Corrosion of the head-neck-taper interface in total hip arthroplasty has been associated with symptomatic adverse tissue local reactions, trunnion fracture, and elevated serum metal ions. The hypothesis of the study is that revisions of these cases have a worse outcome than metal-on-metal arthroplasties without corrosion and revisions of hip resurfacings; this study includes revisions performed from 2001 to 2014 in our institution. Seven patients with corrosion (73 ± 8 years), 32 patients with metal-on-metal revisions (55 ± 12 years), and 40 hip-resurfacing revisions (56 ± 12 years) were analyzed for the study. The time the implants stayed in situ was 5.7 ± 3.1, 3.7 ± 3.2, and 3.8 ± 2.7 years, respectively. The average follow-up period was 4.9 ± 3.6 years. The mean head sizes used for the corrosion group were 33 mm (28–36), 32 mm (28–44) for the metal-on-metal group and 49 mm (47–55) for the resurfacing group. Three patients of the corrosion group (43 %) had to be revised again because of recurrent dislocations and became constraint liners. Five (16 %) of the metal-on-metal group and 1 (2.5 %) of the resurfacing group required a second revision (mostly because of a planed two-stage procedure for the treatment of periprosthetic infections). All corrosion cases showed intraoperatively adverse tissue local reactions with tissue necrosis. The outcomes measured by the Harris-Hip score were also worse for the corrosion group (66 ± 13), followed by the metal-on-metal group (75 ± 19) and the resurfacing group (78 ± 19). Corrosion of the head-neck-taper interface is a rare but severe complication. Because of the high probability of recurrent dislocations postoperatively, we recommend the primary use of constrained liners for revision of corrosion cases. Further analysis, however, is needed to better understand and avoid this complication.