Bevacizumab for recurrent malignant gliomas: efficacy, toxicity, and patterns of recurrence.

BACKGROUND: Bevacizumab, a humanized monoclonal antibody against vascular endothelial growth factor, may have activity in recurrent malignant gliomas. At recurrence some patients appear to develop nonenhancing infiltrating disease rather than enhancing tumor. METHODS: We retrospectively reviewed 55 consecutive patients with recurrent malignant gliomas who received bevacizumab and chemotherapy to determine efficacy, toxicity, and patterns of recurrence. Using a blinded, standardized imaging review and quantitative volumetric analysis, the recurrence patterns of patients treated with bevacizumab were compared to recurrence patterns of 19 patients treated with chemotherapy alone. RESULTS: A total of 2.3% of patients had a complete response, 31.8% partial response, 29.5% minimal response, and 29.5% had stable disease. Median time to radiographic progression was 19.3 weeks. Six-month progression-free survival (PFS) was 42% for patients with glioblastoma and 32% for patients with anaplastic glioma. In 23 patients who progressed on their initial therapy, bevacizumab was continued and the concurrent chemotherapy agent changed. In no case did the change produce a radiographic response, but two patients had prolonged PFS of 20 and 31 weeks. Recurrence pattern analysis identified a significant increase in the volume of infiltrative tumor relative to enhancing tumor in bevacizumab responders. CONCLUSIONS: Combination therapy with bevacizumab and chemotherapy is well-tolerated and active against recurrent malignant gliomas. At recurrence, continuing bevacizumab and changing the chemotherapy agent provided long-term disease control only in a small subset of patients. Bevacizumab may alter the recurrence pattern of malignant gliomas by suppressing enhancing tumor recurrence more effectively than it suppresses nonenhancing, infiltrative tumor growth.

Preeclampsia-eclampsia: clinical and neuroradiographic correlates and insights into the pathogenesis of hypertensive encephalopathy.

PURPOSE: To investigate the clinical parameters that are associated with the development of brain edema of hypertensive encephalopathy in patients with preeclampsia-eclampsia. MATERIALS AND METHODS: Twenty-eight patients with preeclampsia-eclampsia and neurologic symptoms underwent magnetic resonance (MR) imaging. Clinical parameters recorded at the time of MR imaging included serum electrolytes and various indices of hematologic, renal, and hepatic function. Several data were available 1 week prior to the development of neurologic symptoms in 11 patients. Univariate analysis and multivariate logistic regression analyses were performed to study possible associations between these parameters and brain edema at MR imaging. RESULTS: The 20 patients with brain edema at MR imaging had a significantly greater incidence of abnormal red blood cell morphology (14 [82%] of 17 patients vs two [25%] of eight, P: <.005) and higher levels of lactic dehydrogenase (LDH) (339 U/L +/- 65 [SD] vs 258 U/L +/- 65, P: =.007) than the eight with normal MR imaging findings; multivariate logistic regression analysis showed a strong association with red blood cell morphology only. Moreover, LDH levels were elevated before the development of neurologic abnormalities (P: <.05). Blood pressures were not significantly different between groups at any time. CONCLUSION: Brain edema at MR imaging in patients with preeclampsia-eclampsia was associated with abnormalities in endothelial damage markers and not with hypertension level.

Thick-section reformatting of thinly collimated helical CT for reduction of skull base-related artifacts.

OBJECTIVE: Our purpose was to evaluate thick-section reformatted helical CT of the brain base as a technique for reducing skull base-related artifacts and to compare it with conventional CT. MATERIALS AND METHODS: Twenty-three patients with suspected intracranial abnormalities related to the brain base, as determined either by clinical examination or at the time of imaging, were evaluated with contrast-enhanced conventional CT of the brain (5-mm collimation, 140 kVp, 170 mA, 2-sec rotation time) and reformatted helical CT (1-mm collimation, 1.5 pitch, 120 kVp, 220 mA). Helical sections were reformatted to a thickness of 5 mm by a volume-averaging algorithm using a computer workstation. Three observers retrospectively and blindly reviewed the images and qualitatively scored artifacts at the foramen magnum, middle cranial fossa, anterior cranial fossa, interpetrous region, and internal occipital protuberance. Image graininess and observer confidence were also scored. Paired statistical analyses using score differences in each patient were possible. RESULTS: Reformatted helical CT reduced skull base-related artifacts across all five anatomic regions (p < 0.05). The foramen magnum showed the greatest reduction in artifacts and the anterior cranial fossa the least. Image graininess was increased on reformatted CT compared with conventional CT (p < 0.05), but observer confidence remained higher for reformatted CT (p < 0.05). Total additional scan time was 3.15 +/- 0.38 min with 5.3 +/- 1.2 min required for reformatting. CONCLUSION: Reformatted CT significantly decreases skull base-related artifacts in the brain, improving confidence in evaluation of the brain base and adding an average of only 8.45 +/- 1.6 min of scanning and processing time to each examination.

Echoplanar MR imaging for ultrafast detection of brain lesions.

OBJECTIVE: We retrospectively evaluated the use of echo-planar imaging for ultrafast detection of brain lesions. MATERIALS AND METHODS: In our retrospective study, 61 patients were imaged with the following echo-planar sequences: single-shot proton density-weighted, single-shot T2-weighted, single-shot T2-weighted high-resolution, multishot proton density-weighted, and multishot T2-weighted. Lesions revealed in these patients ranged from 0.5 to 12.0 cm (mean, 3.7 cm) and were the result of tumor (n = 16), stroke (n = 21), demyelination (n = 18), and toxoplasmosis (n = 2). Four patients had scans with normal findings. Two neuroradiologists who were unaware of pertinent clinical data reviewed the images. The images were retrospectively compared with conventional spin-echo images for diagnosis, sensitivity of lesion detection, and qualitative criteria: subjective image quality, gray and white matter differentiation, lesion conspicuity, delineation of lesion borders, and artifacts. (Artifacts included those caused by motion, susceptibility, pulsation, and ghosting.) Quantitative criteria, including signal-to-noise and signal difference-to-noise measurements, were also evaluated in 40 lesions. RESULTS: Sensitivity for lesion detection was 97% for single-shot echo-planar T2-weighted MR images and 100% for multishot echo-planar T2-weighted MR images. Single-shot echo-planar proton density-weighted MR images had the highest signal-to-noise ratio (91.2+/-19.3). Echo-planar T2-weighted MR images had the highest signal difference-to-noise (33.8+/-22.9). Echo-planar sequences were superior to spin-echo sequences regarding motion and pulsation artifacts. Spin-echo sequences lacked susceptibility and ghosting artifacts, and were superior in lesion conspicuity and delineation of lesion borders. CONCLUSION: In this study, echo-planar sequences were as sensitive as conventional spin-echo imaging for the diagnostic assessment of lesions. Echo-planar sequences had a strikingly shorter acquisition time and substantially reduced motion and pulsation artifacts. Echo-planar sequences may be a useful diagnostic tool for use in claustrophobic and unstable patients.

Evaluation of intracranial cysts by intraoperative MR.

Eleven patients with intracranial cystic collections were evaluated in the open-bore intraoperative MR system. In each case, the cystic collection or the surrounding cerebrospinal fluid (CSF) space was injected with .02 to .5 cc of .5 mol/l gadopentetate dimeglumine. Serial imaging was performed using T1-weighted imaging. In seven patients, free communication was demonstrated between the cystic collection and the surrounding CSF spaces. In four cases, the cyst did not communicate with the CSF; two of these were drained in the intraoperative MR system with reduction in symptoms. One patient developed an aseptic meningitis 10 days after the study, which was successfully treated with steroids; no other complications were noted. We conclude that the communication of intracranial cystic collections with the cisterns and ventricles can be safely and effectively elucidated with gadolinium injection in the intraoperative MR system.

Half-fourier acquisition single-shot turbo spin-echo (HASTE) MR: comparison with fast spin-echo MR in diseases of the brain.

PURPOSE: To compare an ultrafast T2-weighted (half-Fourier acquisition single-shot turbo spin-echo [HASTE]) pulse sequence with fast spin-echo T2-weighted sequences in MR imaging of brain lesions. METHODS: Fast spin-echo and HASTE images of 34 consecutive patients over the age of 50 years or with suspected demyelinating disease were reviewed independently by two neuroradiologists for the number of lesions less than 5 mm and greater than or equal to 5 mm, and for lesion conspicuity, gray-white matter differentiation, and extent of periventricular confluent signal abnormality. The reviewers also assessed for the presence of hemosiderin and extent of motion artifacts. RESULTS: Per patient, the mean number of 5-mm or larger lesions detected on fast spin-echo images (1.4) relative to the number detected on HASTE images (0.8) was not statistically significant. For lesions less than 5 mm, fast spin-echo images showed more lesions (7.5) than HASTE images did (2.4). The fast spin-echo images were better at depicting gray-white matter differentiation, conspicuity of lesions, and periventricular signal abnormality. Of four T2 hypointense lesions seen on fast spin-echo images, none was detected on HASTE images. CONCLUSION: Although the HASTE technique might be useful for rapid imaging of the brain, our study shows a diminished sensitivity for the detection of lesions less than 5 mm in diameter and for T2 hypointense lesions.

Unusual manifestations of head trauma.

Head injury is the leading cause of death in young adults. More than 2 million head injuries occur each year in the United States. Accurate and rapid diagnosis is imperative for the successful management of the trauma patient. The radiologic findings of common manifestations of head trauma have been well described; however, the patient who presents in the post-traumatic period with atypical radiographic findings is at risk for misdiagnosis and delay in treatment. In this essay, we illustrate some unusual findings in head injury that may complicate diagnosis.

Imaging evaluation of the optic nerve and visual pathway including cranial nerves affecting the visual pathway.

The radiologic investigation of the optic pathways has an integral part in the diagnostic evaluation of diverse lesions, such as inflammatory disease, vascular disorders, and benign and malignant tumors that afflict the optic pathways. These radiologic methods consist principally of computed tomography and magnetic resonance imaging, and, in vascular lesions, magnetic resonance angiography and conventional angiography.

MR imaging of diseases of the brain: comparison of GRASE and conventional spin-echo T2-weighted pulse sequences.

OBJECTIVE: The purpose of this study was to compare a combined gradient and spin-echo (GRASE) technique, which is a rapid T2-weighted imaging sequence, with conventional spin-echo (SE) sequences for imaging brain lesions. The GRASE sequences would allow increased patient throughput with potential cost savings and be useful in uncooperative patients without requiring echoplanar imaging techniques and specialized hardware. SUBJECTS AND METHODS: Conventional SE and GRASE T2-weighted images of 49 consecutive patients (20-86 years old) were reviewed independently by three neurora-diologists for the presence and characterization of lesions (most of which were nonspecific foci of hyperintensity within the white matter), gray-white matter differentiation, conspicuity of lesions, and periventricular signal abnormality. The MR studies were performed on a 1.0-T Siemens Magnetom Impact scanner, with the SE images obtained using a TR/TE of 2400/40 and the GRASE images obtained using a TR/effective TE of 4400/110. RESULTS: The number of lesions detected that were 5 mm or larger in maximal diameter did not significantly differ among techniques. For lesions smaller than 5 mm, conventional SE T2-weighted images showed more lesions (p < .01). The SE images were better than the GRASE images for assessing gray-white matter differentiation, conspicuity of lesions, and periventricular signal abnormality. The two hypointense lesions were better assessed on the conventional SE images. CONCLUSION: Although GRASE imaging may be potentially useful for rapid imaging of the brain, our experience shows it has a markedly diminished sensitivity for detecting lesions smaller than 5 mm in diameter. Currently, GRASE imaging should not replace the routine clinical use of conventional SE sequences.

Preoperative assessment of the carotid bifurcation. Can magnetic resonance angiography and duplex ultrasonography replace contrast arteriography?

BACKGROUND AND PURPOSE: Noninvasive studies are used with increasing frequency to assess the carotid bifurcation before endarterectomy. Therefore, assessment of their diagnostic accuracies is essential for appropriate patient management. We prospectively evaluate two noninvasive tests, magnetic resonance angiography (MRA) and duplex ultrasonography (DU), as potential replacements for contrast arteriography (CA). METHODS: A blinded comparison of three-dimensional time-of-flight (TOF) MRA, two-dimensional TOF MRA, and DU in 176 arteries was performed. CA was used as the standard of comparison. RESULTS: Three-dimensional TOF MRA had a sensitivity of 94%, a specificity of 85%, and an accuracy of 88% for the identification of 70% to 99% stenosis; two-dimensional TOF MRA had a sensitivity and specificity that were approximately 10% lower than those of three-dimensional TOF MRA. DU resulted in a sensitivity of 94%, a specificity of 83%, and an accuracy of 86%. Combining data from three-dimensional TOF MRA and DU, allowing for CA only for disparate results, yielded a sensitivity of 100%, a specificity of 91%, and an accuracy of 94% among concordant noninvasive tests, with CA required in 16% of arteries. MRA accurately differentiated 17 carotid occlusions from 16 high-grade (90% to 99%) stenoses, whereas with DU two patent arteries were identified as occluded and one occluded artery was identified as patent. CONCLUSIONS: Three-dimensional TOF MRA is the most accurate noninvasive test. Combined use of MRA and DU results in a marked increase in accuracy to a level that obviates the need for CA in a majority of patients.

Perioperative imaging strategies for carotid endarterectomy. An analysis of morbidity and cost-effectiveness in symptomatic patients.

OBJECTIVE: To assess the cost-effectiveness of four diagnostic strategies for the preoperative evaluation of symptomatic patients who are potential candidates for carotid endarterectomy (ie, 70% to 99% stenosis): (1) duplex sonography (DS), (2) magnetic resonance angiography (MRA), (3) contrast angiography (CA), and (4) the combination of DS and MRA supplemented by CA for disparate results. METHODS: Cost-effectiveness analysis based largely on published clinical trial data. Sensitivities and specificities of noninvasive tests were estimated from 81 patients undergoing prospective evaluation with DS, MRA, and CA. OUTCOME MEASURE: Incremental cost per quality-adjusted year of life gained. RESULTS: For a hypothetical cohort of symptomatic patients undergoing evaluation for carotid endarterectomy, the combination of tests resulted in the greatest quality-adjusted life expectancy of the four options considered. After incorporating the costs of testing, surgery, and stroke, we found that neither the MRA nor the CA strategy was cost-effective. The combination of tests was more effective but more costly than DS, resulting in an additional cost of $22,400 per quality-adjusted year of life gained. For centers that do not have adequate MRA, CA resulted in an additional cost of $99,200 per quality-adjusted year of life saved compared with DS. CONCLUSIONS: Our results suggest that for the preoperative detection of a 70% to 99% carotid stenosis, the combination of DS and MRA, supplemented by CA for disparate results, is associated with the lowest long-term morbidity and mortality and has a favorable cost-effectiveness ratio. The combination of tests, or DS alone when MRA is not available, could potentially replace the current practice of using CA alone in the preoperative evaluation of patients with symptomatic carotid stenosis.

Cyclosporine neurotoxicity and its relationship to hypertensive encephalopathy: CT and MR findings in 16 cases.

OBJECTIVE: The neurotoxic effects of cyclosporine therapy are well known but poorly understood. Imaging studies typically show subcortical edema predominantly affecting the posterior regions of the brain. We sought to determine the causes for these findings by comparing radiographic data with various clinical parameters. MATERIALS AND METHODS: In a 3-year period, 16 patients with neurologic findings attributed to cyclosporine therapy were examined with CT, MR imaging, or both. In most cases, imaging was performed both at the onset of the neurologic syndrome and after it had resolved. The radiographic findings were evaluated with respect to lesion location and changes over time. Various clinical and laboratory data obtained throughout the patients' hospital course were also reviewed, including cyclosporine levels, blood pressure values, hematologic data, and serum levels of cholesterol, magnesium, creatinine, and albumin. RESULTS: The only major factor associated with the neurotoxic effects of cyclosporine in all patients was systemic hypertension. Microangiopathic hemolytic anemia, thrombocytopenia, and hypoalbuminemia were also common, and patients usually displayed signs of sympathetic overactivation. The onset of neurologic symptoms was unrelated to serum levels of creatinine, magnesium, cholesterol, or cyclosporine. The clinical and radiographic findings of these patients were identical to those previously reported in patients with hypertensive encephalopathy. Findings resolved in all but one patient after reduction of blood pressure, with or without reduction in cyclosporine dose. In four patients, intracranial hemorrhages occurred during the hypertensive episode, resulting in one fatality. CONCLUSION: The clinical and radiologic findings in patients showing the neurotoxic effects of cyclosporine appear to be identical to those with hypertensive encephalopathy. Other associated factors, such as cyclosporine-induced vasculopathy or hypoalbuminemia may also play a role in the condition, and intracranial hemorrhage may occur owing to associated thrombocytopenia. Symptoms generally resolve after reduction of blood pressure, and follow-up is usually unnecessary in uncomplicated cases.

Dissection of the carotid and vertebral arteries: imaging with MR angiography.

Arterial dissection occurs when an intimal tear allows blood to enter the arterial wall, potentially compromising the lumen and reducing blood flow. Carotid and vertebral artery dissections typically occur after major trauma, although they also can arise spontaneously or after trivial injury. Arterial dissection has been associated with a variety of factors, including hypertension, fibromuscular dysplasia, Marfan syndrome, cystic medial necrosis, oral contraceptives, drug abuse (sympathomimetics), and infection [1-8]. It is important to recognize arterial dissection early so that prompt treatment can be initiated to prevent ischemic complications [1]. In this essay, we illustrate the use of MR angiography in the diagnosis of carotid and vertebral artery dissection.

MR angiography of the carotid bifurcation: artifacts and limitations.

Although conventional angiography has traditionally been the preoperative study of choice before carotid endarterectomy, alternative noninvasive methods, particularly MR angiography, are being used increasingly for evaluation of the carotid bifurcation [1, 2]. In this essay, we illustrate the limitations of and artifacts associated with two-dimensional (2D) and three-dimensional (3D) time-of-flight (TOF) MR angiography vs conventional angiography when assessing disease of the carotid bifurcation.