Most reported cases of dislocation of the great toe involve dislocation of the metatarsophalangeal joint due to its greater mobility and longer lever arm. Dislocation of the interphalangeal (IP) joint of the hallux is very rare, and when it does occur it is usually irreducible and dorsal [1, 2]. According to Leung and Wong ^[3], in the past six decades, only 43 cases have been reported in the literature: 13 in English language sources, one in German, and 27 in Japanese sources. A further seven cases have recently been reported [1-7]. The IP joint of the hallux contain a thin dorsal capsule reinforced by the medial and lateral collateral ligaments, a strong fibrocartilaginous plantar plate, and a sesamoid bone located in the plantar capsule in 88% of cases ^[8]. The IP stability is due to the presence of static stabilisers (collateral ligaments and joint capsule) and dynamic stabilisers (flexor hallucis longus and extensor hallucis longus). The collateral ligaments not only confer side-to-side stability but also limit the amount of hyperextension. When these ligaments and the joint capsule are cut, further extension is possible up to the end-point limited by the volar plate ^[3], making dislocation of the joint possible ^[9]. The mechanism of dorsal dislocation of the IP joint of the great toe is known to be a combination of axial loading with hyperextensive force [10, 11]. According to Miki et al. ^[9], dislocation of the IP joint of the great toe can be classified into two types according to radiographic and clinical sesamoid bone interposition. In type 1, the ruptured volar plate is displaced into the joint space between two phalanges, the toe is slightly elongated and swollen, but without clear deformity the deformity of the toe. In Type 2, the volar plate is completely dorsally displaced over the proximal phalangeal neck with severe deformity as the IP joint is locked in hyperextension with a typical dorsal hollow. Reduction of Type 2 dislocation is often insufficient although it can appear to be reduced, due to the incarceration of the sesamoid bones into the joint, becoming a Type 1 dislocation, finding slight elongation, and no dorsal deformity during physical examination. This injury is exceptionally open dislocation ^[4]. In either type of dislocation, the volar plate can be detached from both the distal and proximal phalanges and displaced into the joint, thereby forming a barrier to manual reduction. The authors agree that open [2, 3, 6, 8-12], or percutaneous [1, 7], reduction is mandatory. On the other hand, acute irreducible plantar dislocation of the IP of the great toe has only been reported once recently ^[13]. We present the second plantar case but is the first chronic neglected. The purpose of this article is to present three new patients who show the variability of irreducible IP dislocation of the great toe occurring in sport activities: One open and one closed dorsal dislocation (Miki Type 2) and one neglected chronic open plantar dislocation (not reported to date), and to review the literature pertinent to these conditions, describing the management and outcome.

Clinical Cases (Table 1)
Case 1
A 16-year-old male Basque handball player (traditional sport of Northern Spain and South-west France, played with a hard linen or leather ball thrown against a wall using one’s hand) was diagnosed with open dorsolateral dislocation of the IP joint of the first toe of the right foot after a direct collision against the wall (Gustilo Type II open fracture) (open Miki Type 2) (Fig. 1a and b). After thorough washing and administration of antibiotherapy, closed reduction was attempted on multiple occasions without success, by longitudinal traction and manipulation without traction with pressure applied to the base of the distal phalanx. Apparent reduction was achieved after numerous  attempts, but with a greater toe length and joint stiffness. Radiographic follow-up revealed reduced dislocation with sesamoid bone interposition (intra-articular) transforming Type 2 to the Type 1 dislocation (Fig. 1c and d). A medial approach was used performing an open reduction of the dislocation, separating of the extensor hallucis longus and evacuation of the sesamoid plates. The joint was immobilzsed for 3 weeks by binding the first and second toe together due to stability with flexion and extension. The patient did not present acute or delayed infection and regained 60% of mobility after 4 months. There was no recurrence of dislocation or residual instability, and the X-ray revealed no abnormalities. Fifteen years later, the patient continues to be asymptomatic, participating in normal sporting activities and with −20° active flexion.
Case 2
A 32-year-old male patient, attended in the Emergency Room of our Hospital, having sustained a direct injury to the right forefoot while playing soccer. The patient experienced pain in the first toe of his right foot caused by dorsiflexion after kicking the ball. Neurovascular assessment was normal. The simple X-ray revealed hallux IP dorsolateral dislocation (Miki type 2) (Fig. 2a and b). Sesamoid bone interposition manifested after numerous closed reduction attempts without anesthesia and with digital block anesthesia. Closed reduction was attempted under digital block anesthesia. The sesamoid bone interposition was once again observed (Miki Type 1) (Fig. 2c and d). Dorsal open reduction was achieved by laterally subluxing the extensor hallucis longus tendon and applying pressure to the interposed sesamoid bone toward the plantar aspect of the toe. Due to the medial instability, a Kirschner wire was used to for temporary immobilization of the IP joint with a buddy strapping securing to the  second toe for 3 weeks. Fourteen years later, the patient continues to be asymptomatic, participating in normal sporting activities and with symmetrical mobility compared to the healthy contralateral toe.
Case 3
A 29-year-old male soccer player attended A&E after sustaining a direct contusion to the hallux of the right foot. The patient reported mild deformity progression over the previous months following a previous injury 10 months ago. A punctiform skin lesion could be seen (Fig. 3a). The X-ray revealed plantar dislocation of the IP joint of the hallux with avulsion of the medial distal condyle of IP in non-union (Fig. 3b and c). Attempts to closed reduction under block anesthesia were unsuccessful. Dorsal open reduction was performed, which involved the detachment of the lateral ligaments and capsule. No hematoma or local bleeding was observed, and chronic dislocation was confirmed. Kirschner wire fixation was performed for 3 weeks (Fig. 3d and e) due to medial slight instability. Five months later, the patient had recovered minimal mobility with only 10° of flexion. Ten years after surgery, the patient’s functionality is good when walking or running in sports shoes, but with mild discomfort when running barefoot. All patients agreed to participate in the study. The advantages and disadvantages of the treatment were explained and
informed consent was obtained from these patients.

The sesamoid bones of the IP joint of the great toe (os sesamoidium interphalangeus hallux) measure between 0.05 and 1 cm ^[3] and are embedded in the plantar fibrocartilage. They are macroscopically visible in 95.9% of cases ^[14], but radiographically invisible in 4.3–3% of cases [16, 17], while for Miki et al. ^[9], this figure is 44%, which further hinders their identification as incarcerated sesamoids and their diagnosis. Their presence can only be confirmed in X-ray after reduction of an IP dislocation. It is well known that closed reduction with traction does not usually have a successful outcome in dislocated joints with sesamoid bones, such as metacarpophalangeal joint of the thumb and metatarsophalangeal joint of the hallux, as the traction closes the space and the intrinsic muscles act as a buttonhole. For this reason, reduction without traction is recommended, by pressure applied to the base of the distal phalanx. In the setting of IP joint dislocation is similar. Despite some authors reporting closed reduction of the IP joints of the hallux to be successful in some patients [11, 12], this procedure usually fails, obtaining an incomplete reduction [1, 9]. When performing closed reduction of the Miki Type 2, sesamoid interposition into the IP joint is the rule, transforming the lesion in a Type 1 [8, 13, 15, 16], due to the invagination of the sesamoid-plantar plate complex into the IP joint. As the interposed sesamoid effectively “tightens” the intact collateral ligaments, thereby preventing closed reduction [3, 12, 18-20]. When attempting closed reduction, the hyperextension and lateral deviation of the distal phalanx disappear leaving a straight great toe with apparent normal alignment [3, 9, 14], but with a restricted range of motion and pain, consistent with Cases 1 and 2 detailed above. Ward et al. ^[12] were the first authors to report sesamoid entrapment as a complication of a closed manipulation of a dislocated IP joint of the hallux. Hallucal sesamoid bone interposition as a result of the initial injury at the IP joint of the hallux (Miki Type 1) is a recognized but very rare occurrence [2, 10, 16, 21, 22, 23], normally caused by an active or passive reduction attempt. Therefore, irreducibility depends on the invagination of the sesamoid bones [9, 19], usually with normal collateral ligaments, although the interposition of the lateral collateral ligaments has also been reported ^[2]. We can surmise that Miki 1 dislocations are the consequence of an incomplete spontaneous or forced reduction of a Miki 2 dislocation, if no lateral instability exists. Therefore, after attempting closed reduction, an X-ray must be performed to rule out invagination of the sesamoid-volar plate complex ^[12]. Therefore, it seems clear that closed reduction should always be attempted but this is often unsuccessful (as did occur in our cases 1 and 2) due to the common interposition of the plantar plate and/or sesamoids into the joint. Before attempting open surgical reduction (as we did), percutaneous technique is now indicated. Percutaneous reduction with a K-wire fixation, under fluoroscopy, can be performed in cases of invagination of the sesamoid-volar plate complex [1, 7]. We believe that this  technique should be the current treatment of choice for primary sesamoid bone interposition in the IP joint of the hallux (Miki type 1 or 2) provided that the sesamoids can be seen by intraoperative fluoroscopy. We did not tried percutaneous reduction because it was reported several years after the occurrence of our cases. According to Woong ^[26], to conduct a percutaneous reduction with a K-wire, the sesamoid bones must be evaluated radiographically to assess their intra-articular entrapment and confirm their repositioning intraoperatively. Open reduction should be used in the event of failure and/or absence of apparent sesamoids on X-ray. Dorsal dislocation of the hallux IP joint has usually been reported as a closed injury [3, 10, 12, 15, 21, 24]. The open nature of the injury in adolescence (Case 1) makes it an exceptional case. Open dislocation of the IP joint of the hallux with MTP joint dislocation may also occur, and has been reported on one occasion ^[25]. Only Jones et al. ^[4] reports two open dorsal dislocations in soldiers, consistent with Case 1 reported above. These injuries have been scarcely reported after motor and crush accidents but exceptionally as a consequence of sport combat training ^[4] or sport related injuries ^[20], as all of our cases. Therefore, the spectrum of variants of IP dislocation of the great toe is wide. Plantar, medial, dorsal, and dorsolateral approach for open reduction have been reported, with none of the techniques showing clear superiority over the others ^[3]. Yasuda et al. ^[18] recommend the plantar approach to repair the plantar plate, but other authors do not favor repair of the dislocated volar plate due to its inherent stability after reduction. Furthermore, the plantar approach is a very demanding technique that leaves a hyperkeratotic scar on the loading areas of the foot. Because of the positive outcomes achieved after reduction, there is consensus that volar plate repair is not necessary ^[3]. Dorsal, or occasionally, medial open reduction have traditionally been used [6, 10]. According to anatomical studies of the location of the intra-articular ossicles, the medial approach appears to be the most suitable ^[8] but dorsal approach seems to be easier and safer (6). Various post-reduction stabilization techniques have been reported (3), including bulky dressing, buddy strapping, splinting, Kirschner wire and short leg cast, for 3–4 weeks. In our opinion, binding to the second toe and Kirschner wire in the event of instability, as used in two of the patients, are sufficient. Removal of the sesamoids has ben reported [5, 18, 20] but it is unnecessary [6, 9] confirming our long-term outcome that sesamoids must be preserved. Repair of the volar plate has been done ^[18] to preserve hyperextension of the hallux IP joint avoiding the risk of redislocation ^[19] but our results and others ^[21] confirm the stability without volar plate repair. Prognosis is favorable in the absence of complications, with no morbidity or sequelae reported in the literature, even in the rare open dislocations. Although inveterate dislocation (case 3) leads to persistent joint stiffness, this may not impact on daily life. Recently one case of acute irreducible IP dislocation in plantar direction has been reported ^[13]. Neglected irreducible IP dislocation has once been reported in dorsal direction ^[11] but we have not found any case in the literature of neglected plantar IP dislocation of the great toe consistent with our Case 3 that is the first description in the literature. Theoretically, reduction of a plantar dislocation in the acute situation should be straightforward achieved by traction of the toe due to the lack of sesamoid interpositions, contrary to the dorsal dislocations. Our experience with the Case 3 suggests that the chronic nature of the injury requires open reduction and leads to IP joint stiffness. Although Hatori et al. ^[11] performed arthrodesis on a neglected dorsal dislocation, our patient with plantar dislocation was asymptomatic after open reduction without arthrodesis, enabling him to lead a normal life, albeit with limited IP mobility.

IP dislocation of the great toe is a very rare injury that can be dorsal or plantar. Dorsal dislocation with intra-articular incarceration of the sesamoid bones (Miki 1) ussually are manually irreductible. Dislocations without sesamoid bones incarceration (Miki 2) reduction must be performed without axial traction and only by pressure in the base of the phalanx. Secondary to the trial of orthopaedic reduction or spontaneously, incarceration of the sesamoids bones is the rule (Miki 1). In this situation percutaneous or open reduction must be performed. Plantar dislocation doesn’t provoque intra-articular sesamoid interposition. This article describe the first reported case of neglected plantar dislocation that required open reduction, similar management adding that the long-term functional prognosis is good.

Due to the rarity of these dislocations, we have to bear in mind that the dorsal Type 2 is the habitual injury. The reduction must be performed without axial traction and only by pressure in the base of the phalanx. Secondary to the trial of orthopedic reduction or spontaneously, incarceration of the sesamoids bones is the rule (Miki 1). In this situation, percutaneous or open reduction must be performed. Plantar dislocation does not provoque intra-articular sesamoid interposition.