The Diagnostic Characteristic and Reproducibility of Bone Scintigraphy Single-Photon Emission Computed Tomography/Computed Tomography for Diagnosing Aseptic Loosening of Uncemented Total Knee Arthroplasty

Background: The primary objective of this study was to investigate the diagnostic characteristics of bone scintigraphy single-photon emission computed tomography/computed tomography (BS-SPECT/CT) for diagnosing aseptic loosening after uncemented total knee arthroplasty (TKA), and to evaluate the following aspects: how to manage inconclusive results, the interobserver reliability, and the location of tracer uptake between symptomatic cases with and without aseptic loosening. Methods: In this study, 180 patients who had uncemented TKA and persistent knee pain suspected of aseptic loosening were included. As part of routine medical care, BS-SPECT/CT was used, and its results were compared with the reference standard, which involved revision surgery for aseptic loosening or a 12-month follow-up without revision or imaging. Inconclusive BS-SPECT/CT results were considered either negative (best-case scenario) or positive (worst-case scenario). Sensitivity, specificity, positive/negative likelihood ratios (LRs), and positive/negative predictive values (PPV and NPV) were calculated. Sensitivity analyses were conducted by comparing the diagnostic characteristics between cases with a follow-up of less and more than 2 years of follow-up. The anatomical distribution of tracer uptake and interobserver reliability were also evaluated. Results: Of the 180 BS-SPECT/CT scans conducted, 22 were determined positive, 113 negative, and 45 inconclusive. The best-case scenario showed a sensitivity of 66.7%, specificity of 93.8%, +LR 10.8, −LR 0.4, PPV 54.6%, and NPV 96.2%. In contrast, the worst-case scenario had a sensitivity of 94.4%, specificity of 69.1%, +LR 3.1, −LR 0.1, PPV 25.4%, and NPV 99.1%. Sensitivity analyses revealed no relevant differences in characteristics between the 2 TKA-interval groups. The interobserver reliability was fair-to-moderate (κ = 0.39; 95% confidence interval 0.18 to 0.60), with an estimated agreement of 79% (95% confidence interval 70 to 87). Four prosthetic zones had a notably higher proportion of cases with tracer uptake in those with aseptic loosening compared to those without. Conclusions: The test characteristics of BS-SPECT/CT were deemed appropriate in patients who have complaints of uncemented TKA suspected of aseptic loosening. Inconclusive cases were best categorized as negative, especially in patients who have a short interval between TKA and the first BS-SPECT/CT. Increased tracer uptake in 4 prosthetic zones was observed in cases of aseptic loosening, although interobserver reliability was fair to moderate. Level of Evidence: Level III, Diagnostic Retrospective Cohort Study.