Objective: To assess the validity of the recently developed American College of Rheumatology (ACR) nomenclature for neuropsychiatric systemic lupus erythematosus (NPSLE).
Methods: We conducted a cross-sectional, population-based study covering an area with 440,000 people. A total of 46 patients aged 16 to 65 years fulfilled the criteria for a definite diagnosis of SLE. One control for each patient matched by age, sex, education, and place of residence was randomly identified from the population register. All patients and controls underwent a clinical neurologic examination and neuropsychological testing. The data were analyzed using conditional logistic regression methods.
Results: Forty-two patients (91%) and 25 controls (56%) fulfilled at least one of the ACR NPSLE criteria, which gave an odds ratio (OR) of 9.5 (95% confidence interval [CI] 2.2-40.8) but low specificity (0.46). Cognitive dysfunction was the most common syndrome detected in 37 patients (80%). A revised set of 16 criteria excluding the syndromes without evidence for neuronal damage resulted in improved specificity (OR 7.0, 95% CI 2.1-23.5, specificity 0.93).
Conclusion: The proposed 19 ACR criteria did not differentiate SLE patients from controls, nor NPSLE patients from other SLE patients. The revised NPSLE criteria proposed by us performed well in our population but should be evaluated in a larger patient population.