Antibiotic spacers hold at five years, and the ones that dislocate keep dislocating

Design

single-centre retrospective cohort with Kaplan-Meier survival analysis and Cox regression

Population

302 two-stage revision total hip arthroplasties in 298 patients with periprosthetic joint infection, 2005 to 2022; mean age 65, mean follow-up 6 years

Primary outcome

five-year survival free of reinfection, aseptic rerevision, any rerevision and any reoperation after reimplantation

Effect

89% (95% CI 85.2 to 93.1) free of reinfection, 80% (74.7 to 84.9) free of any reoperation; spacer dislocation to post-reimplantation dislocation hazard ratio 6.2 (2.3 to 16.9)

Two-stage exchange remains the dominant strategy for chronic periprosthetic joint infection after hip arthroplasty in much of the world. This series reports 302 hips in 298 patients treated between 2005 and 2022 with a commercially available, partially intraoperatively fabricated articulating high-dose antibiotic cement spacer, followed for a mean six years. Forty per cent had already had an infection-related procedure.

At five years after reimplantation, survival free of reinfection was 89% (95% CI 85.2 to 93.1), free of aseptic rerevision 93% (89.3 to 96.1), free of any rerevision 86% (81.3 to 90.3) and free of any reoperation 80% (74.7 to 84.9). Reinfection risk was higher in McPherson host C patients (hazard ratio 3.7, 95% CI 1.4 to 9.6) and limb grade 3 (3.3, 1.0 to 11.1). Mean Harris Hip Score at latest follow-up was 76. Spacer-related problems were not rare: 15 perispacer fractures and 11 dislocations managed non-operatively, 5% of patients each.

The finding with the clearest surgical consequence is the link between the interval and the result. Dislocation of the spacer was strongly associated with dislocation after reimplantation, hazard ratio 6.2 (95% CI 2.3 to 16.9) - and dislocation accounted for 15 of the 20 aseptic rerevisions. A hip that has proved unstable with a spacer in it is telling you something about soft tissue and offset that will not be fixed by the definitive implant alone, and the authors' conclusion follows: plan a higher-stability bearing for those patients rather than discovering the problem twice.

• Record spacer dislocation as a risk factor, not an interval nuisance - it raised post-reimplantation dislocation risk more than six-fold.
• Plan a higher-stability bearing at reimplantation for any hip that dislocated with the spacer in situ.
• Set expectations by host grade: McPherson host C carried nearly four times the reinfection risk.
• Warn patients that about one in ten will have a perispacer fracture or dislocation during the interval.
• This is a single high-volume centre with a specific commercial spacer; local fabrication and different antibiotic loading are not equivalent.

The statistics, in plain English

These are Kaplan-Meier survival estimates from a single centre, so they describe what happened to this cohort with this spacer in these surgeons' hands - the confidence intervals, for example 85.2 to 93.1 around the 89% reinfection-free figure, quantify sampling error but not the transferability. The hazard ratio for limb grade 3, 3.3 with an interval of 1.0 to 11.1, only just excludes 1.0 and rests on few events; the host C figure is firmer. The dislocation association is the strongest result in the paper by margin, and its plausibility is mechanical rather than statistical, which is why it can reasonably change a surgical plan on this evidence alone.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.