Patterns of vessel involvement in patients referred for possible Giant Cell Arteritis on PET/CT

Background: Giant cell arteritis (GCA), also known as temporal arteritis, is a medium to large vessel vasculitis that can cause acute irreversible vision loss. Rapid and accurate diagnosis is critical to allow the immediate introduction of corticosteroid-based therapy. Traditionally, temporal artery biopsy (TAB) is used to confirm the clinical suspicion. Problems with TAB are sampling errors as the disease may not uniformly involve the entire vessel, it is invasive and not always immediately available. PET/CT has been demonstrated to have a role in the identification of vessel involvement in arteritis. This has traditionally focused on larger vessels. The utility of PET/CT in the initial work up of GCA patients and the utility of assessing smaller cranial vessels have not been prospectively studied.

Methods: This prospective cross-sectional Giant Cell Arteritis and PET Scan (GAPS) study was conducted at Royal North Shore Hospital, a tertiary referral centre in Sydney, Australia between May 2016 and July 2018. Patients newly suspected of having GCA were eligible for inclusion. Patients with active malignancy, connective tissue disease and corticosteroid use were excluded. Scans were obtained within 72 hours of commencing corticosteroids and prior to TAB on a Siemens BiographTM mCT time-of-flight scanner after a 4 hour fast. 100MBq of FDG was administered 60 minutes prior to imaging. Scan field was vertex to diaphragm with 1mm CT reconstruction. All scans were independently reviewed by two blinded Nuclear medicine specialists scoring individual thoracic and cranial vessel uptake (0=no uptake, 1 equivocal/minimal, 2 moderate and 3 very marked) and giving an overall binary interpretation of disease presence. Final consensus read for disagreement in 9 patients was performed.

Results: 64 patients were recruited. 58 proceeded to TAB. Of these 12 (21%) were positive for GCA. A clinical review panel (blinded to PET results) assessed 21 patients as having GCA. 20% had relevant other findings such as infection (7) and malignancy (5). PET/CT had a sensitivity of 92% and specificity of 85% compared with TAB as the reference standard. No single vessel was always involved. Maxillary vessels were scored at 2 or greater in 42% (reader 1) and 50% (reader 2) of TAB positive patients compared to 4% and 7% in TAB negative patients. Temporal vessels were scored at 2 or greater in 33% (reader 1) and 42% (reader 2) compared to 4% and 7% in TAB negative patients. Cranial vessels (temporal, maxillary, occipital and vertebral) were the only site of abnormal (>2) uptake in 25% of TAB positive patients. Occipital artery involvement was a very specific finding. In the clinically determined GCA group, findings were similar though any cranial involvement was rare in the clinically negative group (2% patients for reader 1 and 5% patients for reader 2). The aorta was never the only site of abnormal uptake in GCA patients. 3 patients with negative TAB had very suggestive imaging and GCA based on the clinical review panel and are likely false negative biopsies. 1 PET/CT false positive patient had malignancy and clearly abnormal large vessels, possibly representing malignancy associated vasculitis

Conclusions: Inclusion of the cranial arteries in the assessment of GCA patients is both feasible and worthwhile. 25% of TAB positive patients only have uptake in the cranial vessels. PET/CT has a high accuracy in GCA diagnosis and can guide clinicians in early treatment decisions for this serious condition.