Lupus patients in an emergency unit. Causes of consultation, hospitalization and outcome. A cohort study

The objectives were to determine causes of consultation, hospitalization and outcome in a cohort of lupus patients in an emergency unit. Patients with systemic lupus erythematosus (SLE) who visited the emergency department for consultation from 1 September 1996 to 17 May 1997 were included in the study. They were evaluated during the visit by looking at 100 variables such as demographic, socioeconomic, clinical, therapeutical, behavioral, (compliance), emotional (Beck depression inventory), disease activity, (Mex-SLEDAI), disease severity (Lupus SDI), chronic damage (SLICC-ACR), and physician's and patient's global assessments of severity. All causes of consultation, hospitalization and outcome were registered. Descriptive statistics, univariate analysis and multiple logistic regression were used for analysis. Significance was set at the 0.05 level. 180 patients were included. 164 were female, mean age 31.7/11.39 y, mean Mex SLEDAI score 3.8, mean SLICC-ACR 1.3. Fever, poliarthralgia and abdominal pain were the main causes of consultation with 26, 25 and 18 cases each. 49 patients were hospitalized and these were statistically different than non-hospitalized patients in level of formal education (10.2 vs 11.8, P = 0.03); compliance (7.6 vs 9, P = 0.0001); malar rash (57% vs 82%, OR, 95% CI = 0.28, 0.13 0.62, P = 0.0008), chloroquine daily dose intake (45 vs 77 mg, P = 0.04); disease severity in physician's global assessments (5.6 vs 2.1, P = 0.0001) and Beck depression inventory (21 vs 16, P = 0.01). Multiple logistic regression identified physician's global assessment, fewer ACR criteria and higher SLICC-ACR scores as the main variables associated with hospitalization. Five patients died; two with community acquired pneumonia, one with pancreatitis, multiple thromboses, and sepsis, one with pulmonary hemorrhage; and one with pulmonary thromboembolism. In conclusion, poor compliance, low level of formal education, severity, depression, lower ACR criteria and higher SLICC-ACR scores were important variables identified with hospitalization. Chloroquine use seemed to have a protective effect. Causes of death were related to infections and antiphospholipid syndrome.