PROPOSAL FOR OPTIMIZATION OF DIAGNOSTIC IMAGING FOR GIANT CELL ARTERITIS USING THREE-DIMENSIONAL COMPUTED TOMOGRAPHY ANGIOGRAPHY IMAGE AND CONSTRUCTING VASCULAR MAPPING FROM VASCULAR ULTRASONOGRAPHY AS REFERENCES

Background: The development of rapid and accurate methods of diagnosing giant cell arteritis (GCA) is critical to prevent blindness and stroke, which may develop rapidly in patients with GCA. In 2018, EULAR published recommendations that the first imaging modality for GCA should be vascular ultrasonography without biopsy. However, many institutions still consider biopsy to make an important contribution to the diagnosis of GCA. Objectives: Our purpose is to eliminate blindness and stroke among GCA patients by optimizing diagnostic imaging and method to diagnose GCA employed by vascular ultrasonography (V-US), CT Angiography (CTA), MRI/A, and PET/CT without biopsy. Methods: We evaluated the clinical and serological characteristics of 20 patients who were diagnosed with GCA at our hospital from 2012 to 2018, and compared the image and biopsy findings of these patients. We then evaluated the effect of optimizing diagnostic imaging and methods for patients with suspected GCA who visited our hospital during 2019. Vascular mapping was carried out using V-US for 3DCTA and other imaging methods as references. Results: Table 1 shows the clinical characteristics of the study population. The sensitivity of CTA for GCA was 85.7% (12 of 14 patients), which was the highest of the studied imaging methods. All biopsy-positive cases were diagnosed as GCA, and we compared these cases with cases with positive imaging findings. This revealed that CTA findings were correct (i.e., positive) in 66.7% (four of six patients), MRI/A findings were correct in in 33.3% (three of nine), V-US findings were correct in 50.0% (three of six). Therefore, CTA exhibited the highest sensitivity for positive findings. Comparison of biopsy-positive cases with cases in which imaging findings were negative revealed that CTA findings were correct (negative) in 33.3% (two of six patients), MRI/A findings were correct in 55.6% (five of nine), V-US was correct in 50.0% (three of six). Thus, CTA had the lowest sensitivity for negative findings. Comparison of CTA findings of positive cases with other imaging modalities which reported positive findings revealed MRI/A findings to be correct in 44.4% (four of nine patients), PET/CT findings to be correct in 50.0% (one of two), V-US to be correct in 63.3% (five of eight). Thus, V-US had the highest agreement with CTA. We carried out vascular mapping by V-US using 3DCTA and other imaging methods and produced references to improve the accuracy of diagnosis. Using these references, we diagnosed five cases of GCA among the 20 patients; the positive predictive value of V-US was 80% (four of five patients) and negative predictive value was 86.7% (13 of 15 patients). The number of biopsies performed decreased from 50% (10 of 20 patients) from 2012 to 2018 to 15% (3 of 20 patients) in 2019. Two cases in the present study had positive findings in both biopsy and V-US; in one case, biopsy, CTA, and MRI/A were negative while V-US revealed positive findings. No patients with GCA developed blindness or stroke during 2019. Conclusion: We propose that V-US should be performed as the first examination for the diagnosis of GCA by the creation of vascular mappings when GCA is suspected in order to prevent blindness and stroke. References: [1]Christian Dejaco et al.EULAR recommendations for the use of imaging in large vasculitis in clinical practice.Annals of the Rheumatic Diseases, 2018 May;77(5):636-643 [2]Kawamoto T et al.Diagnosis of giant cell arthritis by head-contrast three-dimensional computed tomography angiography. Journal of Medical Case Reports 2019 Sep 11;13(1):285. Disclosure of Interests: None declared