EVALUATION WITH MRI, PERFUSION MRI AND CEREBRAL FLOW SPECT IN NEUROPSYCHIATRIC PATIENTS

Background: Neuropsychiatric manifestations are common in patients with systemic lupus erythematosus (SLE), but accurate diagnosis is often difficult due to the lack of effective imaging methods. Recently developed MRI techniques (Perfusion Weighted Imaging (PWI), Diffusion weighted imaging (DWI), Magnetization transfer imaging (MTI) and Spetroscopy) are emerging in order to provide a valid diagnostic support but studies in Neuropsichiatric Lupus patients (NPSLE) are still very few.Objectives: To compare the performance of three brain imaging modalities such as MRI, cerebral flow SPECT and PWI in NPSLE patients by using image co-registration.Methods: We studied 20 SLE patients (18 females and 2 males; mean age 41.6 years, age range 16-67), of whom 9 with clinically proven neurological involvement, 7 with mild but nonspecific neurological symptoms, and 4 neurological asymptomatic patients. All the patients underwent MRI (T2-weighted FLAIR), perfusion and diffusion MRI, and SPECT (after 99mTc-HMPAO intravenous injection).Two operators analysed the images both independently and jointly after multi-modal volumetric co-registration ("Statistical parameter mapping" software, based on the maximization of mutual information method).Results: The FLAIR examination depicted 82 hyperintense small lesions in 11/20 patients. Perfusion SPECT showed 43 hypoperfused areas in 17/20 patients. PWI showed 13 hypoperfused areas in 10/20 patients. After co-registration of images, anatomical agreement between SPECT an PWI was found in 10 hypoperfused areas (8 patients). Co-registration with FLAIR showed some false positives more frequent in SPECT (9/43 areas) than in PWI (2/13 areas).Conclusion: The FLAIR examination confirmed its high sensitivity in detecting morphological lesions represented by subcortical punctiform areas of signal hyperintensity of the cerebral parenchyma in T2 weighted and FLAIR sequences. Perfusion SPECT confirmed high sensitivity for the detection of functional abnormalities represented by hypoperfused cerebral areas in brain parenchyma. PWI results showed fewer areas of cerebral hypoperfusion than SPECT. The disagreement found between SPECT and PWI may be related to the different modalities for disease detection: the contrast in PWI is the result of a dynamic process related to the blood supply to the anatomical districts of the brain during the transit of the paramagnetic contrast material; SPECT images reflect the locoregional distribution of a flow tracer whose uptake may be in part influenced by the metabolic status of the nervous tissue. The anatomical agreement of hypoperfused areas between SPECT and PWI, assessed after co-registration, in the absence of FLAIR lesions, can suggest the prognostic hypothesis of delayed appearance of permanent parenchymal lesions.In conclusion co-registration modality seems to be a valid support for the interpretation of SPECT and PWI findings. In particular, PWI seems useful in the interpretation of uncertain cases.