Low-Dose CCT to Exclude Contraindications to Lumbar Puncture : Benefits and Limitations.

BACKGROUND: Low-dose cranial computed tomography (LD-CCT) based on iterative reconstruction has been shown to have sufficient image quality to assess cerebrospinal fluid spaces (CSF) and midline structures but not to exclude subtle parenchymal pathologies. Patients without focal neurological deficits often undergo CCT before lumbar puncture (LP) to exclude contraindications to LP including brain herniation or increased CSF pressure. We performed LD-CCT to assess if image quality is appropriate for this indication. METHODS: A total of 58 LD-CCT (220 mA/120 kV) of patients before LP were retrospectively evaluated and compared to 79 normal standard dose cranial computed tomography (SD-CCT) (350 mA/120 kV). Iterative reconstruction used for both dose levels was increased by one factor for LD-CCT. We assessed the signal-to-noise (SNR) and contrast-to-noise ratio (CNR), the dose estimates and scored diagnostic image quality by two raters independently. Significance level was set at p < 0.05. RESULTS: The inner and outer CSF spaces except the sulci were equally well depicted by the LD-CCT and SD-CCT; however, depiction of the subtle density differences of the brain parenchyma and the sulci was significantly worse in the LD-CCT (p < 0.0001). The SNR in the gray matter (9.35 vs. 10.61, p < 0.05) and white matter (7.23 vs. 8.15, p < 0.001) were significantly lower in LD-CCT than in SD-CCT with significantly lower dose estimates (1.04 vs. 1.69 mSv, respectively p < 0.0001). CONCLUSION: The use of LD-CCT with a dose reduction of almost 50% is sufficient to exclude contraindications to LP; however, LD-CCT cannot exclude subtle parenchymal pathologies. Therefore, in patients with suspected parenchymal pathology, SD-CCT is still the method of choice.

New MR perfusion features in primary central nervous system lymphomas: pattern and prognostic impact.

PURPOSE: Some MR perfusion features predict overall survival (OS) and progression-free survival (PFS) in glioblastomas. Prognostic value of MR perfusion in primary CNS lymphomas (PCNSL) remains unexplored being the aim of this investigation. METHODS: We retrospectively analyzed 3Tesla dynamic susceptibility contrast MR perfusion in 37 pre-surgical PCNSL for normalized regional cerebral blood volume rCBVmean and rCBVmax and for a PCNSL-typical shoulder-like increase of the time-signal intensity curve ("TSIC-shoulder"), indicating moderate vessel permeability. These MR perfusion features, tumor and edema size, number of lesions and patient characteristics were correlated with OS and PFS. RESULTS: Only patient's age was prognostic for OS (p = 0.0037) and PFS (p = 0.0088). 23 PCNSL had the TSIC-shoulder, a middle-sized diameter (39.5 ± 10.8 mm), volume (15.7 ± 11.3 ml), peritumoral edema (23 ± 8.7 mm) and moderately increased rCBVmean and rCBVmax (1.7 ± 0.5; 3.9 ± 1.2). Seven PCSNL with the TSIC-shoulder presented a sun-like pattern ("rCBV-sun") with a rim of marginally high rCBV. These unifocal PCNSL were larger (43 ± 11.2 mm; 25.62 ± 19.2 ml), with more peritumoral edema (32.8 ± 7.6 mm) and lower CBVmean (0.8 ± 0.3) and rCBVmax (2.2 ± 0.7), compared to the remaining six multifocal PCNSL without the TSIC-shoulder (26.3 ± 8.3 mm; 4.7 ± 4 ml; 16.3 ± 6.4 mm; 2.4 ± 1.6; 4.4 ± 2.3). CONCLUSIONS: Only patient age was predictive for OS and PFS of PCNSL; MR perfusion parameters and features were not. Most PCNSL revealed the TSIC-shoulder, moderate size, peritumoral edema and rCBV increase. However, larger, solitary PCNSL additionally had a rCBV-sun pattern and more edema, maybe due to a centrifugal vessel proliferation, whereas smaller, multifocal PCNSL contain apparently more concentrated and less permeable blood vessels represented by higher rCBV, no TSIC-shoulder and less edema.

A novel reconstruction tool (syngo DynaCT Head Clear) in the post-processing of DynaCT images to reduce artefacts and improve image quality.

BACKGROUND: Latest generations of flat detector (FD) neuroangiography systems are able to obtain CT-like images of the brain parenchyma. Owing to the geometry of the C-arm system, cone beam artifacts are common and reduce image quality, especially at the periphery of the field of view. An advanced reconstruction algorithm (syngo DynaCT Head Clear) tackles these artifacts by using a modified interpolation-based 3D correction algorithm to improve image quality. MATERIALS AND METHODS: Eleven volumetric datasets from FD-CT scans were reconstructed with the standard algorithm as well as with the advanced algorithm. In a two-step data analysis process, two reviewers compared dedicated regions of the skull and brain in both reconstruction modes using a 5-point scale (1, much better; 5, much worse; advanced vs standard algorithm). Both reviewers were blinded to the reconstruction mode. In a second step, two additional observers independently evaluated image quality of the 3D data (non-comparative evaluation) in dedicated regions also using a 5-point scale (1, not diagnostically evaluable; 5, good quality, perfectly usable for diagnosis) for both reconstruction algorithms. RESULTS: Both in the comparative evaluation of dedicated brain regions and in the independent analysis of the FD-CT datasets the observers rated a better image quality if the advanced algorithm was used. The improvement in image quality was statistically significant at both the supraganglionic (p=0.018) and the infratentorial (p=0.002) levels. CONCLUSIONS: The advanced reconstruction algorithm reduces typical artifacts in FD-CT images and improves image quality at the periphery of the field of view.

Perfusion MRI in the Evaluation of Suspected Glioblastoma Recurrence.

PURPOSE: Treatment-related changes (TRC) often imitate tumor progression in glioblastomas. Increased regional cerebral blood volume (rCBV) can differentiate tumor progression from TRC after the standardized first-line radiochemotherapy, but information about diagnostic accuracy of rCBV for patients without any clinical selection criteria is limited. Therefore, we aimed to evaluate if rCBV can differentiate between TRC and tumor progression irrespective of preceding therapies and number of tumor progressions. METHODS: We analyzed mean and maximum rCBV from the enhancing areas normalized to the contralateral white matter in 44 pretreated glioblastomas with MR-morphological tumor progression. The diagnosis (real progression vs. TRC) was determined by histopathology or by clinical/MRI-follow-up. We performed nonparametric tests, receiver operating characteristics (ROC), and Kaplan-Meier analysis. RESULTS: Significant differences between tumor progression (N = 37) and TRC (N = 7) were found for rCBVmean (2.44 ± 1.05 vs. 1.69 ± .56, P < .03) and rCBVmax (3.40 ± 1.25 vs. 2.21 ± .62, P < .0007). A rCBVmax of 2.6 had 78% sensitivity and 86% specificity to detect tumor progression. Neither rCBVmean nor rCBVmax was predictive for the patient overall survival (OS). There were no statistically different rCBVmean and rCBVmax between the first and further tumor progressions. CONCLUSIONS: The rCBVmax differentiates tumor progression from TRC in unselected recurrent glioblastomas, but it is not predictive for the OS.

Distribution and prognostic relevance of tumor-infiltrating lymphocytes (TILs) and PD-1/PD-L1 immune checkpoints in human brain metastases.

The activation of immune cells by targeting checkpoint inhibitors showed promising results with increased patient survival in distinct primary cancers. Since only limited data exist for human brain metastases, we aimed at characterizing tumor infiltrating lymphocytes (TILs) and expression of immune checkpoints in the respective tumors. Two brain metastases cohorts, a mixed entity cohort (n = 252) and a breast carcinoma validation cohort (n = 96) were analyzed for CD3+, CD8+, FOXP3+, PD-1+ lymphocytes and PD-L1+ tumor cells by immunohistochemistry. Analyses for association with clinico-epidemiological and neuroradiological parameters such as patient survival or tumor size were performed. TILs infiltrated brain metastases in three different patterns (stromal, peritumoral, diffuse). While carcinomas often show a strong stromal infiltration, TILs in melanomas often diffusely infiltrate the tumors. Highest levels of CD3+ and CD8+ lymphocytes were seen in renal cell carcinomas (RCC) and strongest PD-1 levels on RCCs and melanomas. High amounts of TILs, high ratios of PD-1+/CD8+ cells and high levels of PD-L1 were negatively correlated with brain metastases size, indicating that in smaller brain metastases CD8+ immune response might get blocked. PD-L1 expression strongly correlated with TILs and FOXP3 expression. No significant association of patient survival with TILs was observed, while high levels of PD-L1 showed a strong trend towards better survival in melanoma brain metastases (Log-Rank p = 0.0537). In summary, melanomas and RCCs seem to be the most immunogenic entities. Differences in immunotherapeutic response between tumor entities regarding brain metastases might be attributable to this finding and need further investigation in larger patient cohorts.

Quantitative T2, T2*, and T2' MR imaging in patients with ischemic leukoaraiosis might detect microstructural changes and cortical hypoxia.

INTRODUCTION: Quantitative MRI with T2, T2*, and T2' mapping has been shown to non-invasively depict microstructural changes (T2) and oxygenation status (T2* and T2') that are invisible on conventional MRI. Therefore, we aimed to assess whether T2 and T2' quantification detects cerebral (micro-)structural damage and chronic hypoxia in lesions and in normal appearing white matter (WM) and gray matter (GM) of patients with ischemic leukoaraiosis (IL). Measurements were complemented by the assessment of the cerebral blood flow (CBF) and the degree of GM and WM atrophy. METHODS: Eighteen patients with IL and 18 age-matched healthy controls were included. High-resolution, motion-corrected T2, T2*, and T2' mapping, CBF mapping (pulsed arterial spin labeling, PASL), and segmentation of GM and WM were used to depict specific changes in both groups. All parameters were compared between patients and healthy controls, using t testing. Values of p < 0.05 were accepted as statistically significant. RESULTS: Patients showed significantly increased T2 in lesions (p < 0.01) and in unaffected WM (p = 0.045) as well as significantly increased T2* in lesions (p = 0.003). A significant decrease of T2' was detected in patients in unaffected WM (p = 0.027), while no T2' changes were observed in GM (p = 0.13). Both unaffected WM and GM were significantly decreased in volume in the patient-group (p < 0.01). No differences of PASL-based CBF could be shown. CONCLUSION: Non-invasive quantitative MRI with T2, T2*, and T2' mapping might be used to detect subtle structural and metabolic changes in IL. Assessing the grade of microstructural damage and hypoxia might be helpful to monitor disease progression and to perform risk assessment.

Proton 1H- and Phosphorus 31P-MR spectroscopy (MRS) in asymptomatic HIV-positive patients.

INTRODUCTION: HIV infection is accompanied by a variety of neurological disorders. Depression of cell-mediated immunity is followed by the development of central nervous system opportunistic infections/tumours, and frequently by the occurrence of the AIDS dementia complex (ADC). However, the pathophysiology of the emergence of neuro-AIDS is still unknown. Despite the development of cognitive impairments, the early diagnosis, objectification and quantification of the existence and extent of this impairment during infection are difficult to recognize in each individual case. To support the early diagnosis of ADC, there is a need for additional, non-invasive diagnostic methods. In this study, it is of interest to answer the clinically relevant question of whether magnetic resonance spectroscopy can detect changes in the cerebral metabolism of asymptomatic HIV-positive patients and is possibly suitable for the early diagnosis and prevention of HIV encephalopathy. METHODS: A group of 13 asymptomatic, HIV-positive patients with combined antiretroviral therapy (cART) and 13 healthy controls were examined with 2D 1H-MRS and 3D 31P-MRS at 3T. The patients were treated with cART for at least 12 months. Changes in the absolute concentrations of phosphorylated metabolites (ATP), N-acetyl-aspartate, creatine, myo-Isonitol, glutamate/glutamine and choline-containing compounds were compared with that of control subjects. RESULTS: Asymptomatic HIV-positive patients had significantly lower N-acetyl-aspartate in the white matter in a frontal and parietal target region. The other evaluated metabolites in the 1H MRS showed no significant difference between the HIV-positive patients and healthy controls. The 31P-MRS detected significant elevated values regarding the choline-containing compounds PEth, GPE and PCho. CONCLUSIONS: This spectroscopic study revealed a significantly lower N-acetyl-aspartate in the white matter in a frontal and parietal cerebral target region in asymptomatic, HIV-positive patients as an early sign of neuronal disintegration. The 31P-MRS detected significant elevated values regarding the choline-containing compounds PEth, GPE and PCho as an early sign of gliosis. Furthermore we could show that with the use of 1H-MRS and 31P-MRS cerebral metabolites can be reliably detected and measured in HIV-positive patients. The 1H-MRS and 31P-MRS is therefore suited as a diagnostic tool for early cerebral metabolic changes in HIV-positive patients.

MR perfusion in and around the contrast-enhancement of primary CNS lymphomas.

Diffuse cerebral infiltration of primary brain tumors may be missed on conventional MRI. In glioblastomas it may be visible on MR-perfusion images as an elevated rCBV adjacent to the contrast enhancing area (penumbra). We aimed to evaluate whether penumbral rCBV of primary central nervous system lymphomas (PCNSL) is also increased and if PCNSL perfusion has different features than that of glioblastomas. We measured dynamic susceptibility contrast MR-perfusion at 3 Tesla in 38 presurgical patients with histopathological diagnosis of PCNSL (n = 19) and glioblastoma (n = 19). We compared normalized rCBV within and adjacent to the enhancing area and evaluated time-signal intensity curves (TSIC) in all patients. Histopathological comparison of patients with different TSIC patterns (with or without shoulder-like increase) was performed. Relative to the normal tissue, rCBV within and adjacent to the enhancing area was increased (p < 0.05) in both glioblastomas and PCNSL. In the penumbra the increase was moderate in both groups, with 1.4 ± 0.46 in PCNSL and 1.82 ± 0.82 in glioblastomas (p = 0.07 between groups). In the enhancing tumor the increase was moderate in PCNSL (1.46 ± 0.62) and marked in glioblastomas (4.13 ± 2.44) (p < 0.001 between groups). A shoulder-like TSIC increase was exclusively found in PCNSL (11/19) and was significantly associated with a less prominent reticulin fibre network compared to the PCNSL without a shoulder-like TSIC increase. The moderately increased penumbral rCBV in PCNSL and glioblastomas reveals tumor-related changes beyond the tumor borders which are invisible with conventional MRI. PCNSL can be differentiated from glioblastomas through their significantly lower rCBV and shoulder-like signal intensity changes inside the enhancing area.

Phospholipid metabolites in recurrent glioblastoma: in vivo markers detect different tumor phenotypes before and under antiangiogenic therapy.

PURPOSE: Metabolic changes upon antiangiogenic therapy of recurrent glioblastomas (rGBMs) may provide new biomarkers for treatment efficacy. Since in vitro models showed that phospholipid membrane metabolism provides specific information on tumor growth we employed in-vivo MR-spectroscopic imaging (MRSI) of human rGBMs before and under bevacizumab (BVZ) to measure concentrations of phosphocholine (PCho), phosphoethanolamine (PEth), glycerophosphocholine (GPC), and glyceroethanolamine (GPE). METHODS: (1)H and (31)P MRSI was prospectively performed in 32 patients with rGBMs before and under BVZ therapy at 8 weeks intervals until tumor progression. Patients were dichotomized into subjects with long overall survival (OS) (>median OS) and short OS (

Metabolic gray matter changes of adolescents with anorexia nervosa in combined MR proton and phosphorus spectroscopy.

INTRODUCTION: There are hints for changes in phospholipid membrane metabolism and structure in the brain of adolescents with anorexia nervosa (AN) using either proton ((1)H) or phosphorus ((31)P) magnetic resonance spectroscopic imaging (MRSI). We aimed to specify these pathological metabolite changes by combining both methods with additional focus on the neuronal metabolites glutamate (Glu) and N-acetyl-l-aspartate (NAA). METHODS: Twenty-one female patients (mean 14.4 ± 1.9 years) and 29 female controls (mean 16 ± 1.6 years) underwent (1)H and (31)P MRSI at 3 T applied to the centrum semiovale including the anterior cingulate cortex. We assessed gray matter (GM) and white matter (WM) metabolite concentration changes of the frontal and parietal brain measuring choline(Cho)- and ethanolamine(Eth)-containing compounds, Glutamate (Glu) and glutamine (Gln) and their sum (Glx), myoinositol, NAA, and high-energy phosphates. RESULTS: For (1)H MRSI, a clear discrimination between GM and WM concentrations was possible, showing an increase of Glx (p < 0.001), NAA (frontal p < 0.05), pooled creatine (tCr) (p < 0.001), and choline (tCho) (p < 0.05) in the GM of AN patients. The lipid catabolites glycerophosphocholine (p < 0.07) and glycerophosphoethanolamine (p < 0.03) were increased in the parietal region. CONCLUSIONS: Significant changes in GM metabolite concentrations were observed in AN possibly triggered by elevated excitotoxin Glu. Increased tCho may indicate modifications of membrane phospholipids due to increased catabolism in the parietal region. Since no significant changes in phosphorylated choline compounds were found for the frontal region, the tCho increase in this region may hint to fluidity changes.

Metabolism and regional cerebral blood volume in autoimmune inflammatory demyelinating lesions mimicking malignant gliomas.

Dynamic susceptibility contrast (DSC) magnetic resonance imaging (MRI) and MR spectroscopy are thought to differentiate tumefactive autoimmune inflammatory demyelinating lesions from glial brain tumours. The aim of this work is to evaluate whether regional cerebral blood volume (rCBV), as well as choline (Cho), N-acetyl-aspartate (NAA) and myo-inositol (mIns) concentrations differ between tumefactive lesions and World Health Organization (WHO) grade II-III gliomas. Five patients with single autoimmune inflammatory demyelinating lesions and nine patients with WHO grade II and III gliomas were examined by DSC-MRI and by two-dimensional (2D) 1H MR spectroscopic imaging (1H-MRSI). rCBV values and metabolite concentrations were normalised to the respective values of the contralateral hemisphere. Normalised rCBV in the tumefactive lesions (mean 2.89, range 1.98-6.74) was in the some high level as in gliomas (mean 2.77, range 1.43-6.22). 1H-MRSI revealed increased normalised choline concentrations in five of six examinations of autoimmune lesions (mean 1.4, range 1.06-1.8) and in eight of nine gliomas (mean 1.35, range 0.92-1.73). Tumefactive autoimmune inflammatory demyelinating lesions not only have imaging appearance of gliomas but may also imitate marked increase of rCBV and Cho in WHO grade II-III gliomas.

Stripe-like increase of rCBV beyond the visible border of glioblastomas: site of tumor infiltration growing after neurosurgery.

We observed a stripe-like pattern of regional cerebral blood volume (rCBV) increase in a defined region adjacent to the contrast enhancement (CE) on MRI of glioblastomas (GBM) that we defined as the "striate sign" (SS). We hypothesized that the SS marks infiltration of GBM outside the CE volume transforming into future CE tumor in the follow-up. T2*-weighted dynamic susceptibility-weighted CE (DSC)-MRI, and T1 and T2-weighted images (WI) of 16 patients with GBM were retrospectively evaluated in a baseline MRI performed before neurosurgery. In seven of these patients we also performed a (1)H MR spectroscopic imaging ((1)H MRSI). The regions of interest (ROI) delineating the SS were defined on rCBV maps for each patient. ROIs were overlaid on follow-up T1-WI and T2-WI MRI performed 3, 6, and 9 months after neurosurgery. Size and maximum signal intensity (max SI) of de novo CE within the area of the SS were analyzed. Statistical analysis was performed with the Friedman test (P < 0.05). In 15/16 patients de novo CE completely covered the area of the SS within nine months. Normalized max SI of de-novo CE of the 3, 6, and 9-months follow-up MR examinations were significantly higher than in the baseline MRI (P < 0.001). Normalized choline was increased within the SS in all patients with de novo CE (n = 6). De-novo CE appeared within the SS in all patients (96% of all slices). This implies that the SS might indicate the site of future CE tumor, which represents the area of tumor growth after neurosurgery.

Is the surgical repair of unruptured atherosclerotic aneurysms at a higher risk of intraoperative ischemia?

BACKGROUND: The incidence of ischemia might be increased in the surgical repair of atherosclerotic unruptured aneurysms compared to non-atherosclerotic aneurysms. The atherosclerotic wall might increase the occurrence of thrombembolic events or its rigidity might endanger the occlusion of perforators within the aneurysm vicinity. METHODS: 87 patients (53 patients without and 34 patients with atherosclerotic unruptured aneurysms, 50.5 ± 9.7 years) were analyzed for severity of atherosclerosis within the aneurysm and the aneurysm bearing vessel, surgical maneuvers, intraoperative alterations in evoked potentials and clinical and neuroradiological results. RESULTS: Temporary vessel occlusion (25% vs. 50%, p = 0.021), repositioning of a permanent clip (21% vs. 56%, p = 0.001) and aneurysm remnants (2% vs. 18%, p = 0.012) occurred more often in patients with atherosclerotic aneurysms. At 6 months, 3/34 patients with atherosclerosis (8.8%) had an unfavorable outcome, all patients without atherosclerosis had a favorable outcome (p = 0.056). CONCLUSION: The surgical repair of unruptured aneurysms is safe but patients with atherosclerotic altered vessels and aneurysms accounted to a minor increase in unfavorable outcome and an increased risk of morbidity at 6 months postoperatively. This factor should be taken into consideration when performing surgery of atherosclerotic, unruptured aneurysms.

Elevated peritumoural rCBV values as a mean to differentiate metastases from high-grade gliomas.

PURPOSE: Increased relative cerebral blood volume (rCBV) was previously found in peritumoural oedema of glioblastomas (GBM). Supposing that peritumoural rCBV is not increased in metastases, we aimed to evaluate whether rCBV values of the whole peritumoural area are accurate to differentiate solitary metastasis from GBM irrespective of the peritumoural oedema. METHODS: Contrast-enhanced T1-weighted (T1-w) and T2*-weighted dynamic susceptibility contrast MRI was performed in 52 patients with contrast-enhancing solitary brain tumours before surgery. In each T1-w slice depicting the contrast-enhancing tumour, a rim within approximately 15 mm was defined in the peritumoural area. The rCBV values were normalised to rCBV values of the contralateral normal white matter. Differences between metastases and GBM for normalised rCBV values for each slice were determined with the Mann-Whitney U test (p < 0.05). RESULTS: Histopathological examination revealed 29 GBM and 23 metastases. Peritumoural rCBV was significantly lower in metastases than in GBM (p < 0.01). Using the cutoff value 1.0 for discriminating metastases from GBM yielded a sensitivity of 96%, specificity of 64%, a positive predictive value of 68% and a negative predictive value of 95%. CONCLUSIONS: The rCBV in the peritumoural area of contrast-enhancing brain tumours has a high diagnostic accuracy to discriminate metastases from GBM irrespective of surrounding oedema and without the bias of slice selection and ROI positioning. Metastases should be excluded, if at least one tumour-depicting slice reveals an increase of peritumoural rCBV compared to the normal contralateral brain (normalised rCBV value >1). Conversely, the decrease of peritumoural rCBV may not reliably exclude GBM.

Influence of iMRI-guidance on the extent of resection and survival of patients with glioblastoma multiforme.

Intraoperative MRI (iMRI) is used in glioma surgery mainly to determine the extent of resection, allowing surgeons to immediately continue resection in case of residual tumor tissue. The aim of this study is to report on the influence of the use of iMRI on the extent of resection and survival of patients with glioblastoma multiforme (GBM). We analyzed our prospectively collected database of patients with GBM operated upon during the initial period after installation of an iMRI; between July 2004 and December 2005, all patients with GBM undergoing intended complete tumor resection were included in this study, while patients undergoing mere tumor biopsy or intended incomplete resection were not. In total, 43 Patients met the inclusion criteria. Of these, 10 patients (23.3%) were operated upon with the help of iMRI while 33 underwent conventional tumor resection. All patients underwent postoperative high-field MR imaging at 1.5 Tesla to determine the extent of resection. Subsequently, all patients received adjuvant treatment. Median patient age was 60.0 years; median overall survival was 70.7 weeks. Radiologically complete tumor resection (P < 0.001) and the administration of temozolomide chemotherapy (P < 0.01) were statistically significant prognostic factors in a multivariate analysis. The rate of complete tumor resections was significantly higher in the iMRI group than in the conventional surgery group (P < 0.05). Patient age was not a prognostic factor in our series of patients (P = 0.22). Intraoperative MRI is a helpful tool to increase the extent of resection in GBM surgery and thereby improve patient survival.

MR angiography in patients with subarachnoid hemorrhage: adequate to evaluate vasospasm-induced vascular narrowing?

The diagnosis of cerebral vasospasm (CVS) following subarachnoid hemorrhage (SAH) is still challenging. We evaluate the accuracy of time of flight MR angiography (TOF-MRA) to assess the arterial diameters of the circle of Willis in SAH patients with suspected CVS. MR examinations (1.5 Tesla) including 3D TOF-MRA with maximum intensity projections (MIP) and digital subtraction angiography (DSA) were performed within 24 h in 21 patients with acute aneurysmal SAH and suspicion of CVS. Arterial diameters of the circle of Willis including the distal internal carotid artery (ICA) were measured as ratios to the extradural ICA in standard projections. The diagnosis of CVS was established by comparing the luminal size of baseline and follow-up DSA. The correlation between the arterial ratios measured on MIP angiograms and on follow-up DSA was assessed with Pearson's linear regression analysis. Arterial ratios on MIP angiograms were categorized as correct, overestimated, and underestimated compared to the ratios on follow-up DSA. Pearson's correlation coefficient between the ratios of MIP angiograms and DSA was r = 0.5799 and the regression coefficient was b = 0.4775. Highest correlation was found for the category of severe CVS (r = 0.8201). Of all MIP angiograms, 34.9% showed consistent results compared to the DSA, while 44.2% of MIP images overestimated the vascular narrowing. Standard MIP angiograms from TOF-MRA are not accurate to assess vascular narrowing in patients with suspected CVS after aneurysmal SAH. The multifocal arterial stenoses in CVS may induce severe changes in blood flow dynamics, which compromise the diagnostic accuracy of the TOF-MRA.

Diffusion tensor imaging in patients with adult chronic idiopathic hydrocephalus.

OBJECTIVE: Diffusion tensor imaging (DTI) parameters were investigated in patients with chronic idiopathic hydrocephalus to evaluate microstructural changes of brain tissue caused by chronic ventricular dilatation. METHODS: Eleven patients fulfilling the criteria for possible or probable idiopathic normal pressure hydrocephalus and 10 healthy control subjects underwent MRI at 3 Tesla, including DTI with 12 gradient directions. Patients were scanned before lumbar cerebrospinal fluid (CSF) withdrawal tests. Differences in fractional anisotropy (FA) and mean diffusivity (MD) between patients and controls were assessed using 2 different methods: manual definition of regions of interest and a fully automated method, TBSS (Tract-Based Spatial Statistics). DTI parameters were correlated with clinical findings. RESULTS: Compared with the control group, patients with chronic idiopathic hydrocephalus had significantly higher MD values in both the periventricular corticospinal tract (CST) and the corpus callosum (CC), whereas FA values were significantly higher in the CST but lower in the CC. DTI parameters of the CST correlated with the severity of gait disturbances. CONCLUSION: Microstructural changes in periventricular functionally relevant white matter structures (CSF, CC) in chronic idiopathic hydrocephalus can be visualized using DTI. Further studies should investigate the change of DTI parameters after CSF shunting and its relation to neurologic outcome.

The striate sign: peritumoural perfusion pattern of infiltrative primary and recurrent gliomas.

MR perfusion depicts angiogenesis as a key factor for growth and malignancy in gliomas by means of increased regional cerebral blood volume (rCBV). The rCBV increase is not limited to the tumour area, but may also produce a stripe-like pattern of peritumoural rCBV increase that we defined as the "striate sign". We evaluated if prior radiochemotherapy influences perfusion values and pattern in and adjacent to malignant gliomas comparing rCBV of treated recurrent gliomas with untreated gliomas. Ninety-three patients with primary or recurrent WHO grades II-IV glial tumours underwent T2*-weighted dynamic susceptibility-weighted contrast-enhanced (DSC)-MRI. Differences of normalised rCBV and rCBV(max) were evaluated using Kruskal-Wallis analysis with post hoc tests. The number of cases showing a hot spot of rCBV (rCBV(max)) and/or a peritumoural striate pattern of rCBV increase (striate sign) was assessed and evaluated by Fisher's exact test. Significance level was determined as p < 0.05. Normalised rCBV, rCBV(max) and number of cases with the striate sign were significantly lower in recurrent (rCBV = 3.24 +/- 1.22, rCBV(max) = 5.05 +/- 2.27 and striate sign = 10/24) compared to primary WHO grade IV tumours (rCBV = 4.44 +/- 1.39, rCBV(max) = 7.31 +/- 3.0 and striate sign = 17/21, respectively). There were fewer cases with a striate sign in treated recurrent WHO grade III tumours than in untreated malignant transformed WHO grade II tumours. The pattern and degree of rCBV increase in and around gliomas differ between untreated and previously treated tumours. These differences might be due to post-therapeutic changes of the tumour-associated microvasculature by radiochemotherapy. Spectroscopic and susceptibility-weighted MR imaging may provide further insights into the tumour biology.

Toxic leukoencephalopathy after heroin abuse without heroin vapor inhalation: MR imaging and clinical features in three patients.

BACKGROUND AND PURPOSE: Toxic leukoencephalopathy has been associated with illicit heroin vapor inhalation. Despite the nonspecific and variable clinical presentation of these patients, they show typical radiologic findings. Previous studies evaluated typical radiologic findings with symmetric infratentorial hyperintense signal changes and similar alteration in the posterior limb of the internal capsule, the splenium of corpus callosum, the medial lemniscus and the lateral brainstem. In context with the reviewed literature, a series of another three cases with toxic leukoencephalopathy after heroin abuse other than vapor inhalation is presented. PATIENTS AND METHODS: All three patients underwent magnet resonance imaging (MRI) including additional diffusion- weighted imaging and apparent diffusion coefficient maps. Clinical and laboratory findings were recorded. RESULTS: MRI of all three patients revealed similar symmetric supratentorial hyperintense signal changes involving the frontal, parietal, occipital and temporal lobes. The cortex was spared and the subcortical U fibers were partially involved. Further, the brainstem and the cerebellar white matter were not affected. CONCLUSION: Toxic leukoencephalopathy without involvement of the cerebellum and brainstem is a rare complication of heroin abuse. The pattern of heroin-induced toxic leukoencephalopathy on MRI might not only be related to an unknown adulterant, but also to the mode of drug administration.

Recanalization results after intracranial stenting of atherosclerotic stenoses.

The purpose of this investigation was to provide a detailed description of the angiographic results after stenting of high-grade intracranial stenosis using balloon-expandable stents. Forty consecutive patients with symptomatic atherosclerotic intracranial stenosis >50% received endovascular treatment by placement of balloon-expandable stents using the concept of slight underdilation and strict avoidance of overdilation. Intra-arterial digital subtraction angiography images before and after stenting in the same projection were reviewed for pre- and post-therapeutic measurement of the degree of stenosis and evaluation of morphologic criteria like plaque coverage, stent apposition, patency of side branches, and signs of dissection or vasospasm. Stenting decreased the mean percentage stenosis from 76.2 (WASID criteria) to 20.8%. Residual stenosis ranged from 0 to 55% with residual stenosis >50% in two of 40 cases. Technical success rate was 95%. There were no major vessel complications, but minor abnormalities like incomplete stent apposition (8/40) or plaque coverage (7/40), incomplete filling of side branches (13/40), and minor dissections after stenting (2/40) were seen. One case with incomplete stent apposition and two cases with side branch compromise were associated with clinical symptoms. In conclusion, intracranial stenting with slight underdilation avoided major vessel complication and created reliable luminal gain. Suboptimal recanalization results were frequently detected and may be the source of neurological complications in individual cases.