Nonunion after arthrodesis of the first metatarsal-phalangeal joint: a systematic review.

Arthrodesis of the first metatarsal-phalangeal joint (MTPJ) has been proposed for treatment of first MTPJ pathology because of the perceived safety and efficacy. Nonunion of the arthrodesis site has been purported as a common complication. The author undertook a systematic review of the electronic databases and other relevant sources to identify material relating to the incidence of nonunion and other complications after arthrodesis of the first MTPJ. In an effort to procure the highest quality studies available, the studies were eligible for inclusion only if they involved patients undergoing arthrodesis of the first MTPJ using modern osteosynthesis techniques (1980 onward time restriction), included a minimum of 30 feet in the publication, and evaluated patients at a mean follow-up of ≥12 months' duration. The studies were also required to include details of any complications requiring surgical intervention, had not primarily involved only rheumatoid arthritis as an indication, and had not involved the use of a structural bone graft. A total of 37 studies involving a total of 2,818 first MTPJ arthrodesis procedures were identified that met the inclusion criteria. The weighted mean age of the patients was 59.3 years, the follow-up was 34.3 months, and the union time was 64.3 days. For those studies that specifically mentioned the indications for first MTPJ arthrodesis, 2,656 joints were included as follows: severe hallux valgus (47.2%), hallux rigidus (32%), rheumatoid arthritis (11.5%), and revision of failed surgery (9.3%). Osteosynthesis involved 3 main configurations: compression screws, dorsal plate and screws, or staples. The overall nonunion incidence was 5.4% (153 of 2,818), with symptomatic nonunion occurring in 32.7% (50 of 153) of all nonunions (1.8% total incidence; 50 of 2,818). The overall incidence of malunion was 6.1% (39 of 640), with dorsal malunion accounting for 87.1% (34 of 39). The overall incidence of hardware removal was 8.5% (69 of 817). The historical comment that nonunion occurs in approximately 10% of attempted first MTPJ arthrodesis procedures is inaccurate. The incidence of malunion and hardware removal is inappropriately high, and efforts to determine effective methods of decreasing their incidence should be undertaken. Additionally, there is still a need for methodologically sound prospective cohort studies focusing on the use of arthrodesis of the first MTPJ for purely severe hallux valgus and specific grades of hallux rigidus, as well as specific forms of osteosynthesis, because this has only been infrequently reported in small series.