부산대학교 도서관

Abstract Purpose Data for ANCA-associated vasculitis (AAV) patients requiring intensive care are scarce. Methods We included 97 consecutive patients with acute AAV manifestations (new onset or relapsing disease), admitted to 18 intensive care units (ICUs) over a 10-year period (2002–2012). A group of 95 consecutive AAV patients with new onset or relapsing disease, admitted to two nephrology departments with acute vasculitis manifestations, constituted the control group. Results In the ICU group, patients predominantly showed granulomatosis with polyangiitis and proteinase-3 ANCAs. Compared with the non-ICU group, the ICU group showed comparable Birmingham vasculitis activity score and a higher frequency of heart, central nervous system and lungs involvements. Respiratory assistance, renal replacement therapy and vasopressors were required in 68.0, 56.7 and 26.8% of ICU patients, respectively. All but one patient (99%) received glucocorticoids, 85.6% received cyclophosphamide, and 49.5% had plasma exchanges as remission induction regimens. Fifteen (15.5%) patients died during the ICU stay. The following were significantly associated with ICU mortality in the univariate analysis: the need for respiratory assistance, the use of vasopressors, the occurrence of at least one infection event in ICU, cyclophosphamide treatment, sequential organ failure assessment at admission and simplified acute physiology score II. After adjustment on sequential organ failure assessment or infection, cyclophosphamide was no longer a risk factor for mortality. Despite a higher initial mortality rate of ICU patients within the first hospital stay (p