Spectral CT Helps Distinguish Periprosthetic Infection

Periprosthetic joint infection remains a difficult complication to distinguish from aseptic failure after hip or knee arthroplasty, yet the distinction directly shapes treatment planning. An original investigation published in Insights into Imaging examines whether spectral CT with quantitative parameters can improve differentiation between septic and aseptic periprosthetic complications. Conventional CT can be limited by metal-related artefacts, while clinical markers, cultures and histological findings can vary in diagnostic performance. Spectral CT adds quantitative measures, including iodine concentration, effective atomic number and normalised iodine concentration, alongside metal artefact reduction. The investigation included patients with suspected periprosthetic complications who underwent preoperative spectral CT before planned revision surgery, with infection status classified using 2018 International Consensus Meeting criteria.

Quantitative Markers Show Higher Values in Infection

The cohort included 59 patients with 62 prostheses. Aseptic complications accounted for 36 prostheses and septic complications for 26. Patient age, sex and body mass index did not differ significantly between groups, although the distribution of affected hip and knee joints was different. Septic cases included total hip and total knee arthroplasties, while aseptic cases mainly involved hip arthroplasties.

Must Read: Low-Dose X-Rays Advance 3D Knee Bone Reconstruction

Spectral CT analysis focused on periprosthetic bone lesions and soft tissue lesions. Bone lesions included osteolysis. Soft tissue lesions included proliferative synovial tissue, abscess cavity, sinus tract and inflammatory pseudotumour. Quantitative measurements were taken on iodine concentration maps and effective atomic number maps during arterial and venous phases. Normalised iodine concentration was calculated by comparing lesion iodine concentration with the adjacent artery.

All spectral CT quantitative parameters were significantly higher in septic lesions than in aseptic lesions. Iodine concentration during the arterial phase showed the strongest performance across combined and separate lesion analyses. In the combined assessment of bone and soft tissue lesions, the septic group had higher arterial-phase iodine concentration than the aseptic group. The same pattern appeared when bone lesions and soft tissue lesions were assessed separately.

Diagnostic Accuracy Exceeds Subjective CT Assessment

Arterial-phase iodine concentration achieved the highest diagnostic accuracy among the quantitative parameters. Accuracy reached 84.3% when bone and soft tissue lesions were assessed together. In separate analyses, accuracy reached 90.7% for bone lesions and 89.6% for soft tissue lesions. Other spectral CT parameters also showed significant differences between septic and aseptic lesions and demonstrated good diagnostic performance.

For soft tissue lesions, arterial-phase iodine concentration and arterial-phase effective atomic number produced the highest area under the curve values. For bone lesions, arterial-phase iodine concentration again showed the strongest overall performance. The slightly higher accuracy in bone lesions was linked in the discussion to infection-related changes near the prosthesis and to the more focal nature of infection in dense bone, compared with wider spread through loose soft tissue.

Qualitative CT assessment using virtual monochromatic images with orthopaedic metal artefact reduction was less accurate than the best quantitative results. A senior radiologist achieved 82.3% accuracy, while a junior radiologist achieved 75.8% accuracy. Inter-observer agreement for qualitative assessment was substantial. Quantitative assessment showed excellent inter- and intra-observer reliability for bone and soft tissue lesions during arterial and venous phases.

Clinical Use Remains Dependent on Validation

Spectral CT provides information beyond visual inspection by adding reproducible quantitative parameters. The findings indicate that these parameters may support diagnostic confidence, particularly in complex cases and among less-experienced radiologists. The technique may also help visualise periprosthetic structures while reducing the effect of metal artefacts.

Clinical diagnosis of periprosthetic joint infection remains challenging because no single laboratory or imaging test reliably identifies infection. Current definitions combine multiple findings and generally exclude most imaging techniques, except white blood cell scintigraphy. In this cohort, serum and synovial markers varied between groups, and some positive biomarkers were also seen in aseptic cases. Spectral CT quantitative parameters therefore add a possible imaging biomarker rather than replacing clinical assessment.

Important limitations remain. The sample size was small, partly because of the inclusion criteria and the relatively low incidence of periprosthetic joint infection. The distribution between septic and aseptic groups was also imbalanced. Comparisons between clinical indicators and quantitative parameters were not performed, and no multivariable model was developed because of the limited sample size. Lesions were selected to avoid metal artefact interference, which helped measurement accuracy but may affect wider applicability.

Spectral CT with quantitative parameters shows potential for differentiating septic from aseptic periprosthetic complications before revision surgery. Arterial-phase iodine concentration provided the strongest diagnostic performance, particularly in separate analyses of bone and soft tissue lesions. Quantitative assessment outperformed subjective CT interpretation in this cohort and showed excellent observer reliability. Larger multicentre validation remains necessary before broader clinical adoption, including evaluation of combined models with clinical indicators and further assessment of emerging CT technologies.

Source: Insights into Imaging

Image Credit: iStock