Background/Purpose: To identify mortality prognostic factors of patients with systemic autoimmune diseases (SAD) admitted in a medical intensive care unit (ICU).

Methods: Retrospective observational study including all patients with SAD admitted to a medical ICU of a tertiary referral centre between January 1999 and December 2012. Only patients with the diagnosis of SAD according to accepted criteria made prior to ICU admission or during hospitalization were selected. Patients with short term irreversible disease and those with an ICU stay less than 48 hours were excluded. The reason for ICU admission, clinical follow-up, immunosuppressive treatment received before ICU admission, and outcome were collected. Mortality prognostic factors were identified through logistic regression analysis. 

Results: Seventy patients accounting for 75 ICU admissions (48 [68.6%] women) with mean (SD) age of 54 (19.3) years were included. Five patients were admitted twice. Twenty-three (30.7%) patients had systemic lupus erythematosus (SLE) (mean SLE Disease Activity Index [SLEDAI] at ICU admission 8.2 (5.6) [range 0-20]); 23 (30.7%) had systemic vasculitis (mean Birmingham Vasculitis Activity Score [BVAS] 14.5 (9.3) [range 0-33]); 7 (9.3%) systemic sclerosis; 7 (9.3%) dermatomyositis, and 5 (6.7%) had Sjögren´s syndrome. The reasons for ICU admission were infection in 26 (34.7%), followed by autoimmune disease flare-up in 17 (22.7%). Other complications related or not with the SAD were present in 26 (34.7%) patients. The mean Acute Physiology and Chronic Health Evaluation (APACHE II) at admission was 16.5 (6.5) (range 5-31). At the end of follow-up, 29 (41.4%) patients had died, 10 (14.2%) during the stay at ICU, 7 (10%) during hospitalization, and 12 (17.1%) after hospital discharge. A logistic regression model showed that multiorgan failure (respiratory failure [p=0.032], renal failure [p=0.017]), and the need of renal replacement therapy [p=0.007] were risk factors associated with increased mortality. In addition, corticosteroid therapy [p<0.005] and the need of intravenous immunoglobulin treatment [p<0.005] during the stay at ICU, and the use of cyclophosphamide in the previous six months [p=0.016] of ICU admission, were also risk factors associated with increased mortality. 

Conclusion: The most prevalent SAD admitted to a medical ICU were SLE and systemic vasculitis being infections the main reason for admission. The presence of respiratory and renal failure, the need of renal replacement therapy, the need of corticosteroid or intravenous immunoglobulin during the stay in ICU and the use of cyclophosphamide before the admission were factors associated with increased risk of mortality.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/mortality-prognostic-factors-of-patients-with-systemic-autoimmune-diseases-admitted-to-an-intensive-care-unit/