Assessing the Southend GCA Probability Score in Patients Referred to a University Rheumatology Practice for Suspected Giant Cell Arteritis (GCA)

Objectives The Southend GCA Probability Score (GCAPS) is a simple tool that can be used in clinic to assess the pre-test probability of giant cell arteritis (GCA), to guide decisions for testing and/or initiation of prednisone.[1] We aimed to retrospectively assess its validity in patients referred to the University of Alberta for suspected GCA. Methods Electronic medical records of patients referred to University of Alberta Rheumatology for query GCA between January 2022 and January 2024 were retrospectively reviewed. Patients with relapsing GCA, or those whose charts lacked sufficient data to calculate a GCAPS were excluded. Data was extracted from the first rheumatology visit and inputted into the Southend GCAPS Calculator. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of a High, Intermediate or Low risk GCAPS score were calculated using both 1) the clinical diagnosis of GCA at 3 months follow up, and 2) confirmed diagnosis of GCA by testing [either temporal artery biopsy, ultrasound, or PET/CT], as gold standard. Receiver operator characteristic (ROC) curves were plotted using numerical GCAPS scores. Results 81 referrals for new onset, suspected GCA were identified, of which 62 contained all data to calculate a GCAPS score. Ultimately, GCA was diagnosed in 39 patients (27 patients confirmed by test, 12 diagnosed clinically) and excluded in 23. Of 62 patients, 31 (50%) had a High risk GCAPS, 23 (37.7%) had Intermediate risk and 8 (12.9%) had Low risk GCAPs. See distribution of results (Table 1). Using the clinical diagnosis at 3 months as gold standard, the sensitivity of having either Intermediate/or High risk GCAPS was very high at 97.4%, with specificity of 30%, PPV of 70.4% and NPV 87.5%. The sensitivity of High risk GCAPS alone was lower at 66.7%, but with higher specificity of 78.3%, PPV=83.9%, NPV=58.1%. Using only confirmed diagnosis of GCA as gold standard, the sensitivity of Intermediate/or High risk GCAPs was 100%, with specificity of 30%, PPV of 62.8% and NPV 100%. The GCAPS ROC AUC was good at 0.805 (95% CI 0.697-0.913). At the previously determined optimal cut-point of 9.5, the sensitivity was 97.4% but specificity was low at 34.8% Table 1. Distribution of GCA diagnoses and GCAPS risk scores. Conclusion In our population, an Intermediate or High risk GCAPS score was highly sensitive for the final diagnosis of GCA, suggesting that for those with Low-risk scores, GCA is unlikely and other diagnosis should be strongly considered. References [1.] Laskou F. Clin Exp Rheumatol 2019;37 Suppl 117:104-8.