Hallux rigidus

Hallux rigidus is defined as a localized degeneration of the hallux MTP joint. It was first described as hallux flexus in 1887 by Davies-Colley.¹¹ In his first description of this condition, he discussed a plantarflexed posture of phalanx relative to MT head. The actual term “hallux rigidus” was coined by Cotterill in 1888 and remains the most common term used today.¹² Numerous papers have theorized the etiology and pathophysiology of hallux rigidus. One such theory is that of metatarsus elevatus, a term describing the dorsiflexed posture of the first ray in relationship to the foot and the subsequent plantarflexed posture of the hallux. This has been discussed by many authors, but the most current data indicate that the elevated posture of the first MT improves after dorsal decompression of the hallux MTP joint.^13–16 Overuse and repetitive dorsiflexion forces, such as those occurring in a runner or kicker, may lead to chondral lesions and other occult injuries¹⁷ or to osteochondritis dissecans.^2,18,19 It also may result as a sequelae to a turf-toe injury. Anatomic abnormalities that may lead to hallux rigidus include the flat or pronated foot,^15,20,21 a long first MT or hallux,²¹ and a flat MT head.²² To this time the true potential etiologies for the development of hallux rigidus remain in question.

Clinical grading from mild to severe (or I, II, and III) has been proposed by many authors. Grading depends on the severity of disease and is based on ROM, pain or crepitus with motion, the size of the dorsal osteophyte on the MT head, the presence of sesamoid involvement, and the radiographic alignment of the hallux (on anterior-posterior [AP] and lateral views). A radiographic classification scheme was created by Hattrup and Johnson in 1988.²³ Their grade 1 is considered mild; the joint space is maintained and there is minimal spurring. Grade 2 is moderate disease in which the joint space is narrow, bony proliferation is present on the MT head and phalanx, and there is subchondral sclerosis and/or cyst formation. Grade 3 is the severe type, with significant joint space narrowing and extensive bony proliferation that involves the entire periphery and includes loose bodies, a dorsal ossicle, or subchondral cyst formation. A more recent grading scheme, proposed by Coughlin and Shurnas,²⁴ combines objective and subjective clinical data with radiographic findings (Grades 0 to 4). Treatment recommendations are made on the basis of grade severity.

Symptoms with which the typical athlete may present include pain that is worse with push-off and more severe after increased activity (i.e., twice-a-day practice regimens), as well as swelling. Although there may be bilateral radiographic involvement, the patient almost always presents with unilateral symptoms. Swelling and the bony prominence itself may interfere with athletic shoewear (especially in soccer and football, sports in which athletes prefer tightly fitting shoewear). A dysesthesia in the dorsomedial cutaneous nerve can result from tight shoewear’s impinging on the bony prominence. Occasionally transfer lesions may develop. This presents as metatarsalgia secondary to the lack of hallux dorsiflexion, causing increased pressure on the lateral forefoot.

In treating hallux rigidus in the athlete, one must consider the sporting activity and position played (i.e., a lineman who requires little hallux MTP dorsiflexion vs. a running back or wide receiver), shoewear requirement, and ROM of the entire foot and ankle. Even more minor or early-presenting cases can be problematic because some athletes create more forceful dorsiflexion, which can limit the function of the runner and incapacitate the dancer. Also important is the fact that if a bad joint is provided more motion, it may hurt more and degenerate more quickly.

Nonoperative treatment options include the use of nonsteroidal anti-inflammatory drugs (NSAIDs) and shoewear modifications. Shoes of adequate size and a more full-fitted toe box or increased depth are helpful and can be modified further with a balloon patch over bony prominences. Turf-toe inserts (Springlite, Otto Bock, Minneapolis, MN) that limit dorsiflexion and subsequent dorsal impingement are potentially useful but may limit performance in the elite runner. Rigid rocker soles function in the same manner as semirigid inserts and, although helpful in the general population, are not popular with the athlete because of the increased weight and excessive stiffness. Orthotic devices can unload the hallux MTP joint, but one must remember to increase the shoe size to accommodate for it. Taping techniques can limit dorsiflexion and provide pain relief. Application is the same as that for turf-toe; however, skin problems such as blistering can occur. Steroid injections must be given judiciously and perhaps only for “big game” situations. Repeated injections may accelerate the degenerative process.²⁵

Surgery in the management of hallux rigidus is feasible, and there are many options. The decision to proceed with surgery requires a lengthy discussion with not only the athlete but the trainer and possibly the athlete’s agent. It must be emphasized to all parties that this is an arthritic process, there is no “cure,” and there is the potential for a lengthy rehabilitation with incomplete resolution of the symptoms. The physician must determine the following: What is causing the problem? Is it the bony prominence over the MT head and secondary shoewear irritation? Is there limited ROM? Are there biomechanical implications such as poor push-off? Does the athlete suffer from transfer pain issues and other compensatory problems? Lastly, and most concerning, is there the presence of global pain and diffuse arthritis, especially in sesamoid-MT articulation?

The most commonly performed surgical procedure in the management of hallux rigidus is a cheilectomy. This procedure can be defined in general as an excision of an irregular osseous rim that interferes with motion of a joint. In this particular instance it is the removal of the dorsal osteophyte of the MT head. As noted previously, the athlete should be counseled that the underlying condition is degenerative joint disease and that full symptom relief is not realistic. A cheilectomy may prolong the athletic life of the individual but probably does not slow the rate of joint degeneration. As a general rule, the dorsal ridge does not recur, but progressive narrowing of the joint is expected to occur.

Indications for a cheilectomy include a lateral radiograph showing that reasonable space exists in the plantar one half of the MTP joint. There should be an absence of pain or crepitus with midrange motion and no sesamoid-MT pain or disease. This procedure allows for complete relief of dorsal impingement. It increases dorsiflexion by decreasing bulk of joint and subsequently relieving dorsal impingement pain. It also eliminates the source of painful shoe pressure. The true advantage of the cheilectomy is that “no bridges are burned,” and even in unsuccessful cases a salvage procedure is still technically possible.

The technique has been described and popularized by Mann and Clanton.²⁶ Their preference is a dorsal longitudinal incision centered over the hallux MTP joint. The joint capsule is incised on either side of the extensor hallucis longus (EHL) tendon and a complete synovectomy is performed. The joint is plantarflexed to permit inspection of the sesamoid articulation. Hamilton²⁷ recommends mobilizing the sesamoids by blunt dissection, for they often are anchored by adhesions and limit dorsiflexion even after removal of impinging osteophytes. The amount of bone to be removed from the MT head is dictated by the size of the dorsal exostosis and the degree of articular cartilage destruction. If degeneration of articular cartilage is not significant and the main problem is the dorsal exostosis, then 20% to 30% of the dorsal aspect of the MT head is removed along with the exostosis (see Fig. 18-2).

It is reasonable to be relatively aggressive with this resection, removing up to one third of the dorsal head to achieve improved motion. The cheilectomy should include removal of all osteophytes and a rounding of the MT head. The cheilectomy should achieve a minimum of 70 to 80 degrees of dorsiflexion because approximately one half of this will be lost in the postoperative period as a result of scar formation. It is Mann’s recommendation that if insufficient dorsiflexion is achieved after cheilectomy, then a proximal phalangeal osteotomy (Moberg) should be performed as described later.

We have modified the cheilectomy technique through a medial approach. This allows for plantar debridement and release of plantar capsule and adhesions, thus improving dorsiflexion. In addition, the incision avoids the EHL tendon and the potential for tenodesis secondary to scar formation while still providing access to lateral osteophytes. We recommend a two-cut technique to avoid excessive resection of the MT head (Fig. 18-3). The first cut of the saw includes the dorsal exostosis and is made flush with the dorsal diaphysis. The subsequent cut removes the amount of articular surface necessary to achieve the desired dorsiflexion while eliminating the risk of excessive head removal that may jeopardize later arthrodesis.

Figure 18-3 The first cut of the saw includes the dorsal exostosis and is made flush with the dorsal diaphysis. The subsequent cut removes the amount of articular surface necessary to achieve the desired dorsiflexion while eliminating the risk of excessive head removal that may jeopardize later arthrodesis.

Rights were not granted to include this figure in electronic media. Please refer to the printed book.

From Adelaar RS, editor: Disorders of the great toe, Rosemont, IL, 1997, American Academy of Orthopaedic Surgeons.

Hamilton²⁷ describes “radical cheilectomy” similar to the cheilectomy of Mann but also removing the dorsal portion of the base of the proximal phalanx, matching the resection performed on the MT head. This modification serves as an option for dancers with end-stage disease and is similar to the Valenti^28,29 procedure described later in this chapter.

A cheilectomy affords a fairly rapid postoperative course and return to activity. The patient is allowed to weight bear immediately, typically in a rigid-soled healing sandal. ROM can be initiated by a therapist or trainer as soon as pain allows but not so aggressively as to create wound dehiscence. Sutures generally are removed at 10 days, at which time active and passive ROM should be conducted at least three to four times per day. Close monitoring is required to ensure that the motion within the hallux MTP joint is at a functional level, a minimum of 40 degrees of dorsiflexion. No significant athletic activities generally are allowed for 6 to 8 weeks following a cheilectomy, giving the joint time to mature following surgery. Athletes can continue to train by bicycling, swimming, running in water, and engaging in other activities that avoid significant impact against the MTP joint. The patient should appreciate that swelling may continue for many months but that maximal motion usually is achieved by 3 months.

A number of authors have provided their results of cheilectomy. Mann and Clanton²⁶ found that 22 of 31 patients had complete relief, 6 of 31 achieved considerable relief, and ROM increased an average of 20 degrees in 23 of 31 feet. Hattrup and Johnson³⁰ reported that 53.4% were satisfactory and 27.6% unsatisfactory. Their failure rate increased from 15% with grade I radiographic changes to 37.5% with grade III changes. They concluded that cheilectomy is the procedure of choice in patients with hallux rigidus and grade I changes. Graves’³¹ experience showed little improvement in motion and stated that satisfaction with cheilectomy was more likely if the patient and the physician had reasonable expectations regarding outcome. He recommended careful patient selection. Myerson agreed that the procedure improves pain, not motion. Easley et al.³² reported on 57 patients (75 feet) with greater than 3-year follow-up (average 63 months). Their cheilectomy was performed via a medial approach by a single surgeon. American Orthopaedic Foot and Ankle Society (AOFAS) scores were 45 preoperative, 85 postoperative, and 90% satisfied. The average dorsiflexion improved from 19 degrees preoperative to 39 degrees postoperative. The majority of patients had worsening of radiographic arthritis, but this did not correlate with symptoms. Three patients eventually required an arthrodesis.

Phalangeal osteotomy has been advocated as a useful surgical adjuvant to a cheilectomy. This technique was first proposed by Bonney and Macnab in 1952.¹³ Kessel and Bonney¹⁹ described its use in 10 adolescents in 1958. Moberg is the name most commonly associated with the procedure, after his case series reported in 1979.³³ The procedure involves a dorsal closing wedge osteotomy of the proximal third of the proximal phalanx. It relies on the principle that the arc of motion of the hallux MTP joint is translated to plantar aspect of head, thereby increasing functional motion. Basically it creates pseudodorsiflexion, which in turn places less stress on the hallux with push-off. Adequate plantarflexion of the joint is a prerequisite. Thomas and Smith³⁴ also found that the procedure appeared to provide dorsal joint space decompression, as well, further relieving stress from the arthritic joint (Fig. 18-4).

Figure 18-4 Space created by dorsiflexion osteotomy of the proximal phalanx.

Rights were not granted to include this figure in electronic media. Please refer to the printed book.

From Thomas PJ, Smith RWL: Foot Ankle Int 20:4, 1999.

The indications for performing a Moberg osteotomy on the proximal phalanx includes grade I or II hallux rigidus, adolescent hallux rigidus, and the running athlete, perhaps regardless of grade. Most authors now recommend combining the procedure with a dorsal cheilectomy.^32,34,35

The technique can be performed through a medial or dorsal incision, extending distally from the incision used for the cheilectomy of the hallux MTP joint. It is important to protect the dorsomedial and plantar medial cutaneous nerves to limit paresthesia and the potential for neuritis or neuroma. Longitudinal reflection of soft tissues at the proximal third of the phalanx is performed, maintaining capsular insertion. The FHL and EHL tendons are protected as a dorsal closing wedge osteotomy is performed with a microsagittal saw approximately 3 to 5 mm distal to the MTP joint. In the adolescent, it is necessary to avoid the physis. Intraoperative fluoroscopy can be useful in confirming proper position of the osteotomy. The plantar cortex is maintained to allow for a “greenstick” effect with manual closure of the osteotomy. Generally 2 to 6 mm of dorsal cortex should be removed, with the actual amount determined by the degree of joint stiffness and amount of plantarflexion of the hallux available. The goal is to obtain 20 to 30 degrees of dorsiflexion relative to the first MT axis. The osteotomy should be stabilized with a suture, K-wire, screw, or staple. If combined with a cheilectomy, stable, internal fixation is mandatory to allow for the initiation of early motion (Fig. 18-5, A and B).

Figure 18-5 (A) Dorsal cheilectomy and dorsiflexion osteotomy of the proximal phalanx. (B) The amount of correction after fixation.

Rights were not granted to include this figure in electronic media. Please refer to the printed book.

From Thomas PJ, Smith RWL: Foot Ankle Int 20:4, 1999.

The postoperative care is similar to that described for an isolated cheilectomy. Immediate full weight bearing is permitted in a hard-soled sandal, with passive dorsiflexion exercises begun at 1 to 2 weeks. In ranging the joint it is important to hold the entire toe as single unit. Plantarflexion exercises are delayed until 3 to 4 weeks postoperative. When a pin is present, removal is performed at 4 to 6 weeks, followed by transition to accommodative shoes.

Published results of the proximal phalanx osteotomy include Moberg’s review of older individuals at short follow-up. Eight patients were noted to have satisfactory results. Citron and Neil³⁶ evaluated 10 feet in 8 patients with 22-year follow-up (minimum 10 years) and identified 5 symptom free, others with progression of degenerative joint disease (DJD), and one requiring arthrodesis. The average postoperative motion was 43 degrees, 22 degrees being dorsiflexion, with late loss of plantarflexion noted. Asymptomatic compensatory hallux IP flexion contracture often was present. They felt that this osteotomy represented an especially good option in the adolescent. Thomas and Smith³⁴ performed the osteotomy with a dorsal cheilectomy in 27 feet, 20 patients. At a follow-up average of 5.2 years, there was a 100% union rate, the average dorsiflexion increased 7 degrees, and 96% of patients were satisfied or satisfied with reservation.

Complications of the Moberg osteotomy include nonunion or malunion, a problem avoided by using internal fixation and “greensticking” the plantar cortex to avoid gross instability. Injury to the FHL and EHL tendons can occur, as can neuritis or neuroma, although the latter typically is transient. The possibility of progressive arthritis of the hallux MTP joint is an outcome that must be discussed with the patient preoperatively. Decreased push-off power can occur and may be of concern in the athlete or dancer.

Salvage for advanced degeneration or for a failed cheilectomy or osteotomy includes either arthrodesis or arthroplasty. Arthrodesis is best avoided in the “sprinting” athlete or dancer. If an arthrodesis must be performed, the toe tip should be at least 10 mm off the ground. Failure to meet this requirement will place significant stress on the distal hallux and IP joint. Slight shortening of the hallux also is of benefit, further lessening the potential of the athlete’s having to “vault” over the hallux during running activity.

Resection arthroplasty, like that of a Keller, is reserved for the older individual. Capsular interposition is a modification of this procedure devised by Hamilton.^37,38 In this procedure the proximal 5 to 10 mm of proximal phalanx is resected, followed by transection of the extensor hallucis brevis (EHB) tendon and dorsal capsule. This dorsal soft-tissue complex then is advanced to the plantar complex. Some authors release the flexor hallucis brevis (FHB) tendon from the base of the phalanx and suture this to the dorsal capsule. Temporary pin fixation is not necessary (Fig. 18-6, A and B). Our own experience with the procedure has noted good relief of pain from dorsal impingement and joint degeneration but a concerning loss of push-off strength. Similarly, the Valenti^28,29 procedure is a salvage technique in which an angled resection on both sides of the joint is performed, preserving the plantar complex and overall length. The result is a “hinge” effect at the level of the joint (Fig. 18-7).

Figure 18-6 (A) Interposition arthroplasty as described by Hamilton. (B) Pin fixation is not necessary.

From Hamilton WG, Hubbard CE: Foot Ankle Clin 5:663, 2000.

Figure 18-7 Resection of the dorsal metatarsal head as well as dorsal proximal phalanx.

From Coughlin MJ, Mann RA, editors: Surgery of the foot and ankle, ed 7, St Louis, 1999, Mosby-Year Book.

Most recently, an “anchovy” interposition of the hallux MTP joint has been performed in those individuals failing a cheilectomy but needing to maintain hallux MTP motion. Conical resection on both sides of the joint is followed by insertion of a semitendinosus allograft rolled into an “anchovy.” We have used this technique on three patients, one a professional football player, with good short-term results. Coughlin and Shurnas³⁹ recently reported on their experience with this technique in seven patients with excellent results. This case series demonstrates that this is a good surgical option in patients who otherwise would be treated with MTP arthrodesis.

Implant arthroplasty has been advocated by some authors; the options described include a silastic double-stem hinge, titanium hemiarthroplasty, or total toe replacement. These implants are unlikely to hold up in the running athlete, and the surgeon is faced with a difficult revision should failure occur. It remains our recommendation to avoid this procedure in the athlete, career or recreational.

Arthroscopic intervention for disorders of the hallux MTP joint has received some attention over recent years. It has been shown to be more of a diagnostic modality than a therapeutic one but may be a reasonable option for the removal of small dorsal osteophytes or loose bodies. It also may be used for debridement of an osteochondral defect on the MT head but is not indicated in advanced hallux rigidus. Van Dijk et al.⁴⁰ performed a prospective study with 24 athletes and found that it was not favorable for hallux rigidus because of “scar fibrosis.”