Search

Background Functional imaging studies on the central vestibular system by fMRI and H 2 15 O-PET in healthy volunteers revealed similarities and differences in cortical activation patterns ( Bense, 2001 , Dietrich, 2003 , Stephan, 2005 , Stephan, 2009 , Becker-Bense, 2014 ). These are ascribed to different experimental procedures and techniques. Whereas PET maps cerebral blood flow, fMRI measures changes of blood oxygenation via magnetic susceptibility changes.Galvanic vestibular stimulation (GVS) is applicable inside PET and MRI; it induces a perception of apparent self-motion mainly at onset ( Stephan, 2009 ). Aim of the current study was to design a protocol for simultaneous PET-MRI of the vestibular system in order to compare the activation patterns between the two methods providing additional information for a deeper insight into the different mechanisms and improve the interpretation of the data. Methods Six simultaneous PET-MRI scans were acquired in a hybrid scanner in nine healthy volunteers. During the stimulation scans AC-GVS was applied at 1 Hz ( Stephan, 2005 ) with eyes closed ( Fig. 1 ). FMRI EPI sequences were measured over 260 s. Dynamic PET acquisition started with injection of 750 MBq H 2 15 O in a semibolus technique simultaneously with fMRI. Statistical direct comparison of PET and fMRI data were performed (paired t-test, SPM8). Results MRI: Onset-related activation was seen of the multisensory vestibular cortical network and ocular motor areas bilaterally, as well as of premotor area BA 6 (SMA). Block-related activation represented a subset of this activation pattern without SMA activation. PET: The PET data of the first stimulation blocks revealed prominent activation of the SMA and cerebellar vermis. MRI vs PET: Paired comparison showed significantly stronger SMA PET signals compared to block-related MRI that was not present for onset-related MRI. Conclusions The proposed procedure is suitable for direct statistical comparison of H 2 15 O-PET and fMRI activations during vestibular stimulation. At comparable spatial resolution, the higher temporal resolution of fMRI can be utilized to interpret the simultaneous PET data in another context. On the other hand, the quantitative nature of the PET data can augment the fMRI data that provide only an indirect measure of cortical processing. The most different effects were seen in the SMA, where the PET signal seems to reflect the onset-related MRI signal due to self-motion perception which is most prominent at stimulus onset causing a preparatory motor response.

Objective Reactive gliosis may improve recovery after acute unilateral vestibulopathy (AUV). In the present study glial activation was visualized in vivo by [ 18 F]GE180-PET in a rat model of unilateral labyrinthectomy (UL) and compared to behavioural vestibular compensation (VC) over time. Methods 14 Sprague-Dawley rats underwent a left-sided UL by a transtympanic injection of bupivacaine and arsenilate, 14 rats received a left-sided middle-ear sham injection of 0.9% NaCl-solution as a control group (SHAM-UL). For functional imaging of glial activation, PET with [ 18 F]GE180, a novel translocator protein (TSPO) ligand, was performed at baseline and 7, 15, 30 days after UL/SHAM-UL. Ex vivo autoradiography was done at the respective time points for high resolution validation of PET and combined with Nissl-staining of the corresponding slices for detailed anatomic mapping. Established behavioural testing for symptoms of vestibular tone imbalance including registration of postural asymmetry and nystagmus was performed at baseline and at days 1, 2, 3, 7, 15, 30 after UL/SHAM-UL. Results Only after UL rats showed clinical signs of vestibular imbalance which significantly decreased until days 15 and 30. In parallel, [ 18 F]GE180-PET and ex vivo autoradiography depicted glial activation in the ipsilesional vestibular nerve and nucleus at days 7 and 15 after UL. Correlation analysis revealed a strong negative association of [ 18 F]GE180 uptake in the ipsilesional vestibular nucleus at day 7 with the rate of postural recovery (R = −0.90, p 18 F]GE180 binding in the ipsilesional vestibular nerve and nucleus 7 days after symptom onset. Interpretation In the rat glial activation takes place in the ipsilesional vestibular nerve and nucleus within the first 30 days after UL and can be visualized in vivo by [ 18 F]GE180-PET. Comparative analysis of behavioural and [ 18 F]GE180-PET data proves that glial activation is beneficial for VC. In human AUV neuroinflammation seems to analogously appear in the ipsilesional vestibular nerve and nucleus.

Background and aim Reactive gliosis may be beneficial for central compensation after unilateral inner ear damage. The aim of the present study was to visualize whole brain glial activation in vivo and over time by serial [ 18 F]-GE180- μ PET in a rat model of chemical unilateral labyrinthectomy (UL). Methods Ten Sprague–Dawley rats underwent a left-sided UL by transtympanic injection of bupivacaine and arsenilate, ten rats received a left-sided middle-ear injection of saline as a control group. Behavioural testing for symptoms of vestibular tone imbalance including registration of nystagmus and postural asymmetry was performed 1 day before and at days 1, 3, 7, 15, 30 after UL/sham injection. For functional imaging of glial activation, whole-brain μ PET with [ 18 F]-GE180, a novel translocator protein (tspo) ligand, was performed before and on 7, 15, 30 days after UL/sham injection. In vivo autoradiography was done at the respective time points for crosscheck and combined with Nissl-staining of the corresponding slices for detailed anatomic mapping. Results All rats showed clinical signs of an acute vestibular syndrome following UL with a maximum at 2 and 3 days. Nystagmus disappeared until day 5, postural asymmetry decreased significantly until day 15. In the UL group, [ 18 F]-GE180- μ PET depicted a glial activation in the ipsilesional vestibular, cochlear and facial nucleus at day 7, which further increased until day 15 and was only slightly visible at day 30. In addition no other regions of the brain showed signs of glial activation. In the sham group, glial activation was found in the facial nucleus at day 7 and 15. The time course and anatomic localization of glial activations were confirmed by in vivo [ 18 F]-GE180 autoradiography and Nissl-staining. Conclusions Glial activation takes place in the ipsilesional vestibular and cochlear nucleus within the first 30 days after an inner ear lesion and thereby parallels the establishment of central vestibular compensation. μ PET is a feasible method to show upregulation of the tspo protein in vivo as a sign of reactive gliosis induced by unilateral labyrinthectomy. Our data suggest that synthetic tspo ligands may be a novel therapeutic principle to augment central vestibular compensation.

Introduction A recent case series study showed significant improvements of symptoms in patients with cerebellar ataxia on medication with the amino acid acetyl-dl-leucine (Tanganil) (Strupp et al., 2013). In analogy to electrophysiological studies in animals it was hypothesized that acetyl-DL-leucine modulates the activity of central vestibular neurons, preferably in the cerebellum, by normalizing the membrane potential of depolarized or hyperpolarized neurons and thereby improves central compensation. However, the mode of action of acetyl-dl-leucine and the site of central compensation remained speculative. To deepen our insights in the central compensatory processes induced by acetyl-dl-leucine we performed a FDG-PET study in cerebellar ataxic patients off and on treatment. Methods 21 patients with cerebellar ataxia of different etiologies were scanned by resting state FDG-PET twice, (A) off treatment (PET 1), and a second time (B) ⩾3months on treatment with 5g/day acetyl-dl-leucine (PET 2; n =18). At both time points the patients underwent a detailed neurological examination and measurements of standardized ataxia scales (SARA, SCAFI). Group subtraction analyses (PET 1 vs. PET 2; responders vs. non-responders) as well as statistical group comparisons with age- and gender-matched healthy controls were calculated with statistical parametric mapping (SPM). Results The contrast on vs. off treatment (PET 2 vs. PET 1) mainly showed signal differences in the mesencephalon (midbrain integration centers) as well as in the lateral medullary brainstem (vestibular nucleus region), thalami and midinsular regions bilaterally. The contrast PET 1 vs. PET 2 revealed signal differences in the posterior insula as well as in primary and secondary visual areas (including MT/V5) bilaterally. Restricting the analyses to treatment-responders ( n =11), PET 2 vs. PET 1 showed similar signal areas in the central midbrain and dorsal medullary brainstem bilaterally that were lacking in the corresponding contrast of the non-responders ( n =6). In the direct comparison of responders vs. non-responders PET 2 signal differences were located in both caudate nuclei, the superior temporal gyrus and hippocampus. The signal in responders – mainly in the posterior insula and the medial temporal gyrus – was lower compared to non-responders. In all contrasts cerebellar signal changes were sparse. Conclusions Central compensatory processes induced by acetyl-dl-leucine in cerebellar ataxia seem to involve mainly centers in the medulla (vestibular nuclei), midbrain (vestibular integration centers), thalami, basal ganglia and insular regions rather than neurons in the cerebellum, the primary site of dysfunction in cerebellar ataxia syndromes.

Background and aim Primary orthostatic tremor (OT) is a rare neurological disease of unknown pathophysiology characterized by a high frequency tremor mainly of the legs when standing. The aim of the present study was to depict the cerebral tremor network in OT by [ 18 F]-FDG-PET during standing. Methods Ten patients (mean age 58 ± 12 years) with neurophysiologically proven OT (EMG, posturography) and ten age-matched healthy controls were included in the study. All subjects underwent a [ 18 F]-FDG-PET paradigm during real standing (i.e. injection of [ 18 F]-FDG during upright stance, PET scan 30 min p.i.). As a control condition in a second session a PET was acquired after injection of [ 18 F]-FDG in lying position. PET images were processed by SPM8 software and compared between groups and conditions. Results All patients but no control subject showed a characteristic high-frequency peak at 12–16 Hz in posturography. In healthy subjects the regional cerebral glucose metabolism (rCGM) during standing compared to lying was increased in the vermal cerebellum, thalamus and primary motor region. During standing, OT patients had a relatively increased rCGM in the cerebellum (including the dentate nucleus), the ventrointermediate nucleus of the thalamus and the primary and supplementary motor cortex bilaterally in comparison to the control group. A relatively decreased rCGM was found in the mesiofrontal cortical areas in OT patients during standing. Conclusions In the present study a cerebello–thalamo–cortical tremor network is depicted in a group of OT patients using [ 18 F]-FDG-PET during standing. The data strongly reinforce the idea that the ventrointermediate nucleus of the thalamus may be a therapeutic target for deep brain stimulation in OT.