Table 1.
Overview of studies reporting diagnostic accuracy of semi-quantitative indices for GCA or TAK. Only studies in which all patients underwent [18F]FDG-PET/CT were selected. † At least part of patients receiving treatment at the time of the scan. †† At least part of the patients received treatment for less than 3 days before the scan. FUO = fever of unknown origin. GCA = giant cell arteritis. ICV = inferior caval vein. IJV = internal jugular vein. SCV = superior caval vein. TAB = temporal artery biopsy. TAK = Takayasu’s arteritis. TVS = total vascular score.
| Authors and Year | Country | Study Design | Patient Population | No. of Patients | Control Group Yes/No |
No. of Controls |
Reference Standard LVV | (Semi-) Quantitative Index | Sensitivity | Specificity |
|---|---|---|---|---|---|---|---|---|---|---|
| Rottenburger et al. 2021 | Switzerland | Retrospective | Suspected GCA † | 17 (GCA) | Yes (suspected GCA but final diagnosis not GCA) | 17 | GiACTA trial criteria | Visual FDG uptake at temporal artery (visible) | 53% | 100% |
| TBR: SUVmax temporal artery/SUVmean liver > 1 | 53% | 100% | ||||||||
| Nienhuis et al. 2020 | Netherlands | Retrospective | cGCA †† | 24 | Yes (oncology) | 24 | TAB positive | Visual > background | 83% | 75% |
| Visual ≥ liver | 58% | 96% | ||||||||
| SUVmax > 5.0 | 79% | 92% | ||||||||
| Imfeld et al. 2019 | Switzerland | Retrospective | Suspected GCA † | 68 (GCA) | Yes (suspected GCA but final diagnosis not -GCA) | 34 | TAB or ACR 1990 or incomplete ACR 1990 with positive imaging findings | TBR: SUVmax supra-aortic arteries/SUVmean liver > 1 and/or TBR: SUVmax aorta or iliofemoral arteries/SUVmean liver >1.3 | 72% | 85% |
| Vaidyanathan et al. 2018 | United Kingdom | Retrospective | Suspected LVV | 17 (LVV) | Yes (suspected LVV but final diagnosis not LVV) | 19 | Clinical diagnosis | TVS (range 0–21) > 3.5 | 79% | 94% |
| TBR: SUVmax aorta/SUVmax liver > 1.0 | 76% | 100% | ||||||||
| Imfeld et al. 2017 | Switzerland | Retrospective | Suspected GCA † | 68 (GCA) | Yes (suspected GCA but final diagnosis not GCA) | 35 | TAB or ACR 1990 or incomplete ACR 1990 with positive imaging findings | TBR: SUVmax supra-aortic arteries/SUVmean liver > 1.0 | 71% | 91% |
| TBR: SUVmax thoracic aorta/SUVmean liver > 1.3 | 25% | 91% | ||||||||
| TBR: SUVmax abdominal aorta/SUVmean liver > 1.3 | 34% | 91% | ||||||||
| TBR: SUVmax iliofemoral arteries/SUVmean liver > 1.14 | 26% | 91% | ||||||||
| Clifford et al. 2017 | Canada | Prospective | New-onset GCA † | 28 | Yes (oncology) | 28 | ACR 1990 and clinical diagnosis | Visual TVS (range 0–24) ≥ 9 | 71% | 64% |
| Castellani et al. 2016 | Italy | Retrospective | New-onset GCA/TAK | 25 | Yes (oncology) | 15 | Unclear | Visual TVS (range 0–33) > 8 | 84% | 87% |
| Average SUVmean > 0.697 | 96% | 87% | ||||||||
| Stellingwerff et al. 2015 | Netherlands | Retrospective | New-onset GCA | 12 | Yes (inflammation, atherosclerosis, normal) | 53 | ACR 1990 or TAB or clinical diagnosis | Visual | 92% | 91% * |
| SUVmax aorta > 2.75 | 83% | 62% | ||||||||
| TBR: SUVmax aorta/liver > 1.03 | 92% | 92% | ||||||||
| TBR: SUVmax aorta/SCV > 1.48 | 92% | 73% | ||||||||
| TBR: SUVmax aorta/ICV > 1.72 | 75% | 96% | ||||||||
| Martinez-Rodriguez et al. 2014 | Spain | Prospective | Suspected LVV | 25 (LVV) | Yes (oncology) | 15 | Clinical and biochemical data, treatment response; TAB (n = 6) and MRA (n = 2) | SUVmax aorta wall > 1.74 | 80% | 83% |
| TBR: SUVmax aorta wall/lumen > 1.34 | 100% | 94% | ||||||||
| Prieto-Gonzalez et al. 2014 | Spain | Prospective | New-onset GCA † | 32 | Yes (oncology) | 20 | TAB | Mean SUVmax > 1.89 | 80% | 79% |
| Mean SUVmax supra-aortic > 1.70 | 81% | 79% | ||||||||
| MeanSUVmax aorta > 2.25 | 90% | 42% | ||||||||
| MeanSUVmax aorta > 2.65 | 58% | 90% | ||||||||
| Besson et al. 2014 | France | Retrospective | GCA † | 11 | Yes (infection, oncology) | 11 | TAB | TBR: highest SUVmax aortic arch/highest SUVmax liver > 0.71 | 82% | 55% |
| TBR: average SUVmax aortic arch/average SUVmax liver > 0.83 | 64% | 82% | ||||||||
| TBR: highest SUVmax aortic arch/SUVmax lung > 7.46 | 82% | 73% | ||||||||
| TBR: average SUVmax aortic arch/SUVmax lung > 6.9 | 73% | 73% | ||||||||
| TBR: highest SUVmax aortic arch/highest venous blood (Right IJV) > 1.53 | 82% | 91% | ||||||||
| TBR: average SUVmax aortic arch/average venous blood (Right IJV) > 1.63 | 82% | 73% | ||||||||
| Lehmann et al. 2011 | Germany | Retrospective | GCA or TAK | 20 (GCA, n=17; TAK, n = 3) | Yes (oncology) | 20 | ACR 1990 or histology | Overall visual interpretation | 65% (data for 13 patients and 16 controls only) | 80% |
| SUVmax > 1.78 | 90% | 45% |
*: average of the three subgroups of controls (specificity: 94%, 79%, 100% in arterial inflammation, atherosclerosis and normals, respectively).