Spettroscopia protonica, perfusione e diffusione RM dell'encefalo nella sclerosi tuberosa

Abstract

Tuberous Sclerosis Complex (TSC) is a genetic disease affecting multiple systems. Brain is one of the most frequently involved organ and the standard intracranial hallmarks are cortical tuberous (CT), subependymal nodules (SEN) and giant cell astrocytoma (GCA). MRI findings are characteristic and are included among the diagnostic criteria set out by Roach. Functional MR acquisitions such as proton spectroscopy (1HMRS), perfusion (ASL-PI) and diffusion (DWI and DTI) are additional tools to conventional imaging which allow a deeper understanding of “in vivo” physiology and pathology of the brain. We have variably applied 1H MRS, ASL-PI and DWI- DTI in a series of patients with definite diagnosis of TSC. The aim was to assess the strength of these techniques in the clinical management of the most impaired patients, i.e. those with refractory epilepsy or with suspected GCA. Combined brain MR studies were performed in 15 patients with definite diagnosis of TSC using a 1.5 T MR system (Signa Horizon LX, GE Medical System, Milwaukee, Wis) with a maximum gradient capability of 23 mT/m and a slew rate of 120 T/m/s. 1H MRS was applied in 12 cortical tubers and one GCA. A short TE stimulated echo acquisition mode (STEAM) technique (TR = 2010 ms, TE = 30 ms, mixing time ? 13.7 ms, 128 scans, VOI dimension = 3.4 cc) was performed. Raw data were analysed off-line with the Spectral Analysis General Electric/Interactive Data Language (SAGE/IDL). The following metabolites were evaluated: N-acetyl-aspartate (NAA, at 2.01 ppm), Creatine/Phosphocreatine (Cr, at 3.05 ppm), Choline-containing compounds (Cho, at 3.25 ppm), Myo-Inositol (mI, at 3.56 ppm) and if detectable Lactate (Lac, at 1.33 ppm). Ratio of signal amplitude of each metabolite versus Cr signal was calculated. Results were compared to cortical 1HMRS data of 20 age-matched healthy subjects. Multislice ASL-PI was used to measure CBF in 6 patients. CBF measurement was obtained using single-shot, GE-EPI images with FOV = 24 cm × 15 cm and acquisition matrix of 64 × 40, resulting in an in-plane resolution of 3.75 mm. CBF values have been compared with those obtained in 46 healthy subjects (age ranging from 4 to 76 years). DWI was performed in 5 patients, using an interleaved single-shot pulsed field gradient spin-echo sequence described by Stejskal and Tanner. The diffusion tensor (DTI) has been measured from seven MR images, the first one with bvalue ? 0 and the other six with a bvalue = 1000 sec/mm2 and diffusion gradient applied along the noncollinear directions (1,0,1), (-1,0,1), (0,1,1), (0,1,-1), (1,-1,0), (-1,1,0). The diffusion weighted images (TR = 8000 ms, FOV = 24 cm × 24 cm, matrix = 128 × 128, slice thickness 5 mm) have been repeated four times to improve signal to noise ratio and processed by a Matlab 5.12 code to solve analytically Dij values and to calculate the Trace, VR, FA and LI maps. 1H MRS acquired on CT presented a metabolic profile characterized by a marked increase of mI, a decrease of NAA and normal Cho. Lac was never detected. On the contrary the suspected GCA demonstrated Cho signal increase. In 4 patients ASL-PI demonstrated a marked CBF reduction in cortical regions corresponding with MRI-FLAIR confirmed locations of cortical tubers (15/28 ml/100g/min). Three of these studies revealed also isolated cortical hyperperfusion (160/198 ml/100g/min) corresponding to normal MRI in two cases and to a small cortical dysplasia in one. All of these patients were epileptics. Global CBF was significantly reduced in two patients without any focal cortical flow abnormalities. Both patients were not affected by epilepsy and MRI demonstrated many small and scattered cortical lesions. Of little use turned out DWI and DTI, since TS lesions demonstrated similar diffusion values compared to other type of glial lesions. Among the techniques here evaluated, 1HMRS and ASL-PI have revealed interesting results for a deeper comprehension of TSC. 1H MRS increases the diagnostic confidence in those cases presenting solitary cortical-subcortical lesions and allows a non-invasive longitudinal evaluation of the potential biological transformation of SEN into GCA. CBF hypoperfusion in ASL-PI is in agreement with SPECT and PET studies of TSC, confirming therefore that cortical dysplasias are brain areas metabolically depressed and functionally inert. Focal cortical hyperperfusion detected in three cases could be related to epileptogenic foci. This hypothesis need to be further investigated by means of multimodality functional neuroimaging (MRI, ASL-PI, PET and EEG) in order to test the ASL-PI capability of visualizing the epileptogenic laterality and of identification of the trigger epileptic foci.