When patients come to me worried that their ankle replacement might be infected, the question they really want answered is simple: “How will you actually know?” Until recently, the honest answer was that surgeons in different hospitals approached it differently. There was no single, agreed pathway. In 2019, my colleagues and I set out to change that.

Why this matters

A total ankle replacement is one of the more sophisticated procedures in foot and ankle surgery. Most patients do well — pain goes, motion comes back, walking returns to something close to normal. But occasionally an ankle replacement becomes painful, swollen, or red weeks, months, or even years later.

That cluster of symptoms could mean an infection has set up around the implant. It could equally mean a number of other things — loosening of the implant, ligament instability, or simply the slow grumble of normal post-operative inflammation. Each of those needs a completely different treatment. Get the diagnosis wrong and you risk under-treating a serious infection, or, equally bad, putting a patient through unnecessary antibiotics and surgery for something that wasn’t infected at all.

What we did

The international Foot & Ankle community held a consensus meeting in 2019 to address this exact question. Delegates from around the world reviewed every piece of relevant published evidence and debated the best diagnostic pathway. I was the lead author of the resulting paper, published in Foot & Ankle International, which set out a stepwise algorithm — in essence, a decision tree — for working out whether an ankle replacement is infected. The process generated 100% agreement among the delegates, which is rare in clinical consensus and tells you the evidence and judgement were aligned.

The pathway, step by step

The algorithm has a clear, common-sense order. The first question is whether there is a draining sinus tract — a small channel that lets fluid or pus track from the implant out through the skin. If there is, the answer is straightforward: it is an infection. No further tests needed.

If there isn’t a sinus tract, the next step is two simple blood tests — ESR and CRP. These flag the body’s inflammatory response. If both are raised, the next step is to drain a small amount of fluid from the ankle joint with a fine needle, and send that fluid to the laboratory. We require the same bacterium to grow from at least two separate samples before we accept the diagnosis of infection. This protects against a false alarm from a stray skin organism contaminating one sample.

If the picture remains unclear and the patient doesn’t need surgery for any other reason, nuclear imaging gives an additional clue. If surgery is indicated regardless — for instance, because the implant is loose — we take tissue and joint fluid during the operation itself and analyse them. That is our final-court confirmation: more than five neutrophils per high-power field on histology, or matching cultures from the synovial fluid, settle the question.

What this means for you

Every step in the pathway is designed to protect the patient from two opposing errors: missing an infection that’s really there, and labelling something as infected when it isn’t. The algorithm is now part of standard practice internationally, and it’s how I work up every patient in my clinic who comes to me with a worrying ankle replacement.

What this paper doesn’t tell you

A consensus pathway tells us how to investigate, not what the result will be. Many patients with a painful ankle replacement turn out not to have an infection — and that’s a relief, but it still leaves the question of why the ankle hurts. The pathway above is the start of a thorough investigation, not the end of the conversation.