Background/Purpose:

The performance of ¹⁸F-FDG PET/CT in the diagnosis and assessment of disease activity of LVV has not been fully established. The interpretation of data from ¹⁸F-FDG PET/CT studies on LVV is difficult mainly due to (1) the scarcity of studies; (2) the heterogeneity of the LVV populations examined in each study; (3) the variety in the qualitative and semi-quantitative methods for imaging analyses. In the absence of gold standard parameters for the assessment of LVV arterial inflammation via ¹⁸F-FDG PET/CT, there is a great need to develop analytical methodologies useful to properly exploit the information derived from such studies.

Latent class analysis has recently been used to investigate alternative ways to classify LVV in few angiographic and MRI studies, on the basis of distribution patterns of arterial lesions. Other classical techniques – such as principal component analysis (PCA) and cluster analysis (CA) – have not yet been applied to investigate the distribution patterns of LVV, assessed by ¹⁸F-FDG PET/CT.

The aim of our study was to investigate whether PCA and CA are useful methods in identifying distinct distribution patterns of LVV, according to the specific diagnostic entity examined (giant cell arteritis, GCA; Takayasu’s arteritis, TA).

Methods:

A total of 135 ¹⁸F-FDG PET/CT studies performed for extra-cranial evaluation of LVV have been retrospectively examined by a nuclear physician blinded to clinical data. Maximum standardized uptake values (SUVmax) in 14 vascular districts including aortic segments (ascending aorta thoracic aorta, aortic arch, descending thoracic aorta and abdominal aorta) and the main tributaries (carotid, subclavian, axillary, iliac and femoral arteries; each bilaterally) have been measured and then transformed into Z-scores. Identification of distribution patterns of vascular involvement has been performed using PCA and agglomerative hierarchical CA (SPSS Statistics, version 22nd).

Results:

PCA and CA performed on the entire population of LVV subjects identified 3 groups of vascular districts with similar trends in terms of standardized SUVmax: (1) epiaortic arteries; (2) aortic arch and ascending aorta; (3) descending and abdominal aorta, together with iliac and femoral arteries. The same aggregation pattern has been observed in the PCA and CA performed on the GCA group, but not on the TA group, where a component including the entire aortic district (thoracic and abdominal aorta) was identified.

Conclusion:

PCA and cluster analysis approach revealed a subtle skewing in terms of distribution patterns of arterial involvement between the two main variants of LVV, assessed by ¹⁸F-FDG PET/CT SUVmax values. The influence of atherosclerosis and immunosenescence on the different trends in the aorta districts of TA and GCA needs to be further elucidated. 

References

Grayson PC, et al. Distribution of arterial lesions in Takayasu’s arteritis and giant cell arteritis. Ann Rheum Dis 2012;71:1329-1334.

Arnaud L, et al. Cluster analysis of arterial involvement in Takayasu arteritis reveals symmetric extension of the lesions in paired arterial beds. Arthritis Rheum 2011;63:1136-40.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/distinct-distribution-patterns-of-large-vessel-vasculitis-assessed-with-18f-fdg-petct-evidence-from-principal-component-and-cluster-analyses/