Multi-b-values-fitting readout-segmentation of long variable echo-trains diffusion-weighted imaging (RESOLVE DWI) in evaluation of disease activity and curative effect of axial spondyloarthritis (axSpA).

BACKGROUND

Disease activity is relevant to the treatment and prognosis of axSpA, and methods to quantitatively assess disease activity and efficacy of axSpA are still being explored.

OBJECTIVE

The purpose of this study was to find an optimal quantitative indicator for evaluating disease activity and curative effect of axSpA, using multi-b-values-fitting RESOLVE DWI.

METHODS

The prospective study included 106 patients divided into axSpA group (n=89) and no-axSpA group (n=17) by Assessment of Spondyloarthritis international Society (ASAS) criteria. The axSpA group were divided into active group and inactive group according to ASDAS-CRP. The active group treated with systematic tumour necrosis factor inhibitors (TNFi) was selected as treatment group (n=20). All patients underwent MRI examination of sacroiliac joints (SIJs), including RESOLVE DWI. The ADC values of subchondral bone marrow in SIJs were measured (ADC was b=50,500s/mm fitting, ADC was b=50,700s/mm fitting, and ADC was b=50,500,700s/mm fitting). By comparing the ADC values between different groups, a relatively optimal b-values-fitting sequence was obtained, further evaluating curative effect of the treatment group.

RESULTD

The ADC values of axSpA group, inactive group and active group SIJs were all higher than those of no-axSpA group. The ADC values of active group SIJs were all higher than those of inactive group. ADC had the largest AUC, relative higher sensitivity and specificity while taking account of the image quality than ADC and ADC between different groups. In the treatment group, there was no significant difference in ADC values between pre-treatment and 3 weeks, 3 weeks and 6 weeks, 6 weeks and 12 weeks (all P>0.0083, Bonferroni-corrected threshold), while the decreased ADC values in the interval of 6 weeks or more were statistically significant (all P<0.0083, Bonferroni-corrected threshold).

CONCLUSION

Multi-b-values-fitting (b=50,500,700s/mm) RESOLVE DWI has a certain advantage in evaluating disease activity of axSpA. It was worth noting that short-term review (3 weeks or less) of RESOLVE DWI was unsatisfactory and review at 6 weeks or later would help to evaluate curative effect of axSpA.