MRI Study Of Axial Spondyloarthritis: Comparison Of Volumetric Interpolated Breath-hold Examination(VIBE) Sequences With Different Slice Thicknesses And Multiple Model Of Diffusion Weighted Imaging In Detecting Sacroiliac Joint Lesions And Disease Activit

Part ? Recognition of sacroiliac joint structural lesions: comparison of volumetric interpolated breath-hold examination(VIBE)sequences with different slice thicknesses to T1-weighted turbo-echoPurpose: To compare volumetric interpolated breath-hold examination(VIBE)with different slice thicknesses to T1-weighted turbo-echo(T1 TSE)for identification of sacroiliac joint structural lesions in patients suspected of spondyloarthritis(SpA)using CT as the gold standard.Methods: 192 sacroiliac joints(including VIBE with both 1.2 mm and 3 mm slice thickness,T1 TSE)from 96 patients suspected of SpA were included.Joint space changes and sclerosis were evaluated for each joint.Erosions were assessed both at the level of the individual sacral and iliac bones and at the level of the entire joint for calculation of sensitivity,specificity,and accuracy.MRI and CT correlation was performed and inter-reader reliability was determined.Fat infiltration on MRI was scored.Results: VIBE with a 1.2 mm slice thickness was the most sensitive and accurate for erosion detection at the bone level followed by 3 mm thickness VIBE and then T1 TSE(p<0.05).At the whole-joint level,only the 1.2 mm slice thickness VIBE was superior to T1 TSE in sensitivity and accuracy(p>0.05).For joint space changes,both VIBE sequences were superior to T1 TSE in sensitivity and accuracy(p<0.05)and had more consistency with CT.T1 TSE was slightly more sensitive for detection of sclerosis(p<0.05).The MR sequences did not differ in detection of fat infiltration.Conclusion: A VIBE sequence with 1.2 mm slice thickness and less than one-minute acquisition time was superior to T1 TSE for detection of sacroiliac joint space changes and erosions in patients with suspected SpA,while the utility of the 3 mm slice thickness VIBE remains questionable.Part ? Comparing diagnostic performance of histogram analysis of different model of diffusion weighted imaging in evaluating disease activity of axial spondyloarthritis.Purpose: To assess whole-joint based histogram analysis of mono-exponential,bi-exponential diffusion and diffusion kurtosis imaging in evaluating disease activity of axial spondyloarthritis(ax SpA).Methods: A total of 82 patients with axial spondyloarthritis and 17 patients with nonspecific low back pain who underwent both diffusion kurtosis imaging(DKI)and multi b diffusion weighted imaging of sacroiliac joint were included.Patients were divided into active,inactive group and control group.Spondyloarthritis Research Consortium of Canada(SPARCC)sacroiliac joint scoring system was used to assess the level of bone marrow edema on conventional MR images.Histogram parameters of mono-exponential apparent diffusion coefficient(ADC),intravoxel incoherent motion(IVIM)-derived true diffusion coefficient(D),perfusion fraction(f),pseudodiffusion coefficient(D*)and DKI-derived mean kurtosis(MK)and mean diffusivity(MD)was calculated,including the mean value and the 25,50,75,90 percentile.Receiver operating characteristic(ROC)analysis was performed to assess each imaging parameter in distinguishing different groups.Regression analysis was applied in assessment the correlation between imaging parameters and clinical activity indices including erythrocyte sedimentation rate(ESR),hyper-sensitive C-reactive protein(hs CRP),Bath AS Disease Activity Index(BASDAI).Results: Each imaging parameter distinguished active group from inactive group with a low AUC(area under the curves)(max AUCs?0.71),SPARCC score failed to differentiate the two group(P=0.08).MD,MK,D,D*,ADC showed good performance in differentiating active,inactive group from control group(max AUC=0.81~0.98).f50 was able to differentiate active group from control group(AUC=0.72)while the AUC was significantly lower than the max AUC of MD,MK,D,ADC and SPARCC score(all p<0.05).The max AUC of MD in differentiating inactive group from control group was significantly higher than that of D* and SPARCC score.All the histogram parameters of D,ADC and the mean,50 th,90th percentile of MD and the 50 th percentile of D* and the 25 th,50th percentile of f were positively correlated with BASDAI,while the mean,75 th,90th percentile of MK was negatively correlated with BASDAI.Only the increase of the 50 th,75th,90 th percentile of MD was correlated with the increase of hs CRP.Conclusion: Whole joint based histogram analysis of mono-exponential and bi-exponential diffusion imaging and diffusion kurtosis imaging showed good diagnostic performance in differentiating active,inactive group from control group,and their performance was comparable.All the imaging parameters were correlated with BASDAI except for SPARCC,only DKI-derived MD was correlated with hs CRP,these parameters can be used as imaging biomarkers for disease activity in ax SpA.