Photon-counting CT in Anterior Cervical Discectomy and Fusion: Improved Metal Artifact Reduction and Impact on Bone Fusion Assessment.

OBJECTIVE

To evaluate the impact of tin-filtered photon-counting detector (PCD) computed tomography (CT) in combination with high-energy virtual monoenergetic imaging (VMI) and iterative metal artifact reduction (iMAR) on metal artifact reduction, image quality, and assessment of intragraft and extragraft bone fusion, in patients after anterior cervical discectomy and fusion (ACDF).

MATERIALS AND

METHODS

Patients who underwent ACDF and postoperative tin-filtered PCD-CT between 2023 and 2026 were retrospectively analyzed. Metal artifacts were quantitatively and qualitatively assessed in standard polychromatic (T3D), iMAR (T3DiMAR), VMI at 120 keV (VMI120), and combined VMI120 + iMAR (VMI120+iMAR) reconstructions. Quantitative analysis included measurement of hypodense and hyperdense artifact attenuation and corrected noise. Three musculoskeletal radiologists independently scored 5 imaging features on 4-point Likert scales and assessed the presence of secondary artifacts. Subgroup analyses were performed for polyetheretherketone (PEEK)-tantalum versus titanium cages and for 140 versus 100 kVp acquisitions.

RESULTS

Forty-four patients (23 with PEEK-tantalum cages, 21 with titanium cages) were included. VMI120+iMAR achieved the strongest metal artifact reduction with hypodense artifacts from -362.5 Hounsfield Units (HU) to -38 HU and hyperdense artifacts from 218.5 to 83.5 HU, as well as significant noise improvements compared with T3D (all P<0.001); effects were most pronounced near titanium cages. T3DiMAR provided comparable artifact reduction to VMI120+iMAR in PEEK-tantalum cages (P=0.55 to 0.83) but inferior artifact reduction in titanium cages (P<0.001). Bone-metal interface conspicuity improved significantly with VMI120 and VMI120+iMAR (P≤0.01). Intragraft and extragraft bone fusion visibility was rated highest with standard T3D images (median 3 to 4), whereas the most pronounced impairments were observed for VMI120+iMAR (median 2 to 3, P≤0.01). iMAR-based techniques introduced secondary artifacts, including pseudo-osteolysis (20.5% to 38.6%), and white zone artifacts (50% to 56.8%), while dark zone and silhouette artifacts were less common (4.5% to 6.8%). White zone artifacts were significantly more frequent at 140 versus 100 kVp (P<0.05), while dark zone and silhouette artifacts were exclusively observed at 100 kVp. Inter-reader agreement was substantial to almost perfect (κ=0.62 to 0.94).

CONCLUSIONS

Tin-filtered PCD-CT with VMI120+iMAR offers the most effective metal artifact reduction after ACDF and improves bone-metal interface conspicuity, particularly near titanium cages, followed by T3DiMAR. However, these advanced reconstructions may impair diagnostic accuracy for bone fusion assessment due to reconstruction-induced artifacts. iMAR-based reconstructions introduced pseudo-osteolysis that may simulate pseudoarthrosis, as well as white zone artifacts that may mimic false bony bridging in a material-dependent and kVp-dependent manner. Therefore, a reconstruction-specific approach is recommended for the clinical evaluation of ACDF patients, balancing metal artifact reduction against diagnostic reliability for bone fusion.