TBA (17A156)

Diagnostic Utility of CT Angiogram in Giant Cell Arteritis


Richard Conway, Anna E. Smyth, Richard G. Kavanagh, Rory O’Donohue, Graeme McNeill, Eric J. Heffernan, Eamonn S. Molloy, Ronan P. Killeen


Departments of Rheumatology and Radiology, St. Vincent’s University Hospital, Dublin, Ireland


Giant cell arteritis (GCA) is the most common form of systemic vasculitis with a lifetime risk of 1% in populations of northern European origin. It is a devastating disease with 25% of patients suffering cranial ischaemic complications such as vision loss and stroke. Temporal artery (TA) biopsy is positive in less than 50% of cases. There are no diagnostic criteria and the diagnosis is based on clinician gestalt. Imaging investigations are potentially useful in this setting; TA ultrasound has a sensitivity of 54% and specificity of 81% for GCA diagnosis. Therefore, almost half of GCA patients have normal imaging investigations; complementary imaging modalities are needed.


The purpose of this study was to investigate the utility of CT angiogram (CTA) of the TAs in GCA.


This case control study was performed in the setting of a prospective GCA registry which has recruited 334 patients since 2011. We included 14 patients who presented to the Emergency Department from March 2013 to July 2015, who were ultimately diagnosed as the initial presentation of GCA and had a CTA performed for the assessment of suspected stroke. We compared these to 14 age- and sex-matched controls presenting with the same symptoms in the same time frame. CTAs were evaluated for the presence of abnormalities indicative of TA vasculitis – vessel wall blurring, perivascular enhancement, and vessel narrowing or occlusion. The primary outcome was the presence of any TA abnormality.


Twenty-eight intracranial CTAs were evaluated in patients who presented with stroke symptoms; 14 ultimately diagnosed as GCA and 14 controls. CTA’s were blinded and evaluated independently by two neuroradiologists Figure 1, Image A demonstrates a normal appearing TA with a well-defined and continuous vessel. Images B, C and D show an abnormal TA with abrupt discontinuation of the normal artery (white arrows) with subsequent abnormal enhancement of the perivascular soft tissue distal to this cut-off point (red arrows) These findings were not seen in control subjects. CTA of the TAs had a sensitivity of 100% and specificity of 71%. Positive predictive value was 78% and negative predictive value 100%. Inter-observer agreement was moderate k= 0.53 (p=0.002).


CTA of the TAs represents a potentially useful diagnostic adjunct in GCA and warrants further investigation in this population. Abnormal appearing TAs with blurred vessel walls and soft tissue enhancement (Cigar Smoke sign) should elicit a work-up for GCA in patients presenting with symptoms of stroke.