BACKGROUND Some autoantibodies have been associated with lupus nephritis but the role of antiphospholipid antibodies (APA) is controversial. OBJECTIVE The present study was to explore the role of APA by comparing demographic profiles and the presence of anticardiolipin antibody (aCL) and lupus anticoagulant (LA) in systemiclupus erythematosus 1 (SLE) patients with and without nephritis. METHODS The cross-sectional study in a tertiary center was conducted in 77 SLE patients. All patients attended our renal or rheumatology clinics between June 2002 and December 2003. RESULTS Sixty-three (82%) of the 77 SLE patients had nephritis. The prevalence of antiphospholipid syndrome (APS) was 10% (8 patients), positive aCL (IgG) was 26% (20 patients) and positive LA was 26% (20 patients). The receiver operating characteristic (ROC) method was applied to assess the significance of aCL in both nephritis and non-nephritis groups. Area under the ROC curve was 0.538 (95%CI 0.312-0.765), a cutoff value of 20.5 GPL had a sensitivity of 75% and a specificity of 53%. In univariate analysis, neither positivity for anticardiolipin antibody nor lupus anticoagulant was associated with lupus nephritis. Analyzed in only the lupus nephritis group, LA-positive lupus nephritis patients had higher systolic blood pressure (SBP) (133.7 vs 121.9 mmHg, p = 0.005), lower platelet count (209.8 vs 264.4 x 10(3)/microL, p = 0.02) and higher 24-hr urine protein excretion (2.6 vs 1.4 g, p = 0.02) than LA-negative lupus nephritis patients. Serum creatinine was higher in LA-positive lupus nephritis than LA-negative (233.0 vs 94.9 micromol/L), but did not reach statistical significance. CONCLUSIONS APA are frequently seen in SLE patients, but not associated with lupus nephritis. However lupus anticoagulant tends to associate with lupus nephritis. Detection of LA in lupus nephritis patients could identify patients who had increased risk to develop bad renal outcomes (elevated SBP and 24-hr urine protein excretion).