Local cerebral blood flow and partition coefficients measured in cerebral astrocytomas of different grades of malignancy

Imaging blood flow in brain tumors using arterial spin labeling

Measurements of tumor blood flow (TBF) are important for understanding tumor physiology and can be valuable in selecting and evaluating therapies. Brain tumors typically present reduced blood flows compared to normal brain tissue. This study shows that the arterial spin labeling (ASL) technique can be used to measure TBF non-invasively in a rat glioma model. Results show that TBF in the core (36.3 +/- 18.9 ml/100g/min, n=4) and peripheral regions (85.3 +/- 26.9 ml/100g/min, n=4) of the tumor are significantly reduced and show considerable heterogeneity compared to cerebral blood flow (CBF) of normal brain tissue (147.7 +/- 31.1 ml/100g/min, n=4), while T(1) in the tumor (2.6 +/- 0.1 sec) is significantly elevated compared to normal tissue T(1) (2.0 +/- 0.0 sec). These results strongly support the feasibility of using the ASL technique to evaluate different cancer treatment strategies, to monitor the effects of agents designed to modulate TBF and oxygenation (e.g., carbogen gas), and to assess and guide the use of anti-angiogenic agents. Magn Reson Med 44:169-173, 2000.

A critique of the role of the blood-brain barrier in the chemotherapy of human brain tumors

There is general agreement that most chemotherapy agents achieve only relatively low concentrations in the normal central nervous system, that the blood-brain barrier is variably disrupted in malignant brain tumors, and that the concentration of chemotherapy drugs in the brain adjacent to tumor is intermediate between concentrations achieved in brain tumors vs normal brain. However, there is substantial controversy regarding the role of the blood-brain barrier in resistance to chemotherapy of intracerebral tumors. Many chemotherapy agents achieve concentrations in brain tumors that are comparable to those in extracerebral tumors, and drugs that cross the intact blood-brain barrier only poorly may be active against intracerebral tumors. Furthermore, the hypothesis that the brain is a pharmacological sanctuary where metastases may grow while tumor is responding in other parts of the body may be flawed: there are only 2 or 3 types of malignancies (out of all those that are sensitive to chemotherapy) in which the risk of isolated central nervous system relapse is moderately high, and even in these 2 or 3, effective central nervous system prophylaxis has minimal or no impact on overall survival. Furthermore, drugs that cross the BBB do not appear to be more effective than other drugs at reducing the risk of brain metastases, and brain metastases at the time of diagnosis do not necessarily convey a worse prognosis than metastases to various other sites. While average drug concentrations in brain adjacent to tumor are lower than those within brain tumors, very small numbers of tumor cells may be capable of inducing local leakiness in blood vessels, and there is little information on drug concentrations achieved in individual tumor cells within the brain adjacent to tumor. Furthermore, any limitation of uptake of drugs into brain tumors could be at least partially due to increased tissue pressure within tumors rather than being due to blood-brain barrier phenomena. This distinction could be important, since strategies that one might use to increase drug delivery to brain tumors might differ depending on whether the reduced delivery were due to barrier phenomena vs blood flow phenomena. The role of the blood-brain barrier in resistance of intracerebral tumors to chemotherapy remains unclear: while it may well play some role (and perhaps even a major one), self-fulfilling prophecies and unintentional bias in data selection and interpretation may have previously made it appear more important than it actually is.(ABSTRACT TRUNCATED AT 400 WORDS)

CBF measured by Xe-CT: approach to analysis and normal values

Normal reference values and a practical approach to CBF analysis are needed for routine clinical analysis and interpretation of xenon-enhanced computed tomography (CT) CBF studies. We measured CBF in 67 normal individuals with the GE 9800 CT scanner adapted for CBF imaging with stable Xe. CBF values for vascular territories were systematically analyzed using the clustering of contiguous 2-cm circular regions of interest (ROIs) placed within the cortical mantle and basal ganglia. Mixed cortical flows averaged 51 +/- 10ml.100g-1.min-1. High and low flow compartments, sampled by placing 5-mm circular ROIs in regions containing the highest and lowest flow values in each hemisphere, averaged 84 +/- 14 and 20 +/- 5 ml.100 g-1.min-1, respectively. Mixed cortical flow values as well as values within the high flow compartment demonstrated significant decline with age; however, there were no significant age-related changes in the low flow compartment. The clustering of systematically placed cortical and subcortical ROIs has provided a normative data base for Xe-CT CBF and a flexible and uncomplicated method for the analysis of CBF maps generated by Xe-enhanced CT.

Role of thalamus and white matter in cognitive outcome after head injury

Local CBF (LCBF) and local partition coefficients (L lambda) were measured by xenon-enhanced computed tomography among 15 patients with remote cerebral trauma resulting from severe head injury. Results were compared with similar measures among age-matched normal volunteers (N = 20). The patients were divided into two groups according to different outcomes based on serial cognitive testing: Group I (N = 10) improved but Group D (N = 5) deteriorated throughout a mean interval of 10 years of follow-up. Initial LCBF measurements were performed at mean intervals of 6.8 years after injury. Cortical LCBF values were decreased in frontal (p less than 0.01) and temporal (p less than 0.05) regions among both groups, but only in Group D were flow values decreased in putamen and thalamus (p less than 0.05). L lambda values were reduced in frontotemporal cortex among both groups but in the thalamus only among Group D (p less than 0.05). Mean white matter flow values were normal in Group I but were reduced in Group D (p less than 0.05). Mean partition coefficients for white matter were reduced in both groups (p less than 0.01) but were lower in Group D (p less than 0.05). Reduced perfusion of frontotemporal gray matter is consonant with neuropathological reports following severe brain trauma of neuronal atrophy, gliosis, and infarction affecting these regions. Group comparisons between patients who cognitively improved versus those that deteriorated demonstrate an association between reductions of CBF in putamen, thalamus and subcortical white matter and impaired cognition after severe head injury.

Stable xenon does not increase intracranial pressure in primates with freeze-injury-induced intracranial hypertension

Stable xenon (Xe)-enhanced computed tomography is a potentially valuable tool for high resolution, three-dimensional measurement of CBF in patients. However, reports that Xe causes cerebrovascular dilation and increases intracranial pressure (ICP) have tempered enthusiasm for its use. The effects of 5 min of 33% Xe inhalation on ICP (right and left hemispheres) were studied in eight fentanyl-anesthetized Rhesus monkeys after right-sided cortical freeze injury. ICP, CBF, and physiological variables were monitored for up to 6 h postinsult. The preinjury (control) right hemispheric ICP was 8 +/- 5 mm Hg (mean +/- SD) and left hemispheric ICP was 5 +/- 2 mm Hg. Postinjury observations were classified into low (less than 15 mm Hg) and high ICP (greater than or equal to 15 mm Hg) groups. Both right and left ICP values averaged 9 +/- 3 mm Hg in the low ICP group. In the high ICP group, the right ICP was 20 +/- 4 mm Hg and left ICP was 21 +/- 6 mm Hg. ICP was unchanged by Xe inhalation under control conditions as well as in both low and high ICP groups postinjury. Postinjury, the MABP decreased 10-15 mm Hg in the low ICP group and 10-17 mm Hg in the high ICP group 2-3 min after the start of Xe inhalation (p less than 0.05). These results show that 33% Xe inhalation does not increase ICP in fentanyl-anesthetized monkeys but could decrease MABP in stressed states, presumably because of the anesthetic effects of Xe.

Neoplastic encephalopathy: dynamic CT of cerebral gliomata

Intravenous contrast enhanced dynamic computed tomography of cerebral gliomata reveals a spectrum of patterns which reflect different degrees of neovascularity as well as a variable breakdown in the blood-tumor-barrier both intratumorally as well as between individual tumors. Phenomena not generally associated with gliomas including intrinsic neoplastic and peripheral cerebral hypoperfusion, hyperperfusion, and indications of vascular stealing are also demonstrated with this technique which conceivably explain and are partially responsible for certain aspects of the encephalopathy accompanying cerebral neoplasia. A comparison of the dynamic sequences with conventional selective cerebral angiography further indicates that the more contrast-sensitive dynamic method is potentially superior in the detection of subtle neovascularity.

Imaging local cerebral blood flow by Xenon-enhanced computed tomography--technical optimization procedures

Methods are described for non-invasive, computer-assisted serial scanning throughout the human brain during eight minutes of inhalation of 27%-30% Xenon gas in order to measure local cerebral blood flow (LCBF). Optimized Xenon-enhanced computed tomography (XeCT) was achieved by 5-second scanning at one-minute intervals utilizing a state-of-the-art CT scanner and rapid delivery of Xenon gas via a face mask. Values for local brain-blood partition coefficients (L lambda) measured in vivo were utilized to calculate LCBF values. Previous methods assumed L lambda values to be normal, introducing the risk of systematic errors, because L lambda values differ throughout normal brain and may be altered by disease. Color-coded maps of L lambda and LCBF values were formatted directly onto CT images for exact correlation of function with anatomic and pathologic observations (spatial resolution: 26.5 cubic mm). Results were compared among eight normal volunteers, aged between 50 and 88 years. Mean cortical gray matter blood flow was 46.3 +/- 7.7, for subcortical gray matter was 50.3 +/- 13.2 and for white matter was 18.8 +/- 3.2. Modern CT scanners provide stability, improved signal to noise ratio and minimal radiation scatter. Combining these advantages with rapid Xenon saturation of the blood provides correlations of L lambda and LCBF with images of normal and abnormal brain in a safe, useful and non-invasive manner.

Two patients with meningioma visualized as high uptake by SPECT with N-isopropyl-p-iodo-amphetamine (I-123)

It has been reported that brain tumors show decreased IMP uptake, however, we encountered two cases of meningioma which had initially increased IMP uptake. SPECT was performed at 30 min, 2 hr and 6 hr after intravenous administration of IMP (3 mCi). The delayed images (2 and 6 h) showed decreased IMP uptake. These were convexity and parasagittal meningiomas which had tumor stains at angiography. Although the mechanism of IMP uptake in brain tumor is still unclear, this finding should be considered in the interpretation of IMP-SPECT.