THU0443 A probability score from a fast track clinic to aid the management of suspected giant cell arteritis. (12th June 2018)

• Record Type: Journal Article
• Authors: Laskou, F.; Coath, F.; Mackie, S.; Banerjee, S.; Aung, T.; Dasgupta, B.
• Is Part Of: Annals of the rheumatic diseases. Volume 77(2018)Supplement 2
• Page Start: 434
• Page End: 434
• Publication Date: 2018-06-12
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2018-eular.2062
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Diagnosis of Giant Cell Arteritis(GCA) is difficult since its manifestations are protean 1 . Under-diagnosis is associated with ischaemic complications whereas over diagnosis is associated with inappropriate glucocorticoids (GC) 2 .GCA is diagnosed by different specialties, including family physicians, who would benefit from a clinical prediction score.A fast track pathway also requires clinical triage in terms of probability of disease. We evaluated all referred patients (08/16–08/17) to develop a pre-test probability score(PTBS) to support a diagnostic pathway and decision-making. Methods: The PTBS was generated from long standing clinical experience. Information collected at initial assessment was given varying positive weightage. This included baseline demographics (age-gender), symptomatology at presentation(onset, headache and scalp tenderness, ischaemic symptoms, constitutional symptoms, polymyalgia), C-Reactive protein(CRP) and examination findings(ischaemic ophthalmic complications, temporal artery abnormalities, extra-cranial abnormalities, cranial nerve palsies). Negative weightage was given for competing diagnoses (infection, cancer, head and neck pathology, systemic rheumatological diseases). The PTBS was compared with the final diagnosis as GCA or non-GCA 6 months after the initial assessment. Analysis was performed in Stata SE, version 13.1. Results: 122 PTBS were collected of which CRP was missing in 1 case which was excluded from the analysis.23 had a final diagnosis of GCA at 6 months follow up. The rest consist our control group(99 patients). The area under the ROC curve for the 121 cases was 0.953 (figure 1). Using the bootstrap method gave an estimated area under the ROC curve (95% confidence interval) of 0.953 (0.911, 0.994). At the point of inflection, corresponding to a cut point of 9.5, sensitivity was 95.7%, and specificity was 86.7%;the likelihood ratio for a positive test was 7.2 and the likelihood ratio for a negative test, 0.050. At this cut point, 88.4% cases were correctly classified. Conclusions: This single centre retrospective study suggests that PTBS is a useful standardised assessment tool for rating pre-test probability for GCA with high levels of sensitivity and specificity. PTBS may reduce variation in clinical assessment and aid decision making.A patient with low probability score(<9.5) can be managed with colour doppler ultrasound examination(US) which if negative will exclude the disease and the clinician can reassure patient.A patient with high PTBS and positive US can safely have the diagnosis confirmed and treated with GC. With intermediate scores, conflicting PTBS and US findings, equivocal US, additional investigations including TA biopsy and/or other imaging scans may be needed. Our results need validation in a prospective study and in internal and external validation cohorts. PTBS has the potential for forming the basis for education programme for the correct and early diagnosis of GCA and limit inappropriate GC in non GCA mimics. References: [1] Murchison AP, et al. Validity of the American College of Rheumatology Criteria for the Diagnosis of Giant Cell Arteritis.Am J Ophthalmol2012Oct;154(4):722–9. [2] Elsideeg S, et al. SAT0293 Association between Clinical Presentation and Blindness in Gca and Effects of Gca-Related Visual Loss on Morbidity, Mortality and Quality of Life. Ann Rheum Dis2014Jun;73(Suppl 2):699.1–699. Disclosure of Interest: None declared