Ankle Fusion or Arthroplasty — Pick Your Poison

A large RCT finds no differences in PROMS at 1 year between patients with ankle OA randomized to fusion or arthroplasty, but big between-group differences in the types of complications experienced

What’s the Claim?

A lovely but short-term (one-year follow-up) large, multicenter RCT of the two main surgical treatments for ankle arthritis — arthrodesis and total ankle arthroplasty — found no differences between the two in terms of PROMS. Not surprisingly, both groups did pretty well. Adverse events were dismayingly common in both groups — a majority of patients in both groups recorded one or more adverse events, with most of them being minor, but about one in five patients in each group had one or more serious adverse events as the authors defined them (these were things like wound complications or infections, nerve injuries, and thromboembolic events). The numbers of these as well as unplanned reoperations did not differ between the groups (the latter occurring in about 3% of patients), but the kinds of adverse events differed, with wound problems and nerve injuries being more common in patients having arthroplasty and nonunions and blood clots happening more in the fusion group. Symptomatic nonunions occurred in 7% of the fusion group, and the reoperations for these happened outside the 52-week study window. An important preplanned secondary analysis suggested that arthroplasty did a bit better in patients who presented with already established adjacent-joint arthritis.

How’s It Stack Up?

There are no real comparators in the sense that this is the first well-done, large RCT on the topic. It generally comports well with the better-designed observational trials, which don't show a difference in outcomes scores even out to five years (which is far beyond the follow-up of this RCT).

What’s Our Take?

The skillful practice of foot and ankle surgery seems to be all about navigating tradeoffs. Got an Achilles tendon rupture? Pick your poison: rerupture (for nonsurgical management) or infection (surgery). Ankle arthritis? Wound-healing problems and nerve injuries (ankle replacement) or nonunion and blood clots (fusion). Fortunately, most patients in both groups here dodged all those bullets. The definitive answers on this question, which likely will come from those comparative observational studies we pointed to earlier before they come from RCTs, should be coming out soon. We're interested to see whether the fusion group comes in with more fusions for symptomatic adjacent-joint arthritis (and progressive worsening of outcomes scores) or the ankle replacement group comes in with more late infections, symptomatic subsidence, and revisions for loosening. If ankle replacement comes out ahead, CORRelations’ advisor in Foot and Ankle, Greg Guyton MD, thoughtfully opines that it will be from “objective improvement in gait and sparing of adjacent-joint DJD over time (eg, subtalar joint), not early pain relief or function as assessed by PROMS. The major caveat is that as the procedure has now become well established, very high-functioning patients likely won’t participate in RCTs as they seek out replacement. The time to perform a truly valid RCT on the topic may have passed.” So we’ll look forward to follow-up reports from those large observational studies.


Goldberg AJ, Chowdhury K, Bordea E, et al. Total Ankle Replacement Versus Arthrodesis for End-Stage Ankle Osteoarthritis: A Randomized Controlled Trial. Ann Intern Med. 2022;175:1648-1657.