High performance of low-dose ultrahigh-resolution photon-counting detector CT for sacroiliac joint assessment in axial spondyloarthritis

Abstract

Objectives To compare the performance of low-dose (LD) ultrahigh-resolution (UHR) photon-counting detector (PCD) CT with energy-integrating detector (EID) CT for sacroiliac joint assessment in axial spondyloarthritis (axSpA). Materials and Methods This was a prospective randomized study. AxSpA patients were enrolled and assigned to LD or clinical dose (CD) groups. Patients underwent UHR PCD-CT of the sacroiliac joints at low radiation dose (IQ level, 30), followed by EID-CT scanning using either LD (matched to PCD-CT dose) or CD protocols (108 quality reference mAs). Image noise, signal-to-noise ratio, interobserver reproducibility, subjective image quality, diagnostic confidence, and structural damage scores were evaluated by two radiologists and compared. Results In total, 40 patients were enrolled (LD group: n = 20, 36.5 ± 9.8 years, 8 females; CD group, n = 20, 35.5 ± 8.8 years, 8 females). Compared with CD EID-CT, PCD-CT delivered lower effective dose (0.63 ± 0.23 mSv vs. 1.79 ± 1.05 mSv, p < 0.001), and exhibited higher image noise (p < 0.001) and lower SNR (p < 0.001). Interobserver reproducibility of subjective image quality and diagnostic confidence on PCD-CT (substantial to almost perfect) was higher than LD EID-CT. PCD-CT provided better subjective image quality, diagnostic confidence, and structural lesion scores than LD EID-CT. PCD-CT showed higher subjective image quality for erosions and joint space changes while yielding higher erosion and sclerosis scores than CD EID-CT. Conclusion LD UHR PCD-CT reduced radiation dose by 65% compared with CD EID-CT, and improved structural lesion visualization and diagnostic confidence compared with both LD and CD EID-CT.