What Are the Risk Factors for Mechanical Failure and Loosening of a Transfemoral Osseointegrated Implant System in Patients with a Lower-limb Amputation?

Abstract

Background: Septic loosening and stem breakage due to metal fatigue is a rare but well-known cause of orthopaedic implant failure. This may also affect the components of the osseointegrated implant system for individuals with transfemoral amputation who subsequently undergo revision. Identifying risk factors is important to minimize the frequency of revision surgery after implant breakage.

Questions/purposes: (1) What proportion of patients who received an osseointegrated implant after transfemoral amputation underwent revision surgery, and what were the causes of those revisions? (2) What factors were associated with revision surgery when stratified by the location of the mechanical failure and (septic) loosening (intramedullary stem versus dual cone adapter)?

Methods: Between May 2009 and July 2015, we treated 72 patients with an osseointegrated implant. Inclusion criteria were a minimum follow-up of 5-years and a standard press-fit cobalt-chromium-molybdenum (CoCrMb) transfemoral osseointegrated implant. Based on that, 83% (60 of 72) of patients were eligible; a further 3% (2 of 60) were excluded because of no received informed consent (n = 1) and loss to follow-up (n = 1). Eventually, we included 81% (58 of 72) of patients for analysis in this retrospective, comparative study. We compared patient characteristics (gender, age, and BMI), implant details (diameter of the intramedullary stem, length of the dual cone, and implant survival time), and event characteristics (infectious complications and distal bone resorption). The data were retrieved from our electronic patient file and from our cloud-based database and analyzed by individuals not involved in patient care. Failures were categorized as: (1) mechanical failures, defined as breakage of the intramedullary stem or dual-cone adapter, or (2) (septic) loosening of the osseointegrated implant.

Results: Thirty-four percent (20 of 58) of patients had revision surgery. In 12% (7 of 58) of patients, the reason for revision was due to intramedullary stem failures (six breakages, one septic loosening), and in 22% (13 of 58) of patients it was due to dual-cone adaptor failure (10 weak-point breakages and four distal taper breakages; one patient broke both the weak-point and the dual-cone adapter). Smaller median stem diameter (failure: 15 mm [interquartile range 1.3], nonfailure: 17 mm [IQR 2.0], difference of medians 2 mm; p < 0.01) and higher median number of infectious events (failure: 6 [IQR 11], nonfailure: 1 [IQR 3.0], difference of medians -5; p < 0.01) were associated with revision intramedullary stem surgery. No risk factors could be identified for broken dual-cone adapters.

Conclusion: Possible risk factors for system failure of this osteointegration implant include small stem diameter and high number of infectious events. We did not find factors associated with dual-cone adapter weak-point failure and distal taper failure, most likely because of the small sample size. When treating a person with a lower-limb amputation with a CoCrMb osseointegrated implant, we recommend avoiding a small stem diameter. Further research with longer follow-up is needed to study the success of revised patients.

Level of evidence: Level III, therapeutic study.

Fig. 1

This flowchart shows the included patients.

Fig. 2

A-F These drawings show the Integral Leg Prosthesis (Orthodynamics)/Endo-Exo (Eska Orthopaedic) osseointegrated implant, including (A) the intramedullary stem, (B) cross-section of the stem, (C) dual-cone adapter with the proximal part of the male part of the osseointegrated implant connector, (D) locking screw to affix the dual cone in the intramedullary stem, (E) distal part of the male part of the osseointegrated implant connector, and (F) the final propeller screw. A color image accompanies the online version of this article.

Fig. 3

A-D These radiographs from a patient with a broken stem show (A) stem breakage and (B) the removed distal part of the stem. (C) The stem was removed with a hollow drill. (D) This radiograph was taken 2 years after revision with a larger-diameter titanium implant (osseointegrated prosthetic limb).

Fig. 4

A-C (A) Stem breakage is shown. (B) This photograph shows a dual-cone adapter with breakage of weak points (indicated by the arrow). (C) This image shows breakage of the distal taper of the dual-cone adapter. A color image accompanies the online version of this article.

Fig. 5

A-B These photographs show a stoma (A) after removal of the broken part and (B) after healing. A color image accompanies the online version of this article.

Fig. 6

A-D (A) This radiograph shows signs of septic loosening. (B) The stem and gentamycin beads in the intramedullary space were removed. (C) This radiograph was taken after removal, and (D) this radiograph was taken at 2 years after revision.

Fig. 7

A-B These photographs show an osseointegrated implant connector (male and female part, left to right), including (A) an OTN connector (Xilloc Nexus BV, Netherlands) and (B) the OPL connector. A color image accompanies the online version of this article.

Fig. 8

A-B These radiographs of the femur stump of the same individual were taken (A) immediately postoperatively and (B) at 2 years of follow-up. At this point, there was more than 20 mm of distal bone resorption (indicated by the arrow).

Fig. 9

This Kaplan-Meier graph represents the two voluntarily removed intramedullary stems (95% CI 58% to 89%).