Should scintigraphy be completely excluded from the diagnosis of periprosthetic joint infection?

Can99mTc-MDP-SPECT/CT Differentiate Loosening and Infection After Hip and Knee Replacements?

Background

Prosthetic loosening and infection are still common complications after joint replacement. Over the past few years, single-photon emission computed tomography-computed tomography (SPECT/CT) was widely used and showed unique value based on the combination of anatomic and metabolic information of foci. However, its performance in differentiating between prosthetic loosening and periprosthetic infection after joint replacement is still the focus of clinicians and deserves further investigation.

Purpose

This retrospective study was aimed to determine whether bone scintigraphy combined with SPECT/CT still can differentiate prosthetic infection from loosening in patients after joint replacement. The differential efficacy in hip and knee prosthesis was also analyzed. Blood biomarkers for the diagnosis of periprosthetic infection were also evaluated.

Patients and methods

Data sets of 74 prosthetic joints (including knees and hips), with suspected prosthetic loosening or infection between 2015 and 2021, were evaluated. Besides the results of nuclear imaging, X-ray images and serum biomarker were also recorded. Telephone follow-up and revision surgery after SPECT/CT were used as a gold standard. The sensitivity and accuracy of different imaging modalities were calculated by Chi-square test. The diagnostic efficacy of imaging methods and serum biomarkers were then analyzed by the area under curve (receiver operating characteristic curves, ROC) in SPSS 26.

Results

In all, 47 joints (14 knees and 33 hips) were confirmed as aseptic loosening, while 25 joints (18 knees and 7 hips) were confirmed as infection. The sensitivity and accuracy of SPECT combined with SPECT/CT imaging were the highest (92.86% and 87.84%, respectively). The differential efficacy of bone scintigraphy combined with SPECT/CT imaging was also better than any other single imaging modality. In the analysis of involved prosthesis, prosthetic loosening occurred more in hip prosthesis and knee prosthesis was easily infected (P < 0.05). Finally, the sensitivity of ESR and CRP were 80% and 84%, respectively.

Conclusions

Bone scintigraphy with hybrid SPECT/CT remains encouraging in differentiating prosthetic infection from loosening after joint replacement. The diagnostic efficacy of differentiation in hip prosthesis was better than knee. Serum biomarkers cannot be used alone to differentiate prosthetic infection from loosening.