Sonication of revised hip and knee prostheses detects occult infections, improves clinical outcomes and prevents re - revisions. A case series study

Abstract

Introduction: Periprosthetic joint infection (PJI) is a devastating complication occurring in 1-2% of primary and up to 10% of revised total hip and knee arthroplasties (THA and TKA) impairing patient's quality of life. Occult infections are underdiagnosed, sub-treated and sub-clinically experienced by patients. This study aimed to correlate patients' clinical outcomes with early antibiotic treatment based on use or non-use of a sonication technique on explanted prostheses.

Methods: 33 patients with revised THA or TKA were retrospectively evaluated. Clinical outcomes were assessed via Oxford hip or knee scores, and correlated with administration or not of antibiotic treatment based on sonication results.

Results: According to laboratory findings the patients were divided in the following three groups: 1. Septic loosening (conventional cultures and/or sonication positive), 2. Aseptic loosening (conventional cultures and sonication negative) and 3. Occult loosening (conventional cultures negative, sonication not performed). The average Oxford score was poor (27.9/60) for the septic, excellent (43.8/60) for the aseptic and intermediate (37.7/60) for the occult group. Additionally, conventional cultures were negative, but sonication-positive, in 6 individuals with patient-related risk factors (male gender, BMI > 30 kg/m², diabetes, hypertension, steroids and rheumatoid arthritis).

Conclusions: Sonication represents a valuable diagnostic technique to guide administration of effective antibiotic treatment for patients, especially for detection of persistent post-revision occult infections. We recommend the systematic investigation of revised prostheses with a sonication technique, but especially in cases with risk factors for infection who it is suspected may have occult loosening.

Keywords: Arthroplasty; Hip; Knee; Oxford score; Periprosthetic infection; Sonication.

Figure 1

Collection of explanted prostheses in a two-stage TKA. (A) X-ray of a patient with loosening if his TKA. The arrows show the radiolucent appearance of loosening around the tibial prosthesis. (B) Explanted prostheses were transferred to the microbiology de? lkjbpartment for sonication in a sterile solid air-tight container. (C) The femoral and tibial tunnels were filled with vancomycin-impregnated cement beads until second stage of revision. The asterisk shows the cement filling the gap (D) The final x-ray after the first phase of revision. The patient received antibiotic therapy according to conventional and sonication fluid cultures. After a 2-year follow up his Oxford Score was found 29/60.

Figure 2

Algorithm for inspected cases. Laboratory findings vs clinical outcomes. The flow chart depicts the algorithm we followed in our retrospective cohort study and the number of evaluated cases in each group. Implant loosening was assessed in laboratory via conventional and sonication fluid cultures and they were correlated with clinical outcomes with OHS and OKS.

Figure 3

Multiple comparison graph of Oxford scores between: (A) three types of loosening, (B) patients with negative and positive sonication fluid cultures (B), (C) patients with negative and positive periprosthetic tissue cultures. (A) Oxford score was measured higher in patients with aseptic loosening, intermediate in patients with occult loosening and lower in patients with septic loosening. These differences were statistically significant (P<0,001). (B) Oxford score was measured higher in patients with negative sonication fluid cultures and lower in patients with positive sonication fluid cultures. This difference was statistically significant (P<0,001). (C) Oxford score was found higher in patients with negative periprosthetic tissue cultures and lower in patients with positive periprosthetic tissue cultures. This difference was statistically significant (P<0,001).