CORR Insights®: Have Surgery and Implant Modifications Been Associated With Reduction in Soft Tissue Complications in Transfemoral Bone-anchored Prostheses?

Where Are We Now?

Osseointegration for persons with limb loss refers to the direct, transcutaneous skeletal attachment of a prosthesis to bone. The advantages that derive from these technologies (compared with conventional, socket-based prostheses) include ease of use and good physical function, patient-reported outcome measures, and even improved quality of life [4, 8, 9]. Other putative benefits such as quantitative gait improvements, as well as novel concepts such as osseoperception and osseoproprioception, remain less-well proven but are areas of active investigation. Concerns regarding osseointegration for amputees include fracture, bone or residual limb length loss if removal should be required, surgical risks, and expense. However, infection is, and for the foreseeable future will remain, the principal concern about the broad adoption of osseointegration.

In this month’s Clinical Orthopaedics and Related Research®, Atallah et al. [2] provide additional insight regarding how such infections might be avoided. Namely, 40 patients were treated using a thicker, “conventional” soft tissue envelope around the skin penetration aperture in conjunction with a cobalt-chromium-molybdenum (CoCrMo) implant, and these patients were compared with 39 patients treated with a titanium implant and a modified, shallower (< 2 cm) aperture. The authors reported a dramatic (> fivefold) reduction in the risk of superficial infections at 2 years of follow-up, as well as fewer problems with redundant soft tissues at the skin penetration site, at the expense of a modest increase in infections between stages. This improvement in superficial infections compares favorably with the results of other recent studies [1, 4].

Currently, we know that superficial infections after transcutaneous osseointegration are expected and (usually) minor, and could to some degree be considered part of the natural history of this treatment modality; we would nonetheless like to prevent or minimize these superficial infections if we can. As I have written in another CORR Insights® commentary [8], soft tissue management strategies for osseointegration have gradually evolved toward a less-is-more approach, what my friend Dr. Jason Souza (a plastic and reconstructive surgeon and former colleague) has termed “trading blood supply for stability” (verbal communication, Jason Souza MD, 2022). The findings of Atallah et al. [2] provide objective evidence that, at least within the constraints of our current state of complimentary infection prevention and mitigation technologies, this trade is a good one.

Where Do We Need To Go?

First, we need to demonstrate (or, more likely, accept with time and at least some evidence) that these reductions in superficial infections and soft tissue redundancy near the aperture indeed were the result of the modified surgical technique rather than the change in implants. It remains controversial whether CoCrMo or titanium has greater resistance to bacterial colonization and biofilm formation [6, 7]. We also need to determine, if there is such a thing in this context, how little tissue is simply not enough (although I would argue there is not, as long as the skin aperture is stable and viable).

Learning curve is another confounder that is difficult to overcome; to the authors’ credit in this study, they acknowledged this limitation [2]. They likely were more facile at the pertinent techniques during the second 5 years of osseointegration practice than they were in the first. I will be the first to admit from my experience with several osseointegration implants that neither the bone or implant approaches nor the soft tissue handling required in these procedures are, at least not insurmountably, technically challenging [3]. However, I would also submit to you that I am way better at osseointegration surgery than I was when I placed my first implant several years ago. That being so, objectively determining how long, how steep, and how important the osseointegration learning curve is requires further study. As has been determined for other orthopaedic procedures such as anterior-approach THA [5], for these osseointegration procedures, we need to determine how great the difference is between novice and expert in terms of patient outcomes, and how much experience is required to bridge the gap.

Finally, we need to better determine the relationship between superficial infections (common), deep infections (uncommon), and implant loss (fortunately, less common still). And further, we need to explore the negative relationship, if any, between these expected superficial infections and patient-reported outcomes, functional measures, and quality of life.

How Do We Get There?

Because the most commonly used implants for osseointegration across all major centers are now titanium-based, I expect the first question (whether titanium or cobalt chrome is less prone to infection) to be answered by consensus and, perhaps, a gradual decrease in the reported superficial infection rates across multiple devices and centers. However, this question of material infection resistance likely will never be fully untangled from the evolution of soft tissue management strategies. Which is to say, I don’t think this question is likely to be definitively answered. Rather, we will accept that this is “what we are doing now,” much like the shift to predominantly titanium implants in THA [8].

The learning curve question should be easier to answer via a graduated analysis of results from individual centers. Did results for superficial infections and aperture-related complications decrease over time? When and where was the inflection point, if any, and was it anchored, to a greater degree, to time (that is, months or years of experience) or procedural volume? The strength of these findings and determining whether the learning curve differs for different implants and techniques would be greatly improved by registry participation. As mentioned in my previous CORR Insights article [8], an osseointegration registry already exists (for more information on registry participation, please email Osseointegration_registry@hjf.org). Moreover, because many surgeons and centers are now seeking to establish osseointegration programs (and for best results, these do need to be programs; it shouldn’t just be surgeons doing procedures), most of these programs will, by definition, be low-volume, low-experience-level endeavors at first. Will we see our collective results fall off a cliff? Only a registry can answer that question.

Finally, we need to clarify the relationship between minor complications (superficial infections), major complications (deep infections), and functional and patient-reported outcomes. No one would dispute that the clinical and functional outcomes of total joint arthroplasty are compromised by a deep infection, even in instances where the infection is “cured” and implants can be retained. A superficial infection in a patient who has undergone osseointegration is not a disaster, but it probably incrementally increases the likelihood that a deep infection will develop, especially if superficial infections are recurrent (not unlike Russian roulette). Despite a profound difference in superficial infection rates between groups, Atallah et al. [2] did not find a difference in the risk of deep infection between their implant and technique cohorts with the numbers available, but they did report an 8% risk of bone infection in the CoCrMo/deeper aperture cohort, which had a much greater proportion of patients with superficial infections, and no deep infections in their titanium and shallow aperture cohort. Furthermore, is there a difference in patient-reported outcomes between patients who experienced superficial infections (without a deep infection) and patients who never experienced any infection at all? To answer these questions, we need thorough, thoughtful reporting of prospective results from multiple centers, in addition to broad registry participation. If we are to learn and improve as much as we can, as fast as we can, every amputee with an osseointegration implant needs to be longitudinally followed.