Rethinking the Surgical Treatment for Ankle Arthritis in Patients With Diabetes: Why Replacement May Outperform Fusion

For decades, surgeons (myself included) have reluctantly used ankle replacement as a surgical treatment for the management of ankle arthritis in patients with diabetes. My hesitation in offering these patients a replacement is attributed to concerns over increased wound complications, aseptic loosening, implant survivorship, and infection. ¹ As a result, diabetic patients in my practice are typically offered ankle fusion over replacement with the consensus that this is a safer and more reliable option; one that is more durable and that carries less risk of harm to the patient. Recently, the notion that ankle fusion is preferable to ankle replacement in a diabetic patient has been challenged in the literature.^{1 -4}

“Why are we focusing on ankle replacement in diabetics?” you may ask. I assume that the growing number of publications on the treatment of ankle arthritis in diabetic patients stems from the increasing incidence of diabetes, coupled with advances in total ankle replacement over the last few decades. Regardless of surgical choice, diabetes remains a significant risk factor for many surgical complications, yet several studies in recent years have demonstrated rates of complications in people with diabetes undergoing ankle fusion to be high, too.^4,5 As it turns out, the “reliable” surgical choice of ankle arthrodesis may not be as safe as once believed.

Meanwhile, the complication rates of those with diabetes undergoing total ankle replacement are only marginally higher than those of patients without the disease, especially with tight perioperative glycemic control. ² Yet, there are very few studies that directly compare outcomes following ankle replacement or fusion in diabetic patients.

In this edition of Foot & Ankle International, the authors of “Total Ankle Arthroplasty vs Arthrodesis in an HbA1c‑Defined Diabetic Cohort: Lower 1‑ to 5‑Year Structural Complication Rates in a Retrospective Propensity‑Matched Study” have published an excellent manuscript that evaluates outcomes of ankle arthrodesis and ankle replacement in people with diabetes. ⁶ Using the TriNetX Research Network, a retrospective cohort database comprising deidentified data from more than 100 institutions, the authors identify 255 persons with diabetes (HbA_1c > 6.5) in each group who underwent ankle replacement or fusion. Primary outcome measures of the study included infection rates, readmission, and structural complications. Interestingly, the average HbA_1c was higher than 7 in the arthroplasty and arthrodesis groups (7.1 vs 7.3, respectively)!

In this study, ankle fusion had a higher rate of structural complications than ankle replacement at all time points, including 39.1% for arthrodesis at 5 years compared with 21.2% for ankle replacement that was statistically significant. Infection rates at 1 year for ankle replacement were only 5.5% compared with 9.4% in ankle fusion, although these were not statistically significant. Lastly, admission rates were higher for ankle fusion than for replacement at all time points.

Now, I am sure many naysayers are questioning the results presented herein. I admit, I am skeptical about indicating any diabetic patient for surgery, especially for a replacement. Yet, after reading this article, I intend to incorporate the authors’ findings into a thorough shared surgical decision-making discussion with patients in the office.

In particular, several aspects of the author’s approach to the article’s methods warrant highlighting and commendation. To begin, this is one of the largest studies to address surgeons’ concerns regarding the ideal surgical option for a diabetic patient with ankle arthritis. Several studies in the literature address this specific question. Second, the authors have used the TriNetX database effectively. Admittedly, there are flaws to any retrospective database. On the other hand, the specific outcomes in this study are ones for which TriNetX can provide high-fidelity data to test the study’s hypothesis.

Recently, I have been involved in a multi-institutional study using the TriNetX database to analyze arthrodesis in an adjacent joint and can speak to its usefulness. ⁷ The size of this database makes it a valuable resource for ensuring a study has adequate power. As with any large database, concerns about data accuracy are reasonable. However, it is important to note that this database is compiled from International Classification of Diseases, Tenth Revision (ICD-10) codes, Current Procedural Terminology (CPT) procedure billing codes, and insurance information. (Remember, these folks try to deny things if the documentation is bad!) Therefore, specific outcomes such as readmissions, postoperative complications, and reoperations are recorded with high accuracy. The difficulty with humongous databases is often how to interpret the results and draw adequate conclusions. To this end, the authors wisely used the false discovery rate correction. This method was popularized in genomics studies to help reduce the chances that a specific finding may have occurred by “luck alone” in a large data set. Using this statistical method when interpreting results helps mitigate this mistake. In other words, the findings of this article are very likely to be accurate and true.

Ultimately, key takeaways from this study are that it serves as an excellent resource when counseling diabetic patients regarding surgical complications and outcomes for their ankle arthritis. Replacement should be regarded as an acceptable surgical option for a person with diabetes with ankle arthritis. However, it is equally important to note that the published rates of infection, readmission, and structural complications for both groups remain higher than those in a nondiabetic population.

In my opinion, the ultimate decision regarding the preferred surgical option for a diabetic patient should remain a shared decision, with a thorough review of all elevated risks and an understanding that outcomes after ankle replacement may be safer than previously believed, particularly compared with fusion. This shift in perspective highlights the potential benefits of total ankle replacement as a viable option for individuals with diabetes, thereby challenging earlier assumptions about the procedure’s risks.

Supplemental Material

sj-pdf-1-fai-10.1177_10711007261424897 – Supplemental material for Rethinking the Surgical Treatment for Ankle Arthritis in Patients With Diabetes: Why Replacement May Outperform Fusion

Supplemental material, sj-pdf-1-fai-10.1177_10711007261424897 for Rethinking the Surgical Treatment for Ankle Arthritis in Patients With Diabetes: Why Replacement May Outperform Fusion by Michael Aynardi in Foot & Ankle International

Michael Aynardi, MD, FAOAPenn State Milton S Hershey Medical Center, USA