PURPOSE: To evaluate which serologic, cerebrospinal fluid (CSF), and neuroradiographic tests alone or in combination are most useful in the diagnosis of neuropsychiatric lupus erythematosus (NPLE). PATIENTS AND METHODS: Prospective study of patients with systemic lupus erythematosus (SLE) hospitalized with neuropsychiatric disease between January 1982 and December 1991. Special tests evaluated as part of this study included serum antinuclear antibodies, complement levels, serum and CSF antineuronal antibodies, CSF special protein studies (immunoglobulin G [IgG] index and oligoclonal bands), serum antiribosomal-P antibodies, serum antiphospholipid antibodies, and cranial magnetic resonance imaging (MRI). Diagnostic sensitivity, specificity, and positive predictive value (PPV) were determined for single tests and combinations of tests. RESULTS: Fifty-two NPLE patients were categorized by neuropsychiatric presentation (32 diffuse, 10 focal, and 10 complex presentations) and compared to 14 SLE control patients. Each NPLE patient with a diffuse or complex presentation had abnormal CSF IgG index/oligoclonal bands, elevated CSF antineuronal antibodies, and/or serum antiribosomal-P antibodies, yielding a sensitivity of 100%, specificity of 86%, and PPV of 95% for this combination of tests. Nine of 10 patients with focal presentations and all with complex disease had evidence of vasculitis/livedo reticularis, antiphospholipid antibodies, and/or a cranial MRI with multiple lesions, giving a sensitivity of 95%, specificity of 86%, and a PPV of 90% for this battery of tests. These combinations of tests correctly diagnosed all nine SLE patients whose initial diagnosis proved to be incorrect based on subsequent clinical course. Abnormal test results frequently normalized or improved with successful therapy. CONCLUSIONS: Specific tests for CSF antibodies are most useful diagnostically in diffuse NPLE, implicating autoantibodies in the pathogenesis of this NPLE presentation. In those patients with diffuse NPLE who present with primarily psychiatric disease, serum antiribosomal-P antibodies appear to be helpful. In contrast, focal NPLE appears to be mostly secondary to vascular occlusion, and the presence of dermal vasculitis/livedo reticularis, antiphospholipid antibodies, and/or an abnormal cranial MRI are most helpful diagnostically. Patients with complex presentations demonstrate abnormalities characteristic of both diffuse and focal NPLE. Abnormal tests can be followed serially and appear to correlate with clinical responses to therapy.
PURPOSE: To evaluate which serologic, cerebrospinal fluid (CSF), and neuroradiographic tests alone or in combination are most useful in the diagnosis of neuropsychiatric lupus erythematosus (NPLE). PATIENTS AND METHODS: Prospective study of patients with systemic lupus erythematosus (SLE) hospitalized with neuropsychiatric disease between January 1982 and December 1991. Special tests evaluated as part of this study included serum antinuclear antibodies, complement levels, serum and CSF antineuronal antibodies, CSF special protein studies (immunoglobulin G [IgG] index and oligoclonal bands), serum antiribosomal-P antibodies, serum antiphospholipid antibodies, and cranial magnetic resonance imaging (MRI). Diagnostic sensitivity, specificity, and positive predictive value (PPV) were determined for single tests and combinations of tests. RESULTS: Fifty-two NPLE patients were categorized by neuropsychiatric presentation (32 diffuse, 10 focal, and 10 complex presentations) and compared to 14 SLE control patients. Each NPLE patient with a diffuse or complex presentation had abnormal CSF IgG index/oligoclonal bands, elevated CSF antineuronal antibodies, and/or serum antiribosomal-P antibodies, yielding a sensitivity of 100%, specificity of 86%, and PPV of 95% for this combination of tests. Nine of 10 patients with focal presentations and all with complex disease had evidence of vasculitis/livedo reticularis, antiphospholipid antibodies, and/or a cranial MRI with multiple lesions, giving a sensitivity of 95%, specificity of 86%, and a PPV of 90% for this battery of tests. These combinations of tests correctly diagnosed all nine SLEpatients whose initial diagnosis proved to be incorrect based on subsequent clinical course. Abnormal test results frequently normalized or improved with successful therapy. CONCLUSIONS: Specific tests for CSF antibodies are most useful diagnostically in diffuse NPLE, implicating autoantibodies in the pathogenesis of this NPLE presentation. In those patients with diffuse NPLE who present with primarily psychiatric disease, serum antiribosomal-P antibodies appear to be helpful. In contrast, focal NPLE appears to be mostly secondary to vascular occlusion, and the presence of dermal vasculitis/livedo reticularis, antiphospholipid antibodies, and/or an abnormal cranial MRI are most helpful diagnostically. Patients with complex presentations demonstrate abnormalities characteristic of both diffuse and focal NPLE. Abnormal tests can be followed serially and appear to correlate with clinical responses to therapy.