We read with interest the recent article by Seo and Stone (1). The authors provide a thorough review of giant cell arteritis (GCA) and briefly discuss new approaches to imaging in large vessel vasculitis. They state that even though magnetic resonance imaging (MRI), electron beam computed tomography, and positron emission tomography have theoretical and technical advantages over conventional angiography, the optimal means of these techniques in clinical practice require further investigations. We agree with this assessment and would like to provide some information on our recent research activities using high-resolution MRI to detect mural inflammatory changes of the temporal artery in GCA (2).
We investigated 20 patients with clinically suspected GCA (mean age 72.4 years), using high-resolution MRI on a clinical 1.5 Tesla system with a dedicated eight element phased-array head-coil. (Magnetom Sonata; Siemens, Erlangen, Germany) Multislice T1-weighted spin-echo sequences with a sub-millimeter spatial resolution of 0.2 mm × 0.3 mm and a slice thickness of 3 mm were acquired, perpendicular to the vessel's orientation before and after intravenous injection of a contrast bolus (0.1mmole/kg, Magnevist, Schering, Germany) (Figure 1). Biopsy specimens of the temporal artery were obtained in 16 of 20 patients for definite diagnosis of GCA.
The mean ± SD thickness of the vessel wall and the lumen diameter were 0.88 ± 0.23 mm and 0.78 ± 0.29 mm in GCA-positive patients and 0.57 ± 0.25 mm and 0.7 ± 0.1 mm in GCA-negative patients, respectively. In 16 of 20 cases, characteristic signs of mural inflammatory changes such as thickening of the vessel wall and/or an increased contrast enhancement were correctly demonstrated on MRI (2). In these cases, diagnosis of GCA was established according to the American College of Rheumatology (ACR) clinical criteria (3), which included positive histologic findings in 14 cases. In 3 of 20 cases, MRI was true negative, and in one case, MRI was false negative. In the false-negative case, histology was also false negative and the diagnosis of GCA was established according to the ACR criteria (3). Possibly, the GCA involvement occurred predominantly in the occipital arteries and was neither in the MRI field nor in the site of the biopsy specimen.
Mural thickening and contrast enhancement on MR images is a direct indication of inflammation of the vessel wall and thus of active disease (4). Klein et al (5) have shown that inflammation of the temporal artery demonstrates a segmental distribution. This might be a reason for false-negative biopsy results. High-resolution MRI may be used to identify segments with the most intense mural inflammatory changes to determine the best biopsy site. Furthermore, the variable extent of involvement of each temporal artery and of each occipital or facial artery can be depicted within the same investigation. Potentially, this may reduce the number of false-negative biopsy specimens and reduce the number of bilateral biopsies performed because the first biopsy was not diagnostic. High-resolution MRI may also be used to monitor the activity of mural inflammatory changes with long-term corticosteroid use. We have shown how mural inflammatory changes decreased over the course of time during corticosteroid therapy (6).
The clinical value of color duplex-sonography in the diagnosis of GCA has been debated (7–10). Comparison of MRI with duplex-sonography including the question of whether MRI adds incremental value to duplex-sonography would be of interest and will be investigated in the future.
All patients investigated in this study were suspected of having GCA by either a rheumatologist or an ophthalmologist experienced in treating patients with GCA. This seems to be the reason for the high percentage of patients with positive biopsy results. Because all patients were referred to us for temporal artery evaluation, a possible referral bias cannot be ignored. Larger patient trials with a higher number of patients who ultimately do not have GCA will be needed to evaluate the reliability and diagnostic value of this imaging technique.
We conclude that high-resolution MRI enables the visualization of the temporal artery and its mural inflammatory changes in GCA. This modality might be useful in determining the correct diagnosis and in increasing the likelihood of obtaining a positive biopsy specimen. We think that this novel, noninvasive imaging modality for detection of mural inflammatory changes of the temporal artery in GCA deserves further exploration.