Background/Purpose: To describe the characteristics and diagnosis of patients with negative temporal artery biopsy (TAB) for whom a final diagnosis of giant cell arteritis (GCA) was excluded.

Methods: We performed a prospective bicentric cohort study from January 2010 until December 2014. Patients with a clinical suspicion of GCA who underwent TAB were included. ACR criteria for the diagnosis of GCA, clinical and biological parameters, final diagnosis and glucocorticoids treatment were collected. Patients final diagnosis was (1) GCA with histological proof (TAB⁺-GCA) ; (2) GCA without an histological proof (TAB^–-GCA); (3) isolated polymyalgia rheumatica (PMR) and (4) another diagnosis (neither GCA nor PMR) (GCA^–PMR^–).

Results: 248 patients with a mean age of 73.7 years were included. Among them, 69 patients (27.8 %) had a final diagnosis of TAB⁺-GCA; 65 patients (26.2%) had TAB^–-GCA; 23 patients (9.2%) had PMR and 91 patients (36.7%) had another diagnosis (GCA^–PMR^–). Among GCA^–PMR^– patients, 9.8% fullfilled 3/5 ACR Criteria for the diagnostic of GCA. GCA^–PMR^– patients less often had clinical signs of GCA: headaches (37% vs 65.8%, p <0.0001), scalp tenderness (18.6 % vs 39 %, p=0.0019), jaw claudication (13.7% vs 30%, p=0.0073) and temporal artery abnomalities (2.3% vs 18.9%, p=0.0006). These patients presented a lower C-reactive protein (CRP) and erythrocyte sedimentation rate (51.7 vs 70.5 mg/L, p=0.03; 53.6 vs 74 mm/hour, p=0.0003 respectively). The proportion of patients with CRP < 5 mg/L was more important in GCA^–PMR^– patients (27.4% vs 10.8%, p=0.0014). The main final diagnoses for these patients were ophthalmologic (17%, n=16), systemic inflammatory (16.7%, n=15) and infectious diseases (13.1 %, n=12). No diagnosis was made in 20 of them (21.9%). Ophthalmologic causes included: non-arteritic anterior ischemic optic neuropathy (n=4), isolated ocular nerve palsy (n=4), papilledema, optic atrophy and macular degeneration (n=2 for each), retrobulbar neuritis and central retinal artery occlusion (n=1 each). Systematic diseases included: connective tissue diseases (Sjögren’s syndrome (n=2), Sharp syndrome (n=1), Antiphoslipid syndrome (n=1)); systemic vasculitis (ANCA vasculitis (n=4), Takayasu arteritis (n=2), cryoglobulinemia (n=2)), Still’s disease (n=2) and a diffuse infiltrative pneumonia. Infections consisted of intracellular infections (n=4), endocarditis, cellulitis, dental infection, urinary tract infection and pneumonia (n=1 each) and three suspected infections without documentation. Glucocorticoids were prescribed in 45% of GCA^–PMR^–patients.

Conclusion: Data available on the clinical characterisitics and diagnosis of GCA^–PMR^– patients are scarse. Ophthalmologic signs and persistent inflammatory syndrome were the two main circumstances leading to a clinical suspicion of GCA. GCA^–PMR^– patients have a significantly lower rate of specifical symptoms of GCA than TAB⁺-GCA and more often a normal CRP. In almost half of these patients, glucocorticoids were initiated in the setting of an ophthalmologic emergency or suspected systemic disease. An algorithm combining GCA symptoms and CRP rate could be helpful to decide if TAB should be performed in patients with a clinical suspicion of GCA.