Feasibility of T2 Mapping of the Sacroiliac Joints in Healthy Control Subjects and Children and Young Adults with Sacroiliitis

Abstract

Objective To assess the feasibility of T2 mapping for evaluating pediatric SIJ cartilage at 3 Tesla (T) magnetic resonance imaging (MRI). Methods Healthy control subjects and adolescents with sacroiliitis underwent a 3T MRI dedicated pelvic protocol that included a T2 mapping sequence consisting of multislice, multiecho acquisition. Healthy control subjects were prospectively recruited from our primary care practices as part of a larger imaging study, whereas adolescents with sacroiliitis were recruited specifically for this study. Regions of interest (ROIs) were hand‐drawn by a senior pediatric radiologist twice and a radiology fellow twice to calibrate and test reliability using the intraclass correlation coefficient (ICC). T2 relaxation time between control subjects and cases was compared using univariate linear regression. We tested the association of T2 relaxation time in adolescents with sacroiliitis with patient‐reported outcomes and the Spondyloarthritis Research Consortium of Canada sacroiliac joint (SIJ) inflammation and structural scores using Pearson correlation coefficients. Results Fourteen subjects were evaluable (six control subjects: median age 13.7 years [interquartile range (IQR): 12.2‐15.5], 67% male patients; eight cases: median age 17.4 years [IQR: 12.5‐20], 88% male patients]. Acquisition time for T2 mapping sequences was approximately 6 minutes, and segmenting the ROI for each SIJ took approximately 3 minutes. The intrarater and inter‐rater ICCs were 0.67 and 0.46, respectively, indicating good to fair reliability. There was a trend, albeit statistically insignificant, in longer median T2 relaxation time in cases (43.04 ms; IQR: 41.25‐49.76 ms) versus healthy control subjects (40.0 ms; IQR: 38.9‐48.6 ms). Although not statistically significant, cases with longer T2 relaxation time tended to occur with poorer patient‐reported outcomes. Correlations with the SIJ inflammation and structural lesion scores were weak. Conclusion T2 mapping of the SIJ cartilage in children was feasible and reliable. Larger controlled and longitudinal assessments are needed to assess the validity and utility of these measurements for routine clinical practice and trials.