Hallux Rigidus—Joint-Sparing Techniques

Abstract

With the rising demands of patients, in cases of grade III hallux rigidus, a fusion may be an unacceptable option. This can be a dilemma for the surgeon, as a simple cheilectomy in those cases may not only result in pain relief; it can result in worsening pain secondary to increased motion of a severely arthritic joint. Many surgical alternatives are available to choose from, all based on considerations of the underlying anatomy, the pathologic changes, and the severity of the arthritis. In our own practices, the cheilectomy, with or without an osteotomy at the base of the proximal phalanx (the Moberg procedure), is the most predictable operation for correction of hallux rigidus. For management of the more severe grades of arthritis, although we perform arthrodesis frequently, we have obtained excellent results with interposition arthroplasty, metatarsal osteotomy, and even first tarsometatarsal arthrodesis.

Key Words

hallux rigidus, arthritis, cheilectomy, interposition arthroplasty, Moberg

Overview of Surgery and Decision Making

Surgical correction of hallux rigidus gives fairly predictable results, and patient acceptance and aesthetic and functional outcomes should be good. However, with the rising demands of patients, in cases of grade III hallux rigidus, a fusion may be an unacceptable option. This can be a dilemma for the surgeon, as a simple cheilectomy in those cases may not only result in pain relief; it can result in worsening pain secondary to increased motion of a severely arthritic joint. Many surgical alternatives are available to choose from, all based on considerations of the underlying anatomy, the pathologic changes, and the severity of the arthritis. Patient needs for activities and footwear will also influence the decision making for the type of surgery. In our own practices, the cheilectomy, with or without an osteotomy at the base of the proximal phalanx (the Moberg procedure), is the most predictable operation for correction of hallux rigidus. For management of the more severe grades of arthritis, although we perform arthrodesis frequently, we have obtained excellent results with interposition arthroplasty. An additional alternative to maintain motion in grade III hallux rigidus without a grind include a shortening osteotomy of the first metatarsal combined with slight plantarflexion. Despite efforts with various metallic implants to replace either the phalanx, metatarsal, or both surfaces, we have achieved less than desirable results with any type of implant arthroplasty. Arthrodesis continues to be a mainstay of treatment in the management of severe arthritis associated with deformity or in cases in which other salvage procedures in the forefoot need to be performed simultaneously.

In planning surgery, the range of motion (ROM) of the metatarsophalangeal (MTP) joint and interphalangeal (IP) joint of the hallux is important. We examine the foot while the patient is seated, as well as standing, because additional contracture, particularly of the flexor hallucis brevis, may become evident with standing. Passive dorsiflexion of the hallux is essential to quantify, since jamming of the joint may be present as a result of a tight plantar fascia or a gastrocnemius contracture. In the setting of limited IP motion or arthritis of the IP joint, we attempt to preserve the motion of the MTP joint if at all possible.

Occasionally, osteotomy of the first metatarsal is advantageous. Elevation of the first metatarsal may not have a significant role in the pathogenesis of hallux rigidus ( Fig. 6.1 ). Nevertheless, a most definite correlation exists between metatarsus elevatus and severe grades of hallux rigidus. In such cases, however, the elevation of the first metatarsal may be secondary to the severe contracture of the intrinsics and retraction of the volar plate, rather than a primary condition. Although osteotomy may be required for correction of primary or congenital metatarsus elevatus, one has to be careful with the notion that an osteotomy of the metatarsal is routinely necessary to alleviate dorsal impingement from hallux rigidus. Clearly, certain deformities will benefit from an osteotomy, for example, a long first metatarsal or one that is abnormally elevated. Other deformities require more care with decision making about the corrective procedure. Regardless of the extent of its deformity, an arthrodesis of the MTP joint will not be successful in a patient with a fixed elevated first metatarsal and hyperextension of the hallux IP joint. The result will be only to create additional load on the IP joint, ultimately causing pain with further subluxation and extension. The hallux will have to be cocked up significantly to position the arthrodesis to unload it from the plantar weight-bearing surface. This cocked-up position in turn will cause rubbing of the tip of the hallux on the shoe. Note that in Fig. 6.2 , the patient had already undergone an unsuccessful cheilectomy. In the standing position, the hallux is rigidly on the ground and the IP joint is hyperextended. Further attempts at passive dorsiflexion of the joint only worsened the IP hyperextension.

What is the condition of the sesamoids? Is there arthritis between the sesamoids and the metatarsal head? It is useful to perform a compression test or “grind test” by pressing under the sesamoids while attempting to dorsiflex the hallux. If such testing causes pain, even if radiographic degenerative changes are minimal, a cheilectomy may not work. A long metatarsal with the hallux in slight fixed flexion is associated with scarring and tightening of the sesamoid complex, and a cheilectomy will not work here either. In this latter situation, we would prefer to slightly shorten the metatarsal to take the pressure off the sesamoid apparatus and improve dorsiflexion.

Cheilectomy

Historically, cheilectomy has been used for patients with early- or intermediate-stage arthritis. An increasing trend over the past few years, however, has been to extend the indications for a cheilectomy to more advanced forms of arthritis. In clinical practice we have encountered many a patient who returns some years after a successful cheilectomy for treatment of hallux rigidus on the opposite foot. Radiographs of both feet typically demonstrate that the operated asymptomatic foot looks worse than the symptomatic foot. This finding may have something to do with denervation of the joint, but certainly, it is common enough that a cheilectomy may be performed for more advanced arthritis of the MTP joint.

An incision is made dorsomedial to the extensor hallucis longus (EHL) tendon extending for 3 cm over the MTP joint ( Fig. 6.3 ). The dorsal medial cutaneous branch of the superficial peroneal nerve must be avoided and retracted laterally. The capsule is incised, preserving a cuff of at least 5 mm medially for later closure. The capsule and periosteum are reflected off the metatarsal neck to expose the hypertrophic osteophytes dorsally. It can be difficult to expose the joint in the presence of large osteophytes, but the entire dorsal head must be exposed. Adequate exposure can be a problem in the foot with a large medial eminence as well as hallux rigidus, in which case the exostectomy will need to be performed with preservation of as much of the capsule medially as possible for closure. Alternatively, a medial incision with a medial capsulotomy can be used to approach cheilectomy in the presence of hallux valgus.

We prefer to use a chisel or oscillating saw to remove the dorsal apical surface of the metatarsal head, because this gives better control than that possible with an osteotome. The chisel is placed in the center of the metatarsal head, and one-third of the dorsal surface of the metatarsal head is removed ( Fig. 6.4 ). The chisel or saw should be angled slightly dorsally to avoid notching of the metatarsal neck. At this point in the procedure, it always seems that too much of the metatarsal head is being removed, but the amount of bone that should be removed is almost always underestimated. When the head is viewed from above, one-third of its volume seems like a large amount of bone to resect until an intraoperative radiograph is obtained, whereupon how little has actually been removed becomes evident. If there is any concern, then a Kirschner wire (K-wire) can be used to mark the level of the ostectomy prior to resection and verified with fluoroscopy, however, we rarely find this necessary. The ostectomy must be performed from distal to proximal, with removal of the dorsal osteophytes, and then the medial and lateral margins of the metatarsal head are contoured. If cysts are present in the metatarsal head, the ostectomy can be performed just dorsal to the erosion, or the head drilled with a K-wire, which may improve the fibrocartilaginous surface ( Fig. 6.5 ). Rounding off of the metatarsal head is performed using a rongeur and chisel, but care is taken not to dissect too far proximally. If the marginal osteophytes need to be removed, this can be done with the chisel, but again, it is essential not to go too far proximally on the lateral aspect of the head, which can result in avascular necrosis. The capsular attachment to the medial aspect of the first metatarsal head should be left intact. Range of dorsiflexion of the MTP joint should be at least 65 degrees after the cheilectomy. Removal of the dorsal phalangeal osteophytes should be performed, followed by the use of bone wax on the raw cancellous bone. Palpation of the skin should reveal that no further bony prominences are present and should be performed before closure. If failure to achieve sufficient dorsiflexion is noted, blunt release of the metatarsosesamoid complex can be performed with a curved gouge and Chandler type retractor. Closure of the capsule should be performed with absorbable suture, noting that distal capsule over the phalanx may not be possible to close because of the thin nature of this tissue. Weight bearing in a flat postoperative shoe, with initiation of aggressive ROM exercises at 1 week, will maximize the ROM for the patient. Expectations of swelling and discomfort for 3 months should be discussed, as patients may assume a much quicker rehab given the simplicity of the operation.

Osteotomy of the Proximal Phalanx (Moberg Osteotomy)

Osteotomy of the proximal phalanx—the Moberg osteotomy—is an easy operation to perform, with a predictable outcome. The hallux is dorsiflexed approximately 10 degrees off the floor. This operation does not increase ROM of the hallux, but simply facilitates clearance of the hallux so that at the starting point, the MTP joint is already in slightly greater dorsiflexion. We use this operation frequently, mostly for grade II arthritis. It is useful in cases in which increased elevation of the hallux off the floor is desirable ( Fig. 6.6 ). In patients with combined hallux rigidus and mild hallux valgus, a biplanar phalangeal osteotomy is performed to adduct and dorsiflex the hallux simultaneously, combining an Akin with a Moberg procedure. The surgery is usually performed in the setting of hallux rigidus, in conjunction with a cheilectomy, and the incision is simply extended more distally over the base of the proximal phalanx. The EHL must be retracted laterally and protected completely during the osteotomy.

The dorsal aspect of the cortex must be well exposed, and two sets of pilot holes are now inserted into the dorsal surface of the proximal phalanx. These are made obliquely at a 45-degree angle with respect to each other. The first set is made just distal to the articular surface and a second set approximately 1.5 cm more distally. These are unicortical pilot holes to be used for later suture fixation, and the osteotomy is planned in between these holes. A 1.5-mm slice of bone is removed with a saw. Once the bone wedge is removed, the base of the osteotomy is quite a bit more than 1.5 mm because of the width of the saw blade, and the osteotomy wedge must therefore be limited to prevent a cock-up deformity. A margin of 2.0 mm must be maintained on either side of the predrilled holes after the osteotomy, to prevent fracture through the hole and loss of fixation. The plantar cortex of the osteotomy is maintained intact, and a greenstick-type fracture of the osteotomy is created by first plantar flexing and then dorsiflexing the phalanx to completely close down the osteotomy. We open the osteotomy first using an osteotome, which loosens the plantar cortex but not the periosteal hinge. The osteotomy is secured with two sutures introduced through the predrilled holes using a curved tapered needle that fits the contour of the holes. These sutures will provide excellent stability, and therefore screw, wire, or plate fixation is not necessary ( Fig. 6.7 ).

Interposition Arthroplasty

Indications

Interposition arthroplasty is a good procedure that reliably increases the ROM of the MTP joint. Regardless of the technical aspects of this procedure, the interposition of soft tissue is a good concept. This technique has been described using autogenous as well as allogeneic tissue, harvested either locally from the dorsal MTP joint or from other adjacent autogenous tissue. In general, we use a turndown soft tissue flap from the dorsal metatarsal neck, but we also have created an interposition graft as a large “anchovy” rolled up into a ball, which is then sutured in place into the joint. Either allograft or autograft tendon is suitable for this purpose. Allograft has also been used as an interposition that is used to cover the metatarsal head and act as a new articular surface.

We use interposition arthroplasty preferably as a salvage procedure when the joint is severely distorted or eroded from prior surgery, avascular necrosis, or cyst formation from a previous implant arthroplasty. In addition, in patient who have pathology of the IP joint (arthritis or arthrodesis) or have a first tarsometatarsal (TMT) fusion, maintenance of motion of the MTP is critical to maximize their function, and an interposition arthroplasty is advocated in those cases as well. In general, an interposition arthroplasty is contraindicated in patients who already have a short hallux, short metatarsal, or adjacent metatarsalgia. Clearly, some weakening and obvious shortening of the hallux will occur as a result of this operation. Regardless of how the procedure is performed, plantar flexion strength is compromised.

Technique

Autograft

The incision is made dorsomedial to the EHL tendon and extends over the MTP joint for approximately 3 cm. The dorsomedial cutaneous branch of the superficial peroneal nerve must be identified and retracted. Once the dissection through the subcutaneous tissue is complete, the extensor retinaculum is cut approximately 5 mm medial to the EHL tendon to maintain an adequate cuff of tissue for later closure. The EHL is retracted, exposing the extensor hallucis brevis (EHB) tendon as well as the dorsal soft tissue and capsule over the metatarsal neck. This exposed tissue is now cut transversely as far proximally at the level of the metatarsal neck as one thick layer. The entire flap is now gradually mobilized and should include the periosteum, the EHB tendon, and the dorsomedial and dorsolateral aspect of the capsule. The flap is gradually dissected sharply off the dorsal osteophytes toward the base of the proximal phalanx ( Fig. 6.8 ).