Article Text

Download PDFPDF

The use of magnetic resonance imaging in the diagnosis of suspected giant cell arteritis
  1. S O Brannan1,
  2. D Cheung1,
  3. P I Murray1,
  4. C Dewar2,
  5. P Guest3
  1. 1Birmingham and Midland Eye Centre, Dudley Road, Birmingham B18 7QU, UK
  2. 2Sandwell and West Birmingham Hospitals NHS Trust, City Hospital, Birmingham, UK
  3. 3Queen Elizabeth Hospital, University Hospitals Birmingham, NHS Trust, Birmingham, UK
  1. Correspondence to: Professor P I Murray Academic Unit of Ophthalmology, Division of Immunity and Infection, Birmingham and Midland Eye Centre, Sandwell and West Birmingham Hospitals NHS Trust, City Hospital, Dudley Road, Birmingham B18 7QU, UK;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Giant cell arteritis (GCA) is a vasculitis of unknown origin that has a predisposition for the cranial arteries in the elderly. It has potentially devastating visually complications and produces a broad range of symptoms and signs that mimic many other medical and surgical conditions. Blood tests reflect the underlying inflammatory process, yet the erythrocyte sedimentation rate (ESR) may be normal in 8% of patients with biopsy proved GCA.1 Nevertheless, making a definitive diagnosis has importance therapeutically as patients are committed to a lengthy oral corticosteroid regimen. Non-invasive techniques, such as colour Doppler or duplex ultrasonography, have been studied in an attempt to improve patient preselection for temporal artery biopsy (TAB).2,3 Magnetic resonance imaging (MRI) has been shown to improve the diagnosis of early Takayasu arteritis.4 More recently several case reports have described the diagnostic potential of MR angiography and gadolinium contrast MRI in demonstrating the vessel changes of GCA.5,6 We compared the ability of MRI to detect changes in the temporal arteries with TAB in patients clinically suspected of having GCA.


A prospective, pilot, single masked study of seven female patients (age range 60–88 years, mean 76 years) with suspected giant cell arteritis, and two age matched healthy controls was undertaken. Local research ethical approval and informed written consent were obtained. All patients underwent a standard clinical examination including a detailed history and clinical examination. Investigations included ESR and C reactive protein (CRP). Each patient was given a GCA criteria “score” based on the 1990 ACR (American College of Rheumatology) classification7 (Table 1). Within 48 hours of presentation patients underwent a unilateral temporal artery MRI scan on a 1.5T scanner using a surface coil and small field of view. T1 and T2 weighted images perpendicular to the temporal artery and a time of flight sequence were obtained. The MRI visualised the location of the temporal artery that was subsequently biopsied in a standard manner within 24 hours of the scan. Two healthy age matched controls also underwent a medical assessment, ESR and CRP, and an MRI as detailed above, but a TAB was not performed. The MRI scans were reported by an independent, masked neuroradiologist.

Table 1

 1990 American College of Rheumatology criteria for the classification of giant cell (temporal) arteritis (traditional format)

Each patient’s ACR criteria “score” and the results of the MRI scan and TAB are shown in Table 2. The finding of three out of five ACR criteria is associated with a 94% sensitivity and 91% specificity for the diagnosis of GCA.7 There were two positive and one equivocal TAB result from the seven patients, but no positive MRI findings were identified. However, when using the ACR criteria as “gold standard,” there were two true negative MRI scan results compared with three false negative scan results. The two remaining MRI scans were described as equivocal, in comparison with the ACR criteria—one patient was positive for GCA and the other patient’s ACR criteria “score” was negative for GCA. From the data the negative predictive values of MRI scanning and TAB for GCA were 40% and 50%, respectively. Of the five patients who showed a prompt response to oral corticosteroid, the MRI scan was negative in four and equivocal in the other.

Table 2

 Summary of results


Although our study sample was small our findings suggest that MRI scanning was unable to distinguish between a normal and an affected artery. We conclude that there is no potential for the use of MRI scanning without contrast enhancement in the evaluation of patients with suspected GCA.