Search for a command to run...
Background: systemic lupus erythematosus (SLE) patients have higher rates of morbility and mortality than general population because of inflammatory activity and high infections frequencies. The infections have been associated with activity index of the disease and the use of immunosuppressant drugs. Objective: to determine the factors associated with nosocomial infections in SLE patients who required in hospital treatment. Methods: we performed a prospective study of SLE patients who met the ACR criteria and who were hospitalized between January 2001 and December 2004. Exclusion criteria were community acquired infections, onset of infections in the first 48 hours of hospitalization, HIV and patients with neoplasm. The SLE activity was quantified with the SLEDAI index and infection was defined according to microorganism isolated or CDC criteria for those without an isolated microorganism. We studied this variables: age, sex, time from onset of SLE, hospital stay, use of immunosuppressant drugs, type of infection, and cause of death. Results: one hundred and twenty three patients were included who had 155 hospital admissions, 106 were women (82.2%). Mean age was 27.7 +/- 12.6 years (7-73) with a mean time from diagnostic of 27.8 +/- 49.4 months (0-300), 56 patients (45.5%) had novo diagnosis, 25 patients had chronic renal insufficiency including 10 in hemodialysis. Respect of previous treatment, 58 (47.1%) patients did not have received steroids, 36 (29.3%) have been receiving 20 mg of prednisone or less, and 29 (23.6%) more than 20 mg of prednisone, 31 (25.2%) patients pulses of ciclophosphamide and 13 others immunosuppressant drugs. The SLEDAI median and mean were 14 (IQR 7-20) and 14.7 +/- 9.9 respectively. Nosocomial infection was reported in 38 (30.9%), 31 patients just one episode of infection and 13 more than one, with a total of 69 events. In univariate analysis was noted that these factors were associated with nosocomial infections: de novo diagnosis, longer hospital stays, use of methylprednisone pulses in hospital or ciclophosphamide, SLEDAI index. In multivariate logistic regression persisted significant factors such as: longer hospital stay [OR= 5.16; (IC 95% 1.5-18) p= 0.01], use of methylprednisone [OR 3.95; (IC 95% 1.22-12.8) p= 0.02], SLEDAI index [OR 4.43; (IC 95% 1.16-16.8) p=0.03], chronic renal insuffiency [OR 5.44; (IC 95% 1.19-24.7) p= 0.03] and 20 mg or less of prednisone [OR 0.16; (IC 95% 0.03-0.95) p= 0.04]. The microorganism was isolated in 81.2% (56 events): baciles gram negative 49.3% (34), coccos gram positives 26.1% (28), fungi, hemoparasites and virus in 14.5% (10). The infections was documented from genitourinary tract in 37.7% (26), blood stream 26.1% (18), respiratory tract 11.6% (8), etc. Fourteen patients died, six from uncontrolled infections, two for activity illness, one for both infections and illness activity, three from cardiovascular causes and one unknown. Conclusion: in our SLE Colombian population almost one third of the patients had nosocomial infections that could be severe and cause major mortality. The infections were associated with higher activity of the illness, pulses of methylprednisone, chronic renal insufficiency and longer hospitalized stay.