Prosthetic Joint Infection: The Two Weeks That Change the Answer
A prosthetic knee infection sits right at the edge of the window that would make implant retention straightforward. Imaging can’t settle it — only the surgeon’s own eyes, once the joint is open, can.
Joan H., a 69-year-old retired accountant who took up watercolor painting after her knee replacement three years ago finally let her stand at an easel without pain, noticed swelling and warmth around that same knee about four weeks ago — not the sudden, dramatic onset of an acute joint infection, but a slower build she initially attributed to overdoing a gardening project. By the time she saw her surgeon, aspiration confirmed infection, growing Staphylococcus epidermidis on culture, an organism with confirmed susceptibility to rifampin on later testing. Her symptom duration — close to but past the traditional three-week window used to define "acute" infection eligible for debridement with implant retention — sits right at the edge of the criteria her team would otherwise use to decide between DAIR and a full staged exchange.
She has well-controlled hypertension and osteoporosis, but no diabetes and no prior joint surgery complications, and a staged exchange — removing the prosthesis, treating with an antibiotic spacer for six weeks, then re-implanting — would mean a longer recovery and a second major surgery she is realistic about not wanting unless it’s genuinely necessary. Pre-operative imaging shows the implant components in reasonable position without obvious loosening, but imaging alone has real limits for detecting the microscopic bone erosion that actually determines whether an implant can be safely retained — the kind of finding that sometimes only becomes clear once a surgeon is looking directly at the bone in the operating room.
The Zimmerli criteria her team keeps returning to were built around three variables read together — symptom duration, implant stability, and organism susceptibility to a biofilm-active agent — not any single one of them in isolation, and her case is unusual for having two of the three land cleanly (a rifampin-susceptible organism, an implant that looks stable on imaging) while the third sits genuinely on the boundary rather than clearly on either side. The framework was never built with a rounding rule for four weeks against a three-week cutoff, and no amount of further discussion will supply one. What will settle it is the moment the surgeon has the joint open and can see whether the bone under the implant has been eroded — a determination nobody in this room can make today, for a decision the team has to plan around before then anyway.
Pre-operative planning conference
She’s close enough to the acute window, and her organism is confirmed rifampin-susceptible, which is the combination DAIR with rifampin was actually built for — rifampin’s real value is biofilm penetration that most other agents can’t match. A staged exchange means a second major operation and a longer recovery in a 69-year-old who’s realistic about not wanting that unless it’s truly necessary. I don’t think a few days past a three-week cutoff, on its own, should force the more invasive path.
I hear the biofilm argument, and I’m not dismissing it — but the symptom-duration cutoff was never meant to be read to the day, and what actually worries me is what the imaging can’t tell us. If there’s meaningful bone erosion at the implant interface, DAIR fails regardless of how susceptible the organism is, because retaining a component that isn’t actually stable just sets up a second infection on a compromised bed. I won’t know that for certain until I’m in there looking directly at the bone.
Agreed: proceed to the operating room with both plans genuinely open, and let direct visualization of the bone-implant interface make the final call rather than deciding from imaging or symptom duration alone.
Explicitly not resolved, and stated as such rather than smoothed into a false consensus: the infectious disease physician leans toward DAIR with rifampin if stability is even reasonably intact; the orthopedic surgeon will proceed to staged exchange on any real evidence of erosion, with a lower threshold for calling that evidence "real" than the infectious disease physician might apply from the same finding. Both agreed the surgeon’s intraoperative judgment governs, without either side abandoning their own reading of how much doubt should tip the decision.