Hallux rigidus is a common entity encountered by the foot and ankle surgeon. It is estimated that 1 in 45 adults over the age of 50 years suffer from hallux rigidus and is the most common arthritis encountered within the foot.1 There are causes of hallux rigidus as the condition can be insidious, traumatic, or postsurgical in nature. Root et al2 provided the most recognized and accepted theory concerning the development of hallux rigidus. They described the condition as multifactorial with 6 specific components: hypermobility of the first ray, metatarsus primus elevatus, degenerative joint disease, trauma, a long first metatarsal, and first ray immobilization.
Symptomatic hallux rigidus begins with mild pain and discomfort to the first metatarsophalangeal joint with limitation of motion. As the condition progresses there is increased pain and discomfort to the joint, especially with activity, in correlation with further decreased range of motion of the joint. With progression of the disease, the first metatarsal head is unable to plantarflex resulting in dorsal abutment of the first metatarsal head on the base of the proximal phalanx of the hallux. Over time spur and osteophyte formation develops along the periarticular ridge of bone to the first metatarsal head and proximal phalanx. This eventually leads to chronic inflammation of the joint and associated symptomatic degradation of joint cartilage. End-stage disease is characterized by a severe lack of motion of the big toe secondary to progressive wear of the joint cartilage and subsequent decreased joint space.3–5
The treatment of hallux rigidus can be separated into 2 generalized categories, conservative/nonsurgical management or surgical management. With either option, the goal of intervention is to manage pain and stop progression of the disease. Conservative care options include wide extra-depth supportive stiff soled shoe gear, over-the-counter turf toe type inserts, custom orthotics with a built-in Morton’s extension, icing, oral or topical anti-inflammatory medication, physical therapy, and injectable medication such as cortisone, viscosupplementation, or an amniotic or placental human connective tissue matrix injection.6–11
There have been numerous grading/staging classification systems to help measure and grade hallux rigidus with Regnauld, Hattrup and Johnson, and Roukis most frequently referenced.8,12–14 The Hattrup and Johnson classification is most commonly used. They described 3 stages of the disease: stage I, mild to moderate formation of osteophytes with no joint involvement; stage II, moderate osteophyte formation, joint space narrowing, and subchondral sclerosis; and stage III, increased osteophyte formation and loss of joint space.8,12 Roukis was the first classification system applied prospectively and expanded on Hattrup and Johnson by adding stage IV, less than 10° of range of motion and loose bodies with obliteration of first metatarsophalangeal joint space.13,14 Dillard et al15 conducted a study to evaluate intraobserver and interobserver reliability using these 3 classification systems. The authors concluded that the Roukis classification system had the best intrarater reliability, although all 3 of these classification systems showed reliability and reproducibility. Grading/staging is important as particular surgical procedures have been shown to more appropriately address hallux rigidus in particular stages/grades of disease.
There are 2 generalized categories for surgical procedures for hallux rigidus, joint sparing procedures and joint destructive procedures. Surgical treatment of hallux rigidus is predicated on recognition of the condition of the joint as one that is still salvageable through primary joint reconstruction or one that would be more appropriately treated with a joint destructive procedure.16–19 A joint sparing procedure allows for maintenance of the cartilaginous surfaces within the first metatarsophalangeal joint and is recommended for surgical management of early-stage hallux rigidus. A joint destructive procedure eliminates the joint construct of the first metatarsophalangeal joint and is considered for cases of later-stage disease.3,14 Examples of joint sparing procedures include a “cleanup-type” cheilectomy or a decompressional (periarticular) osteotomy of the first metatarsophalangeal joint. There are multiple variants of decompressional osteotomy procedures with Mau (Figure 5.1), Youngswick (Figure 5.2), or Waterman-Green (Figure 5.3) identified as 3 of the most common. Such procedures are intended to decompress the first metatarsophalangeal joint through axial shortening of the first metatarsal and to perform concomitant plantar translation of the first metatarsal head.16–18 Joint destructive procedures include implant arthroplasty, Keller arthroplasty, and first metatarsophalangeal joint arthrodesis.
For purposes of the chapter, we focused on implant arthroplasty, cheilectomy, decompressional osteotomies, and Keller arthroplasty as they are common and popular treatments for the initial surgical management of hallux rigidus.16–18 The goal of these procedures is pain reduction and increased function. The cheilectomy procedure is the most often utilized procedure in stage I hallux rigidus, whereas a decompression osteotomy is reserved for stage II.3,6,8,19 However, a retrospective comparative study performed by Cullen et al13 demonstrated that, within the initial postoperative course, decompressional osteotomy resulted in a dramatically lower rate of revisional surgery compared with cheilectomy. Roukis performed 2 systematic reviews looking at outcomes for isolated cheilectomy20 and periarticular osteotomies.21 Although the rates of revision for periarticular osteotomy (22%) were significantly larger than for cheilectomy (8.8%), if the data were stratified only 3 of the periarticular osteotomy revisions were secondary to progressive joint destruction. Therefore, use of decompressional osteotomy procedures in stage I hallux rigidus may reduce the need for subsequent revision.
Implant arthroplasty and Keller arthroplasty are reserved and considered for late-stage hallux rigidus.3,11,14 Despite studies confirming that a first metatarsal phalangeal joint arthrodesis procedure provides more predictable outcomes compared with total joint replacement or hemiarthroplasty procedures,22,23 they are still widely used.
There are multiple types of implants based on implant design, shape, and material that fall into the category of implant arthroplasty procedures for the first metatarsophalangeal joint. The implant may be a partial or total joint implant. The aim of each of these implant procedures is to provide pain relief with also allowing for motion of the joint post implantation. Over the years there have been many changes to these implants; however, the classic implant arthroplasty performed for hallux rigidus continues to be a total silicone first metatarsophalangeal joint implant with or without grommets.24,25 Although attempts have been made using partial implants of multiple materials to the proximal phalanx or first metatarsal head, including joint resurfacing procedures, in an attempt to maintain motion but decrease the overall amount of bone resection, these procedures have not provided long-term success for the management of hallux rigidus. Failure of any of these implants may occur secondary to loosening, fragmentation, bone loss, or residual implant deformity. The joint destructive nature of these implants provides a major challenge for foot and ankle surgeons when revision is required. Resultant bone loss post implant necessitates a first metatarsophalangeal joint arthrodesis procedure with interpositional bone graft to revise these cases.22–26
In 2016, the United States Food and Drug Administration approved a polyvinyl alcohol hydrogel implant (Cartiva, Wright Medical Group N.V. Memphis, TN) for the treatment of hallux rigidus. The procedure involves a cheilectomy along with insertion of an 8- or 10-mm implant within the first metatarsophalangeal joint. The implant is meant to prevent the proximal phalanx base and first metatarsal head from abutting against each other, preventing pain and further joint breakdown. The ease of application, limited bone loss with implantation, and results of prospective multicenter randomized controlled trails27,28 increased the popularity of the procedure. Complications such as subsidence, loss of joint space, and cyst formation with revisional procedures after use of this implant are being identified with increased usage.29 It is the authors’ opinion that over time revisional hallux rigidus repair after failed polyvinyl alcohol hydrogel implant will become a more regular occurrence.
Although these surgical options can provide resolution of the patient’s pain and discomfort secondary to hallux rigidus, there are inherent advantages and disadvantages to each of the 4 procedures. Implant arthroplasty has the advantage of relieving pain with maintaining joint motion; however, fatigue of the implant is associated with synovitis, cystic formation, and lymph node inflammation. Transverse metatarsalgia due to increased load sharing and a cock-up hallux are other possible complications after implant arthroplasty.22–29 A cheilectomy procedure is nondestructive and a relatively simple procedure with a high level of patient satisfaction, especially when addressing dorsal bump pain. However, a cheilectomy procedure has the potential for spur recurrence, progressive joint space narrowing, wearing of cartilage, and scarring of soft tissue structures.2,3,7,11,12,20 The goal of decompressional osteotomy procedures is to improve the cubic content of the joint by shortening the first metatarsal; however, they have been shown not to provide long-term prevention of progressive joint breakdown. These procedures are prone to joint stiffness even with initial improvement of joint space and range of motion early in the postoperative phase.3,14,16–18,21 A failed Keller procedure results in a situation where there is a flail joint or cock-up hallux with bone loss to the first metatarsophalangeal joint and associated transfer metatarsalgia.30–32
With the progressive arthritic nature of hallux rigidus, patients may fail initial surgical management and require revisional surgical management. This chapter reviews revisional surgical management for the treatment of several types of failed surgical procedures for the correction of hallux rigidus. The need for salvage options that address the progression of hallux rigidus residual and any untreated deformities after failed hallux rigidus deformity correction is a necessity for foot and ankle surgeons.
There is a plethora of literature concerning initial surgical management for hallux rigidus; however, articles specifically dedicated to revisional surgical management for failed hallux rigidus deformity correction are less abundant. Most review salvage procedures for failed first metatarsophalangeal joint arthroplasty procedures (Keller, hemi-implants, or total implants) and recommend revisional first metatarsophalangeal joint arthrodesis with interpositional bone graft.32–42
Mao et al32 performed a meta-analysis concerning salvage arthrodesis for failed first metatarsophalangeal joint arthroplasty where they assigned grades to support specific salvage techniques. A total of 12 studies met the inclusion criteria, and ultimately 10 studies were used for analysis as 2 studies were review articles. The analysis looked at salvage arthrodesis with bone graft and salvage arthrodesis without bone graft. Their findings resulted in a grade B (fair) recommendation to support salvage arthrodesis with bone graft and a grade C (poor) recommendation to support arthrodesis without a bone graft; however, both techniques did provide good patient satisfaction and union rates.
Roukis published a pair of systematic reviews concerning the need for surgical revision after isolated cheilectomy and isolated Valenti arthroplasty for hallux rigidus.20,21 In the cheilectomy systematic review, 23 studies met the inclusion criteria. Surgical revision post cheilectomy procedure was required in 62 of 706 procedures for a revision rate of 8.8%. A total of 23 revisional procedures were identified; 3 first metatarsophalangeal joint arthrodesis, 13 interpositional arthroplasty for failed silicone implant arthroplasty, 3 revisions using Keller arthroplasty, and repeat cheilectomy for 2 revisions. There were 17 other revisional procedures that had no mention of the type of revision procedure performed. There were 12 studies where the grade of hallux rigidus at the time of initial cheilectomy procedure was provided: 103 (19.9%) grade I, 210 (40.6%) grade II, 189 (36.6%) grade III, and 15 (2.9%) grade IV. Furthermore, 6 studies provided the grade of hallux rigidus at the time of initial cheilectomy for those cases ultimately requiring revision. Revision numbers and rates were 2 (20%) grade I, 8 (14.8%) grade II, 12 (9.1%) grade III, and 5 (55.6%) grade IV. Although the overall revision rate was low, the systematic review did indicate a higher revision rate post cheilectomy procedure in those procedures performed on stage IV hallux rigidus. In the Valenti arthroplasty systematic review, 3 studies met the inclusion criteria. Of the total of 44 Valenti procedures identified, only 2 (4.6%) required revision, 1 via plantarflexory base osteotomy, the other via Keller resection arthroplasty. The studies did not provide any indication of hallux rigidus grading, including for those cases that required revision.