Method:A multi-center, international, research network enrolled patients within 15 months of fulfilling ACR criteria for SLE and performed annual assessments for up to 10 years. Seizures and other NP manifestations were recorded using the ACR case definitions. Decision rules determined the attribution of seizures to SLE and non-SLE causes. Clinical variables included demographic characteristics, disease duration, educational status, medication utilization, global SLE disease activity (SLEDAI-2K) and cumulative organ damage (SLICC/ACR Damage Index (SDI)) computed with and without NP variables. Plasma/serum samples were available at enrollment for the determination of the following autoantibodies: lupus anticoagulant, anticardiolipin, anti-β2 glycoprotein-I, anti-ribosomal P and anti-NR2 glutamate receptor antibodies. The association between clinical and serological variables and the risk of the first occurrence of seizures was examined by univariate and multivariate Cox regression analysis.
Result:The 1631 enrolled patients were predominantly female (89.4%) with a mean (± SD) age of 35.0 ± 13.4 years and mean disease duration of 5.6 ± 4.8 months. The mean followup was 3.5 ± 2.9 years. Over this period 75/1631 (4.6%) patients had ≥ 1 seizure with a total of 91 seizures of which 78/91 (86%) were attributed to SLE. Multivariate analysis indicated a significantly higher risk of seizures with African race/ethnicity (HR (CI): 1.97 (1.07-3.63); p=0.02) and lack of post-secondary education (1.97 (1.21-3.19); p<0.01) after adjustment for age at diagnosis and gender. In order to examine the effect of prior organ damage and medication use, the analysis was restricted to those seizures occurring after the enrollment visit (n=20). In these patients higher SDI scores calculated without NP variables were associated with an increased risk of subsequent seizures (SDI=1: 3.71(1.38-9.95); SDI=2or3: 1.33(0.27-6.46); SDI≥4: 6.44(0.72-57.2); p=0.04). After adjustment for prior medication, this risk was less significant (p=0.07) demonstrating some confounding between disease severity and medication. Prior use of anti-malarial drugs in the absence of immunosuppressive agents was the most notable treatment effect on seizures (0.07 (0.01-0.66); p=0.02). There was no association with SLEDAI scores or between any of the autoantibodies detected at enrollment with subsequent seizures.
Conclusion:The risk of seizures in SLE patients is higher in patients of African race/ethnicity, lower educational status and organ damage outside of the nervous system. The association with lupus related therapies is complex but anti-malarial drugs may have a protective effect.
Disclosure: J. G. Hanly, None; M. B. Urowitz, None; L. Su, None; C. Gordon, None; S. C. Bae, None; J. Sanchez-Guerrero, None; J. Romero-Diaz, None; D. Wallace, None; A. E. Clarke, None; S. Bernatsky, None; E. M. Ginzler, Human Genome Sciences, Inc., 2, Human Genome Sciences, Inc., 5 ; J. T. Merrill, None; D. A. Isenberg, None; A. Rahman, None; M. Petri, None; P. R. Fortin, None; D. D. Gladman, None; I. N. Bruce, None; K. Steinsson, None; M. A. Dooley, Human Genome Sciences, UCB, Genentech, Cephalon, Bristol-Myers Squibb, Eli Lilly, 2, Human Genome Sciences, GlaxoSmithKline, Vifor Pharma, Lupus Foundation of America Medical Scientific and Advisory Board, 5 ; M. A. Khamashta, None; G. S. Alarcon, None; B. J. Fessler, None; R. Ramsey-Goldman, None; S. Manzi, Bristol-Myers Squibb, 2, Cephalon, 2, Human Genome Sciences, Inc., 2, Immunomedics, Inc., 2, NIH, 2, Cypress Biosciences, Inc., 5, Bristol-Myers Squibb, Human Genome Sciences, Inc., Cephalon, Anthera, Medammune, Inc. .