The Mitchell osteotomy was the first to use lateral translation of the first metatarsal head to correct the alignment of the first ray.
A step-cut was used to provide resistance to medial translation and the osteotomy was stabilised with a suture. Although Shereff showed that this gave very poor stability, some surgeons still use suture stabilisation. Shereff found a K-wire to be superior, and Calder subsequently improved on this with screw fixation, although most K-wires and many screws will need removal. Calder demonstrated in a subsequent RCT that screw fixation allowed early, unprotected weightbearing mobilisation without redisplacement, and this should probably be considered best practice at the present.
Mitchell recommended some shortening of the first metatarsal to relax the 1st MTP joint and allow easy correction of the deformity. To reduce the risk of defunctioning the first ray, he also removed a plantar-based wedge of bone (see operative technique). Not all modern surgeons seem to understand or follow these steps (although Madjarevic's excellent 20-year results were obtained in patients who had plantar displacement of the metatarsal head.
Mitchell recommended the removal of 3mm of bone in the step-cut. In his own incomplete series the average shortening of the 1st MT shaft was about 6-7mm(comparable with other series), and there was no significant difference in shortening between patients with post-operative metatarsalgia and those without. Not all series have agreed with this. Toth et al (using the Lindgren-Turan osteotomy) found a strong correlation between shortening and lateral ray pain.
Many series of Mitchell osteotomies have been reported. No RCTs have compared the Mitchell to any other procedure.
There have been three retrospective comparisons with the chevron osteotomy, with no significant clinical difference; one retrospective comparison with the Keller, in which the Mitchell group were more satisfied and had less metatarsalgia but more MTP joint pain; and one retrospective comparison with the McBride, in which the Mitchell group did better, but with vague outcome measures. Dhukaram (2006) compared both clinical and pedobarographic results of the Mitchell and scaf osteotomies retrospectively. The Mitchell group had lower AOFAS forefoot scores, more lesser ray calluses, more transfer loading of the 2nd and 3rd rays and more MTPJ pain.
A large series of 430 was reprted by Wu, but detailed clinical outcomes were reported for only 100. 85% had little or no deformity or clinical problems.
Two series have reported very long-term results. Fokter reported the results in 64 patients 15-24 years post-op. 64% still had satisfactory results, but this was down from 97% at 2-11 years. 39 had forefoot or 1st MTP joint pain, and 49 had recurrent deformity, which was worse than the original deformity in 14. Madjarevic (2005) reported the results of both Mitchell and Wilson osteotomies at 20-22y, tracing 58/77 patients. Both gave comparable correction of the hallux valgus and intermetatarsal angle. The Mitchell group had less 1st MT shortening (3.5mm compared with 5.4 mm). There were no clinical recurrences in the Mitchell group. Using the Bonney and McNab score, there were 29 excellent and 6 good results in 35 patients. This shows the need for long-term follow-up of hallux valgus procedures.
Overall, the Mitchell osteotomy is arguably the gold standard for distal procedures, in view of its good short to medium term results and the availablity of long-term outcomes.
It is therefore disappointing that no RCTs have compared the Mitchell to other procedures. In particular, one or more RCTs comparing it to the distal chevron procedure need to be done as a priority.