Evaluation of Symptom-Oriented Selection of Computed Tomography or Magnetic Resonance Imaging for Neuropsychiatric Systemic Lupus Erythematosus

Abstract

The purpose of this study is to assess the value of computed tomography (CT) and magnetic resonance imaging (MRI) for patients with symptoms of neuropsychiatric systemic lupus erythematosus (NPSLE) that vary in severity. Seventy-one patients with NPSLE were enrolled and divided into two groups based on symptom severity: (1) those with mild symptoms including mania, depression, intermittent headache, movement disorders and cognitive impairment, and (2) those with severe symptoms including focal/generalized seizures, hemiparesis and delirium. Cranial CT was performed as a primary screening tool in all cases. Medical charts and imaging findings were reviewed to determine whether CT and MRI findings were sufficient to explain the observed symptoms. Statistical significance was evaluated using the Chi-Square and McNemar tests. For patients with mild symptoms, no differences in positive rates were observed between CT and MRI findings (p=0.996). Positive rates for both CT and MRI findings were higher in patients with severe symptoms than those with mild symptoms (p＜0.001). For patients with severe symptoms, however, higher positive rates were observed for MRI as compared to CT (p=0.016). Modern brain imaging permits NPSLE symptoms to be categorized concisely. Not every patient with NPSLE symptoms should receive MRI; when symptoms are mild, the utility of CT is equivalent to that of MRI. When symptoms are severe, however, MRI is superior to CT for explanation of symptoms and guidance in selection of treatment.