Publication:

Comparison of conventional radiograph with low-dose computed tomography of sacroiliac joints in patients with proven sacroiliitis

Background Sacroiliitis is central to diagnosing axial spondyloarthritis (axSpA), yet conventional radiography suffers from anatomical superimposition and low sensitivity for early structural change. Low-dose computed tomography (LDCT) provides cross-sectional detail at a radiation dose comparable to plain radiography. Aim was to evaluate LDCT versus conventional sacroiliac (SI) joint radiography in assessing sacroiliitis.

Methods This prospective, observational, cross-sectional study enrolled 30 consecutive adults with proven sacroiliitis at a tertiary centre. All underwent pelvic radiography (SAPA view) and LDCT of the SI joints (Philips Brilliance iCT 256-slice; 120 kV, 30 mAs; effective dose ~1 mSv). Two blinded radiologists independently scored 60 SI joints using a modified European image-quality system covering technical parameters (joint-margin and soft-tissue delineation, superior/inferior visualisation, absence of superimposition) and pathological parameters (joint-space change, erosions, sclerosis, soft-tissue content). Chi-square/Fisher’s exact and unpaired t-tests were used; p<0.05 was significant.

Results Mean age was 33.1±8.7 years; 96.7% were male; axSpA accounted for 90.0% of diagnoses. LDCT was significantly superior to radiography for soft-tissue delineation (98.3% versus 18.3%), superior/inferior joint margins (98.3% versus 55.0%), absence of superimposition (100% versus 36.7%), and iliac-margin erosion detection (86.7% versus 55.0%) (all p<0.001). Mean total imaging score was 7.6±1.3 for LDCT versus 4.3±1.8 for radiography (p<0.001), with consistent superiority across all subgroups; scores correlated strongly (r=0.742, p<0.001).

Conclusions LDCT delivered substantially superior image quality and structural-lesion detection than radiography at a comparable dose and should be integrated into diagnostic algorithms for suspected sacroiliitis, particularly when radiographs are inconclusive.