A 3- to 4-cm dorsomedial longitudinal skin incision is made over the first metatarsal base. The extensor hallucis longus tendon is exposed so as not to injure the medial dorsal cutaneous nerve or the dorsal digital nerve. The longitudinal incision is made along the medial side of the extensor halluces longus tendon. The extensor hallucis longus tendon is retracted laterally to expose the first metatarsal base. The periosteum of the first metatarsal base is longitudinally incised by lengths of 3–4 cm, and the full circumferential release of the periosteum is performed. The first tarsometatarsal joint (TMT joint) is identified using the tip of an elevator. The osteotomy site which is located 1.5 cm distal to the first TMT joint is marked with surgical marking pen. A crescentic osteotomy is performed with a curved saw blade. The osteotomy is curvilinear, and the concavity of the cut is directed distally. The direction of the osteotomy is perpendicular to the sole of the foot on the coronal plane and is perpendicular to the long axis of the first metatarsal on the sagittal plane.

After completion of a distal soft-tissue procedure and a proximal crescentic osteotomy of the first metatarsal, supination stress of the great toe is performed for assessing intraoperative correction of hallux valgus and metatarsus primus varus, the shape of the lateral edge of the first metatarsal, and the sesamoid position. The maneuver of supination stress of the great toe is as follows: the plantar surface of the foot is placed on the image intensifier. The great toe is grasped, and gentle axial traction is applied by pulling on great toe, and then supination stress is manually applied to the great toe under dorsoplantar fluoroscopic view (Fig. 11.2) [19]. Dorsoplantar fluoroscopic image of the foot is obtained under supination stress (Fig. 11.3). When good corrections of a valgus deformity and metatarsus primus varus, reduction of the sesamoids, and a negative round sign of the lateral edge of the first metatarsal head are observed, the releases of the distal soft tissue and subperiosteum at the osteotomy site are considered to be adequate. The shape of the lateral edge of the first metatarsal head, which consist of the articular surface and the lateral cortical surface of the metatarsal head on the dorsoplantar radiograph, is classified as one of three types, round (type R), angular (type A), or intermediate (type I) according to a previously published measurement system [18]. The round sign as being positive is defined when the shape of the lateral edge is classified as type R, and it as being negative when the shape of the lateral edge is classified as type I or A.

The proximal fragment is pushed medially with an elevator as much as possible, and the distal fragment is moved laterally to achieve parallelism between the first and second metatarsals, and then the distal fragment of the first metatarsal is manually supinated (Fig. 11.4). Temporary fixation with a 1.5 mm Kirschner wire is performed at the osteotomy site. And then the intermetatarsal angle and the shape of the lateral edge of the first metatarsal on the dorsoplantar fluoroscopic view and the sagittal alignment of the osteotomy site on the lateral fluoroscopic view are checked. If the parallelism between the first and second metatarsals, the angular-shaped lateral edge (type A) on dorsoplantar fluoroscopic view, or good alignment of the first metatarsal on the lateral fluoroscopic view cannot be obtained, re-correction at the osteotomy site is performed. If good correction at the osteotomy site is obtained, another 1.5 mm Kirschner wire is used for temporary fixation at the osteotomy site.

The variable angle locking X-plate (VA locking X-plate ) has four holes and is available in extra-small, small, medium, and large sizes. Appropriate locking X-plate size is selected using a template of each plate size, taking into account of the relationship between the four screw holes of the locking X-plate and the proximal and distal fragments in order to achieve bicortical screw fixation in all screws. We commonly used the extra-small-sized titanium VA locking X-plate, measuring 23.5 mm in the longitudinal and 15.0 mm in the width (Fig. 11.5). When the bending of the locking X-plate is needed, the change of the screw direction due to the plate bending should be considered. After the VA locking X-plate is placed on dorsal or dorsomedial aspect at the osteotomy site, the locking X-plate is fixed with three or four 2.0-mm Kirschner wires which is the same in diameter of drill hole for the head locking screw and can be inserted at −15° to +15° deviation from the center axis of the screw hole using the conical drill sleeve (Fig. 11.6). After confirmation of the direction of the Kirschner wires, one Kirschner wire is removed and the head locking screw inserted. The remaining three screws are inserted in the same way. And then two Kirschner wires for temporary fixation at the osteotomy are removed.

Two drill holes are made in the metatarsal neck and head using a 1.2-mm Kirschner wire. One is drilled at the dorsomedial side of the metatarsal head in the plantar-to-medial direction and the other at the dorsomedial side of the metatarsal neck in the plantar direction (Fig. 11.7). A 2-0 braided non-absorbable suture is passed through each drill hole. The medial part of the capsule together with the abductor hallucis tendon is proximally and dorsally pulled to correct the valgus and pronation deformities of the great toe and is fixed with two intraosseous sutures. And then the capsulorrhaphy is made with absorbable sutures (Fig. 11.8).

Intraoperative fluoroscopic dorsoplantar and lateral views of the foot are made to evaluate the hallux valgus angle (<15°), the intermetatarsal angle (<10°), the sesamoid position (<V according to Hardy classification), a round sign (negative), and sagittal alignment of the first metatarsal (no angulation at the osteotomy site).

A short-leg cast with rubber heel was continued for 2 weeks. A partial weight-bearing was allowed 1 day after surgery. Two weeks after surgery, a short-leg plaster shell was applied and active and passive extension and flexion exercises of the first metatarsophalangeal joint was encouraged. Three weeks after surgery, patients were instructed to wear street shoes with an arch support. Four weeks after surgery, full weight-bearing was allowed. Patients could participate in sports activity 2 or 3 months after surgery .

Clinical and radiological results in the literature are shown in Table 11.1. The mean American Orthopaedic Foot & Ankle Society hallux-metatarsophalangeal-interphalangeal (AOFAS) scores after a proximal crescentic osteotomy have ranged from 91 points to 96 points and significantly improved compared to preoperative scores, although there were various fixation methods and differences in follow-up periods among the articles [1–3, 6, 10, 11, 13, 14, 28]. The mean pain, function, and alignment scores on the AOFAS scale significantly improved after a proximal crescentic osteotomy [6, 11, 13, 14]. The mean hallux valgus and intermetatarsal angles after a proximal crescentic osteotomy have ranged from 9° and 5° to 16° and 9°, respectively, and significantly improved compared to preoperative angles [1–3, 6, 10, 11, 13, 14, 28]. The rate of patient satisfaction after a proximal crescentic osteotomy ranged from 85% to 96% in the literature (Table 11.1) [1–3, 6, 10, 11, 14].