What Is the Level of Agreement between Local and Central Readers in the Detection of Active and Structural MRI Lesions Typical of Axial Spondyloarthritis? Data from the Assessments in Spondyloarthritis Classification Cohort Study

Walter P. Maksymowych^1,2, Xenofon Baraliakos³, Robert G. Lambert⁴, Ulrich Weber⁵, Joachim Sieper⁶, Stephanie Wichuk¹, Denis Poddubnyy⁷, Mikkel Østergaard⁸, Joel Paschke², Susanne J Pedersen⁸ and Pedro Machado⁹, ¹Medicine, Medicine, University of Alberta, Edmonton, AB, Canada, ²CaRE Arthritis, Edmonton, AB, Canada, ³Rheumazentrum Ruhrgebiet Herne, Herne, Germany, ⁴Radiology, Radiology, University of Alberta, Edmonton, AB, Canada, ⁵University of Southern Denmark, Odense, Denmark, ⁶Charité Universitätsmeidzin Berlin, Berlin, Germany, ⁷Rheumatology, Charité Universitätsmeidzin Berlin, Berlin, Germany, ⁸COPECARE University of Copenhagen, Copenhagen, Denmark, ⁹University College London, London, United Kingdom

Meeting: 2018 ACR/ARHP Annual Meeting

Keywords: classification criteria and spondylarthritis, MRI

Session Information

Date: Sunday, October 21, 2018

Title: Spondyloarthritis Including Psoriatic Arthritis – Clinical Poster I: Imaging, Clinical Studies, and Treatment

Session Type: ACR Poster Session A

Session Time: 9:00AM-11:00AM

Background/Purpose:

There has been no central reader evaluation of MRI scans from the ASAS Classification Cohort (ASAS-CC)¹ to compare detection of lesions between central and ASAS-CC local site readers. Local readers reported active MRI lesions typical of axSpA in the SIJ of 61.6% and 2.2% of patients diagnosed with axSpA and non-axSpA back pain, respectively. Structural lesions were recorded but not reported. We aimed to compare the frequencies of active and structural lesions from the ASAS-CC according to diagnostic category and agreement for their detection between ASAS-CC local site readers and central readers from the ASAS-MRI group.

Methods:

MRI lesions were recorded in an eCRF that included wording of lesions defining active and structural lesions typical of axSpA that was exactly the same as in the original ASAS-CC eCRF permitting comparisons between central and local site readers. In addition, lesions that met the criteria for an ASAS positive MRI were recorded by central readers. MRI images were available from 276 of the 495 cases who had MRI performed in the ASAS-CC and also had a local rheumatologist expert opinion diagnosis. Image quality was considered sufficient to record global data by 7 central readers in all cases. Lesion frequencies were assessed descriptively according to majority agreement (≥4/7) of central reader data and also any 2 central readers. Agreement for detection of MRI lesions was compared using the kappa coefficient.

Results:

Significant differences in lesion frequencies were observed according to diagnostic category (Table 1). However, the frequency of active lesions reported by local readers (61%)was greater than for central readers (43.2%). Structural lesions were also more frequently reported by local readers (42.1%) compared to central readers (34.7%) but less so than active lesions. Agreement for detection of active lesions was good but poor for structural lesions (Table 2).

Conclusion:

Local readers may have overestimated the presence of MRI lesions in the ASAS-CC, particularly active lesions. Agreement for detection of structural lesions was limited. The impact on diagnosis by the local rheumatologist cannot be deduced.

1. Rudwaleit et al. Ann Rheum Dis 2009;68: 777-831.

Table 1. Central MRI reader assessment according to diagnostic ascertainment of local physician in the ASAS classification study for all 276 cases with MRI scans of SIJ and baseline local clinical and imaging assessment.

Reader	MRI Lesion Type	Local Rheumatologist Diagnosis		P value
Reader	MRI Lesion Type	AxSpA (n=199)	Not AxSpA (n=77)	P value
ACTIVE LESIONS
Local	Active lesions typical of axSpA	114 (61.0%)†	3 (4.2%)†	<0.0001
Central (≥4/7 agreement)	Active lesions typical of axSpA	86 (43.2%)	3 (3.9%)	<0.0001
Central (≥4/7 agreement)	ASAS MRI positive	79 (39.7%)	2 (2.6%)	<0.0001
Central (any 2 readers)	Active lesions typical of axSpA	96(48.2%)	7 (9.1%)	<0.0001
Central (any 2 readers)	ASAS MRI positive	92 (46.2%)	6 (7.8%)	<0.0001
STRUCTURAL LESIONS
Local	Structural lesions typical of axSpA	77 (42.1%)^#	6 (8.5%)^#	<0.0001
Central (≥4/7 agreement)	Structural lesions typical of axSpA	69 (34.7%)	6 (7.8%)	<0.0001
Central (any 2 readers)	Structural lesions typical of axSpA	94 (47.2%)	10 (13%)	<0.0001

† Total with clinical and MRI data = 258

# Total with clinical and MRI data = 254

Table 2. Agreement between central and local readers for active and structural lesions typical for axSpA.

Local Reader		Central Readers
		Active lesion (any 2 readers)		Active Lesion (≥4 readers)
		Yes	No	Yes	No
Active lesion	Yes	85	32	78	39
	No	14	127	8	133
Kappa (95% CI)		0.64 (0.54-0.73)		0.62 (0.53-0.72)
Structural lesion	Yes	58	25	43	40
	No	41	130	27	144
Kappa (95% CI)		0.44 (0.32 to 0.55)		0.38 (0.25 to 0.50)

Disclosure: W. P. Maksymowych, CaRE rthritis, 9; X. Baraliakos, None; R. G. Lambert, None; U. Weber, None; J. Sieper, None; S. Wichuk, None; D. Poddubnyy, None; M. Østergaard, None; J. Paschke, None; S. J. Pedersen, None; P. Machado, None.

To cite this abstract in AMA style:

Maksymowych WP, Baraliakos X, Lambert RG, Weber U, Sieper J, Wichuk S, Poddubnyy D, Østergaard M, Paschke J, Pedersen SJ, Machado P. What Is the Level of Agreement between Local and Central Readers in the Detection of Active and Structural MRI Lesions Typical of Axial Spondyloarthritis? Data from the Assessments in Spondyloarthritis Classification Cohort Study [abstract]. Arthritis Rheumatol. 2018; 70 (suppl 9). https://acrabstracts.org/abstract/what-is-the-level-of-agreement-between-local-and-central-readers-in-the-detection-of-active-and-structural-mri-lesions-typical-of-axial-spondyloarthritis-data-from-the-assessments-in-spondyloarthriti/. Accessed .

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/what-is-the-level-of-agreement-between-local-and-central-readers-in-the-detection-of-active-and-structural-mri-lesions-typical-of-axial-spondyloarthritis-data-from-the-assessments-in-spondyloarthriti/