgical approach is midlateral without dissecting between tissue planes. A single sleeve of tissue, including periosteum, the extensor tendon mechanism, and skin, is raised to expose the fracture site. Clot and early callus are removed from the fracture edges with a finetipped curet to facilitate precise interdigitation along the fracture line. Firm compression must be held between the fragments while placing the lag screws. Screws should be spaced apart from each other and the fracture edges by at least a full screw head diameter, with careful countersinking done to distribute the stress of the screw's head over as broad an area as possible. In addition to a precision fit of the fracture line, the surgeon must clinically check the rotation of the finger (ie, fracture). A standard intrinsic plus splint, as described above, can be used between exercise sessions that include preload-and-release rapid extensor tendon activation.
Comminuted Proximal Phalanx Fractures
Interfragmentary lag screws and percutaneous K-wire fixation both depend on the inherent stability offered by fracture fragment contact. In the presence of substantial comminution, this inherent stability is lost, and the fixation materials need to function in a load-bearing capacity. This usually requires the use of a plate (Figure 7). At the P1 level, the concern with plates is the added potential for extensor tendon adherence to hardware exposed on the surface of the bone. For this reason, dorsal plates are avoided at all costs, which is consistent with the inferior biomechanical stability associated with the dorsal plate position. Biomechanical studies demonstrate that the optimal plate position is midlateral.30-32
The single-tissue sleeve approach described above should be used. Alternatively, the unilateral resection of the intrinsic tendon may mitigate subsequent interference with extensor tendon rehabilitation and PIP joint motion because of plate contact.33 Meticulous plate contouring is critically important, as is the need for clinical assessment of rotational angular alignment during the stages of plate application. High-energy comminuted fractures requiring plate fixation cannot be expected to achieve functional results equivalent to simpler P1 fracture patterns. Only 6 of 16 patients with comminuted phalangeal fractures and associated softtissue injuries achieved >180° of TAM.34
Metacarpal Head Fractures
Metacarpal head fractures are rare. They are best treated by lag screws with deep countersinking of the screw head, especially when it must pass through hyaline cartilage (Figure 8). Careful handling is required to achieve fracture healing and prevent osteonecrosis.
Metacarpal Neck and Transverse (Short Oblique) Metacarpal Shaft Fractures
Metacarpal neck fractures typically angulate apex dorsal; the small finger also usually supinates. The tolerable limit of angulation that does not alter function continues to be debated, although safe angulation from the index through the small finger are 10°, 10°, 20°, and 30°, respectively. Beyond 30° of apex dorsal angulation in small-finger metacarpal neck fractures, there was a rapid decay in efficiency of the flexor tendon system, including tendon excursion, load, and work requirements.35 In the absence of an originally open fracture, substantial comminution, or multiple adjacent unstable fractures, the need for open treatment of metacarpal neck fractures is rare. Wong et al36 detected no radiographic, functional, or complication differences between transverse pins or intramedullary K-wires for fixation of smallfinger metacarpal neck fractures in 59 patients.
The surgeon must be able to control the metacarpal neck deformity primarily in sagittal plane angulation and less so in coronal angulation (more often seen as lateral translation at the fracture site) and axial rotation (primarily with the small finger and sometimes the index finger, and infrequently in the long and ring fingers). A method of precise control over all three planes is transverse pinning of the fractured ray to its adjacent ray (Figure 9). The biomechanics of external fixation apply, with the intact adjacent metacarpal serving as the fixator bar. Three 0.045-inch K-wires are needed; the carpometacarpal ligaments function as the fourth point of fixation. Wires are left through the skin for removal at 3 weeks or are cut beneath the skin, at the surgeon's discretion. Intrinsic plus splinting is important to prevent extension contracture of the MCP joint, particularly in the small finger, where a splint at 90° formed to the radial three fingers will allow the small finger to sit at 60° of MCP flexion. As long as the PIP joints are moved from the outset, there is no long-term difference in motion between either splinting for 3 weeks and then initiating active MCP exercises or beginning MCP motion immediately and returning to the splint in between sessions.
Metacarpal shaft fractures tend to angulate apex dorsal, just as metacarpal neck fractures do, but the tolerable angular deformity is less owing to the greater distance from the joint. Compared with neck fractures, there is more lateral translation between the fracture fragments at the fracture site, with coronal angulation of the distal fragment. The deep transverse intermetacarpal ligaments limit the amount of proximal translation of the central two rays more effectively than of the border rays. The biomechanics of transverse shaft fractures allow the principles of external fixation to be even more effective because of the longer distal segment.
Galanakis et al37 treated 25 transverse small-finger metacarpal fractures by three K-wires (one proximal to the fracture site, two distal), resulting in no shortening, appreciable angulation, or complications and producing excellent functional results. Optimal placement of the first and second wires is as close to the metacarpal shaft fracture site as possible. The third wire optimizes its biomechanical effect when it is placed as far distally as possible in the metacarpal head, but functionally it is better for the patient if the third wire is placed proximal to the sagittal bands and MCP collateral ligaments. The flare of the metacarpal head at the proximal extent of the collateral origin recess is an obvious radiographic feature that easily allows correct placement of the most distal wire. The cortical fracture interface of the metacarpal shaft, which is small in cross-section, heals slightly slower than does the broad cancellous contact of a neck fracture, and pins should be retained for 4 rather than 3 weeks.37
Spiral (Long Oblique) Metacarpal Shaft Fractures
As with phalangeal fractures, the predominate concern with spiral and long oblique fractures is less angulation than it is shortening and axial rotation. As the metacarpal shortens through the fracture interface, laboratory analysis predicts 7° of MP joint extensor lag for every 2 mm of shortening.38 Closed reduction is less able to restore length than it is angulation, and percutaneous pinning (transverse or intramedullary) is less able to maintain reduction with this pattern compared with transverse fractures. Transverse pinning still can be a good option for these fractures, but the treatment required to ensure maintenance of a precise reduction and provide a solid foundation for immediate rehabilitation is interfragmentary 1.5-mm compression lag screws. (Although 2.0-mm plates are indicated for metacarpal fractures, 2.0 mm is too large when used as a lag screw). Patients can start immediate rehabilitation exercises, and office workers can remove the splint and type at a keyboard normally immediately after surgery. With a two-part fracture, the lag screws are functioning in a load-sharing relationship with the fracture interface (Figure 10). An intrinsic plus splint should be worn in the first 3 to 4 weeks during at-risk situations when the hand is likely to have unwanted contact, such as in crowds.
Comminuted Metacarpal Fractures
Without any inherent stability from the fracture interface in comminuted fractures, the fixation hardware bears the entire load of rehabilitation forces until early healing takes over. Plate-and-screw fixation is the standard in this situation. Just as in the proximal phalanx, lateral positioning of the plate is ideal in order to keep it away from the extensor tendon (Figure 11). But, unlike the proximal phalanx, which has an oval (actually, a kidney bean-shaped) cross section, the midshaft of the metacarpal has a more triangular cross section on its external surface. Distally, there is a true lateral surface just proximal to the collateral recess, but that quickly spirals around to a dorsolateral plane at the shaft level. Careful axial contouring of the plate ahead of time to recognize this anatomy is important in order to avoid an iatrogenic rotational malunion. Fortunately, the zone VI extensor (Figure 3) does not have as intimate a relationship with the bone as does the zone IV extensor. Also, the zone VI extensor is much more forgiving of the presence of hardware.
The method that applies in this situation is setting preliminary relationships between major fragments with individual lag screws, then placing the plate in neutralization mode, as with fixation of the long tubular bones. The added soft-tissue crush trauma associated with fracture comminution causes the intrinsic muscles to fibrose and initiates an intrinsic minus contracture, placing added emphasis on the importance of intrinsic plus splinting between ROM sessions as well as at night.
Nondisplaced and inherently stable isolated fractures in the hand do not require surgery. Fractures that require formal stabilization may be treated by several possible methods. No one strategy is universally the best, and the literature lacks any strong level I or II evidence to lend direction. The orthopaedic surgeon should be clinically comfortable with the chosen method of stabilization as long as the strategy accommodates early active motion, especially of the PIP joint.
Evidence-based Medicine: There are two level I/II limited prospective randomized studies reported (references 13 and 22). Most of the references presented are case-control cohort studies or case reports (level III/IV) or are level V expert opinion studies.
Citation numbers printed in bold type indicate references published within the past 5 years.
1. Chung KC, Spilson SV: The frequency and epidemiology of hand and forearm fractures in the United States. J Hand Surg Am
. van Onselen EB, Karim RB, Hage JJ, Ritt MJ: Prevalence and distribution of hand fractures. J Hand Surg Br
3. Henry MH: Fractures and dislocations of the hand, in Heckman JD (ed): Rockwood and Green's Fractures in Adults
, ed 6. Philadelphia, PA: Lippincott Williams & Wilkins, 2006, vol 1, pp 771-856.
4. Henry MH: Fractures and dislocations of the hand, in Heckman JD (ed): Rockwood and Green's Fractures in Adults
, ed 5. Philadelphia, PA: Lippincott Williams & Wilkins, 2002, vol 1, pp 655-744.
5. Ip WY, Ng KH, Chow SP: A prospective study of 924 digital fractures of the hand. Injury
6. Gonzalez MH, Hall M, Hall RF Jr: Low-velocity gunshot wounds of the proximal phalanx: Treatment by early stable fixation. J Hand Surg Am
. Tan V, Beredjiklian PK, Weiland AJ: Intra-articular fractures of the hand: Treatment by open reduction and internal fixation. J Orthop Trauma
8. Botte MJ, Davis JLW, Rose BA, et al: Complication of smooth pin fixation of fracture and dislocation in the hand and wrist. Clin Orthop Relat Res
. Kurzen P, Fusetti C, Bonaccio M, Nagy L: Complications after plate fixation of phalangeal fractures. J Trauma
10. Page SM, Stern PJ: Complications and range of motion following plate fixation of metacarpal and phalangeal fractures. J Hand Surg Am
. Trevisan C, Morganti A, Casiraghi A, Marinoni EC: Low-severity metacarpal and phalangeal fractures treated with miniature plates and screws. Arch Orthop Trauma Surg
. Fusetti C, Della Santa DR: Influence of fracture pattern on consolidation after metacarpal plate fixation. Chir Main
. Horton TC, Hatton M, Davis TR: A prospective randomized controlled study of fixation of long oblique and spiral shaft fractures of the proximal phalanx: Closed reduction and percutaneous Kirschner wiring versus open reduction and lag screw fixation. J Hand Surg Br
14. Ebinger T, Erhard N, Kinzl L, Mentzel M: Dynamic treatment of displaced proximal phalangeal fractures. J Hand Surg Am
. Freeland AE, Hardy MA, Singletary S: Rehabilitation for proximal phalangeal fractures. J Hand Ther
. Brown H, Stutz C, Henry M: Technique for extensor tendon acceleration. J Hand Ther
17. Strickler M, Nagy L, Buchler U: Rigid internal fixation of basilar fractures of the proximal phalanges by cancellous bone grafting only. J Hand Surg Br
. Shewring DJ, Thomas RH: Avulsion fractures from the base of the proximal phalanges of the fingers. J Hand Surg Br
19. Liew KH, Chan BK, Low CO: Metacarpal and proximal phalangeal fractures-fixation with multiple intramedullary Kirschner wires. Hand Surg
20. Safoury Y: Treatment of phalangeal fractures by tension band wiring. J Hand Surg Br
. Al-Qattan MM: Metacarpal shaft fractures of the fingers: Treatment with interosseous loop wire fixation and immediate postoperative finger mobilisation in a wrist splint. J Hand Surg Br
. Margić K: External fixation of closed metacarpal and phalangeal fractures of digits: A prospective study of one hundred consecutive patients. J Hand Surg Br
. Geissler WB: Cannulated percutaneous fixation of intra-articular hand fractures. Hand Clin
24. Chin KR, Jupiter JB: Treatment of triplane fractures of the head of the proximal phalanx. J Hand Surg Am
. Crofoot CD, Saing M, Raphael J: Intrafocal pinning for juxtaarticular phalanx fractures. Tech Hand Up Extrem Surg
26. Vahey JW, Wegner DA, Hastings H III: Effect of proximal phalangeal fracture deformity on extensor tendon function. J Hand Surg Am
27. Widgerow AD, Ladas CS: Anatomical attachments to the proximal phalangeal base: A case for stability. Scand J Plast Reconstr Surg Hand Surg
28. Hornbach EE, Cohen MS: Closed reduction and percutaneous pinning of fractures of the proximal phalanx. J Hand Surg Br
29. Elmaraghy MW, Elmaraghy AW, Richards RS, Chinchalkar SJ, Turner R, Roth JH: Transmetacarpal intramedullary K-wire fixation of proximal phalangeal fractures. Ann Plast Surg
30. Lins RE, Myers BS, Spinner RJ, Levin LS: A comparative mechanical analysis of plate fixation in a proximal phalanx fracture model. J Hand Surg Am
31. Lu WW, Furumachi K, Ip WY, Chow SP: Fixation for comminuted phalangeal fractures: A biomechanical study of five methods. J Hand Surg Br
. Ouellette EA, Dennis JJ, Latta LL, Milne EL, Makowski AL: The role of soft tissues in plate fixation of proximal phalanx fractures. Clin Orthop Relat Res
33. Freeland AE, Sud V, Lindley SG: Unilateral intrinsic resection of the lateral band and oblique fibers of the metacarpophalangeal joint for proximal phalanx fractures. Tech Hand Up Extrem Surg
34. Curtin CM, Chung KC: Use of eighthole titanium miniplates for unstable phalangeal fractures. Ann Plast Surg
35. Birndorf MS, Daley R, Greenwald DP: Metacarpal fracture angulation decreases flexor mechanical efficiency in human hands. Plast Reconstr Surg
. Wong TC, Ip FK, Yeung SH: Comparison between percutaneous transverse fixation and intramedullary K-wires in treating closed fractures of the metacarpal neck of the little finger. J Hand Surg Br
37. Galanakis I, Aliquizakis A, Katonis P, Papadokostakis G, Sterqiopoulos K, Hadjipavlou A: Treatment of closed unstable metacarpal fractures using percutaneous transverse fixation with Kirschner wires. J Trauma
38. Strauch RJ, Rosenwasser MP, Lunt JG: Metacarpal shaft fractures: The effect of shortening on the extensor tendon mechanism. J Hand Surg Am