|Abstract / Summary|
Prosthetic joint infection (PJI) is a rare, but devastating complication following primary total hip arthroplasty (THA). As PJI is a rare event, large cohorts of patients are required in order to study this complication. National arthroplasty registers offer such large and unselected cohorts, but studies have shown that these registers – used alone – underestimate the incidence of PJI.
The aim of this thesis was to estimate the incidence of PJI and the mortality risk following a PJI by combining data from the Danish Hip Arthroplasty Register (DHR), the National Register of Patients (NRP), the Microbiology Databases, the Civil Registration System, the medical records, the Danish National Prescription Registry and the Clinical Biochemistry Databases.
The thesis comprises the following four studies:
Study I: The aim of this study was to estimate the “true” incidence of surgically treated PJI following primary THA. To estimate the true incidence, we developed an algorithm that classified the revisions as due to PJI or due to other causes. The algorithm incorporated data from the DHR, the NRP, medical records, the microbiological databases, the prescription database and the clinical biochemistry databases. The one- and five-year cumulative incidences were estimated to be 0.86% (95% confidence interval (CI): 0.77; 0.97) and 1.03% (95% CI: 0.87; 1.22), respectively. These figures are approximately 40% higher than the equivalent figures reported by the DHR and the NRP.
Study II: The aim of the second study was to validate the PJI diagnosis in the DHR. We did this by comparing the PJI diagnosis in the DHR with the PJI diagnosis derived from the algorithm developed in Study I. We found a sensitivity of 67%, a specificity of 95%, a positive predictive value (PPV) of 77%, and a negative predictive value (NPV) of 92%. When the data from the DHR were linked with data from the microbiology databases, the sensitivity increased to 90% and the specificity also increased (to 100%) along with the PPV (98%) and the NPV (98%).
Study III: The aim of the third study was to examine whether the incidence of PJI observed within the first year of primary THA increased in the course of the ten-year study period from 2005 to 2014. We used the validated PJI diagnosis described in Study II and found that the incidence of PJI did not appear to be increasing as the relative risk of PJI was 1.05 (95% CI: 0.82; 1.34) for the 2010-2014 period compared with the 2005-2010 period. Nor did we find any changes in the antimicrobial resistance pattern.
Study IV: The aim of the fourth study was to estimate the mortality risk following a revision for PJI within one year following a primary THA. When combining data from the DHR with data from the microbiology databases, we found that the mortality risk of patients with a revision for PJI was 2.18 (95% CI: 1.54; 3.08) compared with the reference population, and 1.87 (95% CI: 1.11; 3.15) when compared with patients who had an aseptic revision.
In conclusion, the incidence of PJI is approximately 40% higher than that reported by the NRP and the DHR. By linkage of the DHR and the microbiology databases, the validity of the PJI diagnosis can be improved notably. By such a combination of data from the DHR and the microbiology databases, we show that the incidence of PJI does not seem to be increasing and that revision for PJI is associated with a high mortality.