2412: Evaluation of the Pretest Probability Score, Ultrasound and Biopsy on GCA Diagnose: Data from Real Clinical Practice

María del Carmen San José Méndez¹, Uxía Couto Lareo¹, Vanesa Balboa Barreiro², Francisco J. Blanco¹, Bruno de aspe de la iglesia¹, Antonio Atanes Sandoval³, Diego Dios Santos¹, Jesús Carlos Fernández López¹, Mercedes Freire González¹, Guillermo González Arribas¹, Genaro Graña Gil¹, Natividad Oreiro Villar¹, Jose Antonio Pinto Tasende¹, Francisco Javier De Toro Santos¹, Clara Ventin Rodriguez¹, Maite silva díaz³ and Ana Lois Iglesias¹, ¹Rheumatology department, Complexo Hospitalario Universitario A Coruña (CHUAC). Instituto de Investigación Biomédica A Coruña (INIBIC), A Coruña, Spain, ²Clinic epidemiology and biostatistics department. Complejo Hospitalario Universitario A Coruña (CHUAC). Instituto de Investigación Biomédica A Coruña (INIBIC), A Coruña, Spain, ³Rheumatology department, Complexo Hospitalario Universitario A Coruña (CHUAC). Instituto de Investigación Biomédica A Coruña (INIBIC)., A Coruña, Spain

Background/Purpose: Giant cell arteritis (GCA) affects large and medium size arteries. Biopsy was considered the gold standard in the diagnosis¹, but in the last years other techniques are being implemented like temporal and axillar arteries ultrasound (US) with a good performance being included in the 2022 ACR GCA classification criteria². To improve the diagnosis of these patients a Pretest Probability Score (Southend) (PPS) has been developed³,stratifying patients into 3 different risk groups. The aim was to study the performance of the different tools available for the diagnosis of GCA: PPS, vascular ultrasound, and biopsy of Temporal Artery.


Methods: Retrospective study of the patients referred to our Rheumatology Department with GCA suspicion from January 2019 to October 2022. Demographic and clinical variables including the ones from the PPS were collected.Exploration of the 3 branches of temporal artery and axillar artery bilaterally was performed by grey scale and Color Doppler US (Esaote MylabSix probe 10-22 Hz) evaluating the halo sign and the intima-media thicknes⁴.PPS and US were carried out in all patients referred. Biopsy was performed in the necessary cases. Final GCA diagnosis was performed according to the rheumatologist decision (clinical diagnosis). A descriptive analysis of the sample was performed. Quantitative variables are expressed as mean (±SD) and qualitative ones as absolute frequencies and percentage. A comparative analysis was performed according to clinical diagnosis. The comparison of means was carried out using the Student's T or Mann-Whitney U test. The association between qualitative variables was analyzed using Chi-square test. The agreement between the different tests was analyzed using Kappa index. The predictive capacity of the different diagnostic tests is evaluated using ROC curves, providing their AUC, and the sensitivity and specificity values.

Results: Patients (97) with suspected GCA were included, 59.79% women, with a mean age of 74.4 (±9.69). Final diagnosis attending to clinical criteria and other tests was made in 23 patients (23.7%). In GCA compared with non GCA patients showed more frequently and statistical significance: headache, scalp hypersensitivity, facial pain, jaw claudication and changes on temporal artery. Ultrasound was positive in 91.3% (n=21) GCA patients, from them 61.90% had a positive biopsy (Table 1). Area under ROC curve (AUC), sensitivity and specificity for PPS were 0.83 (0.75-0.92), 78.3% and 78.4%, respectively, for US were 0.91 (0.84-0.99), 91.3% and 91.8%, and for biopsy, 0.81 (0.67-0.94), 61.9% and 100%. Data from the combination of PPS+US showed an AUC of 0.96 (0.92-1) (sensitivity 91.3% and specificity 94.5%). Combination of PPS+biopsy showed an AUC of 0.90 (sensitivity 71.4% and specificity 91.2%) (Figure 1).

Conclusion: The combination of US with PPS value improves the individual predictive capacity of each these diagnostic techniques.






M. San José Méndez: None; U. Couto Lareo: None; V. Balboa Barreiro: None; F. Blanco: None; B. de aspe de la iglesia: None; A. Atanes Sandoval: None; D. Dios Santos: None; J. Fernández López: None; M. Freire González: None; G. González Arribas: None; G. Graña Gil: None; N. Oreiro Villar: None; J. Pinto Tasende: None; F. De Toro Santos: None; C. Ventin Rodriguez: None; M. silva díaz: None; A. Lois Iglesias: None.