Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Thulborn, K. R.
Right arrow Articles by Erb, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thulborn, K. R.
Right arrow Articles by Erb, P.
(Radiology. 1999;213:156-166.)
© RSNA, 1999

Neuroradiology

Comprehensive MR Imaging Protocol for Stroke Management: Tissue Sodium Concentration as a Measure of Tissue Viability in Nonhuman Primate Studies and in Clinical Studies1

Keith R. Thulborn, MD, PhD, Tatyana S. Gindin, BS, Denise Davis, BS and Patricia Erb, RN

1 From the MR Research Center, Presbyterian University Hospital, B855, 200 Lothrop St, Pittsburgh, PA 15213-2582. Received August 5, 1998; revision requested October 15; revision received December 15; accepted March 29, 1999. Supported in part by Public Health Service grants PO1 NS35949-01A1 and RO1 CA63661 and GE Medical Systems. Address reprint requests to K.R.T. (e-mail: keith@mrctr.upmc.edu).


   Abstract
TOP
 Abstract
Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 Appendix 1
 References
 
PURPOSE: To investigate sodium magnetic resonance (MR) imaging for monitoring tissue viability in stroke.

MATERIALS AND METHODS: A comprehensive MR imaging protocol used to measure apparent diffusion coefficient and perfusion parameters was extended to include sodium imaging. Tissue sodium concentration was estimated by using a two-compartment model. This protocol lasted less than 45 minutes. These parameters were followed over the first 6 hours in a nonhuman primate model (n = 2) of acute embolic stroke without or with thrombolytic therapy. This protocol was used in patients in whom acute (<24 hours, n = 11) or nonacute (>=24 hours, n = 31) stroke was ultimately confirmed.

RESULTS: The animal model showed abnormal diffusion and perfusion parameters in the lesion immediately after embolization, and these remained abnormal for over 6 hours. Tissue sodium concentration increased with time (5.7 mmol/L/h) unless halted with thrombolytic therapy. Regions with sodium concentrations over 70 mmol/L were histochemically verified as being infarcted. In patients in whom stroke older than 6 hours was clinically confirmed, sodium concentrations over 70 mmol/L were found in the appropriate brain regions.

CONCLUSION: Tissue sodium concentration provides a sensitive measure of tissue viability that is complementary to the diagnostic role of diffusion and perfusion imaging for ischemic insult.

Index terms: Brain, infarction, 10.771, 10.781 • Brain, MR, 10.121411, 10.121412 • Magnetic resonance (MR), diffusion study, 10.12144 • Magnetic resonance (MR), perfusion study, 10.12144 • Magnetic resonance (MR), sodium studies, 10.12147 • Thrombolysis, 10.1265, 10.771, 17.1265


   Introduction
TOP
 Abstract
 Introduction
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 Appendix 1
 References
 
Stroke, the third largest cause of mortality and the leading cause of disability, represents a major cost of health care in the industrialized countries (1–4). New strategies for the treatment of acute stroke have been summarized recently (5), including the use of thrombolytic therapy (6). The results of the use of one thrombolytic agent, recombinant tissue plasminogen activator (rt-PA), have been reported in open (7) and randomized double-blind placebo-controlled (8) trials. These trials showed that acute intervention within 3 hours of onset improved clinical outcome at 3 months despite an increased incidence of hemorrhage.

Such studies led to the approval of rt-PA by the Food and Drug Administration, Rockville, Md, as an acute thrombolytic agent for restricted use within 3 hours of the onset of nonhemorrhagic stroke. This time constraint was based on concern for the possible promotion of hemorrhage in reperfused, nonviable tissue. As with many such trials (9–11), statistical analyses of generic clinical outcome measures in many patients have been used to measure efficacy rather than knowledge of the specific pathophysiologic conditions of individual patients. A parameter for the assessment of tissue viability in individual patients would offer an opportunity to tailor available therapies that are appropriate to the pathophysiologic condition at the time of presentation.

Authors of excellent reviews (12–17) of tissue viability in focal ischemia elaborate on the classic concept of the penumbra surrounding a focal stroke. This tissue exists between the thresholds determined with two critical flow rates, the higher being that of "electrical failure" and the lower being that of "membrane failure" (14). An important point emphasized in these reviews is that the disruption of ion homeostasis that occurs at the low blood flow rate predicts poor outcome (16). Increasing intracellular calcium is suggested to be in the final common pathway leading to cell death. As calcium homeostasis is linked to sodium homeostasis at both the cellular and the mitochondrial membranes, a measure of sodium ion homeostasis may be a useful indicator of cell viability.

Potential clinical applications of sodium magnetic resonance (MR) imaging, although they were technically challenging previously, were discussed for many years because the information content was very high for specific pathologic conditions (18–26). The major limitations were the low signal-to-noise ratio on sodium images and suboptimal instrumentation, which resulted in images with unacceptably low spatial resolution, even with long acquisition times of 30–45 minutes. These issues were resolved as high-quality sodium images with a spatial resolution of better than a 0.2-cm3 voxel volume have been obtained in patients in less than 10 minutes at 1.5 T, with even better resolution at 3.0 T (27).

With the use of a dual-tuned, dual-quadrature radio-frequency coil, sodium and proton imaging can be performed without moving the patient, thereby saving time and ensuring co-registration of images and maps (28). The twisted projection imaging acquisition scheme reduces signal losses from the very fast transverse relaxation of this quadripolar nucleus while minimizing the total acquisition time by optimizing the k-space trajectory (29–31). The quantitative transformation (32) of the sodium MR image to show tissue sodium concentration (TSC) is now established in the rodent model with the standard sodium 22 radionuclide biochemical technique (33). The use of TSC as a part of a comprehensive MR imaging protocol has been briefly presented on the basis of porcine (34) and human (35) studies.

The purpose of this study was to investigate the role of sodium MR imaging in a comprehensive MR imaging protocol for monitoring tissue viability in stroke. We compared TSC as measured at sodium MR imaging with apparent diffusion coefficient (ADC) and blood pool parameters derived from proton MR imaging in a nonhuman primate model of acute embolic stroke and in patients with acute or nonacute stroke. We suggest that sodium MR imaging provides a clinical measure of tissue viability that is complementary to the diagnostic roles of ADC and perfusion parameters in stroke.

One animal study was conducted to measure the natural evolution of sodium changes in acute stroke and involved no intervention. A second animal study was conducted under similar conditions as the first to investigate sodium changes in acute stroke before and after rt-PA thrombolysis.


   MATERIALS AND METHODS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 Appendix 1
 References
 
Animal Studies
The animal studies were performed by using a protocol approved by the animal subcommittee in this institution. Anesthesia was induced in the male primate (Macaca mulatta, 10–15 kg of body weight) with ketamine hydrochloride (Ketalar; Parke-Davis, Morris Plains, NJ; 22 mg/kg, intramuscular injection) and acepromazine maleate (Promace; Fort Dodge Animal Health, Overland Park, Kan; 0.2 mg/kg, intramuscular injection), along with atropine sulfate (Elkins-Sinn, Cherry Hill, NJ; 0.05 mg/kg, intramuscular injection). Anesthesia was maintained by means of titration with pentobarbital sodium (Nembutal; Abbott Laboratories, North Chicago, Ill; 5 mg/kg, intravenous infusion) to a level that prevented a pain response and blink reflex.

The animal was placed in a plastic cradle lined with warming blankets with a thermostat. The cradle allowed positioning of the head for fluoroscopic angiography and was used to maintain the head position within the MR unit. Physiologic monitoring of heart rate, rectal temperature, peripheral oxygen saturation, arterial blood gases, and arterial pressure was maintained throughout the procedure.

Angiographic sheaths (5 F; Cook, Bloomington, Ind) were placed bilaterally in the femoral arteries. A central venous line was placed via the right femoral vein into the inferior vena cava for the administration of normal saline (at body temperature) for fluid maintenance and for bolus administration of MR contrast material (0.1 mmol of gadoteridol [ProHance; Bristol-Myers Squibb, Princeton, NJ] per kilogram of body weight).

After systemic administration of heparin and under fluoroscopic guidance by means of standard neuroangiographic techniques, a catheter (5 F, Headhunter; Boston Scientific Vascular, Natick, Mass) was passed into the right internal carotid artery. The other femoral arterial line was used to monitor blood pressure. The catheters were continuously flushed with a warm heparin and saline solution to prevent thrombosis. Stroke was induced under fluoroscopic guidance with embolization of thrombogenic autologous clot (1 mL of blood taken prior to systemic anticoagulation) via the catheter in the right internal carotid artery followed by the administration of normal saline (5 mL). The occlusion of the right middle cerebral artery was confirmed immediately at conventional catheter angiography (ioversol [Optiray 320; Mallinckrodt Medical, St Louis, Mo]). MR imaging was performed by using the parameters described later, with the animal under general anesthesia and physiologically monitored.

For thrombolytic therapy in the second animal, rt-PA was administered intravenously by using the manufacturer-recommended dosage of a bolus (15 mg per kilogram of body weight) and then an infusion (0.75 mg/kg over 30 minutes followed by 0.5 mg/kg over 60 minutes).

Human Studies
Patients or their immediate family provided signed informed consent to participate in this study, which was approved by the institutional review board. Although the comprehensive MR imaging protocol now includes sodium imaging and proton conventional anatomic, diffusion, and perfusion imaging, not all parts of the protocol were completed in all patients during the early stages of protocol development.

The patients were referred over the course of 3 years by a wide range of physicians practicing within the medical center for clinical evaluation of stroke with diffusion and/or perfusion imaging. Standard MR imaging exclusion criteria were used and included findings of a cardiac pacer, aneurysm clip, cochlear implant, history of metal in eyes, claustrophobia, or body weight over 250 lbs (112.5 kg).

For the 68 patients who were evaluated at 81 research MR imaging examinations, the sex ratio was 1:1 and the mean age was 58.5 years ± 17 (SD) (age range, 23–80 years). The final diagnoses were acute stroke (n = 15, "acute" defined as <24 hours old), nonacute stroke (n = 41), transient ischemic attack (n = 6), stroke with chronic seizures (n = 1), and other categories not related to stroke (n = 5).

Of the 49 cases evaluated at sodium MR imaging, 14 cases were evaluated within 24 hours of the onset of neurologic symptoms, whereas the remaining 35 cases were evaluated at variable times after 24 hours (range, 1–210 days). In those patients (n = 42) in whom stroke was confirmed, 11 examinations were performed within 24 hours of event onset, and 31 studies were performed after 24 hours from event onset. The cases of nonacute stroke involved various regions of the cerebrum primarily, with only three cases involving the basal ganglia, one case involving the brainstem, and one case involving the cerebellum, with a mean lesion age of 54 days ± 91 (lesion age range, 1–210 days). In six cases, the subsequent diagnosis was transient ischemic attack, as no permanent lesions or clinical findings were found. Diffusion imaging was performed in 46 patients, of which 12 were in the acute group. Perfusion imaging was performed in 18 patients, of which four had stroke in the acute phase.

All studies were performed with continuous monitoring of heart rate and arterial oxygen saturation and with intermittent automatic measurement of blood pressure.

MR Imaging Instrumentation
Both 1.5- and 3.0-T whole-body MR imaging units (Signa, version 5.4.3; GE Medical Systems, Milwaukee, Wis) had broadband capabilities and were retrofitted with resonance gradients and software for echo-planar imaging (GE Medical Systems). The 3.0-T unit was a prototypic clinical 3.0-T unit, and use of it has been reported elsewhere (36–38). Hydrogen 1 images were acquired with the commercial head radio-frequency coil, whereas 23Na imaging was performed with a custom-designed, dual-quadrature, dual-tuned 23Na and 1H radio-frequency coil (28,39).

MR imaging in patients was performed with either the 1.5-T unit (56 examinations) or the 3.0-T unit (25 examinations). Both nonhuman primate studies were performed with a 3.0-T unit.

MR Imaging Protocol for Animal Studies
MR imaging was performed prior to induction of an embolus and repeatedly over the next 6 hours by using twisted projection sodium imaging, as used for human studies as described later. Diffusion imaging with the human protocol was also performed, repeatedly alternating with sodium imaging. Perfusion mapping was performed with the human protocol three times over 6 hours. For the animal in which the intervention (rt-PA) was made, the infusion was administered at 150 minutes after embolization. This process did not alter the imaging.

At the end of 6 hours, the animals were sacrificed by exsanguination under anesthesia, and the brains were removed and cut into coronal slices (5 mm thick) for histologic staining. The 2,3,5-triphenyltetrazolium chloride solution (2%, 10 g/500 mL of saline) was prepared fresh and was stored in the dark. The brain slices were added to the 2,3,5-triphenyltetrazolium chloride stain, covered, and placed in the dark for 30–45 minutes. After the slices were stained darkly, the solution was decanted and replaced with 10% neutral buffered formalin. The slices were digitally scanned in color with a flatbed scanner.

Comprehensive MR Imaging Protocol for Human Studies
For patient studies, the patient was always positioned comfortably and was instructed to remain stationary. Foam cushions on both sides of the head were used to help maintain lateral head support within the radio-frequency coil. Rigid fixation devices such as bite bars were not used for safety reasons. Each study was conducted by an experienced technologist who used preestablished protocols for the unit and was completed expeditiously in less than 45 minutes.

Sodium imaging, which was always performed first in the protocol, involved a twisted projection imaging sequence (previously described [27,40]) with a short echo time (approximately 0.4 msec; repetition time, 100 msec [100/0.4]) to yield images with a high signal-to-noise ratio (ratio > 20). This acquisition involved the use of an efficient k-space trajectory to acquire a three-dimensional data set in under 10 minutes at isotropic voxel volumes of 0.22 cm3 at 1.5 T and 0.064 cm3 at 3.0 T. Other acquisition parameters were as follows: number of projections, 1,596; number averages per projection, four.

The data were reconstructed into images by using a regridding algorithm with Fourier transformation described elsewhere (41). Standard saline solutions with different known sodium concentrations (40 and 80 mmol/L) were used within the field of view and allowed the TSC to be calibrated, as described elsewhere (33). The B1 inhomogeneity effects produced less than 10% variation in signal intensity across the field of view and were not corrected by means of B1 mapping, as reported elsewhere (42). The interpretation of TSC in terms of a two-compartment model is described in the Appendix and in the legend for Figure 1.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 1a. Diagrams of the two-compartment model of TSC. The vascular compartment is a part of the extracellular compartment, as the extracellular sodium concentration (in millimoles per liter) is always maintained because of the rapid diffusion of sodium ions from the vascular compartment that is buffered by the rest of the body. (a) In normal brain tissue, the large intracellular compartment maintains a low sodium concentration against a high sodium concentration in the small extracellular compartment. (b) During stroke, the normal intracellular space decreases and the extracellular compartment expands with loss of ion homeostasis as the cell integrity is lost. As the extracellular sodium concentration is maintained through the vascular compartment even at low levels of perfusion, TSC must increase, which is a direct measure of the loss of cell integrity. Percentages in parentheses are percentages of voxel volume.

 


View larger version (22K):
[in this window]
[in a new window]
  		Figure 1b. Diagrams of the two-compartment model of TSC. The vascular compartment is a part of the extracellular compartment, as the extracellular sodium concentration (in millimoles per liter) is always maintained because of the rapid diffusion of sodium ions from the vascular compartment that is buffered by the rest of the body. (a) In normal brain tissue, the large intracellular compartment maintains a low sodium concentration against a high sodium concentration in the small extracellular compartment. (b) During stroke, the normal intracellular space decreases and the extracellular compartment expands with loss of ion homeostasis as the cell integrity is lost. As the extracellular sodium concentration is maintained through the vascular compartment even at low levels of perfusion, TSC must increase, which is a direct measure of the loss of cell integrity. Percentages in parentheses are percentages of voxel volume.

 
Anatomic MR imaging consisted of sagittal T1-weighted spin-echo (1.5-T; 400/18) or gradient-echo (3.0-T; 500/8) images and T2-weighted fast spin-echo (1.5-T; 2,500/102) or dual spin-echo (3.0-T; 3,500/60) images.

Diffusion imaging (6,000/152) was performed as previously described (35) by using echo-planar imaging with 15 automatically incremented diffusion weightings along the frequency-encoding axis (x axis). The ADC maps were calculated as reported previously (35). In recent examinations (n = 7), diffusion trace imaging was used, in which the diffusion gradient was consecutively placed along all three orthogonal axes (x, y, z axes). The ADC trace map was calculated voxelwise from the reciprocal of the slope of the semilogarithmic plot of the signal intensity as a function of diffusion weighting by using the zero-diffusion gradient and the mean of the signal intensities from the diffusion gradient along the three orthogonal directions (x, y, z) (43).

Changes in ADC were assessed as a percentage ratio of the parameter in the lesion to the parameter in the contralateral side of the brain and by means of the numeric value of the ADC. The ADC values in normal gray matter, white matter, and cerebrospinal fluid were used as control values for comparison between different examinations and patients.

Perfusion was examined last in the protocol by using the dynamic susceptibility contrast material bolus–tracking technique with echo-planar imaging in which the signal intensity loss from the first passage of the intravenously administered contrast material bolus through the brain was measured to derive a relative cerebral blood volume (CBV), arrival time (TA), and tissue transit time (TTT) (35).

Changes in relative CBV, TA, and TTT were assessed as a percentage ratio of the parameter in the lesion to the parameter in the contralateral side of the brain. Also assessed were the absolute TA and TTT. The ratio of gray matter to white matter in normal tissue was used as a control value for comparison between different examinations and patients. The single-dose contrast material bolus (0.1 mmol/kg gadoteridol) was administered manually by means of an 18-gauge intravenous catheter and was recorded to vary between 3 and 5 seconds for all studies.

The temporal patterns in the multiple parameters in all patients were examined by using linear regression as a function of lesion age. Because of the large variation in time points, data were grouped by day up to 10 days, and then all older lesion data were used as a single group termed "chronic stroke."


   RESULTS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 Appendix 1
 References
 
Animal Studies
The time courses for changes in ADC and TSC are shown in Figures 2 and 3 for the untreated animal and treated animal, respectively. For the untreated animal, the ADC decreased rapidly to about half the normal values by the time the animal was returned to the unit after stroke induction and then stabilized with a slowly decreasing drift over the subsequent 6 hours (Fig 2a). In contrast, TSC showed little change initially, and after about 100 minutes it increased linearly (5.7 mmol/L/h) over the next 6 hours (Fig 2b).



View larger version (18K):
[in this window]
[in a new window]
  		Figure 2a. Plots of (a) ADC (in square millimeters per second) and (b) TSC (in millimoles per liter) for the untreated nonhuman primate in the stroke area ({block}) and contralateral normal region (•) as a function of time (6 hours) after embolization of autologous clot in the right middle cerebral artery. ADC and TSC in the stroke area were obtained from a region of interest over the lesion in the right hemisphere that had high signal intensity on the diffusion-weighted image. The normal region is the homologous region in the left hemisphere.

 


View larger version (17K):
[in this window]
[in a new window]
  		Figure 2b. Plots of (a) ADC (in square millimeters per second) and (b) TSC (in millimoles per liter) for the untreated nonhuman primate in the stroke area ({block}) and contralateral normal region (•) as a function of time (6 hours) after embolization of autologous clot in the right middle cerebral artery. ADC and TSC in the stroke area were obtained from a region of interest over the lesion in the right hemisphere that had high signal intensity on the diffusion-weighted image. The normal region is the homologous region in the left hemisphere.

 


View larger version (18K):
[in this window]
[in a new window]
  		Figure 3a. Plots of (a) ADC (in square millimeters per second) and (b) TSC (in millimoles per liter) for the rt-PA-treated nonhuman primate in the stroke area ({blacksquare}) and contralateral normal region (•) as a function of time (6 hours) after embolization of autologous clot in the right middle cerebral artery. ADC and TSC were obtained as in Thrombolysis with rt-PA was instituted 150 minutes after embolization, as indicated by the vertical line.

 


View larger version (16K):
[in this window]
[in a new window]
  		Figure 3b. Plots of (a) ADC (in square millimeters per second) and (b) TSC (in millimoles per liter) for the rt-PA-treated nonhuman primate in the stroke area ({blacksquare}) and contralateral normal region (•) as a function of time (6 hours) after embolization of autologous clot in the right middle cerebral artery. ADC and TSC were obtained as in Thrombolysis with rt-PA was instituted 150 minutes after embolization, as indicated by the vertical line.

 
For the treated animal, the ADC also decreased rapidly to about half the normal values after stroke induction and stabilized over the subsequent 6 hours (Fig 3a). The fluctuation in the value of ADC prior to instituting rt-PA therapy at 150 minutes appeared to be artifactual and related to head motion during this acquisition at the proton frequency. This variation was minimized when the region of interest was reduced in size so that the sampled region remained within the larger stroke region. Such head motion was visualized by examining the head position in consecutive time points of the 1H images. No correction for head position was performed over the time series. Thrombolysis did not alter the ADC time course. TSC showed little change initially, and after treatment at 150 minutes, when the TSC was less than 55 mmol/L, no further changes in TSC occurred over the remaining time (Fig 3b).

The blood pool parameters (Table 1) in this animal model of stroke showed changes smaller than those observed in patients, as the circulation time was shorter for the animals than for humans, but the imaging sampling interval (repetition time) remained at 1.5 seconds. In the region of stroke where the ADC was reduced by a factor of two, the relative CBV was not significantly different between the stroke and contralateral regions, but both TTT and TA were significantly prolonged, as expected for reduced blood flow due to obstruction from the embolic clot. Flow was never zero in this region, which indicated the presence of collateral circulation.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Blood Pool Perfusion and ADC Parameters
 
The increased variation in the relative CBV parameter in the region of stroke was not due to the curve-fitting procedure but was due to the variation in relative CBV across voxels in the region of interest. This variation was suggestive of heterogeneity in perfusion in the affected region. The physiologic condition of the animal was stable over the course of the experiment, as indicated by a stable mean blood pressure and other physiologic parameters (respiratory rate, temperature, arterial blood gas levels).

The histologic results from 2,3,5-triphenyltetrazolium chloride staining for the untreated and treated animals are shown in Figure 4. Multiple tissue slices in the untreated animal showed evidence (absence of pink staining) at histologic examination of a large, right middle cerebral arterial stroke involving extensive superficial cortex and structures in the deep gray matter in the caudate nucleus and putamen. In contrast, a single tissue slice in the animal treated at 150 minutes with rt-PA showed a minimal lesion in a small area confined to the superficial cortex.



View larger version (110K):
[in this window]
[in a new window]
  		Figure 4a. Digital prints of representative coronal brain slices after 2,3,5-triphenyltetrazolium chloride staining for the (a) untreated and (b) treated animals. Abnormal regions (arrows) did not stain with 2,3,5-triphenyltetrazolium chloride and appear brown.

 


View larger version (145K):
[in this window]
[in a new window]
  		Figure 4b. Digital prints of representative coronal brain slices after 2,3,5-triphenyltetrazolium chloride staining for the (a) untreated and (b) treated animals. Abnormal regions (arrows) did not stain with 2,3,5-triphenyltetrazolium chloride and appear brown.

 
Human Studies
The TSC parameter was evaluated in the setting of stroke along with the diffusion and perfusion parameters. Representative images and maps for a 62-year-old man with a 1-day-old stroke are presented in Figure 5 to demonstrate the results from the comprehensive MR imaging protocol. The right middle cerebral arterial infarction was clearly evident on anatomic images and on diffusion-weighted images and maps of TSC, ADC, TTT, TA, and relative CBV.



View larger version (90K):
[in this window]
[in a new window]
  		Figure 5. Representative MR images in a 62-year-old man who presented with left-sided paralysis that lasted 24 hours. A, T2-weighted spin-echo 1H image (2,500/102) through a stroke lesion (arrow). B, 23Na image (80/0.4). C, Diffusion-weighted 1H image (6,000/152). D, ADC map. E, Relative CBV map. F, TTT map. G, TA map. Parameter values for the lesion and the contralateral normal region, respectively, are as follows: TSC, 78 and 46 mmol/L; ADC, 0.21 x 10-3 and 0.64 x 10-3 mm2/sec; TTT, 37 and 14 seconds; and TA, 16 and 11 seconds. The relative CBV of the lesion was just less than half that of the contralateral homologous region.

 
The pattern of heterogeneity of bright and dark voxels reflected the wide range of the perfusion values. On the gray scale used to display these data, typical values of TTT and TA are both in the range of 8–12 seconds, which are shown as gray. When no substantial change in signal intensity occurred during the administration of a contrast material bolus, as for some voxels within the stroke area because of the absence or delayed arrival of a substantial contrast material bolus, the {gamma}-variate function used to model the perfusion-induced change in signal intensity could not fit the data (35). The default values for relative CBV, TTT, and TA for no substantial change in signal intensity were displayed as black. For vessels in which a large contrast material bolus caused large changes in signal intensity, the {gamma}-variate function gave a large relative CBV, which was displayed as white. In the area of stroke, the thrombus caused the bolus of contrast material to have a prolonged TTT and delayed TA. These were displayed as bright voxels.

The mixture of reduced and minimal-to-zero perfusion resulted in the observed heterogeneous pattern of the lesion in Figure 5. This figure demonstrates the application and interpretation of the comprehensive MR imaging protocol in an individual patient prior to considering the group data given later.

The temporal patterns of ADC for the entire patient group are shown in Figure 6 for the stroke region, homologous contralateral region, normal gray matter, normal white matter, and cerebrospinal fluid. For the regions that were not involved in the stroke (n = 32), linear regressions of the data (mean ± SD) showed no marked change over time as follows: ADC in gray matter, (1.12 ± 0.09) x 10-3 mm2/sec; ADC in white matter, (0.58 ± 0.06) x 10-3 mm2/sec; ADC in cerebrospinal fluid, (3.14 ± 0.43) x 10-3 mm2/sec; and ADC in phantom, (2.16 ± 0.16) x 10-3 mm2/sec.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 6. Plots of temporal pattern of ADC (in square millimeters per second) for patients with strokes of varying ages, with values for the lesion (•), contralateral region ({bigcirc}), normal gray matter ({bigtriangledown}), normal white matter ({bigtriangleup}), and cerebrospinal fluid ({lozenge}). All lesions older than 10 days were grouped as a single entity as no differences were found within this group (indicated by zigzag line). Range about the mean for the lesion is shown by dotted lines. Lesion values were connected without fitting because of heterogeneity in lesion size, location, and age. Long, dashed lines are linear regression plots for the contralateral region, gray matter, white matter, and cerebrospinal fluid, which show no significant change over time.

 
The ADC value of the stroke region (number of MR examinations = 25) was (0.38 ± 0.14) x 10-3 mm2/sec and remained below normal values for up to 5 days, had normal values at 7 days, and had values that were greater than normal (n = 8) of (1.98 ± 0.63) x 10-3 mm2/sec after 10 days.

The temporal patterns of the perfusion parameters (relative CBV, TA, and TTT), determined as the percentage ratio of the parameter in the stroke region to the parameter in its homologous contralateral region and the ratio of the parameter in normal white matter to the parameter in normal gray matter, are shown in Figure 7. The parametric percentage ratios of normal white matter to gray matter remained constant for both the acute (n = 2) and the nonacute (n = 8) phases, respectively, as follows: relative CBV, 50% ± 2 and 49% ± 6; TTT, 126% ± 40 and 96% ± 11; and TA, 98% ± 2 and 98% ± 9.



View larger version (21K):
[in this window]
[in a new window]
  		Figure 7. Plots of temporal pattern of blood pool perfusion parameters for patients with strokes of varying ages. Top, Relative CBV; middle, TA; and bottom, TTT are expressed as the mean percentage ratio of the parameter in the lesion to the parameter in the contralateral side. All lesions older than 10 days were grouped as a single entity, as no differences were found within this group (indicated by zigzag line). Dotted lines show the range about the mean of the ratio. Top, plot shows values for the ratio of relative CBV in the lesion to that in the contralateral region (•) and a baseline at 100% from which it deviates. Ratios of relative CBV in gray matter to that in white matter ({Delta}) were fitted as a linear regression line (dashed line) to show that there was no significant change over time (ie, temporal stability) from the expected 50% for normal tissue. TA for the lesion (• in middle part) shows little deviation from normal, which indicates normal cardiac output. TTT for the lesion (• in bottom part) shows variation over only the first 2-3 days, which suggests that occlusions leading to stroke usually resolve spontaneously within this time. {circ} in middle and bottom parts indicates ratios of percentage TA and TTT for gray matter to that of percentage ratio TA and TTT for white matter, which were essentially constant over time.

 
As expected (35), the relative blood volume in the white matter was half that in gray matter. The TTT of normal white matter (n = 2) and normal gray matter (n = 2) measured acutely were not remarkably different from each other (13.3 seconds ± 1.3 and 10.9 seconds ± 2.5, respectively) or from the nonacute values in white matter (n = 8) and gray matter (n = 8) (10.1 seconds ± 2.2 and 10.7 seconds ± 2.4, respectively).

Similarly, the TA of normal white matter (n = 2) and normal gray matter (n = 2) measured acutely were not markedly different from each other (11.1 seconds ± 0.8 and 11.4 seconds ± 0.5, respectively) or from the nonacute values of white matter (n = 8) and gray matter (n = 8) (10.3 seconds ± 2.7 and 10.3 seconds ± 2.0, respectively).

The stroke region (n = 2) showed a percentage ratio (lesion to contralateral region) for relative CBV of 57% ± 22 that was measured acutely and that was markedly less than 100%, as did older lesions (n = 8), with a relative CBV ratio (lesion to contralateral) of 54% ± 27. Not only were the blood volumes decreased in acute lesions (n = 2), but also the TTT of 34 seconds ± 4 and the percentage ratio of 290% ± 39 were notably prolonged, whereas the TA of 11 seconds ± 7 and the percentage ratio of 100% ± 57 were more variable but did not change notably in this small number of acute cases.

Although the older strokes showed a lower relative CBV than did the contralateral control regions, as described previously, the following values normalized with time: TTT (lesion, n = 8), 11 seconds ± 2, with a percentage ratio of 119% ± 21, and TA (lesion, n = 8), 11 seconds ± 2, with a percentage ratio of 106% ± 8.

Although the lesions were heterogeneous in size, location, and age, and in terms of the patient's age, the temporal patterns of the value of TSC and the percentage increase in TSC in the lesion compared with those in the contralateral region for all patients are presented in Figure 8. The mean TSC values of the stroke lesion (n = 26) and the contralateral region (n = 26) were notably different and were 81 mmol/L ± 29 and 37 mmol/L ± 6, respectively.



View larger version (14K):
[in this window]
[in a new window]
  		Figure 8a. Plots of temporal patterns of (a) mean TSC (in millimoles per liter) in the lesion (•) and normal contralateral tissue ({circ}) for all cases and of (b) percentage difference in TSC between stroke region and contralateral region (•). All lesions older than 10 days were grouped as a single entity, as no differences were found within this group (indicated by zigzag line in a and b). Range of TSC in lesions (dotted lines in a) is shown to reflect the biologic variation across subjects rather than the accuracy of the measurement in a single patient. This biologic variation is represented in the scatter of points in b, but all values were greater than 25% (dashed-and-dotted line in b) of normal TSC. (c) Plots of temporal patterns of TSC in acute cases (<24 hours old) for lesions (•) and contralateral normal tissue ({circ}). Line at TSC equal to 70 mmol/L (dashed-and-dotted line) was based on the results from the nonhuman primate and is shown to represent infarction. TSC values for the normal tissue were fitted with a linear regression line (dashed line in a and c) to show no significant change over time. Error bars in a represent the SD of the group data at each time point. As the points in c represent single patients, no error bars are shown.

 


View larger version (11K):
[in this window]
[in a new window]
  		Figure 8b. Plots of temporal patterns of (a) mean TSC (in millimoles per liter) in the lesion (•) and normal contralateral tissue ({circ}) for all cases and of (b) percentage difference in TSC between stroke region and contralateral region (•). All lesions older than 10 days were grouped as a single entity, as no differences were found within this group (indicated by zigzag line in a and b). Range of TSC in lesions (dotted lines in a) is shown to reflect the biologic variation across subjects rather than the accuracy of the measurement in a single patient. This biologic variation is represented in the scatter of points in b, but all values were greater than 25% (dashed-and-dotted line in b) of normal TSC. (c) Plots of temporal patterns of TSC in acute cases (<24 hours old) for lesions (•) and contralateral normal tissue ({circ}). Line at TSC equal to 70 mmol/L (dashed-and-dotted line) was based on the results from the nonhuman primate and is shown to represent infarction. TSC values for the normal tissue were fitted with a linear regression line (dashed line in a and c) to show no significant change over time. Error bars in a represent the SD of the group data at each time point. As the points in c represent single patients, no error bars are shown.

 


View larger version (12K):
[in this window]
[in a new window]
  		Figure 8c. Plots of temporal patterns of (a) mean TSC (in millimoles per liter) in the lesion (•) and normal contralateral tissue ({circ}) for all cases and of (b) percentage difference in TSC between stroke region and contralateral region (•). All lesions older than 10 days were grouped as a single entity, as no differences were found within this group (indicated by zigzag line in a and b). Range of TSC in lesions (dotted lines in a) is shown to reflect the biologic variation across subjects rather than the accuracy of the measurement in a single patient. This biologic variation is represented in the scatter of points in b, but all values were greater than 25% (dashed-and-dotted line in b) of normal TSC. (c) Plots of temporal patterns of TSC in acute cases (<24 hours old) for lesions (•) and contralateral normal tissue ({circ}). Line at TSC equal to 70 mmol/L (dashed-and-dotted line) was based on the results from the nonhuman primate and is shown to represent infarction. TSC values for the normal tissue were fitted with a linear regression line (dashed line in a and c) to show no significant change over time. Error bars in a represent the SD of the group data at each time point. As the points in c represent single patients, no error bars are shown.

 
Although the range of variation in lesion TSC was wide, all lesions in all patients showed at least a 25% increase in TSC (Fig 8b). In the patients with noncerebral nonacute strokes, the TSC values, 50 mmol/L ± 2 and 31 mmol/L ± 8, showed the same trends for stroke (n = 3) and for normal tissue (n = 3), respectively, although the values were slightly lower than those for cerebral nonacute stroke, which reflects the different locations and contributions of gray and white matter.

Of the 11 sodium MR examinations for acute stroke, eight were for cerebral strokes, which were imaged at a mean of 15 hours ± 8 after symptom onset, and three were for strokes of the brainstem (n = 2) or basal ganglia (n = 1), which were imaged at a mean of 8 hours ± 1 after symptom onset. The patients' ages differed between these two groups; the cerebral strokes occurred in a group with a mean age of 49 years ± 18, and the brainstem or basal ganglial strokes occurred in a group with a mean age of 80 years ± 2.

The cerebral hemispheric strokes (n = 8), which were imaged later than the other brainstem lesions, showed an elevated mean TSC value of 76 mmol/L ± 8 and a percentage ratio of 164% ± 30 over the mean TSC of the contralateral tissue (n = 8) of 45 mmol/L ± 5. The brainstem and basal ganglial strokes (n = 3) that were imaged earlier showed no markedly elevated values of TSC (42 mmol/L ± 7) compared with the contralateral areas, which possibly reflected the earlier time but also the different locations of the strokes and ages of patients.

From Equation (A1) in the Appendix, a normal TSC of 40 mmol/L corresponded to a cell volume fraction of about 0.8, which was in agreement with the latest measurements reported in the literature (44). The mean increase in TSC in confirmed nonacute stroke to about 80 mmol/L corresponded to a decrease in cell volume fraction to about 0.5. This decrease corresponded to substantial tissue loss within the infarcted tissue, but this may still have been a conservative estimate because of partial volume effects with surrounding tissue and because of increases in extracellular matrix in scar tissue.

Table 2 presents the technical success rates for diffusion, perfusion, or sodium imaging in patients with acute or nonacute strokes. The technical success rate can be defined as the ratio of the number of adequate images or maps to the total number of acquisitions made. Technical failures were most often due to patient motion during acquisitions that required multiple images to be obtained over time. The perfusion and diffusion studies were greatly degraded by motion to the extent that the respective maps could not be calculated. No attempt was made to correct the data for head motion. No technical failures were the result of equipment failure or inadequate contrast material bolus for the perfusion studies. The important observation was the low technical success rate of perfusion in the acute setting at 1.5 T compared with that at 3.0 T. TSC measurements were robust under all conditions.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Technical Success Rates for Diffusion, Perfusion, and Sodium MR Imaging
 

   DISCUSSION
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
Appendix 1
 References
 
Conventional 1H MR imaging is insensitive for the detection of acute stroke (45–47). Diffusion-weighted imaging and its quantitative counterpart of ADC mapping are very sensitive in the early detection of ischemia. A considerable body of literature exists on the use of this approach (48–54). As ADC and diffusion-weighted imaging show changes well before the thresholds of irreversible tissue injury related to water and ion homeostasis are reached, ADC and diffusion-weighted imaging are sensitive for detection of ischemia but do not provide a complete description of pathophysiologic conditions due to stroke in the primate brain. This was demonstrated in the monkey model in which the ADC remained constant over 6 hours, although thrombolytic therapy reduced the volume of the infarction.

Perfusion imaging with dynamic susceptibility contrast agents (55–68) has been shown to be a powerful tool in the evaluation of acute stroke in animals (58–61,68) and patients (62,66). The technique is semiquantitative, but the ease of intravenous administration of contrast material and the use of echo-planar technology (63–67) for monitoring the first pass of the contrast material bolus through the brain readily allows clinical application. However, the question of using thrombolytic therapy still requires a knowledge of whether the region identified with diffusion mapping as being ischemic contains viable tissue to salvage. The presence of a vascular occlusion does not address this issue.

For any particular set of MR images to be useful in a clinical setting, particularly in a medically urgent situation such as acute stroke, the technical success rate for yielding useful information must be considered. The major cause of technically inadequate perfusion data was patient motion despite an acquisition period of only 81 seconds. Although diffusion imaging takes longer, the use of a single exponential rather than a {gamma}-variate fitting procedure makes it more robust. The acute cases were examined primarily at 1.5 T, whereas the nonacute cases were examined at 1.5 or 3.0 T. The success rate of perfusion imaging at 3.0 T was double that at 1.5 T, probably reflecting the greater signal intensity change from the contrast material bolus that was associated with the enhanced magnetic susceptibility effect at a higher field strength.

Sodium MR imaging has long been proposed as a useful method to assess pathologic changes in the brain (26), but medical applications have not developed because of the lack of an acceptable clinical implementation. The potential benefits of this method in stroke have been overshadowed by the newer proton-based methods of diffusion and perfusion imaging discussed above. Using both 1.5-T and 3.0-T functional imaging units (35), we previously demonstrated that sodium imaging can be performed quantitatively (33) and with high sensitivity (39) and high spatial resolution (27,40). The findings in the current study demonstrate that sodium imaging can be combined efficiently with the proton-based methods of diffusion and perfusion imaging for the clinical evaluation of patients with acute stroke.

Animal models used to study the pathophysiologic conditions due to stroke (68) have shown that the rate of progression of infarction was species dependent; the time course was 3–4 hours in rodents as compared with 6–8 hours in nonhuman primates (12,68,69). Similarly, the threshold blood flow at which metabolic and physiologic changes occurred was species dependent (68,69). Hence, the calibration of parametric maps for clinical applications must be performed in nonhuman primate models. The nonhuman primate experiments of induced embolic stroke demonstrated the natural course of changes in TSC over time without and with thrombolytic intervention.

Whereas the ADC parameter was useful in detecting the stroke very rapidly after embolization, it was less informative over the critical first 6 hours, as it remained relatively stable. In marked contrast, TSC showed a linear increase over 6 hours characterized by a rate of increase of about 5.7 mmol/L/h and approached 70 mmol/L at 6 hours, by which time the tissue was shown at histologic examination to be infarcted.

The progression of stroke to infarction is dependent on both perfusion level and duration of reduced flow. The steadily increasing TSC level marked the changing metabolic status of the lesion. As shown in the treated animal, thrombolytic therapy after 150 minutes when TSC was still below 55 mmol/L markedly decreased the volume of the infarct as compared with the result with no intervention. These results, although limited to two animals because of cost restrictions, provide the basis for the interpretation of the results in humans.

The clinical examinations in which the proton and sodium acquisitions were combined as a single protocol were performed in a clinically acceptable period, with the acute cases completed in less than 45 minutes. Beyond 24 hours, patients in this group had an elevated mean TSC (>70 mmol/L), as was found in the animal model, with all of the expected findings of infarction at imaging and on the basis of clinical criteria.

As there are variations in TSC with location of stroke, another way of expressing these data is as a percentage of signal intensity change from the contralateral region. As shown in Figure 8b, all stroke patients had at least a 25% increase in the stroke region and often had a much greater change. Only three of 33 patients had this minimum value. This minimum threshold corresponded to a TSC range of 46–56 mmol/L, based on a range of normal TSC values of 37–45 mmol/L. The TSC values in the cases of earlier acute stroke showed more heterogeneity and lower values at shorter times, which is consistent with a changing physiologic condition during stroke evolution.

Closely spaced follow-up studies during the acute phase of stroke were limited in these sick patients. The single patient who underwent studies at 5 and 76 hours showed a TSC of 42 mmol/L at 5 hours that increased to above 70 mmol/L at 76 hours and then stabilized at the follow-up examination at 6 months. The question of whether there is a critical threshold of TSC above which progression to infarction is inevitable remains unanswered. The definition of such a threshold requires more extensive animal studies.

The comprehensive sodium and proton MR imaging protocol demonstrates that the diagnosis, perfusion status, and tissue viability can be assessed in a single, rapid study. The gradual increase in TSC for an evolving stroke diagnosed at diffusion imaging allows infarcted tissue to be identified (TSC > 70 mmol/L) irrespective of the time of onset of neurologic symptoms. As the 3-hour window for thrombolytic therapy is established to avoid the hemorrhagic complications from reperfusion of infarcted tissue, such nonviable infarcted tissue can be identified immediately. If a perfusion abnormality exists, reperfusion may be considered only if the tissue is not infarcted. In settings in which collateral circulation is minimal, acute occlusion of major vessels may even produce extensive infarction within 3 hours. Such cases can be identified with TSC.

As new neuroprotective agents enter clinical practice as alternatives to thrombolysis, triaging patients to undergo appropriate therapies may be helped by this comprehensive MR imaging assessment of the tissue physiologic condition and viability.


   Appendix 1
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 Appendix 1
References
 
The model of TSC for assessing hyperacute stroke physiology is described as follows. Figure 1 shows the scheme that summarizes the relationship between TSC and loss of tissue viability for a two-compartment model in which the vascular and extracellular compartments are considered as one compartment. For an intracellular compartment volume Vi and sodium concentration Ci and an extracellular compartment volume Ve, and sodium concentration Ce, the conservation of mass gives the following: TSC(Vi + Ve) = VeCe + ViCi.

For a voxel of volume V, V = Vi + Ve or Ve = V - Vi, which allows TSC to be expressed in terms of the intracellular volume fraction Vi/V as follows:

Thus, for a fixed volume fraction, increasing intracellular sodium concentration Ci increases TSC. Alternatively, as the intracellular volume fraction decreases with loss of cell integrity, TSC approaches Ce. It is important to realize this model is not just a redistribution of sodium within a voxel. The extracellular compartment is buffered with even minimal perfusion and the high diffusibility of sodium ions from the space of the entire body, which is regulated systemically at a constant value of Ce.


   Acknowledgments
 
The authors thank the physicians for referring their patients for these studies and gratefully acknowledge Judy Cameron, PhD, and the University of Pittsburgh Primate Research Laboratory for the assistance with the animal experiments.


   Footnotes
 
Abbreviations: ADC = apparent diffusion coefficient CBV = cerebral blood volume rt-PA = recombinant tissue plasminogen activator TA = arrival time TSC = tissue sodium concentration TTT = tissue transit time

Author contributions: Guarantor of integrity of entire study, K.R.T.; study concepts, K.R.T.; definition of intellectual content, K.R.T.; literature research, K.R.T.; clinical studies, K.R.T., D.D., P.E.; experimental studies, K.R.T., D.D., P.E.; data acquisition, K.R.T., D.D., P.E.; data analysis, K.R.T., T.S.G.; statistical analysis, K.R.T., T.S.G.; manuscript preparation and editing, K.R.T., T.S.G.; manuscript review, K.R.T., T.S.G., D.D.


   References
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 Appendix 1
 References
 

  1. Bonita R. Epidemiology of stroke. Lancet 1992; 339:342-344.[Medline]
  2. Broderick JP, Phillips SJ, Whisnant JP, O'Fallon WM, Bergstralh EJ. Incidence rates of stroke in the eighties: the end of the decline in stroke?. Stroke 1989; 20:577-582.[Abstract/Free Full Text]
  3. Malmgren R, Bamford J, Warlow C, Sandercock P, Slattery J. Projecting the number of patients with first ever strokes and patients newly handicapped by stroke in England and Wales. BMJ 1989; 298:656-660.
  4. Brown RD, Whisnant JP, Sicks JD, O'Fallon WM, Wiebers DO. Stroke incidence, prevalence, and survival: secular trends in Rochester, Minnesota, through 1989. Stroke 1996; 27:373-380.
  5. Barinaga M. Finding new drugs to treat stroke. Science 1996; 272:664-666.
  6. Marder VJ, Sherry S. Thrombolytic therapy. I. Current status. N Engl J Med 1988; 318:1512-1520.[Medline]
  7. Trouillas P, Nighoghossian N, Getenet JC, et al. Open trial of intravenous tissue plasminogen activator in acute carotid territory stroke: correlations of outcome with clinical and radiological data. Stroke 1996; 27:882-890.[Abstract/Free Full Text]
  8. The National Institutes of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med 1995; 333:1581-1587.[Abstract/Free Full Text]
  9. Trust Study Group. Randomised, double-blind, placebo-controlled trial of nimodipine in acute stroke. Lancet 1990; 336:1205-1209.[Medline]
  10. Barnett HJ. The contribution of multicenter trials to stroke prevention and treatment. Arch Neurol 1990; 47:441-444.[Medline]
  11. Marsh EE, III, Adams HP, Jr, Biller J, et al. Use of antithrombotic drugs in the treatment of acute ischemic stroke: a survey of neurologists in practice in the United States. Neurology 1989; 39:1631-1634.[Abstract/Free Full Text]
  12. Hossmann KA. Viability thresholds and the penumbra of focal ischemia. Ann Neurol 1994; 36:557-565.[Medline]
  13. Ginsberg MD, Pulsinelli WA. The ischemic penumbra, injury thresholds and the therapeutic window for acute stroke. Ann Neurol 1994; 36:553-554.[Medline]
  14. Astrup J, Siesjo BK, Symon L. Thresholds in cerebral ischemia: the ischemic penumbra. Stroke 1981; 12:723-725.[Free Full Text]
  15. DeGirolami U, Crowell RM, Marcoux FW. Selective necrosis and total necrosis in focal cerebral ischemia: neuropathologic observations on experimental middle cerebral artery occlusion in the macaque monkey. J Neuropathol Exp Neurol 1984; 43:57-71.[Medline]
  16. Hossmann KA, Schuier FJ. Experimental brain infarcts in cats. I. Pathophysiological observations. Stroke 1980; 11:583-592.
  17. Pulsinelli W. Pathophysiology of acute ischeamic stroke. Lancet 1992; 339:533-536.[Medline]
  18. Berendsen HJC, Edzes HT. The observation and general interpretation of sodium magnetic resonance in biological material. Ann N Y Acad Sci 1973; 204:459-485.[Medline]
  19. Hilal SK, Maudsley AA, Simon HE, et al. In vivo NMR imaging of tissue sodium in the intact cat before and after acute cerebral stroke. AJNR 1983; 4:245-249.[Abstract]
  20. Hilal SK, Maudsley AA, Ra JB, et al. In vivo NMR imaging of sodium-23 in the human head. J Comput Assist Tomogr 1985; 9:1-7.[Medline]
  21. Perman WH, Turski PA, Houston LW, Glover GH, Hayes CE. Methodology of in vivo human sodium imaging at 1.5 T. Radiology 1986; 160:811-820.[Abstract/Free Full Text]
  22. Moonen CT, Anderson SE, Unger S. 23Na rotating frame imaging in the perfused rabbit heart using separate transmitter and receiver coils. Magn Reson Med 1987; 5:296-301.[Medline]
  23. Ra JB, Hilal SK, Oh CH, Mun IK. In vivo magnetic resonance imaging of sodium in the human body. Magn Reson Med 1988; 7:11-22.[Medline]
  24. Perman WH, Thomasson DM, Bernstein MA, Turski PA. Multiple short-echo (2.5-ms) quantitation of in vivo sodium T2 relaxation. Magn Reson Med 1989; 9:153-160.[Medline]
  25. Winkler SS, Thomasson DM, Sherwood K, Perman WH. Regional T2 and sodium concentration estimates in the normal human brain by sodium-23 MR imaging at 1.5 T. J Comput Assist Tomogr 1989; 13:561-566.[Medline]
  26. Hilal SK, Ra JB, Oh CH, Mun IK, Einstein SG, Roschmann P. In: Stark DD, Bradley WG, eds. Magnetic resonance imaging. St Louis, Mo: Mosby, 1990; 715-731.
  27. Boada FB, Gillen JS, Shen GX, Chang SY, Thulborn KR. Fast three dimensional sodium imaging. Magn Reson Med 1997; 37:706-715.[Medline]
  28. Shen GX, Boada FE, Thulborn KR. Dual-frequency, dual-quadrature, birdcage RF coil design with identical B1 pattern for sodium and proton imaging of the human brain at 1.5 T. Magn Reson Med 1997; 38:717-725.[Medline]
  29. Rooney WD, Springer CS, Jr. A comprehensive approach to the analysis and interpretation of the resonances of spins 3/2 from living systems. NMR Biomed 1991; 4:209-226.[Medline]
  30. Rooney WD, Springer CS, Jr. The molecular environment of intracellular sodium: 23Na NMR relaxation. NMR Biomed 1991; 4:227-245.[Medline]
  31. Boada FE, Christensen JD, Huang-Hellinger FR, Reese TG, Thulborn KR. Quantitative in vivo sodium concentration maps: the effects of biexponential relaxation. Magn Reson Med 1994; 32:219-223.[Medline]
  32. Thulborn KR, Ackerman JH. Absolute molar concentrations by NMR in inhomogeneous B1, a scheme for analysis of in vivo metabolites. J Magn Reson 1983; 55:357-371.
  33. Christensen JD, Barrère BJ, Boada FE, Vevea JM, Thulborn KR. Quantitative tissue sodium concentration mapping of normal rat brain. Magn Reson Med 1996; 36:83-89.[Medline]
  34. Hees P, Christensen JD, Boada FE, Thulborn KR. Assessment of stroke by functional MRI techniques (abstr) In: Proceedings of 12th Annual Scientific Meeting of the Society of Magnetic Resonance in Medicine. Berkeley, Calif: Society of Magnetic Resonance in Medicine, 1993; 1399.
  35. Thulborn KR, Uttecht S, Betancourt C, Talagala SL, Boada FE, Shen GX. A functional, physiological and metabolic toolbox for clinical magnetic resonance imaging: integration of acquisition and analysis strategies. Int J Imaging Syst Technol 1997; 8:572-581.
  36. Thulborn KR, Davis D, Erb P, Strojwas M, Sweeney JA. Clinical fMRI: implementation and experience. Neuroimage 1996; 4(suppl):101-107.
  37. Thulborn KR, Voyvodic J, Chang S, Strojwas M, Sweeney JA. New approaches to cognitive function by high field functional MRI. In: Yuasa T, Prichard JW, Ogawa F, eds. Current progress in functional brain mapping. London, England: Smith-Gordon, 1998; 15-23.
  38. Thulborn KR, Gillen JS, McCurtain B, Betancourt C, Sweeney JA. Functional magnetic resonance imaging of the human brain. Bull Magn Reson 1996; 18:37-42.
  39. Shen GX, Wu JF, Boada FE, Thulborn KR. Experimentally verified theoretical design of dual-tuned, low-pass birdcage radiofrequency resonators for magnetic resonance imaging and magnetic resonance spectroscopy of human brain at 3.0 Tesla. Magn Reson Med 1999; 41:268-275.[Medline]
  40. Boada FE, Shen GX, Chang SY, Thulborn KR. Spectrally weighted twisted projection imaging: reducing T2 signal attenuation effects in fast three-dimensional sodium imaging. Magn Reson Med 1997; 38:1022-1028.[Medline]
  41. Chesler D, Vevea JM, Boada FE, Thulborn KR. Rapid 3-D reconstruction from 1-D projections for metabolic MR imaging of short T2 species (abstr) In: Book of abstracts: Society of Magnetic Resonance in Medicine 1992. Berkeley, Calif: Society of Magnetic Resonance in Medicine, 1992; 665.
  42. Thulborn KR, Boada FE, Shen GX, Christensen JD, Reese TG. Correction of B1 inhomogeneities using echo-planar imaging of water. Magn Reson Med 1998; 39:369-375.[Medline]
  43. Beauchamp NJ, Jr, Ulug AM, Passe TJ, van Zijl PC. MR diffusion imaging in stroke: review and controversies. RadioGraphics 1998; 18:1269-1283.[Abstract]
  44. Nicholson C, Sykova E. Extracellular space structure revealed by diffusion analysis. Trends Neurosci 1998; 21:207-215.[Medline]
  45. Kertesz A, Black SE, Nicholson L, Carr T. The sensitivity and specificity of MRI in stroke. Neurology 1987; 37:1580-1585.[Abstract/Free Full Text]
  46. Bose B, Jones SC, Lorig R, Friel HT, Weinstein M, Little JR. Evolving focal cerebral ischemia in cats: a spatial correlation of nuclear magnetic resonance imaging, cerebral blood flow, tetrazolium staining and histopathology. Stroke 1988; 19:28-37.[Abstract/Free Full Text]
  47. Yuh WT, Crain MR, Loes DJ, Greene GM, Ryals TJ, Sato Y. MR imaging of cerebral ischemia: findings in the first 24 hours. AJNR 1991; 12:621-629.[Abstract]
  48. Back T, Hoehn-Berlage M, Kohno K, Hossmann KA. Diffusion nuclear magnetic resonance imaging in experimental stroke: correlation with cerebral metabolites. Stroke 1994; 25:494-500.[Abstract]
  49. Moseley ME, Kucharczyk J, Mintorovitch J, et al. Diffusion-weighted MR imaging of acute stroke: correlation with T2-weighted and magnetic susceptibility-enhanced MR imaging in cats. AJNR 1990; 11:423-429.[Abstract]
  50. Moseley ME, Cohen Y, Mintorovitch J, et al. Early detection of regional cerebral ischemia in cats: comparison of diffusion- and T2-weighted MRI and spectroscopy. Magn Reson Med 1990; 14:330-346.[Medline]
  51. Chien D, Buxton RB, Kwong KK, Brady TJ, Rosen BR. MR diffusion imaging of the human brain. J Comput Assist Tomogr 1990; 14:514-520.[Medline]
  52. Knight RA, Ordidge RJ, Helpern JA, Chopp M, Rodolosi LC, Peck D. Temporal evolution of ischemic damage in rat brain measured by proton nuclear magnetic resonance imaging. Stroke 1991; 22:802-808.[Abstract/Free Full Text]
  53. Minematsu K, Li L, Fisher M, Sotak CH, Davis MA, Fiandaca MS. Diffusion-weighted magnetic resonance imaging: rapid and quantitative detection of focal brain ischemia. Neurology 1992; 42:235-240.[Abstract/Free Full Text]
  54. de Crespigny AJ, Tsuura M, Moseley ME, Kucharczyk J. Perfusion and diffusion MR imaging of thromboembolic stroke. JMRI 1993; 3:746-754.
  55. Warach S, Wielopolski P, Edelman R. Identification and characterization of the ischemic penumbra of acute human stroke using echo planar diffusion and perfusion imaging (abstr) ; Proceedings of the 12th Annual Scientific Meeting of the Society of Magnetic Resonance in Medicine Berkeley, Calif: Society of Magnetic Resonance in Medicine 1993: 249.
  56. Villringer A, Rosen BR, Belliveau JW, et al. Dynamic imaging with lanthanide chelates in normal brain: contrast due to magnetic susceptibility effects. Magn Reson Med 1988; 6:164-174.[Medline]
  57. Rosen BR, Belliveau JW, Chien D. Perfusion imaging by nuclear magnetic resonance. Magn Reson Q 1