Moving Beyond Systemic Inflammatory Response Syndrome and Bacteremia: Are Modern Critical Care Calculators Useful in Predicting Debridement, Antibiotics, and Implant Retention Treatment Outcomes in Periprosthetic Joint Infection?

Background: In critically ill periprosthetic joint infection (PJI) patients, surgeons need to balance the need for aggressive, definitive treatment against the health state of a potentially unstable patient. A clear understanding of the association between treatment outcomes and assessment scores for sepsis would benefit clinical decision-making in these urgent cases. The current study evaluates the effect of critical illness on debridement, antibiotics, and implant retention (DAIR) outcomes, as defined by systemic inflammatory response syndrome (SIRS) and, for the first time, by contemporary markers quick Sequential Organ Failure Assessment (qSOFA) and Modified Early Warning Score (MEWS).

Methods: We retrospectively identified 253 patients who underwent DAIR for PJI at a single institution between 2017 and 2021. The SIRS, qSOFA, and MEWS scores were calculated based on variables on admission. A DAIR treatment failure, defined as reoperation or mortality, was measured at 90 days and two years. Univariate analysis was used to determine the association between elevated critical care scores and DAIR failure.

Results: The DAIR treatment success was 59% at two years, with hip procedures and Charlson comorbidity index (CCI) ≥ 1 independently associated with higher odds of DAIR failure. There were 43 patients (16%) who presented with SIRS, however, only four (2%) had positive qSOFA scores. Neither SIRS nor qSOFA were predictive of DAIR failure. For knees only, elevated MEWS scores were predictive of 90-day DAIR failure (P = 0.019).

Conclusion: Over one in six patients undergoing DAIR for PJI presented with SIRS, while only one in 50 had a positive qSOFA. The SIRS and qSOFA scores were not predictive of DAIR failure. Elevated MEWS scores were associated with DAIR failure at 90 days postoperatively in knee PJIs only, and should be confirmed in a larger cohort. Our results suggest that SIRS is not predictive of DAIR outcomes, possibly because it overestimates the proportion of critically ill patients.