Objective: To determine the role of central nervous system (CNS) perfusion scanning in detecting CNS disease in childhood onset systemic lupus erythematosus (SLE) and serial single photon emission computed tomography (SPECT) scans in monitoring CNS disease activity in childhood.
Methods: The charts of 108 patients with a confirmed or suspected diagnosis of SLE during the period February 1987 to June 1992 were reviewed. Twenty patients with a diagnosis of CNS SLE and 10 patients without CNS involvement had at least one SPECT scan. Patients were divided into (a) focal CNS SLE, when there were clinical manifestations that could be attributed to localized lesions of the CNS (6 patients); and (b) diffuse CNS SLE, when there was a global defect in CNS function including organic brain syndrome, psychosis, and depression (14 patients). If a patient had both diffuse and focal CNS disease that patient was designated as having diffuse disease. Forty-three patients with a diagnosis other than SLE comprised our non-SLE control group.
Results: SPECT scans were performed in 20 patients with acute CNS involvement. In patients with acute diffuse CNS disease, diffuse patchy areas of hypoperfusion were seen in 86% of patients at presentation of the CNS event. In the focal CNS disease subgroup, 33% of patients had an abnormal scan at CNS presentation. In these patients focal rather than diffuse abnormalities were seen. Eight patients with diffuse CNS SLE had at least one followup study at intervals ranging from 1 month to 3 years after initial scan. In 50% of these patients with diffuse CNS involvement, improvement in their abnormal scan correlated with clinical improvement, while in the other 50% clinical improvement was not associated with SPECT scan improvement. The most common abnormal SPECT scan pattern in patients with CNS SLE was one of widespread multiple small areas of decreased uptake at multiple sites, suggestive of generalized patchy hypoperfusion. Although SPECT scans were sensitive to the presence of CNS disease, the diffuse hypoperfusion was not specific for clinically detectable CNS involvement. In patients with SLE, a diffusely abnormal scan had a specificity of 69% and a likelihood ratio of 2.2 to correctly detect overt diffuse CNS disease.
Conclusion: Although we found that SPECT scanning was a highly sensitive method, it was not a specific method in correctly diagnosing diffuse CNS SLE in children. However, the presence of an abnormal SPECT scan in SLE patients with no history of overt CNS SLE may suggest that subclinical CNS disease may be more common in children than previously suggested.