Inflamed temporal artery: histologic findings in 354 biopsies, with clinical correlations.

We reviewed 888 temporal artery biopsies (TAB) performed in 871 patients in a single institution from January 1986 to December 2013. Forty-four biopsies (4.9%) were inadequate, 490 (55.2%) were devoid of inflammation and were considered negative, and 354 (39.9%) showed inflammation and were considered positive. On the basis of the localization of the inflammation, positive TABs were further classified into 4 categories: small vessel vasculitis (SVV), in which inflammation was limited to small periadventitial vessels devoid of muscular coat, with sparing of the temporal artery (32 cases, 9% of the positive biopsies); vasa vasorum vasculitis (VVV), in which inflammation was limited to the adventitial vasa vasorum (23 cases, 6.5% of the positive biopsies); inflammation limited to adventitia (ILA), in which inflammation extended from a strictly perivascular localization to the surrounding adventitia, without medial involvement (25 cases, 7% of the positive biopsies); and transmural inflammation (TMI), in which inflammation crossed the external elastic lamina and extended to the media (274 cases, 77.5% of the positive biopsies). In TMI, inflammation was generally more prominent between media and adventitia and mostly consisted of T lymphocytes and macrophages, with occasionally a significant number of plasma cells. Numerous eosinophils or neutrophils (with or without leucocytoclasia and suppurative necrosis), fibrinoid necrosis (limited to small branches of the temporal artery), and acute thrombosis were unusual, being present in 8%, 1.8%, 0.7%, and 9.5% of our biopsies with TMI, respectively. Giant cells, laminar necrosis, and calcifications prevailed along the internal elastic lamina and were present in 74.8%, 25.2%, and 20% of the biopsies with TMI, respectively. Among the 322 patients with positive TAB on whom we obtained clinical information, 317 had giant cell arteritis and 5 had a different disease: 3 (with SVV at histology) had ANCA-associated vasculitis, 1 (with SVV with amyloid deposits) had primary systemic amyloidosis, and 1 (with TMI limited to a small branch) had polyarteritis nodosa. In none of these cases the biopsy showed fibrinoid necrosis or significant numbers of eosinophils or neutrophils. Considering the 317 patients with giant cell arteritis, those with SVV and VVV compared with those with TMI had a significantly lower frequency of cranial manifestation (including headache, jaw claudication, and abnormalities of temporal arteries), lower serum levels of acute-phase reactants, and a reduced frequency of prednisone therapy at the time of TAB, of the "halo sign" at color duplex sonography of temporal arteries, and of systemic symptoms (for VVV). Polymyalgia rheumatica and blindness were equally represented in all patients groups, whereas there was a higher frequency of male sex and peripheral arthritis in patients with SVV. Patients with ILA were more similar to those with TMI, having a lower frequency of headache, of abnormalities of temporal arteries, and of a positive "halo sign" at color duplex sonography of temporal arteries. In conclusion, the histologic spectrum of inflammatory lesions that can be found in TAB is broad, and the differences have clinical implications.