A fall or punches are frequent accident mechanism of metacarpal shaft fractures [12, 19]. Undisplaced fractures that are stable can be treated non-operatively [8, 10]. Surgical treatment is indicated for fractures with palmar displacement over 30°, with a shortening over 5 mm, with rotation deformity and for fractures of multiple metacarpals [10, 20]. Displaced transverse or short oblique fractures of the distal shaft or the midshaft are indications for intramedullary fixation [8, 12, 18, 19]. In displaced comminuted fractures, plate osteosynthesis or external fixation is recommended [2, 19]. Displaced spiral fractures can be treated by lag screws or plates [2, 10, 19].

Intramedullary fixation of proximal metacarpal shaft fractures requires a retrograde passage of the wires through the extensor hood mechanism [2, 4]. Retrograde intramedullary fixation often caused joint stiffness and loss of extension and is therefore no more recommended [10, 19].

Antegrade intramedullary fixation of metacarpal fractures is performed by using one [4, 5, 8, 10], two [7, 9, 10] or three to four [6, 10] pre-bent Kirschner-wires or ESIN [11, 12]. The technique of antegrade intramedullary nailing was introduced by Foucher. In his “bouquet” osteosynthesis method, three to four wires were inserted [6]. The blunt Kirschner-wires were pre-bent in a long curve acting as three-point fixation. At the tip over the final 5 mm, a sharp so called “hockey stick” bend was made. Pre-bending of the wire facilitates insertion and reduction and avoids perforation of the metacarpal head.

In the following, we describe our personal technique in intramedullary nailing of metacarpal fractures. In fractures of the first, second and fifth metacarpal, we use two Kirschner-wires. One single longitudinal wire controls angulation but not rotation and the risk of secondary displacement is higher. In the third and fourth metacarpal however, a single wire seems to provide sufficient stability thanks to the support by the adjacent intact metacarpal bones (Fig. 15.2a–e).

In fractures of the fifth metacarpal bone; a small incision is made at the ulnar side of the metacarpal base, whereas a radial incision is recommended in fractures of the second metacarpal bone. In the other metacarpals, an ulnar or radial approach are equally appropriate. Dissection to the bone is done blunt in order to avoid tendon or nerve lesions. The osseous entry point at the lateral base of the metacarpal is made with an awl. A pre-bent Kirschner-wire is inserted antegrade with the blunt end first using a T-handle. Depending on the metacarpal width, wires with diameters of 1.2–1.6 mm are implanted. We reduce metacarpal shaft fractures by longitudinal traction on the digit. Neck fractures are reduced according to Jahss’s method [2] with the finger flexed 90° at the metacarpophalangeal and interphalangeal joints; force applied to the proximal phalanx pushes the metacarpal head back into position. After reduction of the fracture, the wire is passed across the fracture site into the metacarpal head. By rotating the Kirschner-wire, the bent at the tip is oriented dorsally, thereby correcting the palmar displacement of the distal fragment. A second wire is subsequently introduced in the same manner. After final control of reduction and wire position with the image intensifier, the wire is cut beneath skin level (Fig. 15.3a–f). Leaving the wire out of the skin is also possible, but requires daily pin site care to avoid pin tract infections. Postoperatively, a cast in intrinsic plus position is applied for 2–3 weeks, followed by physiotherapeutic exercises. Hardware is usually removed at 8 weeks when fracture healing is visible in radiographs.

Biomechanical studies have demonstrated that plate fixation provides more mechanical strength in comparison to intramedullary fixation of metacarpal fractures [20–22]. Firoozbakhsh et al. examined 105 cadaver metacarpals after oblique osteotomies and fixation with a plate and lag screws, two lag screws, crossed Kirschner-wires with tension bands, five intramedullary Kirschner-wires or two intramedullary Kirschner-wires. The plate osteosynthesis was superior in bending, torsion and axial loading tests, followed by lag screws, crossed Kirschner-wire tension banding and intramedullary Kirschner-wire fixation [21]. Black et al. reported similar results after fixation of metacarpal fractures with different plating and wire configurations (crossed Kirschner-wires, one intramedullary wire, one intramedullary wire combined with an oblique wire). Plate osteosynthesis provided significant higher rigidity in torsion and bending compared with wire configurations [20].

Despite higher mechanical strength after plate osteosynthesis, allowing immediate mobilization, several authors describe better clinical outcomes after intramedullary fixation. Mobility deficit after plate fixation is often due to adherence of the extensor apparatus of the metacarpophalangeal joint to the plate surface [5, 7].

Facca et al. found better functional results after intramedullary fixation of metacarpal fractures with a single K-wire and 6 weeks immobilization (n = 20) compared to a second series treated with locking plates and immediate mobilization (n = 18). Active flexion and extension was significantly greater in the K-wire group. The wire group showed seven complications, including three cases of wire migration, three of lesions of the dorsal cutaneous branch of the ulnar nerve and one of esthetic blemish. The locking plate group showed six complications, including three cases of stiffness requiring tenolysis and arthrolysis, two of delayed union and one of head necrosis [5].

Fujitani et al. confirmed the results of Facca et al.: 30 patients with metacarpal neck fractures that had either been treated with two intramedullary K-wires or with a low profile plate, were examined. Twelve months after surgery, the range of motion at the metacarpophalangeal joint was significantly better in patients with intramedullary fixation than in patients that had been treated with plate fixation (Fig. 15.4a–g). With an average palmar tilt angle of 16° in both groups and an average shortening of 1.5 mm in the intramedullary and of 0.7 mm in the plate group, postoperative radiological parameters were comparable in both groups. Complications occurred in three patients. Tendon rupture was seen in one case with intramedullary fixation and transient neuritis was seen in one case of each group [7].