The osseointegrated (OI) prosthesis is a treatment option for transfemoral amputees with a short residual femur and/or difficulties caused by using the prosthetic socket. Implant removal due to aseptic or septic loosening is not uncommon, but the association between implant migration patterns and the need for removal has not previously been studied. We conducted a prospective model-based radiostereometric analysis study to investigate: if the OI implant migration pattern 1) differs between later removed implants and non-removed implants, (2) predicts later implant removal, and (3) if the precision of the method is acceptable.

Model-based radiostereometric analysis of the OI implant migration pattern can be used to predict later OI implant removal.

A prospective cohort of 17 consecutive transfemoral amputees suitable for surgery (11 males), mean age 50 (range 32-66) were treated with an OI implant (Integrum AB, Sweden). Postoperative stereoradiographs of the OI implant were obtained during 24-month follow-up. X, Y, and Z translations and total translations were evaluated using CAD-implant models. Implant survival was followed for up to 60 months.

Six total implant removals (fixture and abutment) and four partial removals (abutment) were conducted (10/17 (59%)), and one patient did not use the OI implant. The removed implants group migrated a mean (±standard deviation) 0.55mm±0.75mm (p=0.009) and the non-removed implants group migrated 0.31mm±0.51mm (p=0.22) in total translations from 3 months to last follow-up. Odds ratio for implant removal was 22.5 (95% CI: 1.6 to 314 (p=0.021)) if the OI implants migrated distally.

Later removed OI implants migrated from 3 months to last follow-up and more than the non-removed OI implants. Distal implant migration greatly increased the odds of implant removal. Ten out of 17 OI implants were removed within 5 years of follow-up. We advise to use OI implants with caution and close follow-up in consideration of the risk of complications.

IV, Prospective study.