Background No study have evaluated if there is a correlation between histological pattern of temporal artery biopsy and survival in patients with biopsy-proven giant cell arteritis (GCA).

Objectives To investigate the relationship between histological pattern of temporal artery biopsy and survival in patients with biopsy-proven GCA in a defined area of Northern Italy.

Methods All patients with incident temporal artery biopsy-positive GCA diagnosed between 1986 and 2012 living in the Reggio Emilia area were identified using pathology register and by reviewing all histopathological specimens. Patients were followed from the time of diagnosis until either their death or until December 31st, 2013. On the basis of the localization of the inflammation, positive TABs were classified into 4 categories: transmural inflammation (TMI), (229 cases, 81.8% of the positive biopsies); inflammation limited to adventitia (ILA), (14 cases, 5% of the positive biopsies); vasa vasorum vasculitis (VVV), (13 cases, 4.6% of the positive biopsies); and small vessel vasculitis (SVV), (24 cases, 8.6% of the positive biopsies). A semiquantitative 4 grade scale was used to describe the severity of mural inflammation (no mural inflammation, 81 cases, 28.9%, mild mural inflammation, 43 cases, 15.4%, moderate mural inflammation, 75 cases, 26.8%, severe mural inflammation, 81 cases, 28.9%). For each case, we identified one control from the same geographic area matched for age and gender. Mortality rates and specific causes of death were reported and compared between cases and controls and between cases with different histological pattern. TAB histological pattern and severity of mural inflammation entered in a Cox proportional regression analysis along with demographical, clinical and laboratory data at disease onset.

Results There were 280 incident cases of biopsy-proven GCA (205 women, mean age at disease onset 74.3 + 7.5 y, mean duration of follow-up 95 + 58 months) during the 26-year study period with complete clinical, laboratory and survival data. During the follow- up period 159 (56.8%) patients died (mean survival time from disease onset 101 + 60 months). Patients with histological pattern ILA and VVV had reduced hazard ratio mortality as compared to TMI type (0.38 and 0.12 respectively). Survival of patients with different histological TAB pattern of GCA was not different from controls. The most frequent causes of death were cardiovascular diseases, cancer, and respiratory diseases. No significant differences in causes of death were observed comparing the 4 pattern of histological TAB involvement.

Conclusions The histologic spectrum of inflammatory lesions in TAB positive GCA have correlation with survival.

Disclosure of Interest None declared