First described in 1887, hallux rigidus refers to the symptoms of degenerative arthritis of the first metatarsophalangeal joint (MTPJ) . It is characterised by pain, loss of motion, primarily dorsiflexion and periarticular osteophyte formation. More common in females, hallux rigidus is thought to affect 1 in 45 individuals over the age of 60 years . Trauma is a frequently cited aetiological factor, especially for unilateral hallux rigidus . Generalised osteoarthritis and inflammatory arthropathies, such as rheumatoid arthritis, can also result in hallux rigidus. Irrespective of the causative factors involved, end-stage disease results in articular cartilage loss and loss of joint space. Subsequent pain and loss of motion results in abnormal gait patterns and can interfere with simple daily tasks such as walking and stair climbing . The severity of hallux rigidus can be classified using the system suggested by Coughlin and Shurnas . Early disease, grades 0 to 2, can respond well to non-operative measures such as non-steroidal anti-inflammatory medication and footwear modification. Cheilectomy to remove impinging dorsal osteophytes can also be successful in early cases as the disease initially affects the dorsal portion of the joint. End-stage disease, grade 3 to 4, does not, however, respond as well to such measures and remaining options include joint arthrodesis or arthroplasty .
First MTPJ arthrodesis is currently considered to be the gold standard for the treatment of end-stage hallux rigidus, with reported union rates ranging between 90% and 100% . Arthrodesis does, however, predispose the patient to interphalangeal joint osteoarthritis and the loss of dorsiflexion can interfere with activities such as kneeling and squatting, and wearing footwear with raised heels can be problematic [8, 9]. Post-operative rehabilitation can also be prolonged and there is up to a 10% risk of non-union, which may require a revision procedure . Thus as an alternative, arthroplasty of the first MTPJ has been attempted.
Silicone implants were initially used following their successful implantation in the hand. Early results were promising with high rates of patient satisfaction . Unfortunately, problems with silicone wear, osteolysis and foreign body reactions limited their longevity [11, 12]. Silicone has therefore been suggested to be a material not suitable for first MTPJ replacement due to its poor ability to withstand the forces encountered.
This led to the development of metallic implants in both total toe replacement and hemiarthroplasty designs . Total toe implants have demonstrated good early patient satisfaction . Early loosening, however, was once again a problem [14, 15]. This is thought to be due to the large dorsally directed shear forces the metatarsal implant is exposed to during toe-off .
Metallic proximal phalangeal hemiarthroplasty (PPH) may therefore avoid both problems related to silicone wear and loosening related to the shear forces exerted on metatarsal head implants. The literature on the results of these implants is, however, sparse and somewhat conflicting. Townley and Taranow  published the largest series of 279 PPHs with follow-up ranging from 10 months to 33 years. This retrospective review reported good or excellent results in 95% of subjects. More recently Sorbie and Saunders  published their series of 23 patients treated with PPH with follow-up ranging from 34 to 72 months. They report an improvement in AOFAS score from 57 to 88 with no cases of loosening or osteolysis. In contrast Raikin et al.  found inferior results for PPH compared to arthrodesis. They conducted a retrospective review of 48 feet, 27 with arthrodeses and 21 with PPHs. Of the arthrodesis group, 85% were deemed to have had a good outcome based upon patient-reported satisfaction, versus 60% in the PPH group, with 5 (24%) of the PPH group requiring re-operation due to loosening. They concluded arthrodesis is a more predictable technique for dealing with hallux rigidus.
The literature is therefore divided regarding the efficacy of PPH for the treatment of hallux rigidus. Studies to date have all been retrospective in nature. The only prospective randomised trial so far has compared total joint arthroplasty versus arthrodesis, and concluded outcomes for arthrodesis were better than arthroplasty . A similar prospective randomised study therefore is required to compare outcomes of PPH versus joint arthrodesis accurately. We therefore propose a randomised controlled trial comparing the AnaToemic PPH (Arthrex Ltd., Sheffield, UK) to first MTPJ arthrodesis. This prosthesis shares similar design features to the BioPro First MPJ Hemi Implant (BioPro, Michigan, USA) but has a better cost profile. Early results reported by Kissler and colleagues are promising, with good improvements in pain scores and revision rates comparable to non-union rates observed with arthrodesis .