Background The role of Fluoro-Deoxy-Glucose PET for the diagnosis of large vessel vasculitis (LVV) remains uncertain. Temporal artery biopsy (TAB) is the gold standard investigation for suspected Giant Cell Arteritis. Biopsy is invasive, may be negative due to skip lesions and does not provide information about the extent of vascular territories involved.
Objectives We aimed to determine the value of PET in the diagnosis of LVV and if any patient or clinical characteristics could help predict positive scans.
Methods We retrospectively reviewed all PET scans requested for the clinical indication of LVV, over the previous 12 months. Data collected were patient demographics, clinical characteristics at presentation, initial working diagnosis (categorised as low, moderate or high probability of LVV), ESR & CRP (at presentation, TAB and at PET), cumulative steroid dose in the 28 days prior to PET, temporal artery biopsy (TAB), reason for PET request, PET results including aortic maximum standard uptake value (SUV max).
Results 20 patients were identified who underwent PET scanning to look for the presence of underlying LVV. The female:male ratio was 3:1. Mean age 67 years (range 44-88 years). The initial working clinical diagnoses prior to PET scanning was GCA/LVV in 15 patients (clinical probability low =7, moderate=4, high =4 based on physician judgement) or PMR with LVV in the remaining 5 patients.
Based on the PET findings a final diagnosis of LVV was made in 10/20 (50%) patients (including one borderline positive scan for LVV). The remaining scans 10/20 (50%) were negative for LVV with 5 of these being investigated for incidental findings (thyroid nodule, lung nodule, cutaneous uptake, adrenal nodule and sigmoid colon uptake). Of the 10 positive PET scans 5/10 had a prior negative TAB. In the remaining 5, TAB was not performed as there were no clinical features of GCA. 7/10 PET negative scans had a prior negative TAB and the remaining 3 had no TAB performed.
Prednisolone use in the 28 days prior to PET varied. Median daily dose was 6 mg per day (range 0-40mg) and mean cumulative dose 192 mg (range 0-1120mg). Median aortic SUVmax was 3.3 (range <1.0 - 9.5) in PET positive cases, and 2.2 (1.5-3.1) in PET negative cases. For the PET positive scans, comparison of the cumulative prednisolone dose against the mean aortic SUVmax showed a negative trend (Spearman's coefficient R = -0.60) although this did not reach statistical significance (p=0.06). There was no difference in ESR & CRP between PET positive and negative cases (Mann-Whitney U testing).
Conclusions PET scanning can be useful in selected patients with suspected LVV. In our cohort of such patients PET was positive in 50% of TAB negative patients and changed their management plan. We also found it useful in identifying whether there was active LVV as opposed to conditions such as malignancy irrespective of TAB findings. There was an inverse relationship between cumulative steroid dose and isotope uptake suggesting that if indicated PET scanning should be done early in the course of the patients' symptoms.
Disclosure of Interest None declared