MR imaging of intramedullary and intradural-extramedullary spinal cysticercosis.

OBJECTIVE: The purpose of our study was to retrospectively review the MR imaging findings in a group of patients with clinically proven cysticercosis involving the spinal cord, the spinal subarachnoid space, or both. MATERIALS AND METHODS: We retrospectively reviewed images of 16 patients with clinically diagnosed spinal cysticercosis to summarize the imaging characteristics. All patients underwent T1- and T2-weighted sagittal, axial, or both sagittal and axial MR imaging before i.v. administration of paramagnetic contrast media. Thirteen patients also underwent sagittal, axial, or both sagittal and axial T1-weighted MR imaging after i.v. gadolinium administration. In addition, all patients underwent cranial CT, MR imaging, or both to reveal possible evidence of cranial cysticercosis. RESULTS: MR imaging revealed isolated intradural-extramedullary involvement (n = 9), isolated intramedullary involvement (n = 3), combined intradural-extramedullary and intramedullary involvement (n = 3), and/or syringomyelia caused by infection and associated with chronic spinal arachnoiditis (n = 2). Evidence of intradural-extramedullary disease included cystic structures within the subarachnoid space or homogeneous sheetlike enhancement within the subarachnoid space over the surface of the spinal cord. Evidence of intramedullary disease included focal cystic lesions or syringomyelic cavitation of the spinal cord. All patients had evidence of simultaneous intracranial cysticercosis as shown on cranial CT, MR imaging, or both. CONCLUSION: In the absence of scolex visualization, cysticercotic involvement of the spinal cord or spinal subarachnoid space has a nonspecific appearance on MR imaging. On the basis of the findings in this group of patients, we believe that spinal cysticercosis is most often accompanied by intracranial disease.

MR imaging of the bird's nest filter.

The appearance of the Bird's Nest inferior vena cava filter on magnetic resonance (MR) images of 11 patients is described. No complication or symptomatic filter displacement was encountered as a result of MR imaging performed at 1.5 T. The filters created significant local artifact and distortion on MR images. However, diagnostic MR images of the pelvis, spine, and brain may still be obtained.

Gradient-echo MR imaging of the lumbar spine: comparison with spin-echo technique.

Gradient-echo (GRE) magnetic resonance (MR) imaging has been advocated as the imaging modality of choice for evaluating radiculopathy in the cervical spine. Axial GRE images of the lumbar spine in 50 patients were compared with similar images obtained using spin-echo (SE) technique on a 1.5 T MR system. The SE images were superior to GRE images in the evaluation of the neural foramina, epidural fat, and disk herniation. The GRE images obtained were inadequate for lumbar spine imaging due to an unacceptable level of chemical shift artifacts. The GRE technique does provide the advantages of rapid acquisition of T2* images and decreased motion artifact. Axial GRE images may play an increasingly important role in lumbar spine imaging with continued changes in software and improvements in technology.

Acute hematomas: effects of deoxygenation, hematocrit, and fibrin-clot formation and retraction on T2 shortening.

Acute hematomas can appear hypointense on T2-weighted magnetic resonance (MR) images at field strengths as low as 0.35 T. Using Raman spectroscopy to measure blood oxygenation and taking T2 measurements at 2.1 and 9.4 T, the authors examined the relaxation mechanisms acting during deoxygenation, increases in hematocrit, and fibrin-clot formation and retraction. Individual contributions to overall T2 from deoxyhemoglobin and the interactions of water with protein hydration layers in hemoglobin, plasma proteins, and fibrin were measured. Overall T2 values estimated by summing individual relaxation rates were in reasonable agreement with the T2 values of clotted blood. Results suggest that deoxygenation may be most important in T2 shortening, followed by increased hematocrit. T2 shortening from fibrin polymerization was minimal at the field strengths used. Effects of deoxygenation and increasing hematocrit are more sensitive to field strength than fibrin T2 shortening. Effects of fibrin may be more significant at middle and low field strengths.

Optimization of gradient-echo imaging parameters for intracaval filters and trapped thromboemboli.

Flow-phantom magnetic resonance (MR) gradient-echo (GRE) imaging at 1.5 T was performed on a titanium Greenfield filter containing trapped blood clots with a high concentration of either deoxyhemoglobin (DHb) or methemoglobin (MHb), simulating acute and older thromboemboli, respectively. Flip angle, repetition time (TR), and echo time (TE) were varied, and a contrast-to-noise ratio between trapped clots and flowing fluid (clot-flow contrast) was determined for each set of imaging parameters. Use of very low flip angles (less than or equal to 10 degrees) rendered MHb clots indistinguishable from flowing fluid. In general, DHb clots displayed greater clot-flow contrast than MHb clots regardless of flip angle. With increasing TE values, T2* effect was observed with MHb clots, and magnetic susceptibility artifacts increased. Overall, optimum clot-flow contrast for imaging of both DHb and MHb clots was achieved with a flip angle of 45 degrees-60 degrees, a TR of 50 msec, and the shortest TE possible. Using GRE parameters similar to the optimum parameters determined in vitro, the authors imaged four patients with nickel-titanium Simon filters and one dog with a titanium Greenfield filter. MR imaging was successful in demonstrating filter location, caval patency, and the presence and extent of intraluminal thrombus.

Ferromagnetism and MR imaging: safety of carotid vascular clamps.

Metallic extracranial carotid vascular clamps of the Selverstone, Crutchfield, Poppen-Blaylock, Salibi, Kindt, and tantalum varieties have been placed for treatment of large, giant, or inoperable intracranial aneurysms. To ascertain what adverse effect, if any, MR imaging would have on these clamps, magnetic deflection at 1.5 T was measured for various carotid clamps. Marked magnetic deflection (and torque) was displayed by stainless steel Poppen-Blaylock clamps. Relatively mild magnetic deflection was displayed by the stainless steel Selverstone, Salibi, Crutchfield, and Kindt clamps. Three patients with previously placed carotid clamps (two Selverstone, one Salibi) and one patient with a nonferromagnetic tantalum carotid clip had cranial or cervical MR studies at field strengths ranging from 0.35 to 0.60 T. No patient experienced any discomfort or neurologic sequelae as a result of MR imaging. Although the ferromagnetic clamps created severe "black-hole" artifacts and image distortion within the cervical and facial regions, no significant image degradation was apparent during spin-echo imaging of the brain. The tantalum clip created a far smaller MR artifact than did ferromagnetic clamps and allowed effective spin-echo and gradient-echo imaging in the cervical region. Our findings indicate that most patients with carotid vascular clamps (and nonferromagnetic clips) can probably be imaged safely with MR.

MR imaging of neurocysticercosis.

Twenty-six patients with neurocysticercosis were studied with MR imaging to correlate their clinical presentation with the location and appearance of their neurocysticercosis lesions. Intraventricular cysts were present in 14 patients (54%), parenchymal cysts were present in 18 (69%), and intraventricular together with parenchymal cysts were present in six (23%). Intraventricular cysts were detected by mass effect, ventricular obstruction, detection of a cyst rim, and/or CSF flow void adjacent to the cyst. The intensity of most intraventricular and parenchymal cysts presumed to be viable was similar to that of CSF on both T1- and T2-weighted sequences. Cysts presumed to be degenerated had increased signal intensity on T1-weighted images, probably resulting from increased protein content. Pericystic high signal intensity surrounding lesions of various ages was seen on both proton-density- and T2-weighted images and represents gliosis, edema, and inflammation. Patients with parenchymal cysts had symptoms of seizures, while those with intraventricular cysts generally had symptoms related to obstructive hydrocephalus. Aqueductal stenosis, seen in 10 patients (38%), was possibly due to ependymal inflammation or adhesions caused by prior ventricular infection by neurocysticercosis. One patient with the racemose form of neurocysticercosis demonstrated abundant cyst wall proliferation resulting in obstructive hydrocephalus. In six patients scanned 1-6 months after oral praziquantel therapy, there was no change in the MR appearance of intraventricular cysts, while some parenchymal cysts showed evidence of degeneration. We found MR to be useful in detecting the cysts of neurocysticercosis and the accompanying signs of cyst degeneration and pericystic inflammation. MR was inferior to CT in the detection of parenchymal calcifications.

Common pitfalls in magnetic resonance imaging of the knee.

Magnetic resonance examinations of 200 knees were studied retrospectively to determine the frequency with which normal ligamentous and tendinous structures mimicked pathological changes in the meniscus or elsewhere in the knee. Although the course of the transverse geniculate ligament simulated a tear of the anterior horn of the lateral meniscus on forty-four (22 per cent) of the sagittal magnetic-resonance scans, no tear of the lateral meniscus was found in the nine patients in this group who had an arthroscopic examination. The normal anatomy of the meniscofemoral ligament varies greatly. The ligament of Humphry, the anterior branch of the meniscofemoral ligament, was visualized on sixty-six (33 per cent) of the magnetic resonance scans. The ligament of Wrisberg, the posterior branch of the meniscofemoral ligament, was identified on sixty-five (32.5 per cent) of the scans. On sagittal images, these ligamentous branches can be mistaken for osteochondral or meniscal fragments. However, a loose body was found in only one of the thirty-two patients who had an arthroscopic examination after the ligament of Humphry or of Wrisberg was seen on the magnetic resonance image. The bursa of the popliteus tendon simulated a tear of the posterior horn of the lateral meniscus in fifty-five (27.5 per cent) of the studies, although an actual tear of the posterior horn was found in only two of the eleven patients who had an arthroscopic examination.(ABSTRACT TRUNCATED AT 250 WORDS)

Magnetic resonance imaging of cardiac sarcomas.

Two patients with cardiac sarcomas were evaluated with magnetic resonance imaging (MRI). In both cases, MRI provided more information regarding the extent of the tumor than computed tomography, two-dimensional echocardiography, or angiography.

CT diagnosis of traumatic rupture of the cisterna chyli.

The CT findings are presented in a patient with chyloperitoneum and chyloretroperitoneum following blunt trauma to the cysterna chyli. The simultaneous demonstration of both intraperitoneal and extraperitoneal water density fluid appears to be specific for this injury.