From: Subject: Can Proton MR Spectroscopic and Perfusion Imaging Differentiate Between Neoplastic and Nonneoplastic Brain Lesions in Adults? -- Hourani et al. 29 (2): 366 -- American Journal of Neuroradiology Date: Mon, 28 Apr 2008 10:22:03 +0200 MIME-Version: 1.0 Content-Type: multipart/related; type="text/html"; boundary="----=_NextPart_000_0000_01C8A919.B3CD9390" X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3198 This is a multi-part message in MIME format. ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable Content-Location: http://www.ajnr.org/cgi/content/full/29/2/366 Can Proton MR Spectroscopic and Perfusion Imaging = Differentiate Between Neoplastic and Nonneoplastic Brain Lesions in = Adults? -- Hourani et al. 29 (2): 366 -- American Journal of = Neuroradiology

Advanced
AJNR - American Journal of=20 Neuroradiology

Published=20 ahead of print on November 30, 2007
doi: 10.3174/ajnr.A0810 =
This = Article
3D"
3D"Right Abstract=20 3D"Freely
3D"Right Figures=20 Only
3D"Right Full=20 Text (PDF)
3D"Right All Versions of this=20 Article:
ajnr.A0810v1<= /A>=20
29/2/366    most=20 recent
3D"Right Alert=20 me when this article is cited
3D"Right Alert=20 me if a correction is posted
3D"Right Citation Map
3D"
Services
3D"
3D"Right Similar=20 articles in this journal
3D"Right Similar=20 articles in PubMed
3D"Right Alert me to new = issues of=20 the journal
3D"Right Download to=20 citation manager
3D"Right 3D"reprints
3D"
Citing = Articles
3D"
3D"Right Citing Articles via CrossRef =
3D"
Google = Scholar
3D"
3D"Right Articles by Hourani, R. =
3D"Right Articles by Horsk=E1, A. =
3D"
PubMed
3D"
3D"Right PubMed=20 Citation
3D"Right Articles by Hourani, R. =
3D"Right Articles by Horsk=E1, A.=20

American Journal of Neuroradiology = 29:366-372,=20 February 2008
=A9 2008 American=20 Society of Neuroradiology

BRAIN

Can Proton MR Spectroscopic and Perfusion Imaging Differentiate = Between=20 Neoplastic and Nonneoplastic Brain Lesions in=20 Adults?

R. = Hourania,d,=20 L.J. Brantc, T. Rizke,=20 J.D. Weingartb, P.B.=20 Barkera,f and A. = Horsk=E1a=20

a Russell H. Morgan Department of = Radiology=20 and Radiological Science, Johns Hopkins Hospital, Baltimore, = Md
b=20 Department of Neurological Surgery, Johns Hopkins Hospital, Baltimore,=20 Md
c National Institutes of Health/National Institute on = Aging,=20 Gerontology Research Center, Baltimore, Md
d Department of = Radiology, American University of Beirut Medical Center, Beirut,=20 Lebanon
e Department of Neurosurgery, Hotel Dieu Hospital, = Beirut,=20 Lebanon
f FM Kirby Research Center, Kennedy Krieger = Institute,=20 Baltimore, Md

Please address correspondence to Alena = Horsk=E1, PhD,=20 Russell H. Morgan Department of Radiology and Radiological Science, = Johns=20 Hopkins University, 217 Traylor Bldg, 720 Rutland Ave, Baltimore, MD = 21205.=20 E-mail: ahorska{at}jhmi.edu

BACKGROUND AND PURPOSE: Noninvasive diagnosis of brain = lesions=20 is important for the correct choice of treatment. Our aims = were=20 to investigate whether 1) proton MR spectroscopic imaging=20 (1H-MRSI) can aid in differentiating between = tumors and=20 nonneoplastic brain lesions, and 2) perfusion MR imaging can = improve=20 the classification.

MATERIALS AND METHODS: We retrospectively examined 69 = adults=20 with untreated primary brain lesions (brain tumors, n =3D = 36;=20 benign lesions, n =3D 10; stroke, n =3D 4; = demyelination,=20 n =3D 10; and stable lesions not confirmed on = pathologic=20 examination, n =3D 9). MR imaging and = 1H-MRSI were=20 performed at 1.5T before biopsy or treatment. Concentrations = of=20 N-acetylaspartate (NAA), creatine (Cr), and choline = (Cho) in=20 the lesion were expressed as metabolite ratios and were = normalized to=20 the contralateral hemisphere. Dynamic susceptibility=20 contrast-enhanced perfusion MR imaging was performed in a = subset of=20 patients (n =3D 32); relative cerebral blood volume = (rCBV) was=20 evaluated. Discriminant function analysis was used to = identify=20 variables that can predict inclusion in the neoplastic or=20 nonneoplastic lesion groups. Receiver operator characteristic = (ROC)=20 analysis was used to compare the discriminatory capability of = 1H-MRSI and perfusion MR imaging.

RESULTS: The discriminant function analysis correctly = classified=20 84.2% of original grouped cases (P < .0001), on the = basis=20 of NAA/Cho, Chonorm, NAAnorm, and NAA/Cr = ratios.=20 MRSI and perfusion MR imaging had similar discriminatory = capabilities=20 in differentiating tumors from nonneoplastic lesions. With = cutoff=20 points of NAA/Cho 3D≤0.61 and rCBV 3D≥1.50 (corresponding to diagnosis of the tumors), a = sensitivity of 72.2% and specificity of 91.7% in differentiating=20 tumors from nonneoplastic lesions were achieved.

CONCLUSION: These results suggest a promising role for=20 1H-MRSI and perfusion MR imaging in the = distinction=20 between brain tumors and nonneoplastic lesions in = adults.

Noninvasive and accurate differentiation between neoplastic = and=20 nonneoplastic brain lesions is important in determining the = correct=20 treatment plan and, in some cases, may avoid the necessity of = performing a biopsy.1 = Conventional=20 structural MR imaging is an established and useful tool in = the=20 diagnosis and evaluation of brain tumors; however, it does = not=20 provide information on tumoral vascularity, metabolism, and=20 cellularity, which are important in grading the tumor, and = has a=20 relatively low specificity for diagnosing mass = lesions.2 In = many=20 instances, reliable differentiation of neoplastic from = nonneoplastic=20 brain masses, or high-grade from low-grade tumors, is = difficult or=20 impossible with conventional MR imaging. As a result, some = patients=20 with benign lesions may undergo brain biopsies to rule out a=20 neoplasm, an invasive procedure with an appreciable risk for=20 morbidity.

Several types of nonneoplastic brain lesions (in particular,=20 demyelinating lesions and lesions caused by infectious = diseases)=20 can be potentially misdiagnosed as brain tumors. The = conventional=20 MR imaging appearance of these lesions can be nonspecific, = and=20 even the use of a contrast agent is of limited benefit = because=20 any pathologic process associated with disruption of the = blood-brain=20 barrier can result in enhancement on MR imaging. Thus, there = is=20 a need for additional imaging modalities, such as MR spectroscopy = and=20 perfusion MR imaging, which may aid in improving the diagnosis = of=20 unknown brain lesions.3 = However,=20 limited data are available on the use of proton MR = spectroscopy=20 (1H-MRS) and perfusion MR imaging in the = differentiation=20 of nonneoplastic lesions from tumors. Although most of the = previous=20 studies used single-voxel spectroscopy with a relatively = large voxel=20 size, resulting in partial volume effects, they have = indicated the=20 promise of 1H-MRS to differentiate between tumors = and=20 nonneoplastic lesions and adding diagnostic = information.1,4=967 = Several=20 groups also compared spectroscopic characteristics of = specific groups=20 of neoplastic and nonneoplastic lesions.8=9611 = Two recent=20 studies using perfusion MR imaging12 = and a=20 combination of MR spectroscopic imaging (MRSI) and perfusion = MR=20 imaging13 = reported=20 individual cases of lesions rather than group comparisons. = Recently,=20 multivoxel 1H-MRSI differentiated 78% of = originally=20 grouped cases of nonneoplastic lesions and tumors in = children.14 = We=20 hypothesized that 1H-MRSI can differentiate = between brain=20 tumors and nonneoplastic lesions in adults and that perfusion = can=20 further improve the differentiation. High-resolution=20 1H-MRSI allowed for evaluation of metabolite = levels and=20 corresponding blood volume in the regions of the lesions = showing the=20 most pronounced metabolic impairment.

Materials and Methods

We retrospectively examined 69 patients with primary brain = lesions=20 (mean age, 45.6 =B1 13.1 years; age range, 20=9677 = years; 35 men).=20 Patients were eligible for the study if they had a new, = untreated=20 primary brain lesion suspicious for a neoplasm on = conventional MR=20 imaging or CT scan. All patients were previously untreated, = except=20 for 3 cases of radiation necrosis in which the patients had = had=20 surgical treatment before with resection of extra-axial = tumors (1=20 meningioma,1 chordoma, and 1 pituitary adenoma) and received=20 radiation treatment; 1 case of a patient with acute myeloid = leukemia=20 treated with chemotherapy with secondary abnormal signal = intensity in=20 the brain that had remained stable for 3 years (at the time = of the=20 MRSI and perfusion scan); and 2 cases of patients with = postsurgical=20 gliosis that had been stable for 2.5 and 9 years. Forty-one = patients=20 subsequently underwent invasive diagnostic or surgical = procedures (32=20 surgical procedures, 9 biopsies).

Thirty-six patients (mean age, 42.4 =B1 15.3 years; 18 = men) had=20 brain tumors confirmed on neuropathologic examination = (high-grade=20 tumors, n =3D 28; low-grade tumors, n =3D 8), which = were=20 graded according to the World Health Organization (WHO)=20 classification scheme. The diagnoses in the group of = high-grade=20 tumors included glioblastoma multiforme (n =3D 13), = anaplastic=20 astrocytoma WHO grade III (n =3D 7), B-cell lymphoma = (n =3D=20 4), malignant oligodendroglioma WHO grade III (n =3D = 3), and=20 IgA-producing plasmocytoma (n =3D 1). The diagnoses of = low-grade=20 tumors were astrocytoma WHO grade II (n =3D 3), = infiltrating=20 astrocytoma grade II (n =3D 1), ganglioglioma = (n =3D 2),=20 and oligodendroglioma (n =3D 2). Thirty-three patients = (mean=20 age, 45.2 =B1 11.5 years; 11 men) were diagnosed with = nonneoplastic=20 lesions. These lesions consisted of stroke (n =3D 4, = no=20 biopsies, and followed up for 4 months to 3 years with = evidence of=20 regression or expected evolution toward encephalomalacia);=20 demyelinating disease ([n =3D 10, 1 biopsy], 5 cases = of multiple=20 sclerosis, 4 cases of acute disseminated encephalomyelitis = [ADEM]=20 followed up between 2 months and 3 years, and 1 case of = progressive=20 multifocal leukoencephalopathy with positive human polyoma = virus in=20 the CSF); benign lesions proved on clinical or pathologic = examination=20 ([n =3D 10, 4 biopsies], radionecrosis confirmed on = surgical=20 procedure [n =3D 3]; vasculitis confirmed on surgical = procedure=20 [n =3D 1]; nonpathologically confirmed postsurgical = gliosis=20 [n =3D 2] stable without evidence of recurrence for = 2.5 and 8=20 years, hematoma [n =3D 1] demonstrating regression = within 7=20 months, ataxia-telangiectasia [n =3D 1] with intracranial=20 involvement, and chemotherapy-induced leukoencephalopathy = [n=20 =3D 1] stable for 3 years; and biologically proved = meningoencephalitis=20 [n =3D 1] that had improved at 1 month follow-up). Nine = patients=20 had stable lesions confirmed without pathologic examination=20 that have been followed up for 11 months to 10 years with no=20 evidence of progression and clinical deterioration. The local = institutional review board approved the study.

We performed brain MR imaging and multisection 1H-MRSI at=20 1.5T with Signa 5X and LX systems (GE Healthcare, Milwaukee,=20 Wis) using a standard quadrature birdcage head coil. Routine=20 brain MR imaging consisted of sagittal spin-echo (SE)=20 T1-weighted images (TR, 300 ms; TE, 9 ms), axial fast SE = (FSE)=20 T2-weighted images with fat suppression (TR, 5800 ms; TE, 108 = ms),=20 axial fluid-attenuated inversion recovery (FLAIR) (TR, 8802 = ms;=20 TE, 133 ms; TI, 2200 ms), and axial and coronal T1-weighted = SE=20 images (TR, 500 ms; TE, 9 ms) after administration of = contrast=20 (gadolinium-diethylenetriamine pentaacetic acid [Gd-DTPA]) at = 0.1=20 mmol/kg. Axial T1-weighted spoiled gradient-recalled = localizer images=20 (TR, 34 ms; TE, 5 ms) at the same section location and = thickness as=20 the proton MRSI scan were also obtained. We performed=20 1H-MRSI using a multi-section spin-echo sequence = (TE, 280=20 ms) with 2D circular phase-encoding, chemical shift selection = water=20 suppression, and outer-volume saturation pulses for lipid=20 suppression.15=20

We adjusted the magnetic field homogeneity automatically, using=20 the unsuppressed water signal intensity. Either three or four = 15-mm-thick axial sections were obtained with an intersection = gap of 2.5 mm. The axial MRSI sections were graphically = prescribed=20 from the sagittal T1-weighted images so that the center = sections=20 included the center of the lesion. However, in lesions in the = temporal lobe, we positioned the sections so that the lowest=20 section covered the main part of the lesion to avoid = susceptibility=20 artifacts. The TR was either 1700 ms (3 sections MRSI) or = 2300=20 ms (4 sections MRSI). Because the T1 values of choline (Cho), = creatine (Cr), and N-acetylaspartate (NAA) are similar, = this=20 small difference in TR between the 3- or 4-section scans has=20 negligible influence on metabolite ratios. The scan time was=20 either 22 or 30 minutes, depending on the TR used. The = nominal=20 voxel size was approximately 0.8 mL. In-plane nominal = resolution=20 was 7.5 x 7.5 mm2. = The=20 multisection 2D MRSI datasets were processed by 3D Fourier=20 transformation, with cosine filters in the spatial = (phase-encoding)=20 domains, and exponential line broadening of 3 Hz, = zero-filling to=20 2048 data points, and a high-pass convolution filter to = remove the=20 residual water signal intensity (50 Hz stop-band) in the = time-domain.=20 No baseline correction was applied. We fitted the signals of = Cho, Cr,=20 NAA, and lactate to a Gaussian line shape using a simplex = routine,=20 and we performed reconstruction of the metabolite images by = numeric=20 integration. The peak area ratios of Cho/Cr, NAA/Cr, and = NAA/Cho were=20 calculated from the peak areas of the respective signals. = Normalized=20 lesion peak areas (Chonorm, Crnorm, and = NAAnorm) expressed relative to the contralateral=20 hemisphere were also calculated.

Regions of interest were selected in the lesion identified on = MR=20 imaging, and in contralateral hemisphere with normal MR imaging=20 appearance. Care was taken to ensure that the region of = interest=20 was entirely within the solid part of the lesion (in the = region=20 showing the maximum abnormality=97elevated Cho, and/or = decreased=20 NAA=97as determined on the metabolite images) to avoid any=20 contamination of the voxel from normal tissue, the area of = necrosis,=20 or cysts. In 2 cases, the lesion was located in the midline = (1=20 demyelinating lesion in the genu of the corpus callosum and 1 = lesion=20 in the superior aspect of the cerebellar vermis); the control = voxel=20 was selected in a midline brain region with a normal MR = imaging=20 appearance (the splenium of the corpus callosum in the first = case and=20 the inferior aspect of the cerebellar vermis in the second=20 case).

We performed dynamic susceptibility contrast-enhanced MR = perfusion=20 imaging in 32 patients (stroke, n =3D 1; demyelination, = n =3D=20 3; stable lesion, n =3D 8; low-grade tumors, n = =3D 4; and=20 high-grade tumors, n =3D 16) using single-shot = gradient-echo=20 echo-planar imaging with 30 dynamic scans and a TR of 2 = seconds and a=20 TE of 60 ms. Seventeen sections were acquired with 5-mm = section=20 thickness/2.5-mm gap for a total of 510 images. The section=20 locations were chosen to cover the lesion. We used a power = injector=20 to administer 0.1 mmol/kg of Gd-DTPA (Magnevist; Bayer = HealthCare=20 Pharmaceuticals, Wayne, NJ) at a rate of 5 mL/s into the = antecubital=20 vein through the catheter. The Gd-DTPA bolus was followed by = a=20 20 mL saline flush at the same flow rate.

We analyzed the MR perfusion data using in-house software = (perfx;=20 Johns Hopkins University School of Medicine, Baltimore, Md).=20 Raw perfusion images were thresholded (typically 5% of = maximum=20 baseline image intensity) to remove noise pixels and were = then=20 converted to images of 3D{Delta}=20R2* at = each time=20 point (3D{Delta}R2* =3D ln[S/S0]/TE). The = average of a=20 range of preinjection images was used to calculate the = S0=20 image. The first pass of the bolus through the brain was then = used to=20 calculate indices of relative regional cerebral blood volume = (rCBV)=20 by integration of 3D{Delta}R2*. After construction of rCBV maps, = several=20 measurements of regions of interest were obtained and = expressed=20 relative to the normal contralateral white matter, and we = recorded=20 the maximum rCBV.

We used unpaired t tests or analysis of variance with the = Fisher=20 Least Significant Difference post hoc test to evaluate the=20 differences in metabolite ratios and rCBV between tumors and=20 nonneoplastic lesions, and among the examined types of = lesions. We=20 applied discriminant function analysis to identify metabolite = ratios that could predict inclusion in the biopsy-proved = tumor=20 group and nonneoplastic lesion group. We used an F test = (Wilks=20 lambda) to assess the significance of each discriminant = function.=20 Discriminant function analysis was also performed on the = subset of=20 patients who also had perfusion MR imaging, with rCBV as an=20 additional parameter. In this subset of patients, we = evaluated the=20 discriminatory ability of MRSI, perfusion MR imaging, and a=20 combination of the 2 approaches by constructing receiver = operating=20 characteristic (ROC) curves for each procedure. We compared = the areas=20 under the different curves with each other using the = technique=20 proposed by Hanley and McNeil.16 = A greater=20 area under the curve indicates a better discriminatory = ability of the=20 corresponding imaging procedure. We also attempted to define = cutoff=20 points for MRSI and perfusion MR imaging for differentiating = tumors=20 from nonneoplastic lesions. We then calculated the accuracy = (number=20 of correct diagnoses/total number of subjects), sensitivity = (number=20 of correctly classified tumors/total number of tumor cases),=20 specificity (number of correctly classified nonneoplastic=20 lesions/total number of nonneoplastic lesions), positive = predictive=20 values (PPV, number of true tumor cases/all cases classified = as=20 tumors), and negative predictive values (NPV, number of true=20 nonneoplastic cases/all cases classified as nonneoplastic = lesions)=20 associated with these cutoff points (all values are presented = as=20 %).

The level of statistical significance was set to P < = .05.=20 Data are presented as means =B1 standard deviations.

Results

The accompanying table = shows average=20 metabolite ratios and rCBV in tumors and nonneoplastic = lesions.=20 Examples of spectra, metabolite images, and CBV maps of a = tumor and=20 proved benign lesion are shown in Figs 1 and = 2,=20 respectively.


View = this=20 table:
[in this=20 window]
[in a new window]

  		Tumors and nonneoplastic = lesions:=20 metabolite ratios and rCBV* =

 = ;



View = larger=20 version (73K):
[in this=20 window]
[in a new window]

  		Fig 1. Axial FLAIR image, = T2-weighted FSE MR image, axial contrast-enhanced = T1-weighted SE MR=20 image, and axial CBV map in a 38-year-old woman with primary = CNS=20 lymphoma. There is T2 hyperintensity involving the left=20 temporoparietal lobe. There is minimal mass effect; the = lesion=20 exhibits patchy heterogeneous enhancement on the = postgadolinium=20 images. The CBV map demonstrates a moderately increased = blood volume=20 in the lesion compared with the normal contralateral side = (rCBV,=20 1.66, which is higher than the cutoff point of 1.5). = A,=20 B, The abnormal signal intensity of the lesion = extended over=20 8 FLAIR sections (5-mm section thickness, no gap; with 5 = sections=20 showing the bulk of the lesion). Images shown of the Cho, = Cr, NAA,=20 and lactate metabolites were reconstructed from the MRSI = section=20 exhibiting the largest metabolic abnormalities (MRSI, = section 2).=20 Axial T1-weighted SE localizer image with proton MR spectra = of the=20 lesion (voxel 2) and the corresponding control spectrum = (voxel 1)=20 are shown. Compared with the contralateral side, the = metabolite=20 images and spectra of the lesion show elevated Cho and Cr, = and=20 decreased NAA signals, with an NAA/Cho ratio of 0.58 (below = the=20 cutoff point of 0.61). Slight contamination of spectra with = lipid=20 signals, most likely because of the patient=92s head motion, = was noted=20 in this examination. =



View = larger=20 version (68K):
[in this=20 window]
[in a new window]

  		Fig 2. Axial FLAIR, = T2-weighted=20 FSE, contrast-enhanced T1-weighted SE MR images and axial = CBV map in=20 a 27-year-old woman with meningoencephalitis. There is an = abnormal=20 high T2 signal intensity in the left frontal lobe with = minimal=20 enhancement on the postgadolinium image. The CBV map shows = slightly=20 elevated levels of blood volume in the lesion compared with = the=20 normal contralateral side, with a rCBV of 1.35. The original = diagnosis on the basis of conventional MR imaging favored a = neoplasm=20 over an inflammatory cause. At 3 weeks of follow-up (MR = spectroscopy=20 was not performed), an overall improved appearance with a = decrease=20 in the size of the frontal subcortical and deep white matter = T2=20 signal intensity abnormality was noted. A, B, The = abnormal=20 signal intensity of the lesion extended over 5 FSE sections = (5-mm=20 section thickness, no gap; with 2 sections showing the bulk = of the=20 lesion). One MRSI section (the bottom section shown in the = figure)=20 covered the lesion. Images of the Cho, Cr, NAA, and lactate=20 metabolites and proton MR spectra of the lesion (voxel 2) = and=20 control region (voxel 1) are shown. The NAA/Cho ratio of = 1.15 was=20 above the cutoff point of 0.61, Chonorm was 0.8, = and rCBV=20 was slightly below the cutoff point of 1.5. On the basis of = MRSI and=20 perfusion MR imaging, the presence of a nonneoplastic lesion = was=20 favored. Discriminant function analysis classified this = lesion as=20 nonneoplastic on the basis of MRSI data alone and MRSI and = perfusion=20 MR imaging data. =

The NAA/Cho ratio was lower in high-grade tumors compared with=20 demyelinating lesions (60.5%; P < .001), proved benign=20 lesions (73.5%; P < .0001), stable lesions (41.0%; = P=20 =3D .001), stroke (37.2%; P =3D .022), and low-grade = tumors=20 (23.6%; P =3D .051). Low-grade tumors also had a lower = NAA/Cho=20 than the proved benign lesions (49.9%; P =3D .001) and = demyelinating lesions (36.9%; P =3D .011). Both = high-grade and=20 low-grade tumors had a lower NAA/Cr ratio compared with = demyelinating=20 lesions (104.7%; P < .0001 and 77.4%; P =3D = .003,=20 respectively). High-grade tumors had a higher Cho/Cr ratio = than the=20 proved benign lesions (181.9%; P =3D .004) and stable = lesions=20 (144.2%; P =3D .026).

Chonorm was higher in high-grade tumors compared with = proved=20 benign lesions (136.4%; P < .0001), stable lesions = (77.6%;=20 P =3D .01), stroke (105.8%; P =3D .011), and = demyelinating=20 lesions (56.1%; P =3D .049). Low-grade tumors had a = 88.4% higher=20 Chonorm than the proved benign lesions (P = =3D .017).=20 High-grade tumors had a lower NAAnorm than = demyelinating=20 lesions (28.9%; P < .0001), stable lesions (17.5%; = P=20 =3D .019), and proved benign lesions (13%; P =3D = .067), and=20 low-grade tumors had a 21.3% lower NAAnorm than=20 demyelinating lesions (P =3D .021). Crnorm did = not=20 show any significant differences among the examined groups of = lesions.

Considering all possible combinations of metabolite ratios, = the=20 best discriminant function to differentiate between the group = of=20 brain tumors (high-grade and low-grade) and the nonneoplastic = lesions=20 was found to include the ratios NAA/Cho, Chonorm,=20 NAAnorm, and NAA/Cr (presented in the decreasing = magnitude=20 of correlation with the discriminant function). The = discriminant=20 function analysis correctly classified 84.2% of original = grouped=20 cases in the tumor and nonneoplastic lesion groups (P = <=20 .0001). The discriminant function analysis misclassified 5 = cases of=20 tumors as nonneoplastic lesions (anaplastic astrocytoma WHO = grade II,=20 infiltrating astrocytoma WHO grade III, gliomatosis cerebri = WHO grade=20 II, oligodendroglioma WHO grade II, ganglioglioma WHO grade = II). Six=20 nonneoplastic lesions were classified as tumors: = demyelination=20 (n =3D 1), radiation necrosis (n =3D 2), = postsurgical=20 gliosis (n =3D 1), and stable lesions not confirmed on = pathologic examination (1.5- to 2-year follow-up, n = =3D 2).=20

We obtained good-quality perfusion data in 30 patients (Table); = in=20 2 patients, perfusion MR imaging failed for technical reasons.=20 High-grade tumors had a higher rCBV (4.9 =B1 3.1; range,=20 1.2=9611.3) than low-grade tumors (1.5 =B1 1.2; range, = 0.4=963.1;=20 P =3D .011) and nonneoplastic lesions (1.0 =B1 0.4; = range,=20 0.5=960.7; P < .001). There were significant = differences=20 between the group of high-grade tumors and nonneoplastic = lesions=20 (P =3D .001), and between the high-grade and low-grade = tumor=20 groups (P =3D .037), whereas no significant difference was = found=20 between the low-grade tumor and nonneoplastic lesion groups = (P=20 =3D .73). In the 30 patients who were evaluated by MRSI and = perfusion=20 MR imaging, the discriminant function analysis correctly = classified=20 93.3% of the original cases with the use of only = spectroscopic data=20 (the discriminant function contained NAA/Cho and NAA/Cr = ratios;=20 P < .0001), and 80% of original grouped cases were=20 correctly classified with the use of the rCBV data only = (P =3D=20 .002).

Using discriminant function analysis, we found that perfusion = MR=20 imaging did not further improve the (already high) percentage = of=20 correctly classified cases using MRSI; 93.3% of the original = grouped=20 cases were correctly classified with use of combined = spectroscopy and=20 perfusion data (discriminant function included NAA/Cho, = NAA/Cr and=20 rCBV) (P < .0001). One case of a patient with a=20 demyelinating lesion (rCBV =3D 1.66) was misclassified as a = tumor, and=20 1 low-grade tumor (gliomatosis cerebri, rCBV =3D 3.1) was = misclassified=20 as a nonneoplastic lesion. In the analyses based on rCBV data = only, 3=20 high-grade tumors (2 cases of anaplastic astrocytoma WHO = grade III=20 and 1 case of B-cell lymphoma) and 3 low-grade tumors (2 = cases of=20 oligodendroglioma WHO grade II and 1 case of ganglioglioma = WHO grade=20 I) were misclassified as nonneoplastic lesions. We carried = out a=20 further evaluation of discriminatory capability using the ROC = analysis. The calculated areas under the ROC curves were 0.92 = for=20 MRSI, 0.89 for perfusion MR imaging, and 0.96 for the = analysis with=20 the use of MRSI and perfusion MR imaging (Fig 3). We = found no=20 significant differences when comparing the areas under the = curves,=20 which indicate that each procedure had a similar = discriminatory=20 capability.



View = larger=20 version (19K):
[in this=20 window]
[in a new window]

  		Fig 3. ROC curves = representing=20 discriminatory capability of perfusion MR imaging,=20 1H-MRSI and combination of 1H-MRSI and = perfusion MR imaging to differentiate between tumors and=20 nonneoplastic lesions. We constructed the curves using data = on 30=20 subjects evaluated with both perfusion MR imaging and MRSI. = The=20 calculated areas under the ROC curves were 0.92 for MRSI, = 0.89 for=20 perfusion MR imaging, and 0.96 for the analysis on the basis = of MRSI=20 and perfusion MR imaging. We found no significant = differences when=20 comparing the areas under the curves, indicating that each = procedure=20 had similar discriminatory capability for these subjects. =

In the daily practice of neuroradiology, it may be difficult = to=20 perform discriminant function analyses routinely. Therefore, = we=20 attempted to identify cutoff values, which could help differentiate = tumors from nonneoplastic lesions. In our sample, it was not=20 straightforward to find the optimum cutoff point for MRSI; = the=20 NAA/Cho range of 0.54 to 0.66 included 4 tumors (all NAA/Cho=20 3D≤0.61) and 5=20 nonneoplastic lesions (4 nonneoplastic lesions with NAA/Cho=20 >0.61). With a cutoff value of NAA/Cho of 0.61 or less=20 (corresponding to diagnosis of a tumor), accuracy was 84.1%,=20 sensitivity 86.1%, specificity 81.8%, PPV 83.8%, and NPV = 84.4%=20 (n =3D 69 cases). All cases of patients with NAA/Cho of = 0.61 or=20 less had abnormally elevated Cho; however, there was an = overlap=20 in Chonorm with nonneoplastic lesions (range, = 0.38=962.32).=20 Similar values were obtained for the subset of cases with = perfusion=20 MR imaging (n =3D 30): for NAA/Cho of 0.61 or less, = accuracy was=20 86.7%, sensitivity 88.9%, specificity 83.3%, PPV 88.8%, and = NPV=20 83.3%. For a simple decision making, NAAnorm and NAA/Cr=20 were not of high diagnostic value because of a considerable=20 overlap between tumors and nonneoplastic lesions for both = parameters.=20 A cutoff value of rCBV of 1.5 or more (corresponding to a = diagnosis=20 of a tumor) resulted in an accuracy of 83.3%, sensitivity = 77.8%,=20 specificity 91.7%, PPV 93.3%, and NPV 91.7%. With combined = criteria=20 (NAA/Cho 3D≤0.61=20 and rCBV 3D≥1.5), we=20 obtained accuracy 80.0%, sensitivity 72.2%, specificity = 91.7%, PPV=20 93.3%, and NPV 91.7%.

Discussion

The primary result of our study was a good specificity of=20 1H-MRSI for distinguishing between high-grade and=20 low-grade tumors, as well as between neoplastic lesions of = all grades=20 and benign lesions in adults. Perfusion MR imaging was also = able to=20 successfully distinguish high-grade and low-grade tumors, as = well as=20 high-grade tumors from benign lesions. However, no = significant=20 differences were found between low-grade tumors and benign = lesions.=20 Adding rCBV measurements did not improve the discriminant=20 analysis classification of lesions with that on the basis of = MRSI=20 alone.

In our group of patients with heterogeneous brain lesions = evaluated=20 for suspected neoplasm, 1H-MRSI and discriminant = function=20 analysis demonstrated a high rate of success of 84% in = correctly=20 classifying lesions originally diagnosed as tumors and = nonneoplastic=20 lesions on the basis of the ratios of NAA/Cho, = Chonorm,=20 NAAnorm, and NAA/Cr. As expected, elevated levels = of Cho=20 were detected in tumors compared with nonneoplastic lesions. = Elevated=20 signal intensity in Cho results from increased attenuation of = proliferating tumoral cells, with Cho-containing compounds = including=20 membrane precursors and products of degradation. Tumoral = levels of=20 NAA, presumably originating from residual brain tissue within = an=20 infiltrating tumor,17,18 = were lower=20 than in nonneoplastic lesions. Thus, in agreement with = published=20 data, our analyses confirmed that highly elevated = Cho-containing=20 compounds in conjunction with a decreased NAA resonance = (resulting in=20 decreased ratio of NAA/Cho) are parameters sensitively = indicative of=20 neoplastic tumoral growth19,20 = and are=20 helpful in differentiating tumors from other causes. There = was no=20 significant difference in the level of Cr among the groups of = different lesions. The levels of Cr in tumors in adults have = been=20 reported to be decreased18=20 or both increased and decreased.21 = In our=20 study, high-grade tumors had a wider range of levels of Cr = than=20 low-grade tumors. The lowest Cr levels (28% of the control = value)=20 were found in a B-cell lymphoma, whereas the largest=20 Crnorm (2.63) was detected in the case of a = patient with=20 anaplastic astrocytoma (Crnorm in anaplastic = astrocytomas=20 were variable, 1.36 =B1 0.72 on average). Two cases of = patients with=20 anaplastic astrocytoma from a previously examined pediatric=20 group14 = also had=20 relatively high Crnorm (1.27 and 1.40) compared = with other=20 high-grade tumors. The presence of elevated lactate was = confirmed=20 only in 2 cases of high-grade tumors (1 glioblastoma = multiforme and 1=20 anaplastic astrocytoma) and 1 case of a stroke. In the = remaining=20 tumors, lactate was either not detected or, in some cases, = was=20 obscured because of contamination by the lipid peaks from the = skull=20 and scalp. Because the voxels selected for evaluations were = from=20 the solid part (often the rim) of the tumor, lactate from = areas=20 of necrosis was avoided.

In our study, in some instances, tumors of grades II and III = were=20 misclassified as benign lesions, and some cases of demyelination,=20 radiation necrosis, gliosis, and stable lesions were = misclassified=20 as tumors. Misclassification of tumors as benign lesions can=20 occur either as a result of high tumoral heterogeneity, when=20 regions of spectroscopic sampling differ from the location of = the highest tumor grade on histologic examination (which = presumably=20 exhibit the highest levels of Cho), or in tumors showing no=20 increase in Cho.22,23 = In=20 agreement with these findings, the 5 misclassified tumors had = the=20 highest NAA/Cho ratio (3D≥0.7) of all tumors. This result is similar to a = previous=20 study (including 86 cases of tumors and 13 nonneoplastic = lesions)=20 showing a NAA/Cho ratio of less than 1 as a threshold value = for=20 tumors.6

The diagnosis of demyelinating lesions, multiple sclerosis, = and=20 ADEM is made on typical clinical presentation, conventional = MR=20 imaging findings, and the evolution and presentation of the = disease=20 with time. However, on occasion demyelinating disease = presents as a=20 single brain mass that is indistinguishable from a brain = tumor on=20 clinical and radiologic examinations. In those cases,=20 1H-MRSI has an important diagnostic role, especially = when=20 it reveals reduced NAA without increased Cho. In these = instances, MR=20 spectroscopic findings may help spare the patient from brain = biopsy.=20 Nonetheless, some cases of active demyelinating lesions may = show=20 similar spectra as high-grade gliomas (ie, decreased NAA and = highly=20 elevated Cho),24 = because of=20 histopathologic similarities, which include hypercellularity, = reactive astrocytes, mitotic figures, and areas of necrosis. = In our=20 study, the misclassified case of demyelination had the lowest = NAA/Cho=20 ratio (0.24) of all demyelinating lesions. The lesion also = showed=20 increased rCBV on the perfusion scan.

Differentiating residual or recurrent tumor from = treatment-related=20 changes is limited on conventional MR imaging as well as on=20 histologic examination; areas of T1 contrast enhancement = after=20 radiation treatment often contain both residual or recurrent=20 tumor and tissue affected by therapy-related changes. Besides = that, the heterogeneity of gliomas before and after therapy = and=20 the inaccuracy of biopsy sampling pose another challenge in = the=20 histologic separation of tumors from necrosis.25 = Low=20 levels of Cho and Cr can differentiate radiation necrosis = from=20 recurrent or progressive tumors.26,27 = In our=20 study, 2 of 3 cases of radiation necrosis were misclassified = as=20 tumors; this could be related to an increase in = Cho-containing=20 compounds after radiation therapy as a result of cell damage = and=20 astrogliosis. In addition, both tumors and necrotic tissue = have low=20 levels of NAA, consistent with neuronal damage that is = present in=20 tumors as well as in radiation necrosis. However, the=20 misclassification of the 2 cases of radiation necrosis as = tumors was=20 somewhat surprising, given their high NAA/Cho ratio (1.35 and = 1.65;=20 in all tumors, NAA/Cho was 3D≤=200.99) and decreased=20 Chonorm (0.63 and 0.58; all tumors had = Chonorm=20 3D≥1.04). However, in the=20 subgroup of patients evaluated by MRSI and perfusion MR = imaging,=20 one of these cases was classified correctly as a = nonneoplastic=20 lesion (perfusion MR imaging showed decreased rCBV). =

Perfusion MR imaging provides measurement of tumoral = angiogenesis=20 and capillary permeability, which may aid in the diagnosis,=20 grading, and prognosis of gliomas. Primary high-grade = neoplasms=20 and metastases are expected to have elevated rCBV because of=20 intensive neoangiogenesis and high metabolism, whereas rCBV = is=20 typically lower or normal in low-grade neoplasms and nonneoplastic=20 lesions.13 = Law et=20 al28=20 suggested a cutoff value of 1.75 for rCBV for diagnosis of = high-grade=20 gliomas. However, rCBV was also found to vary among different = types=20 of low-grade tumors (eg, oligodendrogliomas have been = reported to=20 have higher rCBV than astrocytomas).29 = Although=20 oligodendrogliomas tend to be high-blood volume tumors, with = a wide=20 range of rCBV values,29,30=20 some low-grade oligodendrogliomas may not show increased = rCBV.30=20 In agreement with these findings, 1 case of nonanaplastic=20 oligodendroglioma in our study showed a decreased rCBV of = 0.4, and 1=20 case of anaplastic oligodendroglioma showed a rCBV of 2.5. In = our=20 study, the rCBV of low-grade tumors varied between 0.40 and = 3.07, and=20 the range of rCBV in high-grade tumors was 1.22 to 10.34. = Perfusion=20 MR imaging can also help differentiate between recurrent = tumors=20 and radiation necrosis; recurrent tumors have a higher rCBV=20 compared with radiation necrosis.31,32 = Perfusion=20 MR imaging has been shown to be a valuable diagnostic tool in = distinguishing active demyelinating lesions from brain = tumors, on the=20 basis of elevated rCBV in tumors.33 = In a=20 subgroup of subjects with both MRSI and perfusion MR imaging=20 examinations, the diagnostic accuracy (compared with = discriminant=20 function analysis) of MRSI was higher than perfusion MR = imaging=20 alone, and the combination of MRSI and perfusion MR imaging = did not=20 further improve the overall diagnostic accuracy of MRSI. This = finding=20 was mainly the result of overlap in rCBV between the group of = stable=20 lesions (rCBV, 0.53=961.22) and low-grade tumors (rCBV, = 0.40=963.07)=20 and likely also because of a smaller number of cases of = perfusion=20 compared with the number of patients evaluated by MRSI. = However,=20 the ROC analysis, which has been very useful in radiologic = decision=20 making, did not find any difference in discriminatory = capability=20 between MRSI and perfusion MR imaging.

There were several limitations of our study. One potential = limitation=20 was that the rCBV was evaluated in regions corresponding to=20 those with the most abnormal MR spectra. Although Cho and NAA = are considered measures of cellular proliferation, and = neuronal=20 viability, respectively, rCBV may reflect endothelial = proliferation=20 and angiogenesis. Because these biologic entities are = different=20 from a pathologic standpoint, it may be possible that in some = lesions, the largest abnormalities may be found in different=20 spatial locations within the lesion. We found this to be the=20 case in 1 glioblastoma multiforme and 1 anaplastic = astrocytoma,=20 in which a higher rCBV was detected in a section adjacent to=20 the section included for analysis. Because both types of = tumors=20 had a highly elevated rCBV throughout the lesion, selection = of=20 a different region of interest for evaluation did not affect = our=20 results.

Another potential limitation was that biopsies were not = available=20 for all cases included in the group of nonneoplastic lesions. = Therefore, it may be possible that some of the lesions = (including=20 2 stable lesions misclassified by the discriminant analysis = as=20 tumors) may have been low-grade tumors. However, excluding = stable and=20 gliotic lesions from the discriminant function analysis did = not=20 improve the classification. In our study, we made no attempt = to=20 compare MRSI or perfusion classification with diagnoses on = the basis=20 of conventional MR imaging and clinical information, so we = could not=20 assess the added value of these techniques. Also, we did not = analyze=20 other parameters that may have been extracted from perfusion = MR=20 imaging time curves, such as vascular permeability. Another = potential=20 limitation was that MRSI was performed at long TE for = technical=20 reasons (it is difficult to perform MRSI at short TE with = extensive=20 brain coverage, mainly because of inhomogeneities in the B0 = magnetic=20 field). For this reason, compounds such as glutamate, = glutamine,=20 myo-inositol, or lipids could not be evaluated in our study. = It is=20 possible that future studies of MRSI at short TE that observe = these=20 compounds may improve the performance of MRSI further in=20 distinguishing benign from malignant brain lesions. =

In contrast to several previous studies, our study used = high-resolution=20 multisection 1H-MRSI, which allowed the selection of a=20 voxel with the most abnormal spectroscopic appearance in=20 heterogeneous lesions. It also automatically provided control = spectra=20 from the normal-appearing contralateral hemisphere. Both of=20 these factors may have helped improve diagnostic accuracy; in = a=20 previous study, a normalized Cho ratio was found to be a = better=20 measurement than either the Cho/Cr or Cho/NAA ratios,34 = because=20 the normalized value controls for the intersubject = differences in=20 metabolite levels. We used metabolite ratios to control for = the=20 potential presence of CSF, cystic fluid, or necrosis (though = we=20 tried to avoid these regions in the analyses) in the=20 spectroscopic voxel. Partial volume effects of CSF with the = normal=20 brain was also minimized by the high spatial resolution of = this=20 technique (nominal voxel size, 0.8 mL), which is much better = than=20 that commonly used in single-voxel spectroscopy (typically, 8 = mL).1=20 Although our study used 1H-MRSI at long TE, which = results=20 in lower signal-to-noise ratios compared with short TE=20 acquisitions, it provides less complex spectra that are more = simple=20 to analyze because of the flat baseline, lack of spectral = overlap,=20 and attenuation of water and lipid signals resulting from T2=20 relaxation.

In summary, the results of this study show that 1H-MRSI = may=20 be helpful in the assessment of undiagnosed brain lesions. In = particular, 1H-MRSI differentiated the group of = low-grade=20 tumors, in which it demonstrated elevated Cho and decreased = NAA=20 from nonneoplastic lesions. Perfusion MR imaging and the=20 combination of 1H-MRSI and perfusion MR imaging = were found=20 to have comparable ability to 1H-MRSI alone in=20 differentiating tumors from nonneoplastic lesions. In the = future,=20 technical advances in 1H-MRSI, including the use = of higher=20 magnetic field strengths, may further improve diagnostic = accuracy=20 because of higher spectral resolution and increased = signal-to-noise=20 ratios. In the long term, it is hoped that improvements in = the=20 diagnosis of lesions with 1H-MRSI and perfusion MR = imaging=20 will reduce the need for brain biopsy, especially in cases of = inaccessible lesions or lesions located in crucial brain = structures=20 (eg, brain stem or eloquent cortex).

Acknowledgments

We thank Ms. Priscilla Matthews for her help with editing the=20 manuscript.

Footnotes

This publication was supported by NIH = grant 1RO1=20 NS042851 and grant number P41 RR 015241 from the National = Center for=20 Research Resources (NCRR), a component of the National = Institutes=20 of Health (NIH). Its contents are solely the responsibility = of=20 the authors and do not necessarily represent the official = view=20 of NCRR or NIH.

References

  1. Rand SD, Prost R, Haughton V, et al. Accuracy of = single-voxel=20 proton MR spectroscopy in distinguishing neoplastic from nonneoplastic = brain=20 lesions. AJNR Am J Neuroradiol 1997;18:1695=96704[Abstract]

  2. Burger PC. Malignant astrocytic neoplasms: = classification,=20 pathologic anatomy, and response to treatment. Semin Oncol=20 1986;13:16=9626[Medline]

  3. Al-Okaili RN, Krejza J, Woo JH, et al. Intraaxial = brain masses:=20 MR imaging-based diagnostic strategy=97initial experience. = Radiology=20 2007;243:539=9650[Abstract/Free Full Text]

  4. Poptani H, Gupta RK, Roy R, et al. Characterization = of=20 intracranial mass lesions with in vivo proton MR spectroscopy. = AJNR Am=20 J Neuroradiol 1995;16:1593=96603[Abstract]

  5. Poptani H, Kaartinen J, Gupta RK, et al. Diagnostic = assessment=20 of brain tumours and non-neoplastic brain disorders in vivo using = proton=20 nuclear magnetic resonance spectroscopy and artificial neural = networks.=20 J Cancer Res Clin Oncol 1999;125:343=9649[CrossRef][Medline]

  6. Butzen J, Prost R, Chetty V, et al. Discrimination = between=20 neoplastic and nonneoplastic brain lesions by use of proton MR = spectroscopy:=20 the limits of accuracy with a logistic regression model. AJNR = Am J=20 Neuroradiol 2000;21:1213=9619[Abstract/Free Full Text]

  7. M=F6ller-Hartmann W, Herminghaus S, Krings T, et al. = Clinical=20 application of proton magnetic resonance spectroscopy in the diagnosis = of=20 intracranial mass lesions. Neuroradiology = 2002;44:371=9681[CrossRef][Medline]

  8. De Stefano N, Caramanos Z, Preul MC, et al. In vivo=20 differentiation of astrocytic brain tumors and isolated demyelinating = lesions=20 of the type seen in multiple sclerosis using 1H magnetic resonance=20 spectroscopic imaging. Ann Neurol 1998;44:273=9678[CrossRef][Medline]

  9. Venkatesh SK, Gupta RK, Pal L, et al. Spectroscopic = increase in=20 choline signal is a nonspecific marker for differentiation of=20 infective/inflammatory from neoplastic lesions of the brain. J = Magn=20 Reson Imaging 2001;14:8=9615[CrossRef][Medline]

  10. Vuori K, Kankaanranta L, H=E4kkinen AM, et al. = Low-grade gliomas=20 and focal cortical developmental malformations: differentiation with = proton MR=20 spectroscopy. Radiology 2004;230:703=9608[Abstract/Free Full Text]

  11. Pandya HG, Wilkinson ID, Agarwal SK, et al. The = nonspecific=20 nature of proton spectroscopy in brain masses in children: a series of = demyelinating lesions. Neuroradiology 2005;47:955=9659[CrossRef][Medline]

  12. Cha S, Knopp EA, Johnson G, et al. Intracranial mass = lesions:=20 dynamic contrast-enhanced susceptibility-weighted echo-planar = perfusion MR=20 imaging. Radiology=20 2002;223:11=9629[Abstract/Free Full Text]

  13. Law M, Hamburger M, Johnson G, et al. = Differentiating surgical=20 from non-surgical lesions using perfusion MR imaging and proton MR=20 spectroscopic imaging. Technol Cancer Res Treat = 2004;3:557=9665[Medline]

  14. Hourani R, Horsk=E1 A, Albayram S, et al. Proton = magnetic=20 resonance spectroscopic imaging to differentiate between nonneoplastic = lesions=20 and brain tumors in children. J Magn Reson Imaging = 2006;23:99=96107[CrossRef][Medline]

  15. Duyn JH, Gillen J, Sobering G, et al. Multisection = proton MR=20 spectroscopic imaging of the brain. Radiology = 1993;188:277=9682[Abstract]

  16. Hanley JA, McNeil BJ. A method of comparing the = areas under=20 receiver operating characteristic curves derived from the same = cases.=20 Radiology 1983;148:839=9643[Abstract]

  17. Gill SS, Thomas DG, Van Bruggen N, et al. Proton MR=20 spectroscopy of intracranial tumours: in vivo and in vitro = studies. J=20 Comput Assist Tomogr 1990;14:497=96504[Medline]

  18. Howe FA, Barton SJ, Cudlip SA, et al. Metabolic = profiles of=20 human brain tumors using quantitative in vivo 1H magnetic resonance=20 spectroscopy. Magn Reson Med 2003;49:223=9632[CrossRef][Medline]

  19. Preul MC, Caramanos Z, Collins DL, et al. Accurate,=20 noninvasive diagnosis of human brain tumors by using proton magnetic = resonance=20 spectroscopy. Nat Med 1996;2:323=9625[CrossRef][Medline]

  20. Burtscher IM, Skagerberg G, Geijer B, et al. Proton = MR=20 spectroscopy and preoperative diagnostic accuracy: an evaluation of=20 intracranial mass lesions characterized by stereotactic biopsy=20 findings. AJNR Am J Neuroradiol 2000;21:84=9693[Abstract/Free Full Text]

  21. Nelson SJ. Multivoxel magnetic resonance = spectroscopy of brain=20 tumors. Mol Cancer Ther 2003;2:497=96507[Abstract/Free Full Text]

  22. Saraf-Lavi E, Bowen BC, Pattany PM, et al. Proton MR = spectroscopy of gliomatosis cerebri: case report of elevated = myoinositol with=20 normal choline levels. AJNR Am J Neuroradiol = 2003;24:946=9651[Abstract/Free Full Text]

  23. Londo=F1o A, Castillo M, Armao D, et al. Unusual MR=20 spectroscopic imaging pattern of an astrocytoma: lack of elevated = choline and=20 high myo-inositol and glycine levels. AJNR Am J Neuroradiol = 2003;24:942=9645[Abstract/Free Full Text]

  24. Saindane AM, Cha S, Law M, et al. Proton MR = spectroscopy of=20 tumefactive demyelinating lesions. AJNR Am J Neuroradiol=20 2002;23:1378=9686[Abstract/Free Full Text]

  25. Cha S. Update on brain tumor imaging: from anatomy = to=20 physiology. AJNR Am J Neuroradiol 2006;27:475=9687[Free Full Text]

  26. Taylor JS, Langston JW, Reddick WE, et al. Clinical = value of=20 proton magnetic resonance spectroscopy for differentiating recurrent = or=20 residual brain tumor from delayed cerebral necrosis. Int J = Radiat Oncol=20 Biol Phys 1996;36:1251=9661[CrossRef][Medline]

  27. Rock JP, Hearshen D, Scarpace L, et al. Correlations = between=20 magnetic resonance spectroscopy and image-guided histopathology, with = special=20 attention to radiation necrosis. Neurosurgery = 2002;51:912=9619;=20 discussion 919=9620[CrossRef][Medline]

  28. Law M, Yang S, Wang H, et al. Glioma grading: = sensitivity,=20 specificity, and predictive values of perfusion MR imaging and proton = MR=20 spectroscopic imaging compared with conventional MR imaging. = AJNR Am J=20 Neuroradiol 2003;24:1989=9698[Abstract/Free Full Text]

  29. Cha S, Tihan T, Crawford F, et al. Differentiation = of=20 low-grade oligodendrogliomas from low-grade astrocytomas by using = quantitative=20 blood-volume measurements derived from dynamic susceptibility=20 contrast-enhanced MR imaging. AJNR Am J Neuroradiol = 2005;26:266=9673[Abstract/Free Full Text]

  30. Lev MH, Ozsunar Y, Henson JW, et al. Glial tumor = grading and=20 outcome prediction using dynamic spin-echo MR susceptibility mapping = compared=20 with conventional contrast-enhanced MR: confounding effect of elevated = rCBV of=20 oligodendrogliomas [corrected] [published erratum appears in = AJNR Am J=20 Neuroradiol 2004;25:B1]. AJNR Am J Neuroradiol = 2004;25:214=9621[Abstract/Free Full Text]

  31. Covarrubias DJ, Rosen BR, Lev MH. Dynamic magnetic = resonance=20 perfusion imaging of brain tumors. Oncologist = 2004;9:528=9637[Abstract/Free Full Text]

  32. Sugahara T, Korogi Y, Tomiguchi S, et al. = Posttherapeutic=20 intraaxial brain tumor: the value of perfusion-sensitive = contrast-enhanced MR=20 imaging for differentiating tumor recurrence from nonneoplastic=20 contrast-enhancing tissue. AJNR Am J Neuroradiol = 2000;21:901=9609[Abstract/Free Full Text]

  33. Cha S, Pierce S, Knopp EA, et al. Dynamic = contrast-enhanced=20 T2*-weighted MR imaging of tumefactive demyelinating lesions. = AJNR Am J=20 Neuroradiol 2001;22:1109=9616[Abstract/Free Full Text]

  34. Rabinov JD, Lee PL, Barker FG, et al. In vivo 3-T MR = spectroscopy in the distinction of recurrent glioma versus radiation = effects:=20 initial experience. Radiology 2002;225:871=9679[Abstract/Free Full Text]

Received March 22, 2007; accepted after revision July = 29, 2007.





This = Article
3D"
3D"Right Abstract=20 3D"Freely
3D"Right Figures=20 Only
3D"Right Full=20 Text (PDF)
3D"Right All Versions of this=20 Article:
ajnr.A0810v1<= /A>=20
29/2/366    most=20 recent
3D"Right Alert=20 me when this article is cited
3D"Right Alert=20 me if a correction is posted
3D"Right Citation Map
3D"
Services
3D"
3D"Right Similar=20 articles in this journal
3D"Right Similar=20 articles in PubMed
3D"Right Alert me to new = issues of=20 the journal
3D"Right Download to=20 citation manager
3D"Right 3D"reprints
3D"
Citing = Articles
3D"
3D"Right Citing Articles via CrossRef =
3D"
Google = Scholar
3D"
3D"Right Articles by Hourani, R. =
3D"Right Articles by Horsk=E1, A. =
3D"
PubMed
3D"
3D"Right PubMed=20 Citation
3D"Right Articles by Hourani, R. =
3D"Right Articles by Horsk=E1, A.=20 =

Copyright = =A9 2008 by=20 the American Society of Neuroradiology. Print ISSN: 0195-6108 Online ISSN: = 1936-959X=20

------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/icons/head/anjrInsideHdr.gif R0lGODlhLgFoAMQAAIaUrFlsjQeksA98lRpAa/Dy9QLCxeHk6jpReRVegHeGogS4vhFyjrO8y9LX 4MLJ1hZUeZWhtwuQoitDbqSvwQaut2h5mA2GmxhKcklegwmaqRNoh////wDMzBw2ZAAAACH5BAAA AAAALAAAAAAuAWgAAAX/oCeOZGmeaKqubOu+cCzPdG3feK7vfO//wKBwSCwaj8ikcslsOp/QqHRK rVqv2Kx2y+16v+CweEwum8/otHrNbrvf8Lh8Tq/b7/i8fs/v+/+AgYIqCRcLGIOJggMdjQM4CY2S k5SVjQkjkZaTBgQ2m6CXJAKhpY0CAhoDG4iKVQuSC5CmoZgimqUVnjS0lbYipL29FQO7rk+4jRA3 BAkMAxq9CxIDCcbNAxKwoBI1ztAGoQsXDNYkEBvZ4cKlBo/HThKU3TsE8qEXLQnblu83A+s4+VNh D1SxEeguBLQkwBg8JAQWdjDgg18lAS8IVADFIAeESgNZbKqQAiAoXQ+T/zCw1JEHo00hVySjZGAZ jmiSKMLYhDEFho08UyIBSklDj5mTfrkgWqkTjpenYgQlaJFSS6FDMIByiEPrJq4soFpCaSNZzBVT CYGShXXIBYM9QMnYUIqet0lnVaRVcc/ShrZCJE4iyUNuDKQgbXyUlDfF3hSLLdkF3GNlKJs6DMMg QOvqjKRSGboQHJWyD5yg8u3QvNNUzRqgW190EcyS6R5eO/SlxDbzps+NKlTlBNZF7BePS7K+/VQS geGN/vq2LUOSAAikgxcXKUlpi+QoxFJivmOb0beSV/+uHtXypskvarOf3UJ8LPI5cHWMXGn7/ErA ddATejDJIF9o9IUFHv9+MdzjlAfQdeDZJ+sh2JMHtbEUw4GyUXLhCqhR0hiDLXDWiF32lYbDcsap +Bwor8UnyX+TfKgCU5NgRqIMUEnnQW6VtHIDi9+p6AF2MPpnAofIicaCiQnuKANQOo2AI2M5EMmd gCW4N5aSo8yIoIcteHmclDBEBl+KHfRGIXUWmkBgei0w2WKUKUAHH5ou9KWjB1BW8meAAM5nIzCh jDiCnUXieUKKZPEJwzpujhDiJKq9WWicJrzIEQuMbkmmCkhSEqmkLriXKQlm3jdkhbIdKkKplsSY QqhoOYkCBFdqACaqS0oyKKCg+MgLrE1yiUKrxKmAq166mrCBRO4AG0P/boSdsNske445nqEqzFnJ qSU861i0IhDAQIYTHWQtDFA1Rletv0ILZ6wrsDvPrWJ2yAkqAJPm67sy8CPkCdlN6O0kAcpKgqcF nmAuCuw0dTDBLMzbQbYobCuJUbAhe6fD52TXiMIY9ptsxZJ0izEKqKFMAmKNXOwvw+CywGxOg058 QlrqmuzdyycEyjJeId+bLMkliEsTVz6b8BitoxKt3NHjJr0pvrSFcmrUJSTHZgfGWl1ChEfbfOfW S2eEtm5hNkKjdSdEuEC91vJXAcB8941KdqverHLbL1CdGKJyL2zkCDt3oCjBfT3IgtONVLryt5y+ 0PjJiHcw9+IjRGiA/9ovR9TyC0BSUvbamHMNA+U8p5y44MqaQHMHIJstgpdDO/seoa0TvmEonYBN woKy+6W7CERZvsLOkl+Oc+abvV0BUZ/XbnuSugMZeAsJZ+/5YdPHYLiW9lYtMSi5Ez3nsCvAzjTF IrOQTA2bl8863T6Fsvq7AXEeC/hDCdKdS2n60F8MYBe8RqlPW9x7mcY68L2lREx6g0tgBodXiuzN z3SbaN+7QgQ/BVlCgAd8YH0mYcASvW18tJufB8ZGNowBCYVPKhbqkHc1USjGZDDEoPbyBKMWkmhO FXzBpfinQUeZ0Ic22Fz2cJiJUMiQPCCE4gwa17sT0OxxwcISDhgoPv8XMNBx1hIX3lBgtBq54Isy WIjLDFS/XFmihA+70iRkhh+krJF+F1QBm7rYQ2HlAGIKFNW+WnC+HPHJJIukwQT78cQGrgCSnOBj mgSzGY8hjZEmM8D/AJOAbKBtGqvAYyYGcAE9NoUaXdTXEIvmDEMQTxXlwB/bvPgMDQCxcheoRgsh 8DYwHqNiV8RaEElAmsfdLhQ2EJci2aEoApyxAgz4YyCQ6cCKnSB1jYgeCp6Jvq5tMGxYAyMGBuBK CS2vCClKIhQgJgVsDABgsCDkO3dwJXHu85/20xBAB9qCK82RoAidYSUOmtCBtoqhDf1nq4wZUbMV ZBILUGVFl8cAi7j/Q5sbJRiQPhrSjUZDA5osKUCNqNKWuvSlMI2pTGdK0zAUgAMcUIAQAIDTIvCU A04IQE9hEAAHDLUEQsWpBUaA0wBkAacA8ABUV5BUK1gApxxowE6PulWgNqGqMDgABw7wgBMkNatM 5YBTsTDVqaoArFSgAAeMyoEJpOGnQeVqC5qKgqQaFQEi4CtbORBVF8B1Cje9ak5HwNMHAOCmDZiA AsR6gKWOwAIOKEADMiACoT5gsgCAawYaMNbCToC0HCgABewqVQ5YgLQF0KkK8CoCBFBArJ9Na2Gn 6lnQtva1qZWtBzJA1812Vq8ksC1ZdXpWDpTVBElVAAciEFi1Xjaz/8b1AG1bG9XeHqCwCnjAWCkA WA94N6rExWl2h0vaA0iXsK1dqwfCO97ywpW+ByAvCQBwgAIAQLwBUCxntTtXHEi3AB4Qr1ZFwNMC kBWn/X3ATTlQ3qsWIAI45axQxUpYsGbgphEAcYLn+lgOUKC6B3AAhweMAtoi4KYPCPF0q7tb+G64 rWNVcYY9MAHItve4XjXBi50L4gyId67UhS5Oe4zg+IrAwhjmAGe3y9uxThXDFxZvAQB7Y8L2OKs/ Hi6MZQxfwWI5Alrm8lDPnGYGp7bE1r1pYY1aWBuQVqs/Ze1PAStX6yLArWJdKmlP7FcFIACsci3r gbUb2vmOtbpa/f8zfFNAW7k64Mk4BaxbqzxXQ7c20lNFAAAiMIEJZNq8yBWBpT2gWE1b16w9latO BRtoDwyawEHmtAMMLWnLGpXQON01AkRNalNT2ANyPQCQo8rXXovg16gGqrM9AG0PJJbVfF21pMtb A2PrNAM4lS1tqQxfY9uVp8pOqn2HKmceB0C+QKaxvCk9VLHWGcc17u6p5+1WEvD1sCSwt7wFq2Sg grusfDU3gZVNbn0f29FBbnC0ue1v64o1ydx18nsZm9poQ5zjCAa3V43tVHVLd8E2eC9rxbrgcR+V 01jtKVzB2m8RTOC/Mef3pFs81H7jW96H3fSklYvVkqc64xknOFL/h2pUkRs95j1tuMfdHHGZc5Xo fPV5ma27XbxWtetWz/WrxaoA0gq3BqiFel1xTXXdRhsAcIf71Kfe7gm82wMYPkAE+qzzOredwE0W OI/DjXROVzzfz3buT5+OAoF728mw9qp0+/z0uMtd6vc9at7nTm3FZ33GA7fuxkWw+aqOHu+PFnla 14ph1LK2BpJW+2JdLvaoGpvLAGDuUcFK2ksvmq+jF7rfsW3dO6v60RAH7E1PLGmHHx7oPSU550mQ 7ORDvuA2L8CEnXp78+Ye12q28VElLVuxAjvIVZV+70XwXmb7mfAeML/HyS8C+VvbtcRfq8hNjIM8 BxyttOd2qOcA/zKme+g3VEklY9QlVg7AXz03aTVXf6lFV2s1ZA8Gen0mYe0WdBAIX5LWAEfGeCdg gTiVZEpHAnDVZ6+GYQR4UzolaWSlfeIXZKjnXGK1ZXP3gSFIfA1AZu7Hejj1ADcYfqQXhEPoZt+1 fSNAV5ZlA9XGcXUVgPOmXUa1a/F2hawmXnonAulVWUtWeDs3AgjQADDWhOwVbHU2AXLlX+LlfALo VpM1XeK1VABHAqNFYqvXV+NXdIxVhcKFWWNFX27IWDfYAOtGg3GIZvjHakblhVsnXw6oWYdIiKll iCQQYrFFcBJXUwQDd4AlfceHcpwILL0HAI1IAi44iu9yWjilWhHc1mqqGIuyOIu0WIu2mAYhAAA7 ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/icons/spacer.gif R0lGODlhAQABAIAAAP///wAAACH5BAEAAAAALAAAAAABAAEAQAICRAEAOw== ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/icons/shared/misc/arrowTtrim.gif R0lGODlhBAAJAIABAGZmZv///yH5BAEAAAEALAAAAAAEAAkAAAIKjAMHidsLm2HHFAA7 ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/icons/shared/toc/free.gif R0lGODlhIAAHAIABANYTNP///yH5BAEAAAEALAAAAAAgAAcAQAIjjI+pqwDpTAzzVPjQ3Nl2Boah Q0pkd2YpVq1mW2IypZ5vVQAAOw== ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/icons/misc/permissions.gif R0lGODlhZQAPAMQAACkpqSAgg/Hx+YeHz8nJ6ff3/ENDs9bW7nJyxqys3Zqa1+Hh87y85Orq9/// /1tbvf7+//z8/gAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACH5BAAA AAAALAAAAABlAA8AQAX/oOMEZGmeaKqubOu+qBgAdPMABUEDBl3shsYOICAOjgjAYTAkNgmMI3Og UzKLzsGB9jgIkgaGIOEsU8vY8HhHAhZ6B4ijoAA0mLtFHcCoPvE7WDhNhIWGh4iIbUMGCzkEDV+J k5SVlpdNizsDBUsDDAUMmKOkpZOaAj0Qe0M/NGdNCToMoktNBUcMCwAPDQYIQgOpSgwGdMAAWgAI OgRJbwxkWHdBSRAIyDhhZDSaREkPfQoGBgI3QF4FDWACAgVJCQznNII7BwUFZLC2WEXK5QJsKCkg 4AoNBPi2oKFxryAbGTAiSpxIUaKIEYQUECCQoIepjyBJBcC4Q8GbJgsUsQJxVGCBxybYmkBol48U LET1MGmKNvDGgDq9gAgpSYOAgCoRXAVaCkCBAJfLvDyQNcZoEgTcjDpr+jSJwkYC9hw4sI6GU6jd ZtCj4QAQ02RV2hU5urGIrSERNmZdsDECXIYGs2xhBqjm2kFNdfTzwY2NWgCymkLgezTZrjzchBGz bKVJTgVCjP3tvFhZAjIaH+wqV+bkgTqLVfMQ5A1BBLcPDIfczdvQyMdApPQeTvxhCAA7 ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/math/le.gif R0lGODlhBwAKAMIAAAAAACQkJElJSW1tbZKSkra2ttvb2////yH5BAEAAAcALAAAAAAHAAoAQAMa eKoV+4YMAweh7xRR1h6HEGCQMGWKIQZsICQAOw== ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/math/ge.gif R0lGODlhCAAKAKIAAAAAACQkJElJSW1tbZKSkra2ttvb2////yH5BAEAAAcALAAAAAAIAAoAQAMd GHXc/swMYpoho54i1hsdxBTBEIHakTpC4LrCkQAAOw== ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/math/Dgr.gif R0lGODlhCwAKAMIAAAAAACQkJElJSW1tbZKSkra2ttvb2////yH5BAEAAAcALAAAAAALAAoAQAMn eKqxSqPIKUBRZjDg1ghClzmV0RFWpwxGARCnebiwEjoGeAB87wMJADs= ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/content/vol29/issue2/images/small/zj40020808100001.gif R0lGODdhvwDIAOcAAAAAACCseynmGAgx5hjVKXP/AIvFCP///whK5kH2ABCDtAAIYiCUtEr/AHvV EGr/ACCD3gAQYjnuECCkgwgxQQAIxbT/AKzuAFr/ACnFWlLuCDnFKSneKZT/ABBK/1qsOQAx1VL/ ACmUgwAA9ubmABBqrCDeKRi9WhisiynuACmcxSBz7iDVORikeyCDaimc5ggxlCBi9gggYgAApAAA IAAASimU7iCkrAhB3gAAEDneEBjFICnmACBa9lLmEBhB/xhaaiCcg5S0QSBK/ynuEAAAcxBa1Sm0 aov/AHvuEL3VKRhz1QAAMRiDczm9SjH/AAAYcxhi7hi9MUHuEAharBisSgAA5gAQ/yCL3qT/ABis pHP/CDnVMRBB/wAIKSCUnCnFQSDeGCCDvXPuCBA5/ylz9iC9c1L/CBAx/yBa/yDVGBhq3ggx/1rV IAAI5jmsSghBIAgg/wg55ggY/wAg/xBa9mK9IBhS/zGkUjn2AAAACAAQxSCD7iCcvWr/CAAAYiBq 9inmIAAAKZTVCAAAxQAY/xAp/xCcc0r/CBhzUiCUxZz/ACmspCC0cwgp/xjNQRBi1QggzRCUgwAI iwgpzSBB/0HeKSli/0GDYgAY5kH/ACneOQhBlDHFUmLuEAAAGAAAg3v/ABCLpBg5/xhztCnVQSla /xi0iynFahgx/0H2CGL/AAAAORBK5hh71SDNUjH2ANXV1d7mg73NKc3NzSBS/ykpKQgpACAgILS0 CAggCIvFQbS0tMXFvb3FxTExMTEpMRgYGCAYEFKcYiCkxf/21ZSUg5SUlAAQEFqUYhgp/2JiUhAQ EAAQAFJaWntze3t7e3tzc3t7nDk5OaysrM2UIEFBOe7u7r3mCKSkpL29vcW9QVJSShAg/+bm5jlK QUpKSjE5GIODg5ycnCCL7mJiYgAICBAY/3NzcwgICAAIAFJSUtWsMUFBQd7e3pyUlAgQ/2pqajE5 pBjFQYuLi8XFxQAI/1paWgAA/4vuCM3/ADlqSiB79vb29t5aGBjFWiwAAAAAvwDIAEAI/wAPCBxI sKDBgwgTKlzIsKHDhxAjSpw4cBqAdBcxMovXjpmtYMx+tQO3blo5i+Du3ZuGKxi4diDTMbsnLljG dDjT4bp2wFZGZsFAMtNJb9zLde1GrjuGbiXJadPaiaMFDZdMZkNl5gRA7QC4mxfRxfv4y1a7dWjv rXvJcV25ePfAQcVliyWudTVz6g12oB6Av4ADQ1FmyEapJfAUXKiGzUmQbt34gWEkh0bgy4GhGbTI rBzIYHfVvgRHuvS9cuDKnV77ciRp1utweVyXM1i/ngA+Q1t3D93bklHbQZ0GLh5q4cGF/1q+3Cw4 rDG5egVgi94vXKVJuhWLrrtvpCTjef/vHq+8apLtfq2DBi0YRlwHtAXOISJBiCam+OU3dWnIKENx EEJJKxXMEYMZKpBTiTKjXJJBCYYo8wdmmhVEj2yzoUXSaKKldM8v6ZQVDG9IpUXihtNcl04wtsSz 0wG/gKQaccgRhxRUSZVDizu0xKLNSylGBWRwpE0TW0y8HMALOuLEZSRvHa5TXnflnIRLWW6VI1Z5 ap2mkoZyMSnONP349Zcug+yzgQRTPJCHGMrYYEMl3aTiRD6ooJKHCIbYk0oZjZwCCCN+IKGGGqgE sc8tevxVIUHg4GSWa6bBVpprv5TG2mtodaopbcy0I9A06awlUo3DtSYcOOIcE5c48eD/WGQ7cCWV FHOxMZPkkrbYQtJpVcaVEnfkGScOPW8ZFw805sUVF2/Z3SVOfIBFYkMeWSySBDvXiAHPHKPs8Qc8 MbxADhc+MFKGKUPAU8Q5qZSAjQNlYOGHPllYUgoQADw6kDgr+sqaa65piBN2z6LmXqnoQFPOWrRC 49uM6djSTz+/eEQas8HKdVZS6BxTjq3JrdfwONRkc8zK9IjTXVxWJZkNOu3YAnE7KiUrXqy40Nob e+Kwx2ysUCJV8FpIBaOZL3qAogcU54yChROlfMEAIMrEcQQJ1XiiiSpvCPIGNtYkUoAqRNhxzS6G mJLKOT1woEYKgRwhg78C0VoOUDeS/0Yp0p6mRGI8x7SMsjbawBo0PcdA004604xz20dYYTdcVLIK h1w5x2iTjThr5axayMeIoxpaMGGVZKTijEmjh0dpGU+nX3qZ7D24TNPb7OuMQ8sxZ4HzCz1JmglA o6AoY4YGDWgiAjz2zAHGIxpYokUjbQyyzSp5QOLDGC44IIIlpEQyxRYNcNBJJxjhfUA84lxkyziA gVNPPABUic46AIiVezu5E56G0KIelQAwHeJZmG36sQ564GQa1EgHzlglKXRgTiovadgxaEEN0EVl HLy4B0yEEhNmSKcuDryM7uIRFLiEiHfaMCE46HGReEAuHtn4BQCMdJ14tAwn5oBP4v9a9ZXcgOMY 00CiM+LBjHHYghz8qIQVRlCJSLggBA3oRBdI0QkVjEISElgCDDCSGYS4wxZJooga18jGNrrxjXCM oxznSMc62nEg/cAFZgCzjnqM6SxpOY15HiY8X1FDGz7ZY2COgb89knEdVOkVztizpWBBDBzUqEc7 /oKRTmImGNTAiihHGQx63E87QBqJllrXOo7cgxfZcM8oZwm5A8QPMHogIwCEIQ5t6PB4etELdEDz ymwIAzO6BIA28KjH3PCCHsT5VZXKIx5oHKtwQpPY6VCZnmlowyaXYWRgmOE5Jh0jG9rgBS+oQQ1Y ZuOd1BjHOOiRTXoUJRvjgFXkyrH/R1BiBXLuaBw98Fm6oB1jHO8cR+Hs2TqhHatlzEKHQKgBgFHW koZ/mQYtSJUVnNwkmCAFKTKYgQscYmaZAskjAJbSjnJYk3EIfSc9muKxT90DfvZkqMR8I8J7xBAw 4nQUPc+ZjUMiLp70hMuzTtOwnK5MnvBUGTTuQYtNAsaf6TiGO9jJi3rQoh7aYGc2koqU5QjpLddU mT1RxpTy8GIcWpnGATAKjUhtRZhaGYper7LXTt51Ra0CDEoPkMd0jIOpL2UZNr0DDcYVLmjdyeZ4 GuYyuEBShyL7yy+ooSUNomydj21dyxg6WlYeNBvqTJlCHYYUaqDjL8HIRkjGUY5W/8kTqvgsqliZ EqxVIpSd8QwtZFFTDl5cFAC/iF9HPcrcvIYUpH2FTovkM9h+BOMX4zBSzqrJSkqOh5rURAc9qEEL nvTDG7GIhVfr0cF7yJYejQwZ/KCKUHmWbmhiiYvm3LKsVqFWnV2NhTt4xItW4S+2CKyrlhp7znbG sylpEU+xTpeWt3TMUyOBpi3zJw5RauXDIH6uXn4SYpnYAhppJGwzy5ENnC2YcTrdknE2xd/Gogys 2qgHWN+aMqA28iKurS3KxJrQmDp4noX7bW7b6VUB06LAtChibkJpQoX2ZqD1LWqRFQpRoHmXSUm1 JDh4UY4Hbvgv2TjAPYLyzxE/9/+jOMnKT0JCC34CIMXWpQkz1uEbdDwHAHIJzK/UgpxfCAlawrGF bA6KmcwCQM/gaI+RntydY/0WwCoz3VuaCtGmCOdHnLyqbHERi9eOwyNU+Us8aFEWxd4XLqapplhG Zr978PW4SeTkSHDyEhDDOaSczAku6nGPywz2jshOtrKXvZBrgPMiucHFSjgZDMIFTZ7Q6JWEoZHP cthCJ7p0T6gBw4xs9IMW04pPPO6XOM3Y0x0qOwZhNdMqRsbTHU0a6zjqYc/GukMb1+hHLAjrDcIO XCAHZ0g9+gFvaBR8IOi4jXvF8dbGQoMa/WDPASLObIKU19mwtXWpOCaOcSCuyT7/YqRYjtHVeqSW F4WTcKyi7RONmHsiOHQtNehBC4HU4+EHObc73rjwhFPEG0nyRucO4I1+XOMasRgHT+YYi6dX2ze5 2Q4I8wmNzq13ZrJ5XDugoVuE0pNK5AGHTUbCwmxMXY3/PgAj6aGQa6CX6AJnI49uw8jiGsQ33hBH P7LRWKZ35+0PiYU3vGFN/tliZa2Tp2vRAULgjoOScYFf14FL+CpJTFlisciIqIH4ieydF7dRiDvO /UZa5H2N7li8NoCekHoM3Y0fxwU93VPboTkMALIxSa9+4d7LN/QYO9evCHMD6Mo2TI9jLr1Eqr4j ox9k4KtnOurTa/2GFJ2NsSiT/zdunxC7W98dKYbIx8GJC15s8hfGUc2yWB7WoEG2PCNxmHhKTjwO cqxKFvEL2jAN2eAjt2F39ZBeA9ZzB3ANCXgNt3cNX8UTPPFV3ScQ4zBwEHiBAkELtLB4HpdjQFd0 VTd0HpiA7qBjXtVzB+cNALd3hMV00kcQ3hBwy0R9t0F6RgeBCvFxocJ40FAP0FBR66ZDFcMb8GNO pbM/xHdPjwUOvYIRoFRs5AZCtJANOoaFYcVBvHCFXZhJngNWmXRItGBy7AVW2UALiBMLXbVVHqgN X8ULPqINPpKA5KWG6kUNPiKHXfVkg/dkXwiHaRiGsORyXYVO9YCFZ7hOhpiIcP+ohy63h2xIh3co h5HIC3TYhU8Wh1/1iFNnEX/FXODgGzESZ8N0FcEwDRJDKiL2HjzxbdC1IhiiV2xGUulRQtK1MG/W FV/xXMEQVuBwJb0yDbbgEQ9DF70iSZfnV75ILYoEGBTQDYQACjXgBREAAPsAB8+4jY6yGRlRUt+m HqhEaDjjJKJhK37zGqljMyNmGz0hKbAoFYCEM8BxFvEwDk0RF8uRFCuhHsfSER4mE9LRi8zAWXTx Gg8DSFzSKQUjHg+zGsKBFthhQ3qBC2WyRxSwAYHAAxzQA3fACGlgCGjgCmYQCJbwChsQAo0ADzAQ CCrwAxPACZahSO4jEqDiNzT/hjQOYyvo6EOAw5C3IhPX9RUv8m0gAWEpYmgjsSrd0U1RUQ45Jg4j YUCpEiyjNJCUVw7w9yRI0yXdURajuCUsISVi0ZWqcY7tMFPooBMXCRgGMAZVYAmL0AF+0AH60AmG gAWMQA6pAAqsUAMucAm1YAOMUAt10AlToAnzAAiAMAmCQCEGsQ5YgTPqARvQsiEbcoS2ICyfwhqE lhIcVVKENQ3MQBw4ciPDkRTD0UdbRYdSeTlPcilnoRZBkQ5d8TnpASWroSHAsiwNcw87VU2fxyVH UTTEYQtvQXe+ABgykA8kkA+asAoYgAQbUAjK0A1uMAJzYA9G0AYG4AA3oAyg/zADVnAOtaACjMAI GqAP+UAKIIAFQTAJ7WMQkTIi6diZG1ISQ2JoP4k6coFKEDMUFtMPpGIkpnk5mqM548Aj2YAW8cAL A5Zd/Gg0xeEbrzEUSXJ5uRMVneIpNTZIb0F84bEs2kQiP5kUsTEtfqEHkZAJe9ANtcAIHOA1MfAD inAJozAHc+Afw9ALszALQiAHIzACV9ANgDAKlXAEofAGprACgbAKqsABFOA+4wASNjMa6Qgb26Qd wZMl41VUR/Vqq/hoApExkAMtZoWgS3k5UiIXEYZ8YQVhPrQ/z2Gbc8UUZQFI2gFeqNEb3qEa97Mz U/IwKOMwatEpVzIt8vEXOf8gAxtgAW1gBoAAD3FQCaYQBahADuTQCWfwCnciC0IAA5iwDTyAASGg CQ+gAQKAAaogAZ1gA5agAengPqJXV9cFMTZCKaOBoqsyGg/Dm0Q1Mx8BPBaJGx4BGleKOSkCmuoh MqtCHKBjKz4UK5gpS1bBixfBDCsDQJ6pFlUiO0ylP8chF1qCFm9xqMJTSu+wDnzRC5jBBMqgDG5A CIbwBUcwCspgDyOABpaQD9iQBAHgBA9QAPpQADygAqlgBcrwJ5cARUBARu4DQJykHaR5ETGCCy2R G2alaBhrC/wZjCCBC1pCmswFfDyxHIDxC02xIjkDi3FWi0DRKxjbUUAhFMH/dGcwohM1BE4S04vT Jm06VLHANzLucRcg8mguxYoV2ReBkQ6N8rTHExh60CiTMAcV4AUlEAw1IAMUEAl/MLWN8ozus3FG Cw4dd7Zom7YSEW8r07aKBWM5ZU8LFbeO5bZ262r0EAuugxyac0HKiqBQkaY44rd9W7gcqg08mbiJ C5vIMSTJobiQ21Jz1Yoj1rRSmEyUG0zV1UyAIQ4tNmhTYhy9Kahilh4nFkHj5mOpCwAK1VL1hloq GFbvhE9c1lDfAQ60wLmfRGX/9IvZsG2UlVSVVnmPNWungVKvNUtDIVcYRW6gpg5hm0tx5lzTq1dD wSLUMIR7tLl/4VLFsW1B/7NTjDWcoUtIJ5IUtkAN/ANtAOBof3FQmYSJCRgLX0UNj2VNDWVNvjdQ mJgyU1VyqetPJpRmLrhO66RaCtW2QaNUm8ZgnQY/STeLx/W+pLlcIma901uyFakNusu9v5sl3JVT laUayTJh09Qw3sUes0N5QGVnuFBnNtaHeahWzKISgjQe1nRaDpYyprMOHHxVoYQL2jB/t2Vfbusw clEXOANmT/VYLdNQcfG7FcW8uZENGRyKNFtifQUWQulaf4FnspENxCFI4ZXD9jVTSvUz9wRgb/XE 3fUw2pAO8KVZ2sBgDeZgtHtavBU6YPZZOpZOLec5M+Ve/BRboTIOfPZQOf+FiGKYT7AmKzUmWRNW Eiz2C3MFAGlZUVpcsjcRkNB2xZ0sQcjHTCtFD6kxJXa8hIzlMmfpUnBadRfjdOZXD+MQD9rAP3Ps KAr1X5jWOGJRcux0eSTMJAn8VG04ELEwU8ZVxUMBK71xLFElVdsWY+LrHaJLGqJzU/WQimdmTcvR Eh4Wyh4FGMwYUkEBGtOwasEAxqHWDlF2U4nVb6LbIfkZSN/qJSnxU+9rZ4GBT/fYTus0u0Z8tywz WnU7U+1Qy4EhwP9EccXxUnJ7WyzTOjKmQUUFS9gUD0fkQMt7yYGBDtQgGxoszk3Lvh8WbC0jWKTM Kvx0HaUZDL5BRrb5TDX/AxTehowiQXa8AA0mNG5kFFSA5jAVk0icpB6/cAy0zFo4wiIeexbHUsfb aMileRElkUik8R7E4VWJ0xSeW8MhA4Zkprz/RMUbYRwqyxu4gHW5IUGe42048QvlgLFBkSmSJ2UA YA6BMVgg94y4UDq6i7mXUTHByI0V5XYd13R4x4Fqq35WR85lUVGRBlUl1yOE52eqWCUmd3nc1iRJ 0dNYQc6FnXrKVib9AHMHIBUo5g3lcHmB1xUL8X1rdA0QWg6xoFZYyHQYR0frt1K4s0M5HFr0oA0p SDP3AFPQAK2ZdF/s8WqREipQEdoTcQ0r83RNN2DBXX4D1nqvB3cMeLaK/+cNM4QOoFFys9tf46VO OaYN9KBfbPdfWibMLlVZp6ETvcMLMwgRPEIL9yAydYwQi0d+bLRwAG567sAT8LVu2TDgVGd35UAN XxEMvccs0mbHmPi7T0EXhlYaSO05vowOvsLK92BC02Df4LeA9z0QQ6fYEVFeCh4R4dcjKo5wMf4Q 66cr0IATp0ES4sXLB5VPx4LEqajIolVU+NgUyaJDthBWiMcLtMcQDFfa240Q2Hcb1seDD/F9Uyfa DmF3B4BxMW53NRiDBxB+6LfijU1u9AR8fLY/99hYKdMyqmE6mXJT7tXGrQLneuRePlFuU+eCJ45H UFcPCTF0TC4QSCdHUf/3ehPYbD9XfsJNDe6gXrbXF2NLEP1Aev/2cB74VQPxVW53DR5hwGKMGSIb T25NRsygdsjUIqVDKntUkPeDT+iAid6gYxAY3LFgv65tISjDhgN1EDiUXmOVfnh0DAI37AnRc0FT YOKgYwJWD9cwVctED95AdwUhgUU1V0p3EPh0MVolbwbRD/LWD9Q+DglRdWVK2OR8XTiylDWt7uTG E8+mSDJhFexrE57USSCh7l1hEdxIavWATYEjJV233q5O2MzgjPC+8IRNq+TMXMxgOjGLISQEGkBh FbVpCw2D7yCVG68YyqREUi3jFodkaM5CTVrJV/9kwTjbizmhVw7jsU3/jSIxoqdFUhZ0QRN69FwW aTyKtABlUAumAAipEACv4AIGsAFoYAh7AANYsABh2/De6Fe8cTC80aUlkh76xSmvcQ/F2Bn8M2JF +RMxIRXwdRpOiYR+xhxmhRZQsQ5GyfKrA21CnA0tETzp6JAOKkLbcTpJgR51sT8fBXwHsJwYGQSl oAITEAYBcAlfEAr6wA4XsAM+cAE+cANgEAJ4MAeRkAPc6PAsixVgwpDFmePPYjTu7jfTEBQ9Y0Ku +I5AUTnnoZrEUUBISGs1YlapITxG8vUdha2QY0EB9CvmsSxdoh0nbM0JoxYpoiVL6/MAkAN7oAIZ cAN0kgqpIPTnYA/d/4AFKaAGKlAGawAGYEAMQ1AJL8AIZgAEnr9HDq+VO4SuA7QhUyl/l2Qp2bEO IDIbOOGOvn8WADFt2q92BQ0WnNZuXcJ2v6bdu9cu4Tp00OKBkwhuGrNgzJgBoHagXbBj5QiuW1cu HjqWLKFVfImO3suVKtFBvIexHTiMA8tBE/cR1wFtAIwCuJUkxCsxpkyVETGkkgoVqWggXaEn0hc+ NhhJkFBqghguPjQBOXoU2gG2bcHd88hsGsp1OzGifIsr3S+6v+SW47kOYkG6C9PhCnbPVrBrB8B1 bCdO4i/KCXkKXJfNHa1x6Kh5o6bRoWCJcwGDQzetY7p0IaG9vjdR8P+9deBQlmNZ7t4vXLjA4Xp4 L57FnG8BT65bThy0x0TTNvuDKZQGHprMrKkF4UWZF4wyQOjEpRI5p4xeAcLSSUILQGIqjIKiR23b ttA4MrOFNzDtnHXfgmOmt8T8s00wugpECReOylmnMb9s0U2chUoTaLKMtPFGm1hoWUigCntyqLKG 7uPlAGrQiWennWZbCByVylEOGpNwaceWaXCzKJ54jnuLtgIFoiceepwzyoVZqhnjDUgWuUCMOSqp pYIZuulmF1nsUOUIe2r4o5tRAOHDFDE4mGACH7jggoKrAFiLvgPWSYcZ21YMTL/aFFrHRnBs1G+/ nKbhyTbV0gkmHrb/pjlMnGNS8/Cg0hTShhqUshlHI4YksmugRh9jpsSZKOvxrboE03GlmG4aZxzh WmIpHtpQEm5H2qaxBR1xiKRBDEssCKWDBx5wwAw0qLSHEHjGe6GUUoghRhl8ZqigkjJuyKeaNgJJ IRR92vChCQBudROcOO/EKNA6+wTnFz3xetVAhc69R69f+jlgo7omo2zFg+aihxpedoznmIgK0sm2 WTXFyKMSaYEGF4J+/IWne1TSEcUX41G11JXQgfFOiMyN7Zch6zGqBnjQcKUUTy4YpBQJMDCiG3i8 HGIUfgYxYIovlBnClHN+uIELHOzB5AJLVmBkBwjgoaFN+sqJ812e/+yyreramPklmHYGC8YWXArb GiVAYfULv376QRSXciisENOCxIlFm7UPgtudbFz0D7eBdQomnRJ5QccWW06qa2K6cIRRsQ4Fuosv i1A0HCK8sr61Hhr+QKaNfFDho5ZulOkmlUqU6QELUsbCY5tyOAlHCCV88GQWdgZRoRZTRqmllkAQ WcSCACJxuq1sWBPbRqp7RMnGVwMVjKd+tYGeF+mpyYYecdbei95fxsXXw0svKw1Bhu4Rh55jqKlH m3HEQWe23fw+oB567tHaON1uW0kxXGy5J6Z+l2tVxGpjP4jE407BGBkA/kADQRhhCgUwgxhcEIQe uMEDPpiFAXhghP9RdAIPuIACJiwQgiAowgfYkAALGAGIIWChOyxQgyEEETy2rOQwdzoe8w5Ep9rc gyX0yAYvoLc+i4yDGtdDzDjYUjZmoGQ4dWmbQUAEqIxUZGLQMKKEhISORa1DQZ06APmyQT8VGWdH NpEVRDomIxYp6EbloEf77hKbr2kjfkaZhBdqUAE75EMUoyhDKuLghkiUQhRtIIE+SFANdlTDDt9i Bzv8kYt/vCIAItCAGjhQgjbMAhxsctMB/AKNdjBDRQQLFA/t0g7g0IYlxxCHjsYhvfXhpGvHAEdj bEEoj2htawbUVEKm4SrGrUQjLvKRRtqxRXGII168LNEnmWG+rKH/ki6lYpDBIiIcjdQmJQCT1U0i EoxnUKMc/fBFWgCQAyOo4BWN+IEV5nAFQ/ChczboRCdsUAZXZCAUOpCDJBrAiQV8IRVzMEQrRNEJ HyTCKDSsFwBYkw51GoWiFqVoOqhGLv3YJVb0wwU90tYYXEj0KB+RKGsAwIxxxCNEPjwGL643jnIA CjOkNJiCCNU3AJQIUSq1qDovWtG0eISo/5kGjcxX0qGk0yjAcNW47mENAPzCKLiAQSqGYIg5RCES J9sEBhaxhnMMYZ+1uEIJQoAHc6QFokLqjS2OYZRg1EMc9eCfTX5xDDgNrjeHqVM70jG4YARjGuLY iEpxcY1+rOMY/4LrDS5k0g56/HWYEdEJStoBGIEQpBwtjQsvmTFRkBzARvQ4SjpQBA2rjmOuv0Bs M9sBDaPsqKr/oi00pgEAwQmHtRM9DFHowQzyAWAaxvgGbdsxV3FoQxgy6EEqDDHdPAZjEnwowzlg 4Ao5fGIS6ZBPRSFayriU17y97EhhgyEg9e7vLTJKakfK2xFbkJSoFY3TfXo5Wv721yMqzShwSzsS 9XqtsBxRb2HZm2AGtxcx8W1wMH5xAKfe18JGyQEAPnFhC0P0AP1oBy3a4Y1QltjEJ0ZxilW8Yha3 2MUvhnGMUSyptgRMxjfGcY51vGMe99jHPwZykIU8ZLb0o6QcXv+H+sxHDW1kw64qMZ82xHFkDhvl GPGoslHuwYtjHGMc8suGNqyXKpnuNst0pcZ5zXuYY0j5JUbkskpiuhl69E3Na57GAVDLYZJow0fQ 4MWJbAEMbmBMzCitcomKTOWhrqMeDjEOcWaDG4slk1broAU6iHrRK+PXKOWghi02WxHzuXYmzUQ1 TdY2DS4jmqjByMZ5CXWMbNzIVs3kGIxkQg3qzYRBCokHL9whrvLmNx153rNJjcIMWAJAD8Cd6Gih DVyPlIMWnwxqWuy46LQoR4CCgdHFKoJrl1T6MgP5BS/ukVqTdprdV+7f+XiRDSbXI33Ro9446LHv fZfPejWlFLv/jxKMNMcFANl4zTH2HZPyNdN6isqGwl/DqniIQ9gHuIZeio1sdY7jp6SdNrSdPW1m CMMW8aCFLSq67Q8feWJ4KuDGWuJw673kNbKa450ilmyTBiktXTbiOIJo73oE2rU0f3irbEVrWjpz GtSg8sALng4l0oJ6lEoV0Omd75mw6jXmUzj74tGYN630vxw3Cj3IK21pBzjaxRZwSvM7P6OEl+VG 5i0vtmarUmc9lgwCt+QmdutWyepOrI66uwHQDsDNEnpEj5TCIafG4TTzNeWjNb15oe9yLEqdBB/t PQ4QN17UgxZF57rYJefNwWP+4S9h0DEOwAtjo92HZgcwyflr/+w4CfiiFDUlNbTdFrzbghdvMRU0 wK5vNsJrYjYBiuVjQkxnHqUkR2nHiYBY79PPG5YcY5dKXgOU5WeD3pRy5j3GkRbQp6McYtb89PKt 7/bVaH9zqbiihG5qWn8/J7zYpWPTMy3DspDrvbb7L//qPQMMkGwIBqO4O704EV0bty5bOBTREfK7 QHBCEa9jH8AYh3ZIOywDgGDQhpWAOCHSBu+rOa8bv/FLQekJtOoxIDiSur3ghZ/osh3MOi/Tt1gq jpRQPh/MutdbDldZB17omzwTB96iBtZou5SyqN1rO5JjQN/Qho+IwHTAJeFYDstDtS8Mw8cZN7AL u4kbvx1Zh/+AAQCfuy1bAQqhEyIW1D+JWy3lMx/z4zVqcK1UiaUB0oYHJLhjUxX9S5VUGTqi2zwg NJiKE7p8CzvL45g9mZ8JQy3YWikoHC25AzBOrEK52z1N9Ah6KAdF+7BgmCyDgZGNaaYdtJ5cG7zL a7NY8IZ+cAfU28NjKCIAuD5eXJS+wzpXRLWb6AnksxUj0gZaOD3Tc4cN2TxoCLUSHJ5sYJCf0ECh W8H1qbRkSiVYwY2VOCPJmQaLsEQAqBRjq8JpM6kDXEBQlDaMAgCKMEW8CxjkS76Go4mNyccdITVe oAUSw7hr8IZmtDpouIcsdMPWgCWh2zrqgSVXGY6bw6xYdDj/L6O3enAHb7gGu+EiLIO10WopHyof OOu1iyAV5ZNEiwk3nAsUyUGNcShHXlwH9YI7iQrFdeRE3YuLYBAGaDgf4iupuqK8pds/XQSKNOQP HUQfWjwbgRTIuBEHaqCtNiTBeNQGLjIiXqMUhdu3IIIeaogjU8k8XhMiWnCHxuiHa7hKXkCp9gMH XqAILDK/R5w/89m3Dhw/l0DKcNONe6iH3UK7X3AHetgT+QI5dRyqlArF/FqvX0AHd1C5CDyKaagH 28ARfwMgVykOu6ALNbo1u/Q1bTAzXiyHjpuffvGXq9s/hpxLRNxKfqM1f3nGmWC/gvsIdPgXeuhB fjND1gw7/5ewnlfcGCRUwtFCu08TMVvYxGxDTIwKsE3MKKyhPQgESqFajfwqr+f7mIJBELwoDVuA RnXqRXXyiHs4H/NDT6wrwuW4PLB7uL9jvHWrzeVctpLSxKjZSjvMpm/yySCKuJtoh6vEvXYYQPLU i9QyQJBTtpz0CHMavkWjB6vSqBQJqbmyhWR7GIxQl95wCIMYiN6IBQfMNisrzTbcrQnFNmbwF9sQ kiBqOgNSxdzYjVuhh4xa0EG0BXGwBdcKBl6gh3qoEVpoLsc0wzN0FeTLJsd0B3SQNSY0ribkLZko n4+AiAOzSewMLZG7B2+AmpUjPtVIrwhrsF4Ch9fAiMjSqf8wFdNgGAf7WNMIiwtbAIdxiBQUSSX5 etPCMj6vERDE8NP1Wi809VPyQQeN4FPECNT9EVQDA4cCPTN1uizEsNEzY7lrUM472y85EYf1eQzE 8Ij8KB+X6g1MnS93+DB6OZtTRdVUPRtWddUPg1VXbdVZbdVTtVVZVVVcLbJapVVendVbRVVYjdVh DdZiJdZjNVZ6wbgHTIuLwoV4SS1TCpTRooxf0I2sGaqviTryzAZvsBW2ABxaKAfT6wcNSRVljAd6 cQdtaAwNOYDTi4VyvYZ+8QZe6AdxqFdl1bF66IdY+NF4MFW2QAe2uIZsoIV+cLJ/rIdxuIZ6MFgi QzFauIb/a2BWuhocipJTW1GObNiQWNoTH5olLqtBFBFBieIpdmOGbCC7YYVVSoEGfUVLuGGLWJhY +qAFgvUGjaQXe5UbcaCFIbGVejiAsFsxiQ3YEyM7fcU4+lBaiLXZia1YhzGKw0oV2cRIzui6G4EG r1zBJnucWNolAEAM9lNZpyW+egDIfe3XHMsQofswmapXdFAidEjbHpNYipWomqoqOPIXMtMQ0qOF SAGiQHuNFbSIF7SImawqGlm2sn2xTGMLh/2M1zgAOjWxa/hHHjPaHLtFsw0lmr0G/IARj0AN9pTS NrvFbHAmysgTcBCSQKMGXWwVwFAXlSOIeOgIx8WxpkXa/2bkMX6NBbbl1xQLWLJLW8aKMdD9hddQ jWa6B93kvKgky/QRh46IB6+R3vnLzFXZjRIUFYbNsVioh3gAyGsAB5ajj40M3h2rh2tY3xuLheCN 3bYQh2ygD0M5hmmQG4ahxnqgBkN5MZrl0ushlAw0H4OsuDm0N6ksNxTpFyEqyTVkn2ZyFZWbBnRg V8692aM9MY183xwD3vDNWV5Y2RO7uB0TYGj8JFwwlQhNh4pbiTCbt7CMjXT4JkNTH1RTu52QLJei qOwr4RhrRrUM4lA6yw/GMX7l4BiT2A1Z4hKjBSS+MYn1hgC9B9b4BR8yn6UbXB+lFeOhiJtYCMFZ B4irHv/I+QmWsGASFuH4rdsSC14plrEQxrHgRVs5Zlrf3bG7DYZStCr+ERLnTYfdsBHgkF6+EpzI 8EOOkYgm0pqFkxGKGgeJ5dx+QNsndpPgdQd6iYVbVNabdbGDxWMWs0WicN8U28gljt8iVjE+rqrQ WCmF6B+UpIcZlBG9dCkX0c2Z6MomY59cEywA8AxqWNl+MEsU+2BaNFX7XeUTC16NzNnPTVoVU+Ih JgpM9obMdRONNJED2GQUg+ZmDN6NtGRtZjFXviq4LEGbiEHpmZ+JWwmFqIilmzdtKLpX/CRb0IaR KNvNOAbe1YaMbAsBPktq0Nd/3Ei2IMiMfONWhldQHr3/Oz5YjIyfbk2xm8XjtcU4iDY9b9ZX0CA+ b17Z+KXZ02tYb9iQd51YU7qGLqOFWtvnEuyfv/s615K4cgiG2igHelsfV7myI9NNlEpZ3vVctR3l opaxuzWlojCKWChZ4jpBZ92shNONu4EjXdSNUroqXfy9ZUMfn9w3WPrCfvCGdTDVY5CfEvPREkHr IQmlMBuSH3VrN/kXtpDrExPa4ATr9rTrxuAFaBDaUNo3EvtriKYPWJJfn4Vi+zURxTYxStY3hZPs HaTs5fOXZNTIiaWXp0S9yKbszx4H8T29ZSS60iY60jZt0zvt1F5t1nZt1Gbt4F1G2G5t2l7t0VZt 01bG/9t+bXvDbdcO7MoNw+Em7qM8yi+8POJWbh1mC7FeblT7QYeT7uembpn10c/G7jbjw2Sc2Mzu 1yab7OzGbiVaUlYx7/NG7/RW75YwxbYws0mdVKwRDt1Il3QBFYrZCHaD76vSpZyMTt44sF7yGoew M8ZUl3QUuZD4pMM0Kdx0rYVwiLFJnv3hC82INblbRwBrqirTA0H4hPCquwx7VHWCqBNVx2B4jbKJ k+vsGtZYjSYyyI9wO9LCBXpRuQWcqAD3CN+AkYa4YJ8UB05Z8f8ywNJacJIjRUXtjcgoiLXpUFoJ VAgJCmqDtqEgmQszhxgwBHuwhzlQAVFgBQMYBlZgBf890IMIEIRHLXFOtEmDFFvlHKysGVVr9aL9 UpBCqV4MV6z+rsLQMiV60AZSEgyJ3KyBwZgrxlIqbw2RAMU4QaJ1UJd3wYu5sEwfqZGc6BpwmDL/ FlsisbAiAAQVIAYsUIYY2AALwIYj6AEvIQUdwIE/OLM1B7mL+FTMKIyIyI9R6aacqBHW8CHfE1td irakGq3MkDIf/xDcaIhqdQj1q7/Ruk4oNHK5W17dkhMfNyNeL475DhR1EQivoYiQGwqmJiooCAJi MIUb6AQt/4BFkoV86IA3cIUyIIYUcAFCYIUqk3UsjgdoF5UdIo2HuJc5AZQ5ka9nksLFMi3W8Jq4 qMb/iGCOCoEjagSRDs0JghCONVMpBbdJQGelubAT1y1UjAjCVTmN5sGXwMlE1qjxK1cnQQiCEOCB RoCAI5iAS1ABLjiDJAiFApACO9iCf2iCIJCAWgAEF/CCC1tziaINsW0X75Rl2tiaVFolgqGNwtoT DXcQ/pKvYXoXkE8IiAAiwJgVgiiI0aipugjFj4gmk40NwwqMjhFCGO3M27CYHSmXurAR3Up4Tz8K JnCCECgFRZiAIUgFPriEOQCFc1AEMLiBRiiFV5ACBrgBRSgDUwCTBQBxEg8lMxMIQhmgcimXFUmJ sPmYU3oXhdieJgrAYF94jhis4aAQs2cIiXGmR4mi/xrJF9VIwGlPB4MEjrnICR+SOR3xJjNqCWzC uRpR9pRqeXWiAD5YoVEwfEBwilEYhdBJBREQgDJQhlFQg3lQAangB68wAlD4g81/qM6vqnV7Vjzp k9B/vvtJJZ0gGIVAlGBIF5M9G+Xsm2UCiHbTwK0rt64dQoED70EzOO0hwmkJI/6atg5XMGbM0gHg dQAcAGjQbC28t27dvXjoVqKDFs8kunv3WrKMF8+gQXA629lCFw9kumAH6gEoWvTDA01m+JUxNUqZ qTK1zv1hlYqSoAqibPBToehSmUoT8nBhsECP0aLQDrBt+ytYPI0EEe4kCM5ku7vrbOEFF28dQcAE p//9+rWRWbC9AHBdO2BLIzOThR9WnAiO8sOH5bJBsyjups6U5cTdQ6yx48dpLn+ho5vSZrmVBmna 9PlwJeB77VDiFciQHkhcB3ylxZRFxxdSbxpZ2uKk2wQzQ4oA0ANl0pBLL1TYsNQgRINGP2IoApJW bdu20+L9SpfuF+DdJ1Hq7Lxb5+Bf4kIflBj4ni1B4RJPOowdMI1GtsSzWzsV+XfZNOLEkk05CJWj jTvjrCPRPRUS9uFAGGmUDjUHrCPbL3nRJ9hdLW2IC3wvttOhT+XohNJB94DToDgtYTRUWom4UMID CWjghx+qfDEKMS+8YIMp50BiSS38jMIHI2Ex8Er/KROYQgkArOSQ1lrpHRAPgunwZddJBOkYWHu4 HJTONH3hON+GPAUVzEDBNGZYZK01CNFchI7jjTva0BILmoPmJdBu8dBDj0GQkXjAMdDQ05dd9wxU kGwG2YILOLjMWI5NL51UTjm90SWOS7j0Q1RRzSSSzhtI6LMIBmLAI0aWqdTSiSaoWIKKIaZkZ8MR TmhRyhSeoKLIEEMQQqaZmDKDCzMz3hgffTfqWCAuuJj07bmCAQYYYnD98qdG4oxzjI0RSaSQQOtg WI87vOwI4r8fOgiOLcEEk45HvECDzkC//Mciqyu9NA0u09jSjk0uofomQTntyJc49MTTD3FF7VMN /xdJeCIJBlP0kCw/3eyRyhJK7DKPJq+cM0MFo/DzwiUqICJBCWZgsEgHiRhVZnqlqWkLfCa1eS5+ JwVTcZ07BqbXtzu2c3Aw7UDT2GMz1qlqZpCmOJA23vCiTT2eZvaQ1xBVpJNpHo0DzT0On9Rhq6Gh gyo0n1GsG8YiDX4Tx3+/2dM42QBZlAyD+KMED1IkQYIQmaDRjRVWdDNHDErQgs0E8NRQgz3KVPIC U1ysgsoA5yjSBCjomblOOszYJR9/LNrFjC28z2WXm/Pdp1vvuNDDVnu/1Ds3RBJRVk49tIQcCzUQ 4rvOTXnm+RYzJWYjzkMFyVRvx6iq5BI69FAzDv+qLbkEm0nre0xSyJPnsAQXQpGHBCwCCRLAQhzi YAg3WMEe51gBCizhBH5cwh74sMI5yvACRmhACRfg1RjG4AEZBCMk2WqHe+6TF3G1iU0CqVsL8SOT GbFoTsxYxwH6MQ065cVGaYuIvcCRjXHoyCU/HEhCKOOgi5TvANSARmF6KBN1uW9hqSrHMcQRm/ex pFW7kUnVSiWpyX3CCEfIhz4s0IFB0OINhpiDIeBBCHuYwkmbCMSTatGNIkwiFaOYgx224QRG+MEC JLBAFvAAAGmcMCj40drWhIefBqloXZa8JJ4ONo1+HMh3gyLM2iYSkXtk0STxgMZlHqUiSALMYB7/ OUY8YIQn5d1FJT65ScY2laravCSVgZkaSY4xOUEYwgyaCAUSNDGGfGwAEvAYRQVmoAw+2OAFnkjC P7BwCVMoYwaRiEE+8kEJNHQCEVN4xD/4sQcAiCNb4OhdXXbUpkvW5RdQQxeeHIUuXKiJk+3pkG5A JEqBQCMWtKCFS46hDWpkI0+ipB7zmKE3cWDtl+CwEUpiE6q/rEMcxyhI/QYXOKrJ8zLOO4A2igKK CrDiHOT4RxL00QYjWIILlehGHJRRi0qsIBeDSIIIzpGKVMxhFGVAwzla4QAliCAVR+DAEWLwB6a1 5R4A8N3vqjZPgpCkYnmpjLfANUvAcIt4/nwP/zgc9KEkWmhRDbUeOurhjVgIqj8z+stufHIwj1AD HaNaYVot4qaQsqog7cCFLXZ0LpHglTdTbEeFcNFObdCAFTIYRhu+0I1LKGIFruBDKpQBiKlUQhmO 2IUslMCOfuyCCkbAgwiWUoYbMAIQQ5iAJrKQBQ5wIhrZQkfv7gPJFs7HFgAIxi/Wd4+DFQhHW7vL FPdi1gPEqSVoiiIQJSKOegAHX3nJRiyOcQ960K8gLwFoXviZMCgmFrIyeW+HVMKei/KmJR26aFoZ F7hvAYZia+nFJ3KXiBgYghxAU0YqDBFaUmBgEFnQxBbyQQEgAOEWuwhFFi7gj2p8IBWXGMU5ev+Q gBSEIgmv6G22jqGRFPkthuv6Wtik5qZpGGwzvBhHS8Th0WNkqkL9pK6aUoleSNkIuviaWw/pcYxx UIMX9dAGL47hE5QEyCPa2E8w2BTfLfrEFiSB7MIKJxIjJrdNHGqVbg6CC3SgVA+gqGwlzDAFWMBC E5oIgj3m4AR2UIMHEyDFDhbwDSHY4g0NCIIY7FAMB6ygKakYAiMywAU1IEAQVGVLNg6GvEfqBM1p 3qq4JMVQJ2tjodnIom0A0I4cAoouAimMWv2jEP/EJiHrgMYxsgFlKb/EfRe5FDXogY5gXOyiUkOJ SsoxqhPJZhwH+IzEOgQYaNw4Jm86iC2cTZT/T3yCCaIIRSjEMNQrdGMUitDAKo6ABRso4oyDEAIJ HIBuIoQAAwI4wyYY0Yk8vMIGN3hFALxw6QNQSk0sossj/ya8T5lXHNQo9UJ7tBJ69GiHZdqhmtC0 EB9CZCKTOQlCYKMpXcPSJ+jYm49Rg+tx2NPMMpGvyRd3S3CM4yYL2wvg6MGL8yoPGOvwCK2qAwpK DAIbnQhtKroBj0wEoAxH0MCuLFEKIRgDAMmoxgWw0QEeqEICPNBAI1ShikYk4QMmNFNsuCWneAIW T1oFTDwyRY9svI0X9MC1OOi+KWb8Yhz+PBgz7pUZB6kyIuDr0V9mQqO50wt8ViwNYjySF58g/3Yg F23VqqbMG1SxykOFMW5kcgLZCukGF9BwdkoBEAlm0MAKDB4DB17RCnvkWQ2lSAI29JEFdlzOE0AQ gQX0kYtskAIMTRhAAzDQiTu0wQC3OHt6CAONqx08ITeKpEm/SPGQHePhnDEI4DDSErIhpndI1Bqk XJNWyMIqIp0+CMYidwxK9c09Ev0IAIRYDv+Aq5Y20RFxsYqOSA0uqco4IBRXfcMQzcp5sMIoMAIq nIDsxQE8xAEYNIINEMMQkAEWkEIbTME8fIETTEEVJEIJhAEOwMMETEEDqAEjiAINQJ965B86IEZ8 cBoOrlD6dIxPdNo4QFmUaVE6lMNIkM1x8f8TYr0GKFmPRTiEw9QLru0f+ODHTbAKxmVEibRDUawD LPGdfKwK54VPYYEUJL2GeZ1LenHhOAhH6qWFF5jCDSwB7c0BGXRDJTzFJXiCBUyBBHAAF+jDA7yB DmgCKVRAuSmDG8BDLShCAcCB7kQfAKRDO6ADt0hEZaSNJTrMJPWX8rRJ3MHKO8XP1RihgDSPQsDa KWXGyTXe2fiHRXSX1/DTtjADAGRhJKoaqxBb9TBIXiDOLMEIYeRHnYDRYcWPOCwGSp1HUQiCFwiC IMgAHIQDFDABKMBDN9wBFTjAB9ABGWwCBmRBKagAIKTCOXRDBUQHI9QCdTxiVY2Djl2UUTD/QzYY 142Mw6TAiPVtiJsQRIrkD4AEAz2sw0YwRj+0QxbRw3ooGWK5I3AAx4foCDooDNwcRFrRXGuY1GGU FWrwGDVQlDJ+ZFEEnlHED70YRTn8Qv4dxC84j4ZEoqwQBZ2MgzWsRDtkgzW8Aze8g3EZgzwMgRlc Qjd0gyHMACeEQgiMASXYAz+gAiO8wgRMgCU4IragHS/YBj0YxTHQHABEDI5Rg2FYzLbwCdW8xUV9 je8cQ4C4h4GAgzbolZLFRTu0JZosiC/RzRQ91jSMgzuIg8FsREZsBEd4BElcZVGQi7CBg45hTDlo iI0cA0egUjrcRLb9glfmHziEzcLwE0fI/4o2HEMwoE9RhA013IMxlIYxcMMywIANCOUcOFAUbIAk cIERLIHYrcEM4IAL1MAnKGM7mckpFQUu8MK0xUP+8U1sgERs8EmDlAvUyEc7aERGtAP6ACZHBEM/ 9MM9HCMAUOZ6QENNIoTanciYoYNhrIM9okOOtId77AlgAmaJ8M9V1WBR6Nj7ZCc4oAPNCREt4kJD 3UM9TENW8gJKZplIoNB6Fkg/aENpmEP80MNCagouAECEAEAERMAkdENoeUE66EEEkAMjGIIVgAIh CEJ1gOTAoSRIpuhxQQ23rOTViAipxAQ/3SJHHBcnRaiKhuaIrOeIUOdh7ElGHOh6ooZxGf/FZRbF keaoknaIZh4oAAhDMiqpMrrZDOTAHmwoWkgpO7IForgDLbgDmIapmLoDonTpmMYCmHrDNVynmY4p mKIpWzwZxM3pnMppPdwpnuZpnsINncrpnDbGNXiDoJbpoBaqoF4Dohqqoqppoi5qoR7ANbwNnU4q pVaqpFYqxPGCO2QLW4iD9HAqqIaqqI4qqZaqqZ4qqqaqqq6qqmpDO2Cns7GqrM4qrdaqrd4qrfbD L9ACovzCpuIqsAarsA4rsWaLmrZFoBarsi4rszarsz4rtEartE4rtVYrqF4DmmarO2hrLGSrt25r t3YruG7ruHJrucZCevRDCSmpJN5LCbX/Q+Q0BI1pqVFQCr0CKS1KKKVcZr7S61Vlg5AGrHtIaP6s w9WggzbYo43QosAK7HYSnJZaxHwYTDQkLKUIpL9qQ1v0A46CJH9qwzqkSKQ0BBiBEjTQglVJqcgU 5i0qI38Kp6nMxPzFB92IAy0QiIrW6HFlA2T07I6uQz2MQ4Os5Jc1YUso2du8U+/4bM/SCcSmhR7o LADcgzZAQ79e1branzBMwzvUwzqwrDLWqMayBceCbY3qXN/chQBm1BapCkLYU1x9rYqWJNgaRTr8 4IZk3qSEmcTAxkGUA0KlaI0GAzUwrV8mbC3VyHwQoZLNX0MQDGXSQnsY7kZMw9MqIzNc/xl1Hujm 9s7Sxks9AIOKjm0OGZfOyqM27B902RKahVQVoqHFTAMtdGzLFgXdgm0wjEPcOB55UUOTUQOPBe+k 6NjC6AhlooPd1u5xFW7TtgMt8AKrnJKO3d1NtASTzc/dxQSy+V09fG7TWi5hpkVc0gMtqoOQ/qjA HoZGUO156Czplq1RtEM99KL+XFFI9e3/0SxhBJvUYmU5KCP2sF82MBQ11INB0UI98AIvMNSpzd/2 wcpPiAMvXG3LEm7TQkM9oIOOwYpIbPDChAyTnRoHX1GJsAXv+KzThu9jisOQFibnem7A/ui2gEMs fG2Niu3GRmh1xs2n1K/7aDDFhZkV7f9XRMhJ+KYF3dYoLtSDSsydAvuuAkfxE88LxU0KxS2ZlG1I pn2kBZ/GOtDCaIyDPcIKEAtv4wYxSyhOTKADJ73T915u/mkD+jYsDLNswF4NZcpty3oE2eLoPWSD 5blJbbAE9znumIUPf5lUzdUtLCHxlQXvw5XaAg9R4/JY9tqSR40DL2TDYd4DNUhtdfIsLUrO/Ijx vBxk5PyuPQbxmBVO8PYINHBSPzyGpYCvUeDYjlInjUYiHets1AaFLQho8r6vH+Nn59lSzOVYyFiR xHjRF+5IRXzyeTQyVubaExuwQS3UGIfM8JKxzTnxJsOSQYIE7jIvADib2zBwNgxR5Kz/8xA5bgEW zhlbES9w0lXW8uXSw3N2bvp6rvoOrAuvJymBLR/nkA5Tw1/k2KQ48BDXj37FxsJ0Hpr9SznwgtRS 83Fxl669jQEj8CZfcsyJ2UqsHBSfWkzUQ5EahQWrSWM8b6YyFBV7s6qsjzwrmTe3BLpCw1VV7tMG 518ObMM2rQwDdCT+MjMIW1oQMwCY5zQQIQiLMSVrr3wc20yEDAc3s44AaMra7v+qRaZcL0ebGqph 3stp8AYrGVT7rknHQzbgLs8CAD3MSpNlz5MtsD26BBoiBLJpivBynxadrDsEiEZYrnai0lU5qf1x rqWob+e6MPFog+gWhVL/gr9ANPe1/7Owsc8v7TU3b98Qh2z3YCVxFkU8PNGS0R3c2DXFae+NEM70 onU7r7M7MoR2VmfhTi0tpDKfLvD8ZBFe2VNFgEomi/GkMFmm9Jo2BKktTy1xwjCPOvdQN+1zD+xg W7Rk5/A5j8PlvU8hQ/Bdbtkxj9lVF690jjYAUDOJSBl5vc2XPm9U71d8jbQrpzVMkwY4HENaEG6B 0EIW5doAu7MYO65urGSKHB5sr7KSrbZu8MJjOO0xBsN/8vKQPrePJvYcpy9/4qhSwwrgcLcVS1z1 KnONHDNLyPfgsCWKonc2SC+T3SktaHMQwwbhlDhfD7AUczJ6ZgPyhrJGrLimUHHwBv/5zVHSjJSx JXu2jg3OiUxwCkeiijW2Y1/4+dLxtqBDL1x3H6dDNqTESkxvX9+SBs/LXZscMqdxkv8Fj9nuaFP2 TIR5k6k2kp/1A3OfjU+yOybebS8v39HPjok5O0P1vHxG/jBuCMO04xJvLymohEKsavjzhPPyhW+E hMew+lYMyAIAMd+tcFt1FZOx/Xz4wigz8aZx4fiEjqwsNVM2d+fa755aFVsy8dY0j62zWmOxFuUI L4TmJ+flaOSaKadyWN+YoO81q0/y/Hnwif/CPBK2hOq0w/aykE5652oEYgloQXPs3e6GRoW6QoP6 iEvvabPz3RXOCMvEOJz3mletjp3/dhS/s5KJ8OCkhDxHjhQzsLCDBsst73sQ0Y8HOd3h6QJLWb2o yCmdcayHuo38gkrYAsTqx6Q7rB07+oQ7N0cs7RBGZA5PAyCz7Q/zNXFn9ntJrxMf1Is3MI+54+C0 NTXHQ9WitTsz8CkvmciAi1PnWqnRtaIY1EeztVXpt6Yc3mtbcwKfWkTriPUNVpnb72W0xC88GwDY xAtXuM5G/cT/6KQvNbVtbDBgkXxEr4d/OOJ5fZjpXCzIcizcKcSdz4WkAzWjvLz8el8fcn+dWUtY 80G5+JvSQrBRiL7D0kWZNXnZeEeGz5a9l2CoiPswjuHPCMtBLKX8847yKGBKex1L/76kVzyd6FwO Q7iNxHcHn3mqjPQtpUreyRXZdqlBdeQfH/VoY48Gh/sAB3oV1khWC7JIuPw6KwqZxoLdrax+g0M9 tEjIDDCUnVq9bLo3o1lGqQpF0sdu0EPTX6XsFkwuQ3e0B/TlU7jPRsi1R+g6AMf9ch/Fhf6YSfTI 8QKv5hCiBqqX1tzpoftInlqdP/Fxx8bJz/zLif/80XssNEY/ZAhAlNM2DUAwaswAiIu3Dp24Y+Oy UaOWjd7Ce+UaHhMHDRo6jx7jxSt3b107cODWXbzXLtu6XwfoAQAArRwuXMFwMku3E8BOn+lkyvw5 FCgzZsGM4rLFax2vA0/74QIAjv8auHsh0UF7OG4cPY7QxIVFd49sPGjjeNVzd+1AP7fXrsWq55UX 0GPxgkLLJg4iNV4Ss2nkeMwvr73lysVDdyxb44hpY7FtSwsdr3IyDeoUR29dPIeOI2YbFzYsPa9Y O378WC7lRcT3skEDMA2mTGjUVtrCZRSoUKI8e/b8fTSY0qZ4tT1tKxXAuIWIsx57eMxr1o8kUV7d nI0XrVjuYoWPpY2auHrMj13GrK2c1nGARUtnXJjasa/03jeWqM27t37eaOHqmKCCycaoYCxrj57G +ppItHGkqy4rjlQTC7GRSEKHmp1oi6mnWGhZ5ybehPuNKOF0+unAX9CSKTmomAv/ph50wEmsoc3o 2QgdkUYqqZ11gITuM8cmGq2edoK6KygAyqnnKsLgY5DIiaTjisEq/dKmnnp4kY6aJTPTqR13QsKP qzPRPJO6HL/iiLQdE2OtKaTS6TCoX7wRRzfefArqt+BUHMqoXzZ0Ubl+bFlyHGisUgws0zaKh6x1 TjKJUpRSstE6dMoBR5t1ksQrOACWWuc2vxxk0Mr40swRRyu7SuxLAg8y6ihaFMPRNNOo2+pMCaOL cCMKgayHTjuDYoaWe3bjqbcS+wRKWmlTHAoYbXp7cbklExqn0azC6iikHi9F6SQg0TW3nV+YWlJJ bnHRxizuyINPvl1z3HVNfDdK/8lYMA9KhzdmqGH0ngk320o6HTFMTKtstNHGMK8uOk/gOmtbMh1q yqkWUGifFco3Z3v6pZdggtI2qnTiMS2YdhLlVqh1Lryn0mlMyo7SdX/RjZpfuH2XZdOYE4ca/KrU TyJquBLnI2joEbajctqJZ1aRMwtGHKDuAQqamHERB4BfAos6R07BsTQxh7oS555fwKnLVozFxgVU mUcGDlBn+R6qHV64VTkYeoCWaTHFQhLn7jrT5dkWnDEFZ5pfgqmsLpHtEpUeggynZre/UkIHP2oi 1iYwmsjCSKyFwLFabJkzs2WzX+hhBp11jtHpmHXECTvq6ewTCcirJPUxG1rEtv9VJztpl+kXjkTy DEmSd9LJ+oujDWYcWmQL/NBgwDnmF5DAsSXHaUSSiZl4pmm/55uCscWWX94P5hfq8JbpXXDGsSWr hZZED4lJLCzUgcZC2oezlJSkfdq4B96AEruYxCMW66gdOHK0jnSApR5MMw2FPEIS4iUuRABQnq2Y R43J0QMxXmlHOcQRGmogCUUCG5hPmBE3cHCrN06BUf6AiAt0RAQa2rgGLyD0kIrgImR4YyEQgfi8 DrqtHrHIFz0MI45pIASKBeIiFP1Up2OkhRpwoUVjeMELetxDJz253vVo8zowcusXMTyaO7xRH9OM g0ZfzJ+2DnANb8AFLoMcpCD/C4lIQnrDHXg0pCIJGclFCrItiXzkITEpSW9skpGd5KQmJSnJfiQy lKVUpCcPCUlTRvIA3ihdPUoXS1hqKWKzjFgaYxnLXtSSlrY0onJauUk8OnKYnmxkP55iyE1KRpCM LOYwi9lJdwCTmtW05jWxmU1tbpOb3fTmN7l5DT/ijRntsA/ngMiMaZRkHeMkp198yUt50vKVubQn L2GZz3zeM5e+9KeW+sELeuJzoL3k50ERmlCF8mKUKBuVyMK2QwI1CzO7sQVSkvUL5oiMo8zIBjKZ KRpa9GOaT3EHMtuSzf8cCpix6EcZwYnNepD0GseoBzB9WM1YeCOmPeXmTq/h/9CQBQMaUtmJLSjl GbBclDXgEB3VHCqTF8qsNwW6xtGe4sBxNIk62WgZWI5xj6eUYzRqjAcv7jGRzXCkbFFzBy8k41Ng 0uIaJ6VMXKHxFK18NT4CrIi8UCpXn0YmqKNKxy/ahxm4lSNq4+ggNGAGw/d0KR7tmMY9tiiTdZLz o7EYxwEiRo0D5IciZH0rNeLBFsauKR4RcUdgGMOLjlRkHO7QRj/O441YiAMmtDmAO3BBC27ONBZg rYdwD7ARwvBFG5uhxlygUY9sjDawgo0pUKMaDMwiJBgwdIjotGHbjoADFydprTbocbvN9DFZMvNo XK3pDV7wtJr0BacRTZeNxf/Q4oXoOEA70NGO6l6TuNYN6AHqAbV+jNS6P4VLVOuEkOdphXTcAVF6 xwIOrRhNYvkZy1Uymw4mLum9Df5mPQQp3PG8qIM89cZNtUnXksr1GtogjHLAEg9toGNRJtbpg4XS Du62Lmrcea53DLOodtzjLPQ0zEPCEg9wcHc3ycoGfL15DQhxspsLnrF1RxoL67rSG2kdIj28ERgN +dfHcwUyAH4EgLDVJyzj4I94AGTnesRQoGjRRleq8xohp0N+Di0xOGPB5oAyOTBOsy81r0GLLwt2 ppOO6Yvb/E26FtYW90jU8xyyFdGoxTLikBg9/AJZyznNTfhJL0HmBw6UHRr/nOX4i9Wo4Q6nzUjM 1qxrrxtcYMG6A8aZ5uamg5GOiyTbVFvh1THCW49j0Oiy7KuROPrsFQqtBGY7WddlaG1sX4PHx8KW 64sHLG5rbjqHIklHd/EjGkg51jt/jgflphGMzoyxSF/xyEgWYpQftSMdV7aubbv5ax/LeMwzfco1 0n0A/iUXLraAMUZ6umkh3q5OSy7yAY0m0FuiwyjtpB18IOSRjqDubaQChzhYFPFvEns03hgjLOPL yIWT9OBhjkcgHSsO+0YXLNdgcjtGS5KcepOu3miHQnRDGK9shOrcSQst0KvRdcxPQ38BjNMU054d kRfOTjX4sCkj3DPm1de6/905sAfrUsoEUitwF20/oC1cm4v2ANMFZ9PLkR6jqO4hO+IL6brDJXFg ZEci8fPSpt3CqHGkHEhphwCxHFN36J0aLr7mJuFO6X6E/rprQeYgs0mLRwu26S0DlezS62pw0Su0 OhKJR+w4EbM6SiQiBAehBZJ5cG6+7zJXzlqQG+zRWxeo2li9NcO7cEGeZYe2AMtYSjKuMaaxPhSC Wa4WJBE+glB+GPH0TuIm/G+G54jGf4pLSe9Tcw+2rvV4fjUl7ePIeGOI1TdLqG0hHhZlMVDlw2CG GeAmJUytg6iDjxLjcRbvFxCiHWBKsMLjxdRPOeCv3JbPAkfvO7YpPPRPkP/SqmtmwzPCAheYDCzQ ghfEoR0sS37WxRZ4po784kEIziQoBGUIJQO76TtA5P5a6gBcysTmr6fETC3iD5g+UP82aRqyoWPW BxouAmoUoiEgpB1wAbFQMFzCAkhQIisupSHEBR0SZRoqUK7WIjwsjZrErAhbCcHacJsqjfkQzO1U ityYUAi1ibBwIWqkYhpazW0M4m1kDRemAdqc4yJEZyJux33+EKkGMCsSJR7o6uCuwf6WUAN/C5k2 LzJ+aw5lqgN9iuecbxMfDg+PL/nWDy4QsB5AhRkwQlIesGeGx0wgpBy26OWogzSokH5oUKloBCjS Ar5uC9H6wYhQkQhbiS3/IkNbTurEeI7GpqmMllGQolEDN+m6BgllFgQh1qlRoCa2GqMjNmJSXqZN BiMiNGJHSELIpoI84EstvonnRGsZxWyQSFE5WHGbZOylnmK+ronvqGkb79H9mmnBhIukFuwYvYmw UCYdyqFQLotCROI2/qxNQuLeWkcstALVpEtcEmOH0oEXWiuufs07dMq2JCOgVO8A7jH6fC0fVZEJ UcyVIG0blaPShukAJG0tHs6kVBGlUM8psvHz4IKh8pEteEEPf/DNJDIbemKdaKYhIOYvnAMkdiSp 7kEhzCQtpC01oAFlNgMXHGvBfBKRfsuZaKHYkCsuHBK0wiPdFgylBPKa/9xi+XhKD9VuweLiDVMs 0mDsP4QrLnzSywBJJ6kJpYAymX4JJjtRA4st3bBrSWBjJ1aQTaxkIsACGjojwDCkZQxDP0ZDUhBi UYzC4F6KGtzv/fhDuNpQoGiBrkaPweoR4gBppbCuH8aBp7qED03KKfEvGXEO4lYMDtti9SItvI7p uLxjpNoyCR+su0LibVrChGjGgipijyjiUeLBbg6GzyqiK0VCKoLBgYri7GLKB6upNXuqDr9p6bCJ p1QMmeRimzIPjygpmeLCFdtBG2TCFmKBGdZhIAqiMySlIbwCUqgQfFpGrRRHEDUIzuqhcEwoG0wP 4jR0lDh0QzX0Q0cJRP/hIi9JtERN9ERRNC9naqQ8NERbtEU7tEMJiUNTtEZtNC/ZrR3yJOQQYn20 gYaCYZ3MUToOqH3a4Sw4ZUe0MD3RIWTei5L8I5gkQyVt7v6kqzCPoQ3rwbFa6Rjub/T20kuxaaTs zzhH6ZiACThZSptGCT/Zwv0Iq5zcAQbRwR0gbBpQjYbebX7Iwj+mobxCAh57Yh04xp0IZktpoRwC BB3OaByKSyCSCyZ87SEklR6syR3oYZrooR8stT15S1I/1ZpuytScwx20QkuO4QAsg6fE4Ro61Q0Z leJe9VBowTBgwlWtyQWfgh5w9ZrYbY4w40iJFEMYEU7M4iHuDYoKTh5StcSUOqlMVYnYnjWVHglE viMTM+mRignFTimRbusYRiobUomQ3MGxTCceuHWRvAEsBMqrbnKRyGSMzjVdMem1xsNdSQmTHlPd +LVf/fVfGywgAAA7 ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/content/vol29/issue2/images/small/zj40020808100002.gif R0lGODdhyADHAOcAAAAAADnuGDnuKSCcvSnNWiCUpCmLtP///0H/ACnFUpT/ACCLvSB7/wgp5imk rCm9cxha3imU1SBi/2ruGCCD7ggx7iCDxSDNYggppBhK9iCspCnVQUrNWhBS5imscyCLlHP/AAgQ 9mL/AAAAzRBK9iDVSiCkrCBK/yl7g1r/ADnNQbT/AM3mGAgx3imspCBz5gAYQQAIOQggOQAQSiB7 5gAA9kq0agAAYkruGGLVMTneMSm9Ygg5eynmEBBzpDnuCAAQxQgp/wAgtCmU5gAAQRhqiymkiwAA MSC9cyCUxRBavQAY1RA57gAIpDnNWgAAGAhKeymUixha7iBa/2LuGCBq9gAApCm9gwAAczGse0rm KSDVKVreOQgYxQAAWgAY9nP/CBBK5qT/AEGscyBzzSnNQQAAvSCsgyCklFruECmkxSCLrCC9Uine IBBB/xAp5imL5iBa3il79jnmIFr/ECm0exBKxQAAlCCLzQAI9hhS9gAACAAQnAggexA5/wAIcwAg 7gg5rErVQZT/EAgY9inFYhhqzQAAKRAx/wAYeyB7rAgg/0r/AAAASgAQ9iBS/yBz9jG9czneSgAA 7imUxXveMRBalCCktCnVShikiwAAgyC0lAAI7imc3hhK/wAQ/wAAICCcpHv/AAAY/wAA1Rg5/xAp /znVSim0nAgpvSnNaggx/0r2ECCD1QAI1QAQ5gAAEAAAahBB5gAY5mr/ABBixSCcxSnFcxAg/wAA 5jn2ICDFYgAAOQAA3hAx7imD7hhB/xiDpCnmMQAArHu9Uhgx/+60GM3NKaSkpKSknKysajmke6y0 tN7e3ubmAJycnHNqc2JiYnt7c3t7e729vSlq/+7u7hAQEBAQCBgYGKSkrCkpKc3NzWpqaggICAgY /7S0tIODg5SUlMXFxXNzc+bm5iAgIFpaWqysrDExMQgQ/zmLg4uLi3vuGNXV1VJSUjk5OQAI/ykp QUpKSnOcYrTmCAg53kFBQQAA//+UEMX/ACli/xBi5vb29v9iIDHFUiwAAAAAyADHAEAI/wAPHGjm rqDBgwj79bNG8GCzctb6HSiHsCLCcuW8AdjIsSNHb9dCehu5ceTIbCJJfvSocSO8ZtdMypzpbR4y cunSzXt3rue7e/DmzesGjtw1ADSTAjiX0aPTp1CjSr1mTeCBftmQkrx2zpy3bD7PRTt3b6fYd9Hm wTtHDh21Z+S6Rbt3DWVZrRo1zrN2lCRIkdvU/vyZLujOefdykqM2bhy3Z+HIRds2MqVJl+76ltzp 1VxOeJ/hiS5cePS2dNtOfzZ3zdy2bttWajzXrKXUI1HE6ONQBUkKBSsK4FqUYVgNXJEk9ZHKkarV fuYApIsWc+2760KzZ2d7Tu5Odm6pUf8DJ04cOsjTppGLfu1dbADvysWs2xrxO7XwToPOme7euevv FBZNOM+Id544zyAjTjjhTDNPSCIBcE9mAJDzDkqflSaaaEBtqKGHpPGHGn/bXAPPO+S0RJttHfVx hSzqGCKGDfjgsg8ehIRjDAcScELII9X88kgpcJBhSgSdzPHPEU85J1A/0aBkIk9kCVWWdj39lx2K yCADzpfiffklOgdGlg5I84RTTol1kUMWiCL2dw9QOZX1TjfkMMgOMujwCQ465P13XTp11WRNOPeU qBpoG24455yNRjoaPGUltp+IrK01VlMczcABHXUoogcuk+TSSwhrhJLHN4KsMMEi6gz/wEgd1Tzy SwT7FOPjN8DIYUgtdsDCUTZVPRmbOT4hduWW2P0UDU/X/XeOUPdFIw5jjY33TDf3leUNU15BKiKj pXWI4lxAzRMNtx6uNQ07DE7z7H6i1VQOPN7ol6GkhZUVVKM/dRjpvpOids024HAKwBGGOHELKr+c sAghvZiaxzJijEEJJnRI4sEmpwQQhyaThFBKKSfsM8ULEaByixNr8CGskwK94w2y95k1LbPQUovW s1sKmvNOPD0bLTwARGONV4GJdppq+jVKmmj30Ss1f/fwdNic2XhzTz/hYEiwpAJ36C+HaAtc2ris EVUbR4E8Eo8BY6CRBQS5mJrLJMOM/9ALLq1MYIwxK1yxDyc3xMJJCL30sogRIoCRhj5U0JJDDAAQ +5w52UxzjnRaXpn1f1fep7WW2Ak1lrIA9gQPSuHsJF9rJ+FHKaNqLXolpUAllphOavG34Txdt3bN hN6kkydKqu07NbkfDix9YV7BI845377d0SEWdGAGJ6Z8Y+MJjRDySycq6ONMDwqkYUIxpphSAxNg cDDMG63AAUknRXik+ZPXcNBG3sGOkJxDHNtQzzzc1A3pIPAn2UhH64R2n3SgpDuEusw8moEnl6Bj GwfcBjqcZiKBbQMo2cjGlYSXn65VJhuskYmEKPSMdADAHO/QCjLmcY5piGM63TiH9f/OgYyseAMZ FxIHO+RyD3Bw5E5ClMlRaDMWAIhjLh8UBzTm8Yx7ACAZ6NjIH/hghhHcwRCnsAQfYgCKjeyBOcMq llWaAQ90zMMqeMyjHvfIxz768Y+ADKQgB0nIQhrykIhMpCIXychGOvKRh/zcSjqSjnHUMGuIIdph qGWldJjjPHCsUAOjMqF5SHBdcelGN6bRILkkKzsTikZJkPIUb4AjOh1hkTmogQx30MUcwHRNalCz jRRmjh3UOMoknfItFtGSlgh7hji6QZ8UwjA1wyzNO9whS6jYZhtWkWRHyqMucrDjnOwQh5fGURBu UEMc03ClWKJBz1WG4xyAyopT4uL/FGSegx0JQsYzzqlOt4zDQAuSzM7MEg1yuCka3JjGLD9CDX1y 5B7o8AzvMqkdWOZnG5yLCUj+cg5u6NM2eelmR7JRjmY4CDWu4VwK6wLDYJ5mHtBIT1z+E45mpOgp 4BSIOO/BwXWlR08JAhQ4GOMOjLhjHOCAjEKjZU8lEqgZFhUlR8zBDXb48BkDjadc8PQuPomHGufx qmTGUk9yqOeoPSVHLiu6kWnQkVmHGRpeq4YalIwUQjOFhxNzqdKglqMbwSMmNlPjmnGB6HagQZE0 rtMMXG7EHOHcyDa4gZZwEFSJBH0GoLCVrYOCQ6CRGSueVMlah3ZjHEhTZUnG4Sa3/+o0lah8V1LJ xFsyDRReegItcBtEDgUhha4AQMc7zGEYos1zLFrSTs/ymp3RCE9540gpR8IhEHBwi5gxlakwgQm1 a4LUpot1TdbcMY8bZrZC1OChPdOZIHBwwx3NaAY3oCrQpFKDGwA+KKDIpKBztlKdG+EnAKbxjAUi CKzSLM+DvfTfgjRkv+MR6HBDG1BpjiM6towOOM6hoTlRS2hWguWjTMxQsURXKPAIhzhKgr2l4JEa 4egWaBbLWJBuw7rBDGYKyQvMebCjH9zQCGaFypK3dOddEi4PvNJDZVayEqBd8u1AI4PPwXJEwRxJ RzPGIuMBd+m3DJKxaP+0IPWY8//MAlWivHyZy1t25BzU6NeKH8XRSimLQyau1KNOlORc1rg57rBK OcgRlKqdF5g1DXJMT+OfbhCIGxKxrHtrtpErwgMZ8GBHNlD0w2nkcCOWFsd1wBEObtgwgs585lNk i5QFsWM9EnXKNLiBxFXu14vRKIdj0OGOZ3gzl0gBRzbgIdd3UKM7ALgGPNu7DQQJ277usMYzstFT ZDjGu6EsiSwH6qZnvAMdA+rGtjkyD2oopB/CLpY73IpjcTLTI0sOd3OcRl7DeHLIQbGgvmstR0ga 3I/WQMfBF+5HTNMy1n7RSUNXiSB4pSUa72JHWvADlBKVJC8TLYk4ysHwQBZLIlf/QfkBCt7HhJf8 5Xh0uG0OZkQJrhKgAOVtQu9EjvLEhTChgWwxkeIVlI4c5nqE6iNdjvSXyzzaOUHKdBo0IGR0BYbd 6O8SGxqvIGLnHqwJyQnzk5OjZAOkIzl6061SEJYn0hrIcGQ0EhSNhRT8UAKJ+wGmgfI/JXIcByA6 LkdNVlYiKM4SRmc4uOXDdCJIuJFZaxDpOZZ3HMUzalckMrhxAL0jPNtuRyTTB4mMabiDQHxX0DQO S/I/EvuRDgfAjy/7Dp3C1fEQRoaBw/EsjJeHQNJ0fIPEytrlbsSUmUdkU90B4Gb8Eb/WAHwj4c7I gjQmv398COcFwvztH/LpSBFK/0m2cad4utWVNze/hRaonnPKxaFqxdM7Xqfkc/RFHKEHZDPGofZm gOMA0ncA96Vw3FAO/YAO0kcNVqGAeRRvlpR/AjF6f2QN1GAN4NAP3laAepQweeR3AJhoFNgPDAh3 ztdUKwcO/jcOC8FHDpcOcPcM9vUeKjQPbKISfzFT1zAP4tBef2GDf3FClMFMogUe6IBMugdQB3B6 K7cgiaZHCwJ43EAge9QMnmWAC+J8eqSEIhgOnrdHCadtvSRa3PBgVcEOBohjDJhH4xAO6BAR5RF6 5QF41MAOAZhHz2ANbigODScRWPFwIAcAayElXwFTKOQfpkQ9f9Ee9vdweKER8f/RFzQBIbXzH2y1 RNOySpV3Dy4kiZdxGTOkTH/oNd3gV2dXGNTyLKGxE2lRdOZATyfFiLOhMOF2CI2QCDAABSqAA6ig BI0wcBtBM9BBS3MSEtOBOt3CQ+uyLq6EgGjlW/HEE5pYTOdgRAMkH3lBitRiKfRCJdPyEzpIJuBw UDmGO5QBIc80IcqUDThRF/QSPUG3NhlyXaDBOenQE7LGFLHWKUbQBmAwBofwj0cgC1FQP4jQCE9A BLgADMVABnUACTkgBmFwCFEBjGABD3URXX7GM1qTFvfQDeZBWuMBDrdmTmDFDoSSDVEyD+UgJdeg E5cSJx9iYjlxJ/QVjiGpcav/lCDTQCiFMiHpEDswtDZkIz0o9GOAJjD7gU3AdA9VtCIekQiBowrV MAS28A29wAnqEAsA4Ar0YAwoUAxA4ArFICTAUA2lUAylAAkugAaQYArq4AUeQTMHYJHJs3Fbsiw5 IxYn5I2rMxTgcVBnhVbx5CWTMRLdUA6skQ22c10a0i9Zd3EWQidrQZJiElWehW7qEhL3YA3RQCj6 0i6OwjvvyDuQAj05ATX8cXbngIIsAgSc8A0NUAEOUAcU0zi4sAsuUAq4QAkJAAkJKQha8Ajx0wlw oJvf8A2mUArA0AR/4Ave8Eb/cxWUsRaa1Cw+gxYzqSyBhiIAop0rxkN66Q3v/2ANXaMu3WA2L0ka q4cTjHkP2wIizbMoJ9Jem4kvZ7co0dMozJU27YKUUvOSFmkOCdMSMTAG+lAJHaAO6tALk/ANjOMI i2AARTAJxVANpyACV4AIpAAM/7AD4cMJpBACb1AKppAHk1AFu0ALlXANTxCd1sAWXyEt09ITLKZJ JgaNlDcYORNoLNYTZ3IPyLAmf2Elpdk8hfFjiaITO4oacEIpQpk187FB0WCS+TIiLCQp7RINjeKf j7UfKKFqTskRGAAJBsAKjJACtJkLI9AIXWAKj8ABywAMv+AA/8AItPAP37A3I0AKveALKkAJ++AA tcAJobAMTJI5coQV8KAe3f9QGB9BTbIHD+bAqMu2Sc51ivOnLmKRQbx3FO/QDO+ADAySDl/1OqLK DkI0KVZTKe0yfyNSFpQBEjB0FD65J9OUj95kQxuRDlIGiqMkHQL3T92AS/h4fAMlDt4gDb+1DeGw EdIgDzfwB22kCTRwCwnQAE2gCXzwCAYwAW0QCRDwC4HwBE8RndCRFFJUU30VEtbUrpzDXP8hqX5V GSM1Eu/QD3JBefqqr3gSF/WkjMmYr23lWvuajJRnhtDlE9zIjQ71VjqlSg4Vsa7lWmGRJRbLEyPH pVi6sRzbsZEyDyp3ABhnFZu1diZ7sij7cva2TLJmDvd0TjQ4Dd42LUFYS8j/hmq/CkfZUCBtxg5P FTvhYCBFNHAhxrKydrQd0RrI1hIoxRJIcWj3xogPd1LmMA+a9ofPFGtBdQArK4yc1R1u1bDq1CW6 568V62LPMA7KhLQdxEzdwF4DQlAExScDllCqBF2r5FDRQFshR1FZNU7ksGw8ymIA0nEyBSHk4A5I ExXNdLMAIA3jQBiPJmTW5Bn+IQ1qBQ3S0AzGJmsgt7X2RhTI2FBQ5ieMUYAK0Qy85FXqUXlUYrDu oKstAWZLMQ5T6iV9YlXnJFqOgV/ccB6Lx1eioTM9wUXIwCLegFyZww1qUTqwRKSlmRNhh1KJeA3d QA1OwRUoBXjX40mS1q41/8VjdQKerstD3LC4HAG6HDENtqtbiJdlS/Ul7JRf+oVWwes7yOhmQYRM HNG28DEOaVIe0kRlbrW7ShUmvFVgjNqvVRZPtLG4yatPMstchPu6lKewWdIt3itpktZzH1FYAhFf iXJe/8I5RgkadvJ1pLkfHfdj0MANyrRkXNtpyIBxjqe7CGJf+dVSgBmYf9JfSXVmVhVPzWBDCsYO RbEncTZlwVVQS9UY2XJaVtUgVPxWSqwgbpFs0aFclts6E5Q610mkZoEn9ERBVQNujdsNVrGDnqSU dYGa2HSkQvauwsRYntEd6JAMsvdeGyFNAxJQQJxlBBZlYKUgSiRhEeZZ5/9EDkXxZTnrxwSyVIzx wxFWX+HYVPu3ulUcXAG1RIpbEsiVvEAjXVtiqZm0LETTnd7JlDAsbrNEDnNEDqbEwWO3Y+ObGCBl Ta3BXPMwDQKBmJf1Xu55TtdgFFHxM2XcOg2FW+rGuX/ryBuxs+tyvNKBRGLhUHpiYHfbHTf3YEok GaE6ZrWkbC8RHch6DdwgVzdUDl5kJaYBaeTVH6qoSrsGDu/BEuNmWc7EDoo2DsgQRAqrLD20Zq2X d8+8aSmb0Aq90I6UtU2ytkhxdqJDaWdTTCrhtLWEfwegxop2DmsoqtNAgVLIhpzXDx/5DNxgDXly AFwYVdMEKAfnckpUcHX/J7KQEVVoJRDsu3nPsHdHpk5AHQ1FOBCSIaqn1XPwwtLssF96OEjg1xxd A3UokicEglryVMDyUkHDlDXu0RdRzRIafbLjoHAbjYGVNHcN8mxI1IQtR9YMvUjgd5/S4VZULGMC pXPiYCE2nFrTIplTQyJdM6t5EdZOzdHWwA338C7UEBd1uEeOsXRd+NaIFHutcSbSMSANMh1uBn+G vHj0fMO3xpGjURas4R8BclklknyKNA0M2A+MvEf94BhYyEgSKEhr+H+znc7kQBCNPW8szUU9DRnd oBDnELJO3Q9IIalIQX7KWMCedcgDnHGeg9np8XigFS/vwtc/oRE5SNiI/xTbBdiGE/gQbL1ItY1I 4hDb/zeBjYFH6Z1IT6eYfTEdZktWuKd6/noiyrgn8HJr9hSx8qKJu3om3n1IgOcObdjYeoRpB5CG K4cO3KBfhXTehMTDB4DSSPZHB14sCqiC42FIDpcNeTusJ4RLX3HHiAHQysgticHN9PwsqpQYUT0S s/cRBZ5H/cB8Dm7cMUeBDI5HBR3Cz8DjemSBcUcmfNjh4uBO8du7QW5I7kYRTw7kLAgRCEfWsUev cFQXISUS1VS5kIarT+Gzv6xIRN505YAMDDERECgREFgVfNhH7xaBgXTmVqGCV0GNjouSsFE8MNSR yxVprtEd97xMLfGIsP/oh4I4EuaQGPORQn6RiDKUF5QOiG8TirTUyzTo5yxJrzfY6LHjF36oFUsh i7546tmbqEYk6ulATSABGupSNDSaFlmySvH0OtUxjY146PJB6nkhEpVae66qOvNCKb3cJ0GELz04 E7T0EiDXEtuAEyCBmtalqkmJpGxCeKIOcsUaSk8ABe2gAPqQD/5gDAlQCArAAs5ABTzQi+EGjCCm Ef1hPKfjE2XxXM/ingKGDhImL3T8IKQOH73+6yJlHUFBJ+MrKEARDn8yyY1hasqeEs/k7LRkvTzp Gc/TsSSiWPxRF9PhRfeoPVDx7RNwCp2ACFZwBxJgARMgCMAwDBsBBED/sBGWIArtwAPMAe8uQTwq 9CyvZB93K0/4lC3nQU88gYmzanzVKFKA8SZ6to02GhRcOB7oMI7q4lnkUI4Sj47RNg2auGxCKT1Y Oi4hYqUtqS7P1O0esQdQIAxiUA9qcAmpcAN8kAVaYAP1QAku0AW4UAyKgAokKgcpwAhlIAAcIANv lOrPgRKB6w3WYe+krKlQdB3i8MRQnGHwhM33ABKTIfCANWq9Y8uMsh9mQScLlE5+MmCL1xPdgKr0 cRTOjiIlEhrRE/pkF5oDA5PYtGwwuhQizxE84DBtIADeYwakkDfq0ALK0AweYASrgAvxcwKQ8AjF IAed4AGXgAoDAJbC/5K0cmQNPKGZCyU62lF59iEUbSEejeFOVE+2ZLKD0/4siElTvUMnzcPC2sgo lgZQ9YVWX5IgAEFt3Lhn565lu3btnrtu0e5lSxcRXjp4FS3ew1jx3rZs5jp6A2Du3rx3Jd/Nuxcx 3TaWEO9N2/auHEgANWGR4hNIVohviziR+rYPAyxAxPIR46TJ1BQ5v+CokYMr1zc9IXqF+FNTK4Bs 1g58PUAuYbZzJaNFKztP7dp359KiPBetWzhxyMBRw3sXHDhk09CSm+ftILJy5hKmuwdP7buUE1VS VNuN8dt757pNE4cO3V50yJ51I9ntnmBv8/qRg9eRokqLrTO2vph4ov9jx9se286W7V43ds1o1uRj KA0YOmniNMrFKQSZFIbUSdFCh0kvQO24VDshB06pEMXgnDBViguHJrC0XvP6tV86wSUXjzxJ0v28 svBMpmsb/104ZHjH4QUQGbTeSQeAc6wxLJ154EnsMYkqougecgAbiRzQUopwLnQEogYdccKJZkF4 tlGonHdye5Ai2DSSTUQGGZTNottUsi0iEu8BZ6atYOBACwyIGCGXXnpJzhQtJKnin0jQcMAA4W5R B59ccJkCF1wWMcWUYjj5po5TEqmpK7D62cabexhD6cxz5FvMpHfIiY+kssoyaR5y2AnHwmgYU/Md bw5EqLEGI8RIxYr/MhONtWjCOac1yAa1LbGNzCwHHm9qZG1FGRWEbcEYNV1NJcPMQUdHAI4whYwP aPBlGHUAyXKRb/LYB4dOXCgkAFbKgAMYLib4ZkhOThiiEzWc2MUIT6b4R4UZuErvADK9MeekOtva E76T8FtsrTb3PGkkjEjytiRv3rEmG28QY9BblGarCMTZEhORIsY2RcxTDFMy0xo/IbJRotVWjFGy F2WT1NFQZzNzHN9q2kOWB0SYIAoIFDnFgjxy4SnLLlwxIA19RNHFgl6+OWGKUkwhJJdHnODiCjWM OKEVAbLZQ0z1wkFoPje1ZZOtNc/BSNv4xqWzrD0TQgZBwQ52D8PZ/1RUCeFCDXVQMbXOtBSAd5qZ 5pyQsjb0IkdHgi1h2Khe6Zp0wgnn4ZoaiaKNUBxQYQBghOzFCiv20ASfKa4IQJBIRJlDkmJ6uYMU jockpJo6hNlAlCiwCDNaMskRe5s1tx4XPpTkhMstNL0d9yy3Vr/nGnPY6aYwhC6dR2DaUMK0UHEn qj3ClfItGiNzAEhooXS6ESfBxyAEFV7Xmp+6zHfEwc9UAPaIweIuCCHEKk0C+ccQXEqJJA0FIgnF mXZMuEIJAA7ZR5FhXCmGkk3aYEWUSMCsCb0xhzfHNO4REveg6UzhEtqa2mI6lORnXGopyzW8IRmv lUOCB1HXulS0Df9zbMM+80JJvk4ntQV5sEUZzA0AFnINAGwjGuQoUzZYMkMarmSGv7Ph72xTI5bM xhzegIc4osHCc8htK0+wAwY00QQreOEIgbBDI4ZRjVNEAREOUEMVODYEO3hhHfogxiL0cIoNlAEK oNgKtMZ0jWmEIyXvmIZWGCUOr8EpHOmIBkyC5q62LIZo2fDaNOAhQa81IxrgmAY5ssGOZyhyG+OY xjQiQi/GSCRcEVJMZRo1kh96wxsdyaAKH0Y9rXjSG+joxoTCFjZ2yGUe6BheOp4BAPt4LmzgqIlg rgGPaYiGhTU5hzuigYx5sGMrz3iGOMDhDbTMAxoo0IEEuFcNO7j/4g1ZCAAegqEII3AABS3oAxHS uBWdSQuQ40TnOFciGHPYqySDFExMGhVKT9ZEJjT5TTolKJjcJMSf/OynJxNiDsPU05QggYcRAcBM rUzjl+mE6Dgvhc+FVrSI+YwoREHxhIxC1H9g+Qo15hEtkJbUpCdFaUpVulKWttSlL4VpTGU605Qi RpHioGlOdbpTnvbUpz8FalCFOlSiFlWn7+hoUpWazm50Y6lPhSoAwHHOqFYVnX+yalbHaQ6wIDWX XyUnOZqBDI7IMBsekmFZr0EObvBMqRbq6D24YQ1umAM2jbIRBzl4DmrMMqneAMfwIEoTczxjHOKQ zDy4UY5w2Kkz/4adx1XBWspoKDUdyBhHNCbyugNAo4PQIEc4uBHYUlY0jTTZRlchmg5qbIM+TZVL aOGWyNWt7iTwiMY4qJpOuFYUJDTJxjhGMhdxsMO47BDHMzrzoUQ2BC1yecdedmvaXFJDsFv5LTWQ 2s52xUhciSEoQiSIDnKU9qrnwOhvnkHWvBI0vLlRjX2kwQ7OhWMcEk1jar/i1a1cAxyAyWMbw2Fc 5f6nQ+LoJVrmFNtwTMMddDxtTXpr2lPC6U7JPG5yN8MZBFvILfR5ITkSCQ5uYPQ8pE2jYk/UrtGN pE1pWkk//ZkbZIRjspRFZzrK8YxKctAj8AUyQe11p0SSA7TUcP8HRPV7AP7WRLn2kUskB1yXu7ij Gc1wBzWYK+LmukUuA2pGZNM44ZqwI7OZ0YxnxFEXu4yDG9z4DzLY4ZdwXSQ/jHnGMk9Ljd1moxnv 6C7oyJWmBnFwdr8RqDfCQY2H4rOy2B3HAZ6Rjg7O8McIga97e/jaCTVkTdAoh9jyq9r+UcPLqYwk IzVzFzc3wxr9aMY40IEncqiuLakMbSTJ4Y5tbGXC3qBGHuuCjmfM2UKppEt/xuEOWT+jl+GCT1ws NKFaIxK7fNaKgMY1n621OHiJMfQ+F6rL2RVRsL99tIQPUA7A2NDHoNQ0phzk4sU0lST9SHcLSQ2A dKDDMphZM3L/17th/zSjHOVoxmhr3I3P3XoayKV1OQYoYaeGCR1vQi5zcf3w5NplM5rpjJobXGRV injA3aAGLqt7zmy4o1FtWqDp3BTzsqTEMGPJjXsNQw5q/OYaoq4JOg4wjksKudA0vKG4cGgj+9xL GtaYeEjAAvTQ5nHAGF7zwAn+5jePgy/0zZacLHMWsrvDqxMu5jwEPNtjXybZG+ZMmovd4AnB7bhz 9ktnVl6TXrpk215WMJ1MYtt31caGld7GPdDhaK3odxySiQhBd+gR24yohw1yjA5ZEvkOslbMXP2K qMcRDoxnvbh4YoeyrcxsAOVFzcfFMNxkT190VLa3/XnhyCNJ/3IpIxcZyFj1XsBB7Luvubi9Ny7c xuFJ0iLTJTMHPJ1a3C0/noXh0C7aghq5UFFHQz3iADTSdT5+Huac/DS0a1zCgQ6ph74mHXoHqu0+ 7GUbPOGyBv5mqMF1rwPf/8hYM7rjOZDorW7ghrkAPmT6kPkbOIHghtWDsw55vQaTvSnrDA+hhvIC rOF5h2fgrgeSDwZKk7X4tmgjO5mLk9bivprQsa8oB3bAKwjpvNrwMQdBv0wLL/x4BiZrPx6sifWb h7lgJGRSLlb7j+Ezvt/Ti+EDucdCvQZDuUebMGSyE0bKv+H7PWQCPv94QIRzB4UrLrsTOM8AwDyp q4UiLW9wB/8BIjT5cBN3+S56KwnbQiC0WbRcSjdyAAtrQAaDWYkOOj9DYYkf84hCrBHF+wp3wIYe BDqueLxF6TgFJMKAq8C7y7hkSibjy5PR87WKCxNukIvUu4uUW67k8z1wWDZZ+5C6My7jU0D64gag 20CtgAd0iMPs65Y48TZc3J3WmAdeK6VGBABuAAt26J3b8DEfkzxMKavOOxNyELoDKJCaAL0DEMYw QQZAk5MTxDVq80a6WC+Rq7VnaIZp7MQbaws8GsIsfD3UMz5T7DhxPIdpaAZZRMPr6h/QMD/3Uht8 qbwH+hnMwkfuM7F0oLZuQIdmoAYYHBGds7R48xxq6AeQOsP/rahGsZGlrUCGeYwGdngJoPMgxNgI ghIXzykHAByidKKJ3uKcxkMGwVA7OCuHEHmHZmpIZ6zJcwiHRgIHhjCxUpqq3WAlABAidCgQcIgu c0ilhgi/Q+sIxWCH4Rsrc0SnnwOAZ+g1r5mHjRSHc4GHnZxIk+oHbti/ZjApaxCH6cqvsGwyrcqq pnLLRWIHtXyqqdIqkvhJpcIqrXIdrQI94QPMwBTMwByIB4Qzr+OMwVRM4UMy/3BMDoFMA/uPyXzM 1rPMy7RMyaxMzMRMN+PMzwTN0BTN0eTMSKMwjyKRG/uqeiotU/orcbAGHgMpQ2oGRoo0vOCGacgR RjqAMMsy/7qihgMYrWYoLmsolb0QKuPsB3qMHbCIhonsh2cQCDjSQ2uYB77IMtgkiOErojxrq96c htgch34AvoSrTSvzjLB0KW7oh9NsTXyqvCCku3fwHLfzi3fhiHxKr/MQh3KQlv9UD/UYy0YaT7Dw z5JCOABdN2swOJL6KeMU0JLqh4l8tTE5qQMtqVdTz69Ijwp1wZxiT+oqpd+olt7rOHGotZokMoZz kYngIHz6SW+ATaO6UHf4CjghTmqwBrmYSMRCBpaCUBql0RD9rTD5IeLZDSjEjL1AhrazjDu5voow oTpTlzBhzVyaUSGlyOBUqQ1dqSDVUqIi0lzqIOLBD1b8P//NKC7AmEfkqrU1YR5mjLFPKigsxdCX ik7YhLWYIssH/dEwFdP2BAnXSYeDeBNVGrDf+78APIvQQjAWDR7mgZAXvRSEWKgsfSlrOIcHPCSc iiRZY4eVwgsvzSk+BNRA7R+7AiTPiYtowCNxqIjWOgfPeIY8eSGOe9NwkaF5uAbcGCTcIB4SkdE7 1akdBZE/Pal+WDYHLVVk1SlwCNXeXLOScocJqdawaFLhFLFIm6kQ5QqOWCh4uIw2Ajh2zMQFtJA2 akUMk4wCmsNWnQdAWhdz6E+eegYb9YtoGCuzPEtuMDifAlOXYiv1SKVoyM0Ga6m5+iki5aBwDTBy nb/iqov/PBkwzhlXRz294+qlpkokuQANFtKNet0pdGBPcSDVDLUyfuWpgN2psWKpfkCynzJNWuq1 dYmL3au75HLFAJyzspCy0DpX4wtDavOLaYwJkc2pcmA2smRWBHWHPe0plp0phYy1cWhakGJQG3VB cFBZmfJWhZBXc5iH3MMTKdNZZCrbhqgMVFszz8A7EWuqCTmL1iGeQUJamporskxQlVJYbl3ZaNQp qyVZBl2pvA3L8eTSbhXUmlCM/kGMkziHaWsugKM7cHFUz0jX5hIxPXmImvAcT8LUmJpIdwAH4RQ6 YnRBnCopbkXdA7AGv32zk0Upqe1SkBrLDVm3ifRblIq0/38tB2I0zAMoXZga01waCZpwnZqErT2h tnBgOM4JQrIFkRfCkz2BhyMNiXM4p2Fdz/Gcq8QtKc1YrOEdh9/VUbBQ2N8V1QhFh6dFS3cAXLGM NXTYUK1NqcVCX7PskDsVB5XV2n5QJr+NtaEjWThLxB/ltySriWZoHZQgpFySIA8Kl9saW4fQmpQI qFz6pHsYSBnVi7xAxWX7Qix7wC/8QsM0THcwYRN+MxUu4Rcm4RNe4Rd24Rqe4S+8VxW2YRrW4R7e YR/mYR7+4Rsm4iCG3wPgt2YYFW4or1Jqpxjzp30yqPfUJe4SVvPKpVpdM3IoNnGIs0Y6kAdzB2cF C7wIh/8D6F92YNZymIZIezBuIGOwOOMD6A12kN11a99p6Is1czBx6Cs4Cc5w6Ic5Lil/JQeY1YuT AsCJtGNCLilyFF5Z291CvuNTteRLxmSf6ocfiuJOviB4ELHaiRS1sLwzebhoUBdP9uRzsaCDcmVT imJXluIL+qeDGrcpBomFQDT33IZU2gj4SjwCMb9ImZCadU/quii31Kpy2uRbdmV+64ZBSghdkuKA AigXCgxnNiiQkAkpfuVadpsgdNNs+ErkigZhpeZXfuaEYqFtNqUgzE9qsYzb6oiQNDSB2gbOaWdc 1opkzqoYwIAZOISqYubhcWdaMghTMgfVOQtAO8GGvt7/aSYJeXVNmuhmdxaoTw5lrYkGBEsTBom/ hijUdBY311wh36ooUO6kxDM8lmiMgDGhQnWdyK3SIqUJf04qWOCBKLABYjiGY8CBORADFhiDPoCq gvYtkFhoQKqWuGAgbBnb+FgUzA2Rg3Abgxi3rL5oZz4IwUCbM9mXpiMQBZkGdPgPRjkH9kjnZz5p 37qGaE6IkBwYScWazZMIEsmNskDpm1YoiJqBIqiHQbgFLjgGY9CHSJCAUJiDZViGWzAAo04qZq5p T5qHH4MHmZvbM1GdNbmHzBi+WoXbQjWMdzBorW7lg5rmbLiXo7GNOVSLiCAHVuOGNAORaZ5mW76H h/kt/9zyCLfJPIWh66X7nUHqiHuIBn4+B+tJpycoAlbQggKoAFPYgR4YhEHAA3Wwghv4A0I4hESw AR3gAY6KKMmuqGs4EQnCrTn5HPggO6bErA5JMwTbk35KjFsGgJHa565eFwtul0OEoKjx7FFcSKKZ FF3CbXdo5/t+BwmCiKlZmxdhmxmpDRkS1+0FCZwepz0IBCcYBAN4g1mwAgCwAl+gA0l4BE0AAFjA hS7AAibIBB1wBmKAgT2AKPL2JFTODVt7D83+shNERVJspD0hCVf1CPsw7X36pw/aGgixF+obCc9O OR6TlDfxk1rOZQSviSA0DHmDl0GRVNpYnuV5ShgyLf8M3wpQQAEckAQPUIE+oHEMwAE6EINfMIVG aII8KIZOAAZ1OAELOBJISAQaRyfyfmvDUO230MWxbVUWndWB6JDP4NiyDYxeRSpubuV/IgnZSJEH iYtK2g1lwgshyiOIc17xEreTlhAS4S6seRFNgfBQcRAbKW7UsKi+rgleMAApiAJ9qIcsyAN1iIM5 qIUiOIZlcABCwIV9GAIXkIN92IA5KIBnF4V1EO80Iu+0ppYB6SM4rOAgTKz1g0wQdrZIGqJePe4K UhfxghFSDvOIcEMR4Y3MgO/4tlUYwiAW0uVtECTXAXOzafWyUZhYb6/Eo3UDUe6aeAIlOIVTyIJe iIP/OBgGTuCEYqgAZTCGdkADQviGUgAGYDCFbxiffVADB0gAB+gCLxgnyb5sCSKXo8lFCAoRIT8H 6XRAA5u1WnNUsXAbpxopTOsIQKMIu86Ux90WBYl35Uq54bu+uXC23pagFQobjsgU11CbB4ER54H1 vNq83ei15MaoQxgDUTiFS3ABSugFx0mKWPgCZTiGDtgHQCCEYqgG7XCBaoCDX/iFTiiDHQCGeBAn cooWazgehMBmEqS+/CAXognCPKt51yMwtLUU3x4pQyfn2qHB5cG8q5nH5EL6lKOGvtjJ4ZOksVAI sxMkCgf42IgRd+NkESI0F908jxiJafgThdoDXwCC/0AohmSXg2Dpjj+AhRCgB2NYh3iYhKt4hJXx +G9AmU7wBU7IA1doBPPQinKyBl+6BvowCRfjfu3Xxf0YPgAxMFJkwt9TpISQi8IYC0iVt6C/eqox bjGksgsMf2JzXhIpnmZoNxmaerN5jasHCHjZsqWDB+/evXfv5iFMl27bNofmzMHrNu9cM28ANgIA 5SsLHRWQTOXq1UsdHCWxhFRyVmmWKwOnHMAZcqVTMVOjFoXgtOjPkSMcAVyzduDogWnXlt67uHAh w3lSE547J1XquW7hxKEDB44ata9ewYnrFu3sPW/X0j0rl22puXnwrsp1WNAhwoT35jY9F43cNHbP 0P8R9oosnFWp6a5l83av37R7Ax1GhFfQIGaEBzFn5nzZLuiH265tm0euW8ahMLTggMCkAqcRJtVR wlHqGyU6E5jkCQPGiZxSwKoB6wnsRLUBrCWA4pjNKNJ33rxts1o1WuKmC6lGmyv3nd9z5J4hAzcO rFhwz6J1E3fumlpx1t5ea+oQs13L+hO6n4edoUHvaEVeV+AgM81C4YTzzmL1WXMOPNeYc99dnBmU l34ZamahZZSJJtE18LyDTGob7dFFCB9wIUlIrpiUiz1atKLOIoUIIIErb3DBRSnF7NOJHLgA8wgu uGSghSV77OEcdEdFeM08UNF1lUJYnZPOVfD8FyX/OeWdB1ZYyKBDDWLzwNdNOeaQVtdeFH5m2T3T PPNOQux001A690QzDVdggoPOYVGpec+Dat5VIYd5PYRfQXt59pCHEdllTjbzUFPiRo0AI4EXRITg Swi9uJIHLqc4AQwwcLgwhSnxOIEDMLh8M8UvxSzyDSfftMqJIoLAsGRRSPWzDQDmKCRlU01FWaVV x1YZjZTnsDMmNeeNM4562NEJwDlpOnYOnvllaJBpFiHkF50ZziNnV+g8E45FnNXXjJkT3mfQZRbu 1RRmUl1oYWigdbgWOJjCUgQtoqggAaiciNrLN06wgsoc/2xQyC4BiFDHN8NMggurxQxZTCmlPHIL /xlEbPScsMTCkx2VUjlbJTnOWtesQt084xWg4tRMpTfnWDNoaAcpylk0z5h70DvkRHNPowaFd85C HeYJj2PlpLVNh/iN67VcmG3nKIf46kcZaeiUoxEAsDQRDy6mhOAKLoTkwgknrtRggRhjdLKBCHSo MoABokQhWy647FPNPnJEssEGDgxRiAVCsXxUP+Z4I3OVMle1bJUvJyuzf1R3bjqAF1W1WLeaMx02 h5e94w46DFLG9TThwA6ppBEpG808ANxTDtaSUmi2hb7/ix/ZnnktmoRqc8RLBoOI4gEw6uRaitwh LGJIHcXI0ck/rCCRvQUikIE3KSFI8AslcmRAiP8ebKSRQwxENTmsN8bek5iU6PKOaJhDLwwZ3T3M EqUDHjAvUbKKN97hFsdAZUP+qpA4kGE7RqXjHOFwlF0qYzV8OcYaabGXm76GvL2ki3llQ959vJGO S7HtEFEQBSvCUINcTEI4rijJrLqANzdwQQFzwAMQXFEKNUiAJ6TIhSlQpYeSlaIOE5DBHi7nJOqY biFnoRrpoAKVqoDRc8eKikK6uEBuzYc6sDvgCI3WNWXhK2qHOghDFIKQeRDrMeTwBkEKIikVboYz itnMvhJlGeOlAwDvGAemfMGEZeiDFrqgBQ5kURJCvMJ7iUMDFZzhDFEU4g294EQIekSIXshiAvr/ SIMW9MGFU2gBBvpDygHCAYBsVCkvowNdAAeoxzw6iy5kLN05zAGAcAytf7Ar5PEqhBd+QapRn2nU vpQFn+GFYzTG61q+XpiQfCWybKFhTDhIxDYAJGIIBtADIhBRClxw4omTmMQNsBACTzgADPpwRj02 AYlvYIEXJXFYL3CxC1YgQBTtSEARDrHL/ZHjHgB42Rixs0cpbUd1L6vKUwK4x+tU5R6MoUaa4FMa ZV2Qa/n6kGagkqGA3WePmttcM8gRDsbQ9IWFTCA5fwqwSQXtNJh6gg8oUQo0IEAHkAgVKZ4AC164 4lNyQIUEJNADBdBCk5MwySmLwYRiPCIDYXCC/wGesJFgYe4e4VDm/7pIJ5aikSHo+hxLR1fMhLyj MblLU2O8MRfQfBNqXGNIow5I0zzRaYGbmU427kEvdkSjf0V7Uzg7o6+hnnMb4vjf2lTTCRf8Ywc7 oEFsbkCGLFRDDu3QRw7gsAsxiOACEQDGMIZBCFc0ohhluIUczqADDvxjDRKdqLAKKI6B6DGvTnEg eKyElahYZY/YIek2itqtgTSmPs/ES12CypfNXCa82BRROqbDmHu4g5eIoU5+pPkmFxLyUWi7Rjim YQ6MrHMjlojCH+6QiwETohNkuIMrHlGHLSjgGKhgQSU0UIdENCIWtZhDBngIDBd4IAFs6MNQtP/Y j8bMgx2aW2Avn8u5mUVXjNByYJSikd1tmFiCcHmLDOWimGneC49Tico8wAsPEV5NvZqTrFrOIQ7p BBJR8/3MhsIpZbu8JXeN7FZ/N5KIX3zDFaZQhyZSUY0ueFkOTliBIHaBiTqUIgRwOIEmAlGPSgDB ED2IRALqMIMlcYStBxjxRqYT6KVMRy2FFrR3n3IdMVJtLml8hzLhwQ5i2Vi9K2ObegVbFznmqU1M K0iUHjK67KpFTY5xB3wAIOihsHoo/cX0NVzNEeoMeSm5Cx63MCUh52Aj0LMGBSEw0IhZwIESDzAF FlJBCCuMYRC2oEEIXnAFS6iV1SLGbzcg4kH/jsCjGedwJIImHQ1xbOMpy1rgA6kLIfh4ML2ObEY3 yLEReKADAOmQsTg44umzXci8cIKad0wNSFNfFNXeeEZaWJ0OhSQ5eNl42jbGQaxrIINYVHNk0gIt EMdM406Cxkh1zoGMWF+JHY1MxjaeYQ5sJKMPcrDAGzSBgUgYIAnFiIIBssAKLYxSFaiwg1BafUvM XYMc/XnHNNbKDXJwo1I1Q0c3oF4acMXMZiVtDNIjxLZ3NCMab50IpdqT8mwYqhvv8MxeuOadvPjl HqMpNNnhw7Zua0Qc8HDOM6TFGGTkm8b3EId+823vZ9g7HOx468H7nMDuxJpt53AHtMxB+ItK/xpr 0jhHNsQxaShEARicwEUp/uALYNSBEoR4wg1C0IhqC51JwrrGlTbClvySg1jZ6IaI5rJLchwEPnEJ oEHefm9Iy13VjizHQ/Td8aWcxSKGpsiF3pKNbVAk7YsxdKkFbmkkCy8b9pb3Rp4xDXJIqOOCnUZl IbSyWy/mNLgOGqQFXehcT4Qo7CFWB2NfLGJxZA+wAEAfFEE2sF7rtZ6IeV8BJqA5QAREfAilEETb qYnxaYSgzUNoJWABUqCqzd+sMUaWsRo8uEOr8R8GlqAJcoQEnaAKriAAYEOTHIA7xKA7NIMM1uAM NkM5lIM1lAMN2mAMNgMQ8qAPDuEM4pIRHv8hEiahEi4hEzahEz4hFEahFCJhXyHDFF4hFmahFm4h F3YhF+bXAXSbF44hGZahGZ7hFnJDyj1DRaGhG74hHMbhFzYDUjwDNcghHuahHu4hH/ahH/7hFk6D Tw0iIRaihWieZjlKIi0iIhVSIzIiJP4UNxgWA1aiJV4iJlbihzQgJ0ZEAwqiIYaiISqi1CDFt7Gg 0MXaCWbZabBgddwDOWDe7dnhkH1gCXoDNSCgrM0aq9kiKs7aOWigL6IinxVjAZqDKWZgiIUDNdjJ WXCDO4TDvREGMlDD3fEiNg5FK+5if5nDNLjDci3FUnSdqoljNjwDN7zHCV4DOChTq9liOvj/DHl0 hziAgzugn0CwYNC84y5uxDwYyJggQ5d4BTWIn3Ss4Db0w1GcIj8ShVdYwzuoHSMWEteICab1ojZ2 QzYORTR4m56ABxlhx3/4EsCNAzv44jrhIgK+mnNMQzlU3EDAhzgsl0M0Bre4g4ltJDZ6QzT4mgGK Qz+Mg5qI4zuoHESA3T2ggztAg9B9IDIuJEYGGjXUnuqAR/OZBbQ8GoCkwzgwpPFh4za2HjJkG9Ip SO7kDjvMZDichlmQEdWQAzXcYi765FCEYDhQRwMKDGZQH44BwLR8YJbBXutpxDa4wzNQ30SQ3UCA 3QJGRIy9AzQ0g0VNIEY+5QF4Jaul5f/E/9tp6FRaPgM7dNxZXFeVvIM4qGFTbkRYttq06ElgaF5a igOBoMPhkYPTiORfmB1hpKaqtSNdzhsy9M9hJcse0cVDPOBScCUJ0iUF9uRXDsU1cMOdiAb18SV3 kV2enAOfwMs5mGSvHWMyYiM8cAOCAMY0nGdgDAbPAF5bUk1VdEPuwCX4Cd1qss08pKN2piU76Cd5 jAU6AN5pXFdWoGd7jEM/9qY7fuXmqGGeRMW+mBGAQAR9eCDT6aKCqlowZuM8dB1iMeZ1cteE/I9O jZ808Ik1OKfQWSZDzh04iKjh7adsjgk3cAO2+Iy2hORf5Jcd+uJqboQ5UAMsCgbf7aeQpv8H362l tjyQeaLfOPwRN/qmglJDNBCEd0zJlCzSRIijWrxFQWbgKfYXOljDNEBIYybmdTJmQTRNbXYD1VzK TvpoeDoH5PkHYChIWorJOMggNyBDaLJlgGZFvEGLmKjiV3aDRoYYTo4HoCDDM8xkf1IDjYYFaDrN 2GhHeESJO7yDrHkDlHLEu3DNHv1SBbVJOlAKoVGgoU2DxDVlhvpaNxwANdQFA5rpmU4ERPwPNJCD NDiNX7BDTmVjQkKlc3AD8PiFbRoeeXjJDO7gOPznWo4fpfrFaIocMrTaakbnWu6MgTTqfvZnnroD jf5nx1VNZoCk6bjDZM5apwLAMzxDg17/xS85KL7wZfEFmqFdQzNo6nMGTX+RwwGUAzmIBmM+IIja agNShaGeBVR4Raup6FAciLTGW3oWSLXMoEI2Q7MCqFkoGnvYZjcgwzhY6GrOZDewQ3n85+GdJ7JC agxyg3rAC7nikV9g5Vk0Q9Kpa4J+bKeBDrwSZ5SZKqGql5YCaS96JTz8KzkEmZkOLNk1picOEl89 GoPI5nduRLBeJkegxgAZqnkKaYFcizvooDtQA9+JZkjC54vyCTiwWljOw4nyCXkA6GjGWzp9RbiK A3eOUVbYJnrmjjs0Es76aItqxxnRRbzCEfXJndAuBX2gJjBi2h0CbJBNCqUwZgOOS35s/4h2lkm+ DsU2xGk22OyevGi3iklXgAU34CANUoO4smka0e1MhiagaKChDsU4jKXmad74cWZ6iolYuAu3rqXH 2ibx5k5+NQPgIui8ZaqysNgZzQwYDRmOmeN1huBy8mugWaE7VBd13o7T5qXRQMq8/g/wsBc7+Nrn CisAdAM1iMdWtKuj+i5BQhIQNoPLHobZ6dgA6ZRZJo0uqeahAoClxMlM6m7xImthhMWfiMmR7m7x oudasq8qcqo73sNhFuexhCQZgYe5pss2fGhiMmamPu5aucMBjMNWLmCnVeftJFa64MUcyUW5kS1H pC/WboTO7IlgNGr8FohXjEPqxuB58P+p2T3o1v4pOdQbAKPgM6xLAfcp3/buVygwNQ6pA5unWXbc ORCt8i4TMlRf84LktD7aokXoUZ5xmZ7D2gZaq6Zv+xqWCkMNpajdopAO2ZBqCBHEfbrj1eorAKDD mHomkX7mYJwuWFwLtgBKaLquVYpkxLLDGoflVjQNenps8wGGYBRGFQNKo5ql4RWw5tFeN3BDI1Hw RowpQWCwtH5RaaZRGTVQHXUvV35pq6bDUTQphJSqh2LnQ0yfILEwYn5v5dLYGtvwKYpDrOrUEw/y zgDxjIIJeqhHyqJlbUbxWkopAGwjRuzteQIG8XKtgrSr6bbLn3QykcaugthpfqnqdPj/Jjy0Y4Mq 2nsq7FyxlLmlUZJuyL+AQ/Bgb+j+2Xq43awOLNj5cqku5sBW4kQsnJL1XXRsBDrE4mvG7uGlJTrk KRC6w5f4CScXMDIwKmymczSAQ0/G20YIBnzqZxZD8A4nMJhwckXbaWiO31bI5uGNQzsrEzgkE5aw mLSWkXagm7K8Z1sGkL/Yiao5p7/+GTucncE+oNOCnUt9cOU2rTBPiF+kU7GEp6W47yDf6ZjcoP0q 8BRz9EczMEjX5hZ7X+3aW5NmUM/sp0y36+k68zNvK96uc99uxUf/JzgknUou01v5GIpxcIrZMXTN rIAW0+zcHU9aLXRIqUBLhK12yBnf/0UwE/RRZsPLuGRPWqa+UoPZ/QUmCynL5qBGb+tHm8eMgivr VnG7pmx7/O82uovXDQbfhXJsrjbLZrRGhwVMp7OdxihhPIM7EIspU8M0iBB0aeWP+dKFWOq0ytU8 dANOJ/VGhIOwQF05MdJxqjBhVaepTt8Dagk3uAUAGPNalYNFpHT82q0Qs26jEki1bLR/Bm9+lawS t/U2wJsHySajnjWj0rV5APEPCrGBqOWLOip5gGY4NINjz+VIY8hGGbZIwbI9/3QFHRAzsvHK0OFR gOOLYal12qo1IaZiVu73RsMdzkd6x6k3jAN78PWA97WR8nCycnK7ivOQHl5+sQOq4f9wAG9DOuqU eqJHRxMIfGd0xvr4cO84n3aJEpsyAExDrP7YipkbvDIQlYDkGYXqPYg2G7PNPx+FaCvtAkp1wPwL ioNoZQjNUZzvix+FHwsPP5ODyQ44lH80t54zRcfmjsNmyrpDgq5md2rnYNQ36w5pD0Pq6jqryj75 judOXAZap+6nshSnAKGbgwIZ50xFNsHDgZCw7CnkUehOqSp072him6C4if/PMxyF4M35DWctkGqF bPLwMq+zcH8ygAc63uYONxBqNh8q28gm0j1qOTv4ed6pebxsxynznoPmNFRjSs7lWknpI8KYlUK3 A3m7Zfy1q6HoRkQDUkRjkIWvqxf/tMHqcohehNFxwwHIudU+NKuZw7UIyETDJrQuKXpmcrs0sMgd N9sGMEd8o6FuRRWn9YIzsI/nTgbx+X7GIjmg96xhO0cACpg7KEnCkYb4bLKwhRLPGoqyzWMgRfv+ BYRsdorTqq3qSTpB0i1XbXqbOkMyhgkCqiWzh3vH7u7mabpmZJ9RDaZVt0zHbij3rfF+MvwuMtJ5 2zvm4q4B0qVxhDI5xqME8+V6zhcpZYJO4LdpTgGywy3rzgeTnfEg5lF6oxE2w9fXO50TBToo09OU RWXd3kGqGkYp1n4wxFZMA43WOV1q880OGd7OqEBmxV+oLJv+mF+g5e560Hkkby9S/0OI0KY3iMPm zKRGlOxdbrAMDyXaO9regkM5TANKfpvKbYS0KEWJGR0S7iA30Gazm/ARAiwvrpN6m0O12ts9nOQ8 bMVF/W9RPcM4iOn6lgOkVsvNDqbQO9L/skOJhZg5kAO9qF1iqpc1Wb96TIOF/qZKwuI1AMqPGt00 pAM4RN14RLC2nL1/AI985FfQ9+M+PoP3eQM7KFk0nO9byN4LMiFARINnDkBBgwa9Fdx2gOG5gte6 TUsHAJ43c9vMkXuX8GBBbxGzdRSJkCQAct0Optv2cCSAdOyeIUN2Lhy7bs0qprsXUiTHntSyebtW MNs1jtmCEvWWrlvTe/DAIRtH0P8bz5YjvTnMli4d0oFBd4o0pzKktn7UCF71WBLAQobs5r17F1fu XLl069alGxdv3r55/cY8dy4aYcOFoyFWbHjw4HeHE0eWrHhyNHfdDhNOrJkz4sOPNW+uPFqyuLuA Uae+i3evatTzujGUPZt2bdu3cefWvZt3b9+/gQcXzlvtVYs6VW7jCg/e04HFW4pzZ416devXsWfX vp17d+vNnnkXP558eWvUDkBHONYoAKEWCQ91743+taI+i3sTV+5As9n92GEHGWvGma3AA/rphyF0 +LMtQdma8W843KxB5gBu2EFPNgtpE4cbaxiS6rZ+JDygHHAOsAad2srhZkKGuOn/B6u13EsKIaSQ oi8bc8wxqr218OuII/2saSY22cqZ55loqAnnGXHIsQmZmiw08plu0HHymXPQmYYdcZARB0sNX6yt QgSp6UZBho78kpxnwgmTzTTHCYebcKYBZxo4qdkSHHZMnMbJbth5ExxxpnFnGnHEeeaZ32Kc0aMf hTIHnne6Oekec+YhrKlz7knHHBuDxEocEKc5YBz+mgnnAGoYBSecc8AE8M8UnwEHHEXDuZWdLt88 Bxkyy5ytwn6i+ZKbA6fxUtZwmkGnGXIYesYdcPrBExx0+kFnnCilXLFCWcUZR1Bw+tyTGm6uhVRG tTi6ZptOCZtmymlAPYdQm+Zx/w6erkoV0r1Td1PxwGIRNpbDhBl2tyX66APgGtjImabimPCMZp5o yBEnnGjmYq45gOcLeGAQG/4t2QOeKTAaVw9wJx2Uczsz5Zt1i7QkoYaq6h6OBY0pzHDIKSzTcMLp JuTktgmVp2vaSyje/XDOjZsVD5jm23KRwQy9edh5p+CFqy5bNp0PqqpnczwVVEB0kLG4G6OjLPqd UNNh7q5t2oM6SP0aNJu2cThkB0QuAT0AT3deJds2mwU3G22PeHSP04gQ7SZXcOCcBrNougkHSpDx 1qnfe+7Z5mlST46cNm5QTBhy13GevCpzErqcnCiRQQdu3+FUOlNGpyH93+OX4/9qm4SQkk9igmln iBpiy5w9+oYnl3eibB4jB2mYZOodHUbJ6XRRKDGbh7m8VUpeVPpGlY/I4DRHhkloGWomcNy4cVF2 rK+XMrQdJyjw8F6AAhSTZ/gOGc9gx6w4VhOLnSNkIkNe+5Z3DZIBDjjcoFZ/bmWxfihKN+M4WLFU FMCboS0b26gIpwhTMS8tUCbocFTnjsaO4s3ldKiz4L9ERZRRtQ447tidO8pRIHCUQ1Br4h81nFg9 AJapH0+iGe1YGMRswOMc5NjJmzZFDu7ZMHig89LHQDYPNdonHd5wIdQ0uBKL9Gh+CevHOBiXG3d8 6H8Io0aqUoSOefgvf+awUGz/qLGN2FFwQjGSGnvaUrFMpYMm96gKALjkKKIBTRwItFtYhtK0TV3j HvMgyDVGVcdiOQpBhOvG/hhyRyTKznHDscY59hiNRtHGHeFwRzSmUQ54CMQa4VAjIIOjM3nJER6Z emAzBRQ+GzKKaJly25ccqJG9qK8q2dhYRQCwlWtA70X8ydKFvJRH2vSDXSWqXi11QydyZIkaEiJk bUgEI3coiJ153KdwImWR5UkMNs0SnaNkglBGBShpwEQgo55UqMhoZm5zmwff2rINqpUJHXmMooP0 d8UXWW83N0GQolRUkwOQ4563sRbDCuQejNCHKXjaXbNgcsNG6RBpmJHh+RzY/0kdyo1uu5tHSLZI zglRY5/iKBg3ygHLCZF0N1d61DhGGK1uwLM26Gjpi1zkniAuJS4RRJrbnkS8ihVKX83yXqOEetZm NQt0hbnHUCb2DKkC545QHZBuilQOdb4TYY+6WsHGcU+RdlBG3ngOAMzxDknuTnRfgqjHHqjDwQgK T3BdKAK/hKemRMOS4XyHUoWTRKguUTflaIY1BjtSrv5GRYLFKmDdcbAK7bU3AVWJTCOzu5PUBKI6 rcmsYHNWJ4WJUdS02O7m9o60oG6jE4pQYlelmwgRiJbFakb/+rdY2jTDHWRKYmwdBtmNuMeAFhOd UJvbKAcizW5d9BKjGuhAHf+eZK67iwZVdoLa4LwWdnvUrjXY2d0yQRWPEcpZi9YE1a/6Bm3M8Qgl oZspzD0pJjrEDKhAt+EGismtJ5nbOQbCvDZWdzgxKm+LSgiiEzJkHLGj7Wx9o0QC64Zx40CZiwq0 rmS+yz2mLAgMKQi6myINrp6z6MY2TLRmNYW+HOtGOnp2DtyxWDgFItwdZyNV/90THCizcW+o6ptt FQgZjKNebdAzS1WpqkV4ZOxB2MYTc9xDXyeRYUVh0t97YKomMCmacJPm5HMMFADqGxhvddMPcTSj zDF7FJncgY4rbkt6ZwvcHbkVS8Di+HXhMVGKPhTFNws2lgUiL537Mw7/8FH/NtQQ79mITBQtz2fP IZ5bYiIiKCVT0HPgKh7HJIqZTXHkbh4RMG98fEIFXY0/BSoHOsAha3Xe9lW2IZw1YDyiaxcM2yBy BzJaNEUIIcOJd/zuAWxMDXCUyHD96TE32E0NbBfIHYMNa0cmNhGiNI1jcumiNclhmGFGhLIgk6T6 cPeQUCHk2bVJ7InwPWPZPIMbsHoUhJwKIVmzO55R9VYz9C2TZvTuiv0At7x9HJwKretEre7QYgE0 LHzmscZSQY9LJC2OXTXtqAaB2kUo+ZgPg+xfnZLsiUs5kKhZBB49ewg7dhWhCOV2etN7Lc6SiGYU fTSQKhSgjNJRjrlliHmo/wtJxHitHOSYIxoNxAiPepQ2oTQtLT7Rzz6nA1vqfKhFrqVGi17b7yJN Z1nWWGJuy0GgIh2e0uSVdWKtwQ39ybpFSFxix6O6qmasqt+C1SuBIu9jwRbpQ42/uObP2/lyoBvl RZJ1x3ObWF1Rml3UwOO6rrXHNo+DqcTPfW7ZhSKXuMMgYTK6TlRXlB/RCCk9kVpVmGMjIcXNHeKI BhPJERUP4ak/ev04bcrxqwMYjsv9KQe0wAMe2xjxLdZgB9lloyIMUeMc24qGntBB3qJBQcKhmGwD 5ahF3uQNn2DnUQoQZmjDGl4J3hQQN2JqGwSr357B7zxiR7Yh+nwkanykKIx2ZCz+JZUexj2QAeus QRyg6EH4KeumRLwcqH9aRd1oLBy8ZfbqzTawhal6pWY6ahpgRzoQLPKoo806LhzeDN4ERPMCZLGs 4U10jwlvo8zw6Evwj8ayZq68UFC+8AuJy7K+BIGQBk/CcK7QEA0NSgG77g3hMA7lcA7psA7tkA6J z/fucA/5sA/38AACAgA7 ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/content/vol29/issue2/images/small/zj40020808100003.gif R0lGODdhyACqAOcAAAAAAO4QGO4gMfaktGp7/9UAAPasvf///6ys/2pzzd5qe//m7r3NxUpa/+7u 9u6LnP/u9s3V1WJzc0oAOeYxQUpS/4OL/+5BUvZzg+YIEEExKe5zi9UQGKwACOYQIAgYxZystN7m 5gAQEP+svfaDlAgY/9UIGO729hgx/zFB/+YIGAAQ9u5SYu727s3N/zE5QdUACAAAMSlB/3N7/9Xm 3r0AAOYgKQAI5gAQ/6y0tMUACPbF1d6kvUpSc4OLpBAp/5QAGO4pMe6kvfbF3v/m9s0AACkxKSA5 /3ODrJSk//Zqe3MAMfZqc/+ktKwAKf/29lpiYuYgMfZqg8UAAPZ7i//F1fZBWikpKe5qeykpIAAY 9u4IEDk5Of+9zRggGBgpINXV1QgAlO4AANXVzaSkpJycnINKi+bF1Sk5OSkxOYOLiwgICCAx/wgQ CIuLi5SkpIucnBAYGGJSStXV9oODg5SUlLS0tAAY/wAACHt7e96s1ZR7rEFBQaysrINzvay0pKy9 vUpKQfYgMXNzc7S09kpKSikxMaScnDExMVJKQZSclL2kpMUAIEpSUlogi1JSUvacrM3NzWp7e83F vebm5jlBOd7e3r29vZSUxXODg3uDg/9qe3N7gxgp/4uUi2pzaoOLgwgQEN7e9qSspJyknFpaWgAA EFpKSr0AGAgg/1JaUt7m/0pBiwAg/xAgg+7u7u4YIO7u5mJiYnuLg//N1SkpMd4AABgYGNXe3hAY CNXe1RAQEO7u/82sxUFKSr3FxYuD9rTFxebu7t7u/yk5tLSDpJSc9ubu5ubu/8XFxSAgIL3FveYA AEpSSr3F/3uDe5wQQc3V/2Jz/xgACFpaxay0/0FKQdXe9uYACPb29vb27mJqalpqampqamJqWvb2 /7S99ubm/yAg/xAg/wAIAO72/3tzc4OU//YAAJys9lJi//bV3v/u/zE5KSAAAM0gOVpq7jEYxUFK 7r3N3v+LnOb29hAI1e4xQXOD//be5jlK//a0xfZaau7/9v/2/9Xm5iwAAAAAyACqAEAI/wAPCBxI sKDBgwgTKlzIsKHDhxAjSpxIsWC2PgfIWOpD5xVBbh4rihwp0B+EfOuqjICEgQmLC0ECqBCDbUsA WFHusdhED9KILusgaDv4hAUzZrY8jJjoz0EIML9IPfPW6IWXUDHCaLXn5Cizd8CSgBP1TSCyeE5o MgNCjWTBbSEPQgHitYOsg9yAZDgKzVo/g5Xa7NqVKxQntwlDeBkcZ1ecZQS/hUNiqrLlDyj0SbOA IBovf4hDD8S1TTRBWScMdsHWRGCINclML+RVTR2KEinwRQN9wIW8VKlSIJBNvPi10hLzkAHz6AAl PqXyfAzpTwA/gpQYJPPSKBIuh9Es6P8r0UkdgnDF06sPTVo2ODPoikArK5A2vE44Oklz9SjW0PUA BijgAe2JVg4B4ywRDzyrINTPLZoMKOGExF3TDUJ87CKLHdlIRAsMkCQkQS1/UWjiiSIVaFAefXQY 0Qw4tAIcDuoUBM4akaCo444QHXeQG1eU0g0dLkI0hA5CFHTNGiDw6OSTCalIUDZXKHMFInZIdGSS A6WBHJRghonLhRZ100cZXEhnEB11JFPKAW4gogyZB/SCpEDbKBOmQhXgkEorNBYkyy3hkAGAG93c cosbbgy2iym3DLKLJWQoc4kytdxSRiTKWCLpGnw4MMgtyqxRh0B24AFAHYUAoEwkAND/AUY3pvSx hjmmSUnSBthgY0uvv1JAxJ4J8cJLOcaWQyyAFhI3RBHuqLLstE9a8mVEYMQmkCVZHjALAOCC+wK1 BV2ASg1ToFtDDSbQMlCii+6yxiDd7MJHhmRccgsAiNBhyS50CJQMIqM+kgcAgxZyCQCPcAGAtmUA wEcha9ixyyOFXOEGv4kSDIAldeCxcSl85LFwpbWUggcZXNwii8gG6TqSM3cAZ/MdxpCrM4XWEsfF FFzuLPSAMrvVC9BDJx3gmBS5kUc2ZVA6CHryOIGOLczA4AgyppUom9ddK30A2CTlQQcdboA8iNMk WcJHMnyAYUkppVhCUjZ8XMJHJJGU/1LILXaPVMba/pbRjR2FkHRJKX10s1EpkSBSpEh13AWGLGRs KPbmnHcum8GIDdqt5w3lwcW2hTxyKukS7rOFV8ywwI56VPh6FDb0GFTNEsBio7WyoUHhAEK0wAK7 GB7sA8HyC0BASxTHcwAJJCRgsIkVVrgTLrhLwKITFiRAIsQ+6xDxGX0CeV3iX2T/hUso2wOQQ/oD kaHq9ooAokgeqmSRSxvkaEMo4nCFZgzCE3UYBSHScQ58qOMIbGADCmTQAAIYgxDXeAXZ3FI0t4AE IZDABi1osIZLPEkdNbMZcMZRgh+kAB4zSIILwoE+1h2EacT54EDysQUqCIQT8WsEYv9EkYQKjGMc FTDGHGw4oZ4RBwotOIAV7pEQKFRiIN9wxgyOUAIUSIMQNWSizpyIGBZg4yhYYwY2POAuggBxe80R I+vI6BaarRA4d0jCQXzRAznKkY4CAYMyAEAqAIxuIIO4BCXeVAdElGJ1dbrTQTwRiuH5UYw4JMgr wOAGMvAtLgz5RgpSCJxOiGIgcyDH/C75x2sR5AqDUsYhHbIlgpQDYKy8JCAHQodd+I0SWpLkARrB h1yysmh1GMQr8lCGiRwpRJ7YBfB2Bg8/pQIH0igIR3hzgEnIohtguEQennAAOmSpcnUgUh1KYQ5h WMIc2ZiVLExIhkPAiQyBzAMlzJn/DDpQYhCTcE43ymCyXNEpNNWgAAw4EAN8Ds1+8VslnqoEr24M gpC7UMYk3AaANUAOl0nbpVvOoAMNHGNz7JBCPozJEJEypA+lGEQeIqEvN4BBIBboxDj0MzmhUaEI PjRINobnEQdk46gCyQYwlQrMA4DyFcB8RTY8MtVtOXUgdQBAJJrqnKaGxKgHUOpVD2AJrwqEEl4F zSvIWZAOjkQDBVgKS43p0pF0A2lzzaVbFQKGiWWJEoXgwyH6wQqrYUMMWhtGXv1YV4o4gBcD4YUl 3RI4qyKmsklFzCvi4g/MikSsAvmGZ0dSJEqEcbGoTa1mk3GLOlzkEgPjl96uQIdQ//SyXnWAZSno IKRHIEJfZVgDHfhgB43plgtl2AUoVctckuwDFizg5nowEIAHJKQcxjBDAMzgAom8Io51KEM2EicQ YfjiE5rgxDPykAn25iG9620vfNVL3/i6V77qva974avf9vr3vfn9L38FXF/9aiITmjjwLGahCQZr whzdkIUZTuEIaCwBCBcGQgdqkEavYM0WUYhCEEI8YhGT+MQmTnGJRxyEFt/jHrAQw4ePIoYiOOHG N2aEjH3VK2ZcJzQ6PAggimEDZnhgD384SA7cIAhmBGEPwTjILwBxCSrn4DslOck6aLESDCiBHzCB xRa2oIIteAAnF2ABBsIHCaCwg/8d/vAHIH4xDzrnYJoDuYYaFFEHPkPSIP7YhFyd+hT9ccIbj7DK YBS1hhjE4ANhgDQOtJCKCaoDH8YYSw1pwYJ1LIQHFFipbOByEH9cAHZeCYICVq2ADVAA1cwQwwU4 8QkJ1FoCuYifET7Ba1vXehuf6Eawhx1sYHejGxIQ9jaOzWxlfwIKawgir4/Na1/cD1zkGEUIYhGL EzjAGUkAxgN1Oo7MbKYac1gu0VwZPHUPhB4B4MESJrFBgkCABGxdzwzugAN+VwCykUEGuM+hjhSw 4Yid0MelETDDniZNFOxGDBRSY5ARYGMpulgDIHRUjmpIgw0lsNkPUnGHCjRXIJn/NA1qCrIOMYTo E6bAg8wZVpxwIKABPyiBPs6xxJNDZA4SoEg4BwKGWU58IBAIgBQG8sZwCXEiczCGPlLxgwok4ZQ+ D01jE9KHRwyCDslwQynKQF6BQMEj/rjHBRCiCl8kpB8uwEcKSsAGeFRjslln1kFFw41+8CMK0iUI L6jxAZAfYQbOOG3eT5RykoTDETT5lRiAoFN9WCAaAtFE/Jqx+DBtXSLGWIJaULGHg3xie+SQVueh 9PmfqxCP1TgIKcjB1dVDaa9Ql8EdWnEHNoDjIKJYwyhsv6cODqJKg3jEFQwyiFKExA2P0NYBhsCI ASAEGbtQBPH3tEs7RIcL4J+l/3MC28/F8aGZAtEDEHiAEC9tv/gRd2rEAIARh4CjZuO4Wc4Gooor vt/z7JYNIJMMlkAHOQIR4TAD8UANPTcQUIAI/1d8e3cAlzAkj1AK0acljBA0P+QF9RaBOtJBdIBc heBuC1FLA6EIoYAeIAgmLoUIa6AMtyB+DIGCBwAIaxABLSgm8ccHMngL0odV3VAHZEIGODIQKBgB qvRwWFcQZDcx6HcQdZAmCeECicCCChEJhXAhLlAIsiAKj3Avj2AH4VAIDqV17FYGqnMLr4AIZ6hJ UXUQtRQOoeAJQoMAKZR/d/B777ILdpAMr2AHACAkspAHuyAQysAiH3MItxAJ3f+QIWO4CxETCU+w C/SSOFdAXn2gDFmlDHRwBcmAB3WwMWkDAIOQK/FXEecwAb2CDUBwA1AgNt8AcCfXeg8RDtDwK5Ln BGfwUNcGAHggUQcgCwBwOnyABxHGL/tCChuTIUAIAGoyNLhHEdTHCl6AZzpDCr8oPwRBMnmAB6Yj C7JwOlmVfKb4CACQXNEoNLYYEUMAA+5QfztILO3oEJrwi0bwgcRCBBSgA+sCC1UgKHiQOKUAAMcG ALuAB7twCbayL3GzBuu4ELBiQg+RIRFJEJewC0HIEBA3EczEHAcABjRVTE0HLlegj3vSBCawLugS VAMhC3xwUXnAB99UTN2wKlz/4IcHUwe3oCYb00gls3wYYyiV8zACETF5sJMIw1uNAkywBCuVYyqP 4AbKIEursgaYIzJc4E+xEjOpCHX5lwphWQKxN488aBoucAR/cgedQAhmCX/EoQF49ZZiMoFucQrC RJe395UVUQdTYH16CSaN5xblUAy2oACKFZhPMo0HMQh2QAnNEV7ZoAr80DutCAOnUAm+4AuauZm+ YA2+wAebKZqhOZqmWZqVoJmkuZqnyZql+ZqumZquOZutWZuwyZmemZu6uZu8yZup2ZvA2Zuk6Zkv cBim4Q96wAKQQIuK6UfdQAZlEDBo0wenOBKR0AyvwAWv4ABu0IY3JRKxoJ2F/5AjYIAId0ESjNI4 GVEyGzkRdjAIkdAcbrNObtEHV6AmpZAMgyCPzdmf/vmfJGEHa+AGWLUGV7AG/JkQsuAGdTAvhfCd AkFcQwigJvIKwuVU2XCTFEMHj7AGLbOJymcHPugyj2BRhfAvfXALiJCOV1AHfHALYNAqB0ihqDUs ErIADWGjEwGh3NKcI6ACZxRdAIIBrWhdBtEPKSB66IAN8YAPEJENeUBe6fQcA3ECUHEJv4ClWpql XLqlXtqlYPqlYhqmZDqmZlqmaIqlv5AMDPALkcAAcDoPDMA3tAAJHoAUeMoMF6AEfNqnfvqngBqo f2oUsPMrarYJ1oMB7wADrf94WCACEZaQOHZABl53ULKgeEzkD0SQD7TQBdSzCZvAAvcgCGN2E4Lw YlbADxjQE03AAjPGDCqwD6BRFv5Aq7Z6ALWKq+tDP0llCdnSB55wCjUgBrADBPaQCp2QAi9EDTr2 K2cEDeuAkhGxcgqxDkoQkNWKAaImGiaRD/mgEizxZWE2ZmV2ZlFwAfxABT0xArSwDnA2EUVhCxmw DwrRD68QAroQFaAwCN7QDIawGI3maFpxAzfQQkegDwRwDosQBbYQBJ52ENVADSYgBtAADPOgclGE EHJAF7HWAadwEHLQAV4BBMVUEP6wotujARjAD1MUAAEwZh5wEwKQZmsGCU3/0AX58GaBZxCWEArX hgcbVxCjED8AMAYFoQ2UYAmRkAOz8AmqgAZx0AaCcQtZwAXWsA2cAAXysLVqWWlHIA/ScA4yREMk cQ0SIK0UEWQEsQAMi6dpZAscYAIcMLcm4LZuOwUd0AEYtgTTED9rkAVpkAZoMLhoUAmBawiIG7iB iwiKy7iGwLiMq7iPmwZXQLResLiK6wXbuAaDkQuB8AmgoAh/kAwRMAfRgADnMAMVcASd0LpHoA5h 4Qzh8B/rwZgTcXSlRgG2MGPvsK0C8QTv8KrvUA+AET/VWRyUAD/b8wsGIYjxYwGYIA3yQG4/8Loz YAyyy5whaJdugbsHYQkc/+ABYnAPtEC8BlEPY3ABtsACumC+BdECIUAJ8RsCtDsR4IAPM5AOCeEA xxALlIALEXC650AA8oACP7BTKUANM3AOlyAK2ksttisRpFZq1kEDIhBlTlIN1jQjBHAN4GAMBFAB bEBubNAAYTsWWGhDETytFGcQOnEAEdAGkMEjvCADISeWwPEDCHANXKNaKwwRajsQF0BFMBwKMzwh 9gFyuUEIKJBCNVOWzVWPQKxuLCAABBIVa+AJvwChNYcA6tAJd2B5mHcQ0eAC2MhcUvwQZ1cQSuAB s1OSAKB6oUFEDZB/+mAMY/yWusCXFTHBAkEFHmCjcBxHFXENxlABYml1Tf/YnD8sEJEwMRgBWKpz Gi38ANjwsAdweuSAbXIMHucgDy3UAOmwyBTayAfgmJbQHHbwCHRAJ//wCBQXQthaEFzwJgwRDfgw HuVRDcJAowmRxkdJJBpBUJh1dPuADfQqF/4XGS4wA3NXd4TwwL6MEKbcEKhRBdgAmAexDZWQRR+X CmxAAOCAqdO8ENW8EP5gBpCQAblTELxACB+nBZ2ADy5AzuX8EMCsEEnAOzQBDTOADF7cCbhxDj1g CqWCkLF4zyMxB58gGhtAE72DDozQC2AzyAo9Eud8EPkADT12FIygBwYBx5x30RWR0QgBDu+ADSaw CAhBCb+gDG+zDFxM0hH/YdIGUU2t8AOAUgGKBwYaR9Mjkc8nTUp4dA4HkQceCNQpwscQ4Q/6UDMj 13sP3A/KcLxKPRGDKRCKRBG8AAzx4AekPBCksAYNctUU4VKvUAe7AABXQIMKQX0cWBC1wNRmjRDN YhCUcAWDYAksEoUCQYRESIGFQAdRaIM2koN1TRFFgzhs8oa85AaLcwBESAmnIxDPos2CMi6JPRG7 BAZ1QAZnkwej5VRxKIADYQG3MQ7SoL28UIebzdnxNyr3aYIHMQd/okJ7SBBqsAst/NoPkdUH8KKy MKMNAQ43jMM4g0VeYNW+7RAywzKyIC9+CBHOwAZ38APXPRwpuAY93Ny//81ufWWgreLYDVEFBdAa kaEMkuDdEdGRBdFPdCCOwx1MIUIQZLCC7A2p8dcHdvBNDleDNRDXvkDI+d0QRZMNiFDLfEDeNZiX EfDTBe7c8ddPbuAvwcSB21ALEf4QPnIQXGAHm+VMwhQOa6B9G94QLmUHZ1MHtJ0QNqgJcWDPJz4Q IkUlgzAIXMDggyALcWHhli1J3xAHETLjDNFBlnBsCSoQjgmZISkkFHkAJFXfpBAKtUct4aAORwQP 3S0QsmAKpYAIppAlc3BTkmEJrFULfXBK37BEvDAJpxQJXABMc3CA4eAAltCAfRBznOQqC8MFsqAM u9AHBmkad00Qgtigj/+wkHhNfnkAiJrz41zCBQmtM3ZkMyXQXe+ikGvghuBYCLtwMIcABi5jP5TS WgxTCnZwCWtgKG6QDLHSDa+iDImAKrfAB1xwJVcwCWuw6yITCXjA3G3DbgK6CxalVRd+AJFADsy7 My5A1OPQgAdQCmtA4/bCBYONB1JF7JOAB1wgg27gMrtQCLZiB1mgDDDoJsV0C5q46hLTS65+Kokg ioMuGkKdEOlgBmLwDsYABRo+NN/wcWyADzKuV3TtEKvgCOgQeegABJhA5BIuGljgrI0aBL5LLTlQ lRhvKQRBCAVlB3mQ6pxgDixylLLACXYDdkpT7weRDo5wRmvRAEkzCtv/KIw3qZFrTS/aGQkeJQtr UAZk0A0ZGTBJA9wi0Q/rwAgKEAcYLDQyHz+HJAuJUJAYc2yIEFxc8A1qUAeRQAercgtCL40FbyS2 MAH5mDTJ0AiP8Aiq8AiNoIMD4Qbw+TZrE9jDOJ6yUAtcwOKPwODU0uGmMQQm4A4E6vAPLxvvOAG6 QPiFHxqe8AKI4A7TUAmTLjT8OAUUgMlE1wd2AwZ9gAtFx9+BEwlk0CIUONoIkQ0cAqlk0J6RZQfS HCVh3xBqED8QSPkUwJI10AEeYAAEcZMwui/0ggiUkAweVQp+OIUW8/XDVQqcnwfdMJ4HUAZ80A3k QJHJYFHEdQDJkDpm/x5hnt6i3eAAbrJPhcD8faW0JINPdJAHJfrARM9XkIzsfKAKATP721P7O9MF HID7NaADG8BNACHr0aM1fAp166ZMFoBCBy6RKfMIQJ1bdA4coATl0S4+eXYdKFWIDoADdgBculgG ACUwePLgMcjFzS5KBxANSraSzJo+XG4hKmNyJMor3QoN+obH4kWmlrYxhRpV6oFBdig9upiMD5gD kTxxcUetzqipZc2eRZu27IMiNXRQkTrokZ1bfQrlGdQwGyIudris2SXL0hU3FyNduXXlLp8D3R5l 23hFWbKUeJTd6nYgz61becgoq8kHKZddXBDZQaRMWZ1Lt5JR3FXqFf8fN6+UFY7qVO3UOnmykemD t84BaThSHcfRYPdy5s2j7nMePW0yWTWlX9d9XSq4O8dTjUt1x5h28uXNn0e/G9dT9N9S3AH/4w6b cOnt38efHy2uzPZFdXKEGlH0I7BAA81b775+uChCiAMfhDDCsxK0rx8NphhAQg03fJDC9CwswgAO RyTRPv7wK2UKB0tkscXoTrxvjxpWdLFGG83yML1FVLyxRx8vyhG9M2b8scgag1zOEsouomTJdIhh BBtGHEHCSCtHhLG5PMqIBKs66KCkED2ixKZMbGDYYBhh1lwznDbZhDNOOR2Qs04771yTTjz35JOS Nf0UBlBB/yQ00EL/B2UTUUUN9bNRQxNd8xf2muujlDzySIaMR+j45AwysbGFGTS/KecbU7/J5oRs Uv3mhFNPLfXVU4XB5dVYSZU11ltxldXUciLI1VZTXUUV1XKyaTWbY1VlNhtcmoU2VWlVlXbaVaPN RpgQUMU2Wm+vTVbaV2iwDwQoxbCFyivXZbddd99tLhKDLDlAlEC4iMS8Qbjw5KJu+Ojmm/K+GoSp Ovojz5JC+Mj3FVkKIaO8Vx7p64ATZOGjDPNe2WYpvx55pbxSLrGjMHO4NKe8ZDJ7xLpu8ijPEqy6 QekurMqjo49LGiKjI4SvI4MOSxA5YLZHZDmPjP4KAaOOiMl7pZTq68DsRhXryCujkEsucaOUUlou b2cywLDIkuHKc0CWUigRWi6uyhskEVzcACOPROw47yFK3JD5Z3j/BjxwwQcnvHDDi2SAXqYyzVet ZMDIRudLsoHqkqsPx7wsMG6ZiKlSlLEjjz7o4AMROp7I46aS+EqGDjeu2KVStq9oWZYy1lCGjz4o 2Spz3w/IZpc6ItlFmVIOicSQQo4+4BZZYjP9lkuUfOTStZWh45ZskokEKC7qsOQWZVL+3Xcuoapj EI1No4OyS8yJmG06XrEjmWT6yKaO7bc8oAxLIsnDJR5xhfIVsCyrOk82Qia4gAAAOw== ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.ajnr.org/icons/head/ajnrSearchGoInside.gif R0lGODlhGQAQAPcAAKGyu0RTdqmyyaKzw6SuyVNnhKKuvKK1wamwwqCuzaitwZ6wwqGrwaGwzqWw wqmxx6ixvaawvKGzv6Kqv56ru6qyxKezxp2xyaOvwqGsvqGtvR81YUJVeae0vaqvw6GzwaWyu6ex u5+wuiEzYaeuwaCwv6SvwaOwwaGxvqauw6etw6auwaevxqauuaKwvaSswaOwwKevxKSwwKOvyaOv vqWxv6euvqOww6KuvqOxvqKxxKiuxKitw6iryqWwxKavwKWuvaGxwKasxKStvKKtv6WsvqWyw6ax w6ivzKewv5ytv6atwKatyqCvwqOyyaSrvqWxwWF0lKevwqWtwKCut6awyaSwvqWsxqSxwqavvqe0 xGN3kKSux5+wxJ+vxk5jgktefo2dt6WpxoCOqKSzxpanw6etvaKzx6W2yqurw5+sv192laGyxJmv x6Sow6Wrw4qYsz9ZfKGsyH2Sr6auxaeyuKetuaCqxYqguKKvv6evvCc2Xxc5ZiM0YqOuxDBGdyIx XKiuuqSsvy1FYWpzmniImFhwjJ2pwaCuySxCcamuwoaWuJKhuKq3wKusyqyrypuywCw7Yqm1zaGq yVBYfae1wqWyxaaxx6OqtKeryKGzyyY7ZqSxupysu6ewuaO0yDxNeKarwaeuwJytvaeswqKpu52u vp+vviIwVyk9XqGwxaSvw6GyxnyGoX2OqqGywqSqwHJ+lqaqw46Ys0FQeZelyENWfkZWeKqwxqm0 uJelv6WtxKWsyZustoKRpIKRrp+uzamxvamzvKuyvKOuwKGxyKGxyqGnv6CrvaStzKWtuqi0xKWv u5yrvp2wtqSrvaCuu6aovaGoxaatv6KpxqOovImjvqm0tpyxxJyzxaS0xKW0x5KjwairujlQeaGt u6ewt6KvuFJmiZ+vyKSsw56sxpqyyqaswjBHZ2hylWx3l0hYeThLc6Ott5aovJ2xuGV5kaCrxyUy Zl99oaOtxqKuxputxR00XTRFcCE0XqayyKSst6utuqStvixBYJOuwSH5BAAAAAAALAAAAAAZABAA AAj/ABtksCNqRYoV00hMUZDChA2GdEjE8GHEhQtWF6CQOjGJRYsWpXCFCoYFRQgIxDD8SOLARQ0V S4LAUACjBgMuWA68g4SN1y4k5HCYENKDCIAjK3yUOMEDBo5KNDRgadeICj8P1GboI9IiAoYcFfoJ wsL0RAcKkghckeUmU5oiKD6QWdBsgB8Nq0ikKDFTBosm7sCMkLdh3gsnc9gBQrUOjwUTGmy8eHVC wY0QCWjd0pXD1SZfYVIZKmYswJ87y26QgIVCxo4SV+D4K5MlUI1yJeB9AUaCUxd0rZCpIaEiCI0V OZaMGWROU4ANfAiJi8IimocZHGJpsfAkBoopQ3Ts1JsVqdaPXAsKqEtHqYqeDojwFTogAoQJU0gU TQBhzxYHRqf84k0U1vizRTZtxJGIHC888MQ5lhyjAAkNXPBPAfn0cc8abHixCCgj7HELN0yQ8EwR FXRyAw7amDEMJkm88YAUUnQTTDgDJBAPDS2wUMIrLuTxQTIqTEBDFjuoAMU3QEDwgBUAOCOBDgZ4 EgMFAhhxiBM91HHNGePUc4kSaKAwhDRCIPACPQTIAI4NwijDTAw0MNGECEA88YgYFUBTjSMEVKEB FVaYcIQqM4zyiRLb9BIQADs= ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: text/css; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Location: http://www.ajnr.org/css/ajnrInteriorTemplate.css BODY { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; FONT-SIZE: 11px; BACKGROUND: = #fff; PADDING-BOTTOM: 0px; MARGIN: 0px; COLOR: #333; PADDING-TOP: 0px; = FONT-FAMILY: Arial, Helvetica, sans-serif; TEXT-ALIGN: center } P { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } UL { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } OL { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } H1 { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } H2 { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } H3 { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } H4 { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } H5 { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } H6 { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } TABLE { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } TD { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } INPUT { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } SELECT { FONT-SIZE: 12px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } P { MARGIN-TOP: 0px; MARGIN-BOTTOM: 1em } UL { MARGIN-TOP: 0px; MARGIN-BOTTOM: 1em } OL { MARGIN-TOP: 0px; MARGIN-BOTTOM: 1em } H1 { MARGIN-TOP: 0px; MARGIN-BOTTOM: 1em } H2 { MARGIN-TOP: 0px; MARGIN-BOTTOM: 1em } H3 { MARGIN-TOP: 0px; MARGIN-BOTTOM: 1em } H4 { MARGIN-TOP: 0px; MARGIN-BOTTOM: 1em } H5 { MARGIN-TOP: 0px; MARGIN-BOTTOM: 1em } H6 { MARGIN-TOP: 0px; MARGIN-BOTTOM: 1em } TABLE { MARGIN-TOP: 0px; MARGIN-BOTTOM: 1em } INPUT { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 0px; MARGIN: = 0px; PADDING-TOP: 0px } SELECT { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 0px; MARGIN: = 0px; PADDING-TOP: 0px } FORM { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 0px; MARGIN: = 0px; PADDING-TOP: 0px } A { COLOR: #00326d; TEXT-DECORATION: underline } A:hover { COLOR: #6999d2 } A:active { COLOR: #6999d2 } SUP { FONT-SIZE: 0.8em; VERTICAL-ALIGN: top } IMG { BORDER-RIGHT: 0px; BORDER-TOP: 0px; BORDER-LEFT: 0px; BORDER-BOTTOM: = 0px } .alt { DISPLAY: none } .clear { CLEAR: both; DISPLAY: block; WIDTH: 1px; HEIGHT: 1px } .noBreak { MARGIN-BOTTOM: 0px } UL { PADDING-LEFT: 2em; MARGIN-LEFT: 0px } LI { PADDING-BOTTOM: 0.5em } P { PADDING-BOTTOM: 0.5em } H2 { MARGIN-TOP: 0px; FONT-WEIGHT: bold; FONT-SIZE: 17px; MARGIN-BOTTOM: = 10px; VERTICAL-ALIGN: top; COLOR: #009999; PADDING-TOP: 1.5em } H3 { FONT-WEIGHT: bold; FONT-SIZE: 12px; COLOR: #034469 } #subline { MARGIN-TOP: 1px; PADDING-LEFT: 2px; FONT-SIZE: 10px; MARGIN-BOTTOM: = 2px; COLOR: #1c3664; FONT-FAMILY: verdana,arial,helvetica } #subline A { COLOR: #1c3664; TEXT-DECORATION: none } #ajnrWrapper { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 0px; MARGIN: 0px = auto; WIDTH: 675px; PADDING-TOP: 0px } #ajnrFrame { PADDING-RIGHT: 0px; MARGIN-TOP: 50px; PADDING-LEFT: 0px; = PADDING-BOTTOM: 0px; WIDTH: 675px; PADDING-TOP: 0px; TEXT-ALIGN: left } #ajnrFrame H1 { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; BACKGROUND: #1c3764; = PADDING-BOTTOM: 0px; MARGIN: 0px; WIDTH: 675px; PADDING-TOP: 0px; = HEIGHT: 134px } #ajnrHeaderSearch { CLEAR: both; PADDING-RIGHT: 18px; PADDING-LEFT: 5px; FONT-WEIGHT: = normal; FLOAT: right; PADDING-BOTTOM: 0px; WIDTH: 300px; PADDING-TOP: = 13px; HEIGHT: 50px } .text-input { MARGIN-TOP: 0px; FLOAT: right; MARGIN-BOTTOM: 0px; WIDTH: 90px; = MARGIN-RIGHT: 6px; HEIGHT: 16px } .submit-button { MARGIN-TOP: 0px; FLOAT: right; MARGIN-BOTTOM: 0px } #ajnrHeaderSearch A { CLEAR: right; PADDING-RIGHT: 0px; PADDING-LEFT: 5px; FONT-WEIGHT: bold; = FONT-SIZE: 9px; FLOAT: right; COLOR: #fff; MARGIN-RIGHT: 0px; = PADDING-TOP: 3px } #ajnrHeaderSearch IMG { PADDING-LEFT: 5px } #navContainer { BORDER-RIGHT: #1c3764 1px solid; BORDER-TOP: #1c3764 1px solid; = BORDER-LEFT: #1c3764 1px solid; BORDER-BOTTOM: #1c3764 1px solid } #ajnrNav { BORDER-RIGHT: #fff 2px solid; PADDING-RIGHT: 0px; BORDER-TOP: #fff 2px = solid; PADDING-LEFT: 0px; BACKGROUND: #ccc; PADDING-BOTTOM: 0px; MARGIN: = 0px; BORDER-LEFT: #fff 2px solid; WIDTH: 669px; PADDING-TOP: 0px; = BORDER-BOTTOM: #fff 2px solid; HEIGHT: 19px } #ajnrNav UL { PADDING-LEFT: 6px; MARGIN: 0px } #ajnrNav UL LI { BORDER-RIGHT: #008890 1px solid; PADDING-RIGHT: 10px; MARGIN-TOP: 3px; = PADDING-LEFT: 10px; FONT-SIZE: 12px; FLOAT: left; PADDING-BOTTOM: 0px; = PADDING-TOP: 0px; LIST-STYLE-TYPE: none; HEIGHT: 13px } #ajnrNav UL LI A { TEXT-DECORATION: none } #footerNav { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; BACKGROUND: #1c3664; = PADDING-BOTTOM: 0px; MARGIN: 0px; WIDTH: 675px; PADDING-TOP: 0px; = HEIGHT: 25px } #footerNav UL { PADDING-LEFT: 6px; MARGIN: 0px } #footerNav UL LI { BORDER-RIGHT: #008890 1px solid; PADDING-RIGHT: 10px; MARGIN-TOP: 6px; = PADDING-LEFT: 10px; FONT-SIZE: 12px; FLOAT: left; PADDING-BOTTOM: 0px; = PADDING-TOP: 0px; LIST-STYLE-TYPE: none; HEIGHT: 14px } #footerNav A { COLOR: #9bcdcc; TEXT-DECORATION: none } #mainOutsideContainer { BORDER-RIGHT: #1c3664 1px solid; BORDER-LEFT: #1c3664 1px solid } #main { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 0px; MARGIN: = 0px; WIDTH: 675px; PADDING-TOP: 0px } .ajnrGreyBack#main { PADDING-RIGHT: 0px; PADDING-LEFT: 20px; BACKGROUND: #eaeaea; = PADDING-BOTTOM: 20px; MARGIN: 15px 0px; WIDTH: 655px; PADDING-TOP: 0px } .ajnrGreyBack#main H2 { FONT-SIZE: 19px; COLOR: #009999; PADDING-TOP: 15px } .ajnrGreyBack#main H3 { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; FONT-SIZE: 15px; PADDING-BOTTOM: = 0px; MARGIN: 0px; COLOR: #009999; PADDING-TOP: 0px } .ajnrNoBack#main A { COLOR: #009999 } .ajnrNoBack#main H2 { FONT-SIZE: 19px; COLOR: #009999; PADDING-TOP: 15px } #mainContainer { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 0px; MARGIN: = 0px; WIDTH: 600px; PADDING-TOP: 0px } #mainContainer H3 { PADDING-TOP: 20px } #main H1 { PADDING-BOTTOM: 1em } #main H4 { CLEAR: right; PADDING-RIGHT: 0px; PADDING-LEFT: 0px; FONT-SIZE: 17px; = PADDING-BOTTOM: 0px; MARGIN: 0px; PADDING-TOP: 0px } #main H6 { FONT-SIZE: 12px; BACKGROUND: #99cccc; FLOAT: left; WIDTH: 320px; COLOR: = #fff; PADDING-TOP: 5px; HEIGHT: 20px; TEXT-ALIGN: center } #main A { COLOR: #1c3664 } #main UL LI { PADDING-BOTTOM: 0px } #main OL LI { PADDING-BOTTOM: 0px } #main A:hover { TEXT-DECORATION: underline } #main A:active { TEXT-DECORATION: underline } #main #editorailTable { MARGIN: 0px 12px 15px; WIDTH: 650px } #main #editorailTable TR.trHeader TD { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 0px; MARGIN: = 0px; PADDING-TOP: 0px } #main #editorailTable TD { PADDING-BOTTOM: 10px; VERTICAL-ALIGN: top } #toc { PADDING-RIGHT: 0px; BORDER-TOP: #fff 2px solid; PADDING-LEFT: 20px; = FLOAT: left; PADDING-BOTTOM: 20px; MARGIN-LEFT: 2px; WIDTH: 651px; = PADDING-TOP: 20px; BORDER-BOTTOM: #99cccc 5px solid; BACKGROUND-COLOR: = #eaeaea } #toc H4 { WIDTH: 400px } #toc H1 { PADDING-TOP: 0px } #toc A { COLOR: #1c3664 } #tocSearch A { COLOR: #1c3664 } #tocLeft { PADDING-RIGHT: 20px; FLOAT: left; WIDTH: 173px } #tocCenter { PADDING-RIGHT: 20px; FLOAT: left; WIDTH: 173px } #tocRight { PADDING-RIGHT: 20px; FLOAT: left; WIDTH: 173px } #tocLeft { BACKGROUND: url(../img/ajnrPubIssue.gif) no-repeat; FLOAT: left; WIDTH: = 138px; HEIGHT: 184px } #tocLeft UL { CLEAR: left; PADDING-TOP: 10px } #tocLeft P { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 0px; MARGIN: = 0px; PADDING-TOP: 0px } #tocLeft P#ajnrPub A { DISPLAY: block; WIDTH: 135px; HEIGHT: 178px } #tocLeft P { MARGIN-TOP: 7px; MARGIN-LEFT: 3px } #tocLeft P#issueNav A { PADDING-RIGHT: 6px; PADDING-LEFT: 6px; FLOAT: left; PADDING-BOTTOM: = 0px; MARGIN: 0px; PADDING-TOP: 0px } #tocLeft P#issueNav SPAN { PADDING-RIGHT: 6px; PADDING-LEFT: 6px; FLOAT: left; PADDING-BOTTOM: = 0px; MARGIN: 0px; PADDING-TOP: 0px } #tocLeft A { MARGIN: 5px 10px 0px } #tocSearch { PADDING-RIGHT: 10px; PADDING-LEFT: 10px; FLOAT: left; PADDING-BOTTOM: = 10px; MARGIN-LEFT: 4px; WIDTH: 600px; PADDING-TOP: 10px } #tocSearch #tocSearchMain A { PADDING-RIGHT: 6px; PADDING-LEFT: 6px; FLOAT: right; PADDING-BOTTOM: = 0px; PADDING-TOP: 0px } #tocSearch #tocSearchMain INPUT { PADDING-RIGHT: 6px; PADDING-LEFT: 6px; FLOAT: right; PADDING-BOTTOM: = 0px; PADDING-TOP: 0px } #tocSearch #tocSearchMain P { PADDING-RIGHT: 6px; PADDING-LEFT: 6px; FLOAT: right; PADDING-BOTTOM: = 0px; PADDING-TOP: 0px } #tocSearch #tocSearchSm P { PADDING-RIGHT: 6px; PADDING-LEFT: 6px; FLOAT: right; PADDING-BOTTOM: = 0px; PADDING-TOP: 0px } #tocBody { PADDING-RIGHT: 10px; PADDING-LEFT: 10px; FLOAT: left; PADDING-BOTTOM: = 10px; MARGIN-LEFT: 4px; WIDTH: 600px; PADDING-TOP: 10px } #tocBody IMG { PADDING-LEFT: 10px } #main #tocTable { PADDING-TOP: 35px } #main #tocTable H2 { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 0px; MARGIN: = 0px; PADDING-TOP: 0px } #main #tocTable H3 { PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 0px; MARGIN: = 0px; PADDING-TOP: 0px } #main #tocTable { WIDTH: 600px } #main #tocTable TD { PADDING-RIGHT: 5px; PADDING-LEFT: 5px; PADDING-BOTTOM: 10px; = VERTICAL-ALIGN: top; PADDING-TOP: 10px } #main #tocTable A { COLOR: #009999 } .copyRight { CLEAR: left; FONT-SIZE: 9px; COLOR: #1c3664; PADDING-TOP: 10px } #loggedNav { WIDTH: 675px } .smalltext { FONT-SIZE: 11px; COLOR: #1c3664; FONT-FAMILY: Arial, Helvetica, = sans-serif } .pap_note { MARGIN-TOP: 2px; FONT-SIZE: 11px; COLOR: #1c3664; FONT-FAMILY: Arial, = Helvetica, sans-serif } ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: application/octet-stream Content-Transfer-Encoding: quoted-printable Content-Location: http://www.ajnr.org/javascript/ajax/xmlhttprequest.js /*=0A= =0A= Cross-Browser XMLHttpRequest v1.2=0A= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=0A= =0A= Emulate Gecko 'XMLHttpRequest()' functionality in IE and Opera. Opera = requires=0A= the Sun Java Runtime Environment .=0A= =0A= by Andrew Gregory=0A= http://www.scss.com.au/family/andrew/webdesign/xmlhttprequest/=0A= =0A= This work is licensed under the Creative Commons Attribution License. To = view a=0A= copy of this license, visit = http://creativecommons.org/licenses/by-sa/2.5/ or=0A= send a letter to Creative Commons, 559 Nathan Abbott Way, Stanford, = California=0A= 94305, USA.=0A= =0A= Attribution: Leave my name and web address in this script intact.=0A= =0A= Not Supported in Opera=0A= ----------------------=0A= * user/password authentication=0A= * responseXML data member=0A= =0A= Not Fully Supported in Opera=0A= ----------------------------=0A= * async requests=0A= * abort()=0A= * getAllResponseHeaders(), getAllResponseHeader(header)=0A= =0A= */=0A= // IE support=0A= if (window.ActiveXObject && !window.XMLHttpRequest) {=0A= window.XMLHttpRequest =3D function() {=0A= var msxmls =3D new Array(=0A= 'Msxml2.XMLHTTP.5.0',=0A= 'Msxml2.XMLHTTP.4.0',=0A= 'Msxml2.XMLHTTP.3.0',=0A= 'Msxml2.XMLHTTP',=0A= 'Microsoft.XMLHTTP');=0A= for (var i =3D 0; i < msxmls.length; i++) {=0A= try {=0A= return new ActiveXObject(msxmls[i]);=0A= } catch (e) {=0A= }=0A= }=0A= return null;=0A= };=0A= }=0A= // Gecko support=0A= /* ;-) */=0A= // Opera support=0A= if (window.opera && !window.XMLHttpRequest) {=0A= window.XMLHttpRequest =3D function() {=0A= this.readyState =3D 0; // = 0=3Duninitialized,1=3Dloading,2=3Dloaded,3=3Dinteractive,4=3Dcomplete=0A= this.status =3D 0; // HTTP status codes=0A= this.statusText =3D '';=0A= this._headers =3D [];=0A= this._aborted =3D false;=0A= this._async =3D true;=0A= this._defaultCharset =3D 'ISO-8859-1';=0A= this._getCharset =3D function() {=0A= var charset =3D _defaultCharset;=0A= var contentType =3D = this.getResponseHeader('Content-type').toUpperCase();=0A= val =3D contentType.indexOf('CHARSET=3D');=0A= if (val !=3D -1) {=0A= charset =3D contentType.substring(val);=0A= }=0A= val =3D charset.indexOf(';');=0A= if (val !=3D -1) {=0A= charset =3D charset.substring(0, val);=0A= }=0A= val =3D charset.indexOf(',');=0A= if (val !=3D -1) {=0A= charset =3D charset.substring(0, val);=0A= }=0A= return charset;=0A= };=0A= this.abort =3D function() {=0A= this._aborted =3D true;=0A= };=0A= this.getAllResponseHeaders =3D function() {=0A= return this.getAllResponseHeader('*');=0A= };=0A= this.getAllResponseHeader =3D function(header) {=0A= var ret =3D '';=0A= for (var i =3D 0; i < this._headers.length; i++) {=0A= if (header =3D=3D '*' || this._headers[i].h =3D=3D header) {=0A= ret +=3D this._headers[i].h + ': ' + this._headers[i].v + '\n';=0A= }=0A= }=0A= return ret;=0A= };=0A= this.getResponseHeader =3D function(header) {=0A= var ret =3D getAllResponseHeader(header);=0A= var i =3D ret.indexOf('\n');=0A= if (i !=3D -1) {=0A= ret =3D ret.substring(0, i);=0A= }=0A= return ret;=0A= };=0A= this.setRequestHeader =3D function(header, value) {=0A= this._headers[this._headers.length] =3D {h:header, v:value};=0A= };=0A= this.open =3D function(method, url, async, user, password) {=0A= this.method =3D method;=0A= this.url =3D url;=0A= this._async =3D true;=0A= this._aborted =3D false;=0A= this._headers =3D [];=0A= if (arguments.length >=3D 3) {=0A= this._async =3D async;=0A= }=0A= if (arguments.length > 3) {=0A= opera.postError('XMLHttpRequest.open() - user/password not = supported');=0A= }=0A= this.readyState =3D 1;=0A= if (this.onreadystatechange) {=0A= this.onreadystatechange();=0A= }=0A= };=0A= this.send =3D function(data) {=0A= if (!navigator.javaEnabled()) {=0A= alert("XMLHttpRequest.send() - Java must be installed and = enabled.");=0A= return;=0A= }=0A= if (this._async) {=0A= setTimeout(this._sendasync, 0, this, data);=0A= // this is not really asynchronous and won't execute until the = current=0A= // execution context ends=0A= } else {=0A= this._sendsync(data);=0A= }=0A= }=0A= this._sendasync =3D function(req, data) {=0A= if (!req._aborted) {=0A= req._sendsync(data);=0A= }=0A= };=0A= this._sendsync =3D function(data) {=0A= this.readyState =3D 2;=0A= if (this.onreadystatechange) {=0A= this.onreadystatechange();=0A= }=0A= // open connection=0A= var url =3D new java.net.URL(new = java.net.URL(window.location.href), this.url);=0A= var conn =3D url.openConnection();=0A= for (var i =3D 0; i < this._headers.length; i++) {=0A= conn.setRequestProperty(this._headers[i].h, this._headers[i].v);=0A= }=0A= this._headers =3D [];=0A= if (this.method =3D=3D 'POST') {=0A= // POST data=0A= conn.setDoOutput(true);=0A= var wr =3D new = java.io.OutputStreamWriter(conn.getOutputStream(), this._getCharset());=0A= wr.write(data);=0A= wr.flush();=0A= wr.close();=0A= }=0A= // read response headers=0A= // NOTE: the getHeaderField() methods always return nulls for me :(=0A= var gotContentEncoding =3D false;=0A= var gotContentLength =3D false;=0A= var gotContentType =3D false;=0A= var gotDate =3D false;=0A= var gotExpiration =3D false;=0A= var gotLastModified =3D false;=0A= for (var i =3D 0; ; i++) {=0A= var hdrName =3D conn.getHeaderFieldKey(i);=0A= var hdrValue =3D conn.getHeaderField(i);=0A= if (hdrName =3D=3D null && hdrValue =3D=3D null) {=0A= break;=0A= }=0A= if (hdrName !=3D null) {=0A= this._headers[this._headers.length] =3D {h:hdrName, = v:hdrValue};=0A= switch (hdrName.toLowerCase()) {=0A= case 'content-encoding': gotContentEncoding =3D true; break;=0A= case 'content-length' : gotContentLength =3D true; break;=0A= case 'content-type' : gotContentType =3D true; break;=0A= case 'date' : gotDate =3D true; break;=0A= case 'expires' : gotExpiration =3D true; break;=0A= case 'last-modified' : gotLastModified =3D true; break;=0A= }=0A= }=0A= }=0A= // try to fill in any missing header information=0A= var val;=0A= val =3D conn.getContentEncoding();=0A= if (val !=3D null && !gotContentEncoding) = this._headers[this._headers.length] =3D {h:'Content-encoding', v:val};=0A= val =3D conn.getContentLength();=0A= if (val !=3D -1 && !gotContentLength) = this._headers[this._headers.length] =3D {h:'Content-length', v:val};=0A= val =3D conn.getContentType();=0A= if (val !=3D null && !gotContentType) = this._headers[this._headers.length] =3D {h:'Content-type', v:val};=0A= val =3D conn.getDate();=0A= if (val !=3D 0 && !gotDate) this._headers[this._headers.length] = =3D {h:'Date', v:(new Date(val)).toUTCString()};=0A= val =3D conn.getExpiration();=0A= if (val !=3D 0 && !gotExpiration) = this._headers[this._headers.length] =3D {h:'Expires', v:(new = Date(val)).toUTCString()};=0A= val =3D conn.getLastModified();=0A= if (val !=3D 0 && !gotLastModified) = this._headers[this._headers.length] =3D {h:'Last-modified', v:(new = Date(val)).toUTCString()};=0A= // read response data=0A= var reqdata =3D '';=0A= var stream =3D conn.getInputStream();=0A= if (stream) {=0A= var reader =3D new java.io.BufferedReader(new = java.io.InputStreamReader(stream, this._getCharset()));=0A= var line;=0A= while ((line =3D reader.readLine()) !=3D null) {=0A= if (this.readyState =3D=3D 2) {=0A= this.readyState =3D 3;=0A= if (this.onreadystatechange) {=0A= this.onreadystatechange();=0A= }=0A= }=0A= reqdata +=3D line + '\n';=0A= }=0A= reader.close();=0A= this.status =3D 200;=0A= this.statusText =3D 'OK';=0A= this.responseText =3D reqdata;=0A= this.readyState =3D 4;=0A= if (this.onreadystatechange) {=0A= this.onreadystatechange();=0A= }=0A= if (this.onload) {=0A= this.onload();=0A= }=0A= } else {=0A= // error=0A= this.status =3D 404;=0A= this.statusText =3D 'Not Found';=0A= this.responseText =3D '';=0A= this.readyState =3D 4;=0A= if (this.onreadystatechange) {=0A= this.onreadystatechange();=0A= }=0A= if (this.onerror) {=0A= this.onerror();=0A= }=0A= }=0A= };=0A= };=0A= }=0A= // ActiveXObject emulation=0A= if (!window.ActiveXObject && window.XMLHttpRequest) {=0A= window.ActiveXObject =3D function(type) {=0A= switch (type.toLowerCase()) {=0A= case 'microsoft.xmlhttp':=0A= case 'msxml2.xmlhttp':=0A= case 'msxml2.xmlhttp.3.0':=0A= case 'msxml2.xmlhttp.4.0':=0A= case 'msxml2.xmlhttp.5.0':=0A= return new XMLHttpRequest();=0A= }=0A= return null;=0A= };=0A= }=0A= ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: application/octet-stream Content-Transfer-Encoding: quoted-printable Content-Location: http://www.ajnr.org/javascript/ajax/utility.js /************************************************************************= *****=0A= * javascript/ajax/utility.js=0A= *=0A= * Utility functions for working with XMLHttpRequest data.=0A= *=0A= * Copyright 2006 Board of Trustees of the Leland Stanford Junior = University.=0A= = *************************************************************************= ***/=0A= =0A= /*=0A= * Copy XML nodes into an HTMLElement. This effectively=0A= * clones XML markup which uses XHTML naming conventions=0A= * into an HTML DOM.=0A= */=0A= function copy_xml_to_html(src, dst) {=0A= if (src.nodeType =3D=3D 1) { /* Node.ELEMENT_NODE */=0A= var e =3D document.createElement(src.nodeName);=0A= for (var i =3D 0; i < src.childNodes.length; i++) {=0A= copy_xml_to_html(src.childNodes[i], e);=0A= }=0A= for (var i =3D 0; i < src.attributes.length; i++) {=0A= var n =3D src.attributes[i].name;=0A= var v =3D unescape_xml_string(src.attributes[i].value); =0A= e.setAttribute(n, v);=0A= if (n =3D=3D "class") {=0A= e.className =3D v;=0A= }=0A= else if (n =3D=3D "style") {=0A= set_css_style(v, e, "");=0A= }=0A= }=0A= dst.appendChild(e);=0A= }=0A= else if (src.nodeType =3D=3D 3) { /* Node.TEXT_NODE */=0A= dst.appendChild(document.createTextNode(src.nodeValue));=0A= }=0A= }=0A= =0A= /* =0A= * It is unclear that this is the right thing to be calling=0A= * from copy_xml_to_html, but it appears that Safari decides=0A= * to convert & to the NCR #, and then encodes that=0A= * NCR to &%26%2338;. So, I'm going to treat the DOM Attr=0A= * value as a plain string, and run our XML string input=0A= * through the decoding routine below.=0A= */=0A= function unescape_xml_string(s) {=0A= return s.replace(/'/g, "'")=0A= .replace(/'/g, "'")=0A= .replace(/"/g, "\"")=0A= .replace(/"/g, "\"")=0A= .replace(/>/g, ">")=0A= .replace(/>/g, ">")=0A= .replace(/</g, "<")=0A= .replace(/</g, "<")=0A= .replace(/&/g, "&")=0A= .replace(/&/g, "&");=0A= }=0A= =0A= /*=0A= * Parse set of CSS rules and apply them to an element.=0A= * This is quite horrifying, but I'm unable to determine=0A= * how else to handle this with IE 6. FireFox and other=0A= * sane browsers let you simply set the style attribute=0A= * or use e.style.setProperty(rule, value, priority),=0A= * IE 6 appears to have neither of these capabilities..=0A= */=0A= function set_css_style(css, e, priority) {=0A= var rules =3D css.split(";");=0A= for (var i =3D 0; i < rules.length; i++) {=0A= var nvpair =3D rules[i].split(":");=0A= if (nvpair.length =3D=3D 2) {=0A= try {=0A= var name =3D nvpair[0]; /* style attribute */=0A= var value =3D nvpair[1]; /* attribute value */=0A= =0A= /*=0A= * For each possible style attribute, set the=0A= * appropriate style property in the element.=0A= */=0A= if (name =3D=3D "background") {=0A= e.style.background =3D value;=0A= }=0A= else if (name =3D=3D "background-attachment") {=0A= e.style.backgroundAttachment =3D value;=0A= }=0A= else if (name =3D=3D "background-color") {=0A= e.style.backgroundColor =3D value;=0A= }=0A= else if (name =3D=3D "background-image") {=0A= e.style.backgroundImage =3D value;=0A= }=0A= else if (name =3D=3D "background-position") {=0A= e.style.backgroundPosition =3D value;=0A= }=0A= else if (name =3D=3D "background-position-x") {=0A= e.style.backgroundPositionX =3D value;=0A= }=0A= else if (name =3D=3D "background-position-y") {=0A= e.style.backgroundPositionY =3D value;=0A= }=0A= else if (name =3D=3D "background-repeat") {=0A= e.style.backgroundRepeat =3D value;=0A= }=0A= else if (name =3D=3D "behavior") {=0A= e.style.behavior =3D value;=0A= }=0A= else if (name =3D=3D "border") {=0A= e.style.border =3D value;=0A= }=0A= else if (name =3D=3D "border-bottom") {=0A= e.style.borderBottom =3D value;=0A= }=0A= else if (name =3D=3D "border-bottom-color") {=0A= e.style.borderBottomColor =3D value;=0A= }=0A= else if (name =3D=3D "border-bottom-style") {=0A= e.style.borderBottomStyle =3D value;=0A= }=0A= else if (name =3D=3D "border-bottom-width") {=0A= e.style.borderBottomWidth =3D value;=0A= }=0A= else if (name =3D=3D "border-collapse") {=0A= e.style.borderCollapse =3D value;=0A= }=0A= else if (name =3D=3D "border-color") {=0A= e.style.borderColor =3D value;=0A= }=0A= else if (name =3D=3D "border-left") {=0A= e.style.borderLeft =3D value;=0A= }=0A= else if (name =3D=3D "border-left-color") {=0A= e.style.borderLeftColor =3D value;=0A= }=0A= else if (name =3D=3D "border-left-style") {=0A= e.style.borderLeftStyle =3D value;=0A= }=0A= else if (name =3D=3D "border-left-width") {=0A= e.style.borderLeftWidth =3D value;=0A= }=0A= else if (name =3D=3D "border-right") {=0A= e.style.borderRight =3D value;=0A= }=0A= else if (name =3D=3D "border-right-color") {=0A= e.style.borderRightColor =3D value;=0A= }=0A= else if (name =3D=3D "border-right-style") {=0A= e.style.borderRightStyle =3D value;=0A= }=0A= else if (name =3D=3D "border-right-width") {=0A= e.style.borderRightWidth =3D value;=0A= }=0A= else if (name =3D=3D "border-style") {=0A= e.style.borderStyle =3D value;=0A= }=0A= else if (name =3D=3D "border-top") {=0A= e.style.borderTop =3D value;=0A= }=0A= else if (name =3D=3D "border-top-color") {=0A= e.style.borderTopColor =3D value;=0A= }=0A= else if (name =3D=3D "border-top-style") {=0A= e.style.borderTopStyle =3D value;=0A= }=0A= else if (name =3D=3D "border-top-width") {=0A= e.style.borderTopWidth =3D value;=0A= }=0A= else if (name =3D=3D "border-width") {=0A= e.style.borderWidth =3D value;=0A= }=0A= else if (name =3D=3D "bottom") {=0A= e.style.bottom =3D value;=0A= }=0A= else if (name =3D=3D "clear") {=0A= e.style.clear =3D value;=0A= }=0A= else if (name =3D=3D "clip") {=0A= e.style.clip =3D value;=0A= }=0A= else if (name =3D=3D "color") {=0A= e.style.color =3D value;=0A= }=0A= else if (name =3D=3D "cssText") {=0A= e.style.Sets =3D value;=0A= }=0A= else if (name =3D=3D "cursor") {=0A= e.style.cursor =3D value;=0A= }=0A= else if (name =3D=3D "direction") {=0A= e.style.direction =3D value;=0A= }=0A= else if (name =3D=3D "display") {=0A= e.style.display =3D value;=0A= }=0A= else if (name =3D=3D "font") {=0A= e.style.font =3D value;=0A= }=0A= else if (name =3D=3D "font-family") {=0A= e.style.fontFamily =3D value;=0A= }=0A= else if (name =3D=3D "font-size") {=0A= e.style.fontSize =3D value;=0A= }=0A= else if (name =3D=3D "font-style") {=0A= e.style.fontStyle =3D value;=0A= }=0A= else if (name =3D=3D "font-variant") {=0A= e.style.fontVariant =3D value;=0A= }=0A= else if (name =3D=3D "font-weight") {=0A= e.style.fontWeight =3D value;=0A= }=0A= else if (name =3D=3D "height") {=0A= e.style.height =3D value;=0A= }=0A= else if (name =3D=3D "ime-mode") {=0A= e.style.imeMode =3D value;=0A= }=0A= else if (name =3D=3D "layout-flow") {=0A= e.style.layoutFlow =3D value;=0A= }=0A= else if (name =3D=3D "layout-grid") {=0A= e.style.layoutGrid =3D value;=0A= }=0A= else if (name =3D=3D "layout-grid-char") {=0A= e.style.layoutGridChar =3D value;=0A= }=0A= else if (name =3D=3D "layout-grid-line") {=0A= e.style.layoutGridLine =3D value;=0A= }=0A= else if (name =3D=3D "layout-grid-mode") {=0A= e.style.layoutGridMode =3D value;=0A= }=0A= else if (name =3D=3D "layout-grid-type") {=0A= e.style.layoutGridType =3D value;=0A= }=0A= else if (name =3D=3D "left") {=0A= e.style.left =3D value;=0A= }=0A= else if (name =3D=3D "letter-spacing") {=0A= e.style.letterSpacing =3D value;=0A= }=0A= else if (name =3D=3D "line-break") {=0A= e.style.lineBreak =3D value;=0A= }=0A= else if (name =3D=3D "line-height") {=0A= e.style.lineHeight =3D value;=0A= }=0A= else if (name =3D=3D "list-style") {=0A= e.style.listStyle =3D value;=0A= }=0A= else if (name =3D=3D "list-style-image") {=0A= e.style.listStyleImage =3D value;=0A= }=0A= else if (name =3D=3D "list-style-position") {=0A= e.style.listStylePosition =3D value;=0A= }=0A= else if (name =3D=3D "list-style-type") {=0A= e.style.listStyleType =3D value;=0A= }=0A= else if (name =3D=3D "margin") {=0A= e.style.margin =3D value;=0A= }=0A= else if (name =3D=3D "margin-bottom") {=0A= e.style.marginBottom =3D value;=0A= }=0A= else if (name =3D=3D "margin-left") {=0A= e.style.marginLeft =3D value;=0A= }=0A= else if (name =3D=3D "margin-right") {=0A= e.style.marginRight =3D value;=0A= }=0A= else if (name =3D=3D "margin-top") {=0A= e.style.marginTop =3D value;=0A= }=0A= else if (name =3D=3D "min-height") {=0A= e.style.minHeight =3D value;=0A= }=0A= else if (name =3D=3D "overflow") {=0A= e.style.overflow =3D value;=0A= }=0A= else if (name =3D=3D "overflow-x") {=0A= e.style.overflowX =3D value;=0A= }=0A= else if (name =3D=3D "overflow-y") {=0A= e.style.overflowY =3D value;=0A= }=0A= else if (name =3D=3D "padding") {=0A= e.style.padding =3D value;=0A= }=0A= else if (name =3D=3D "padding-bottom") {=0A= e.style.paddingBottom =3D value;=0A= }=0A= else if (name =3D=3D "padding-left") {=0A= e.style.paddingLeft =3D value;=0A= }=0A= else if (name =3D=3D "padding-right") {=0A= e.style.paddingRight =3D value;=0A= }=0A= else if (name =3D=3D "padding-top") {=0A= e.style.paddingTop =3D value;=0A= }=0A= else if (name =3D=3D "page-break-after") {=0A= e.style.pageBreakAfter =3D value;=0A= }=0A= else if (name =3D=3D "page-break-before") {=0A= e.style.pageBreakBefore =3D value;=0A= }=0A= else if (name =3D=3D "pixelBottom") {=0A= e.style.pixelBottom =3D value;=0A= }=0A= else if (name =3D=3D "pixelHeight") {=0A= e.style.pixelHeight =3D value;=0A= }=0A= else if (name =3D=3D "pixelLeft") {=0A= e.style.pixelLeft =3D value;=0A= }=0A= else if (name =3D=3D "pixelRight") {=0A= e.style.pixelRight =3D value;=0A= }=0A= else if (name =3D=3D "pixelTop") {=0A= e.style.pixelTop =3D value;=0A= }=0A= else if (name =3D=3D "pixelWidth") {=0A= e.style.pixelWidth =3D value;=0A= }=0A= else if (name =3D=3D "posBottom") {=0A= e.style.posBottom =3D value;=0A= }=0A= else if (name =3D=3D "posHeight") {=0A= e.style.posHeight =3D value;=0A= }=0A= else if (name =3D=3D "position") {=0A= e.style.position =3D value;=0A= }=0A= else if (name =3D=3D "posLeft") {=0A= e.style.posLeft =3D value;=0A= }=0A= else if (name =3D=3D "posRight") {=0A= e.style.posRight =3D value;=0A= }=0A= else if (name =3D=3D "posTop") {=0A= e.style.posTop =3D value;=0A= }=0A= else if (name =3D=3D "posWidth") {=0A= e.style.posWidth =3D value;=0A= }=0A= else if (name =3D=3D "right") {=0A= e.style.right =3D value;=0A= }=0A= else if (name =3D=3D "ruby-align") {=0A= e.style.rubyAlign =3D value;=0A= }=0A= else if (name =3D=3D "ruby-overhang") {=0A= e.style.rubyOverhang =3D value;=0A= }=0A= else if (name =3D=3D "ruby-position") {=0A= e.style.rubyPosition =3D value;=0A= }=0A= else if (name =3D=3D "scrollbar-3dlight-color") {=0A= e.style.scrollbar3dLightColor =3D value;=0A= }=0A= else if (name =3D=3D "scrollbar-arrow-color") {=0A= e.style.scrollbarArrowColor =3D value;=0A= }=0A= else if (name =3D=3D "scrollbar-base-color") {=0A= e.style.scrollbarBaseColor =3D value;=0A= }=0A= else if (name =3D=3D "scrollbar-darkshadow-color") {=0A= e.style.scrollbarDarkShadowColor =3D value;=0A= }=0A= else if (name =3D=3D "scrollbar-face-color") {=0A= e.style.scrollbarFaceColor =3D value;=0A= }=0A= else if (name =3D=3D "scrollbar-highlight-color") {=0A= e.style.scrollbarHighlightColor =3D value;=0A= }=0A= else if (name =3D=3D "scrollbar-shadow-color") {=0A= e.style.scrollbarShadowColor =3D value;=0A= }=0A= else if (name =3D=3D "scrollbar-track-color") {=0A= e.style.scrollbarTrackColor =3D value;=0A= }=0A= else if (name =3D=3D "float") {=0A= e.style.styleFloat =3D value;=0A= }=0A= else if (name =3D=3D "table-layout") {=0A= e.style.tableLayout =3D value;=0A= }=0A= else if (name =3D=3D "text-align") {=0A= e.style.textAlign =3D value;=0A= }=0A= else if (name =3D=3D "text-align-last") {=0A= e.style.textAlignLast =3D value;=0A= }=0A= else if (name =3D=3D "text-autospace") {=0A= e.style.textAutospace =3D value;=0A= }=0A= else if (name =3D=3D "text-decoration") {=0A= e.style.textDecoration =3D value;=0A= }=0A= else if (name =3D=3D "textDecorationBlink") {=0A= e.style.textDecorationBlink =3D value;=0A= }=0A= else if (name =3D=3D "textDecorationLineThrough") {=0A= e.style.textDecorationLineThrough =3D value;=0A= }=0A= else if (name =3D=3D "textDecorationNone") {=0A= e.style.textDecorationNone =3D value;=0A= }=0A= else if (name =3D=3D "textDecorationOverline") {=0A= e.style.textDecorationOverline =3D value;=0A= }=0A= else if (name =3D=3D "textDecorationUnderline") {=0A= e.style.textDecorationUnderline =3D value;=0A= }=0A= else if (name =3D=3D "text-indent") {=0A= e.style.textIndent =3D value;=0A= }=0A= else if (name =3D=3D "text-justify") {=0A= e.style.textJustify =3D value;=0A= }=0A= else if (name =3D=3D "text-kashida-space") {=0A= e.style.textKashidaSpace =3D value;=0A= }=0A= else if (name =3D=3D "text-overflow") {=0A= e.style.textOverflow =3D value;=0A= }=0A= else if (name =3D=3D "text-transform") {=0A= e.style.textTransform =3D value;=0A= }=0A= else if (name =3D=3D "text-underline-position") {=0A= e.style.textUnderlinePosition =3D value;=0A= }=0A= else if (name =3D=3D "top") {=0A= e.style.top =3D value;=0A= }=0A= else if (name =3D=3D "unicode-bidi") {=0A= e.style.unicodeBidi =3D value;=0A= }=0A= else if (name =3D=3D "vertical-align") {=0A= e.style.verticalAlign =3D value;=0A= }=0A= else if (name =3D=3D "visibility") {=0A= e.style.visibility =3D value;=0A= }=0A= else if (name =3D=3D "white-space") {=0A= e.style.whiteSpace =3D value;=0A= }=0A= else if (name =3D=3D "width") {=0A= e.style.width =3D value;=0A= }=0A= else if (name =3D=3D "word-break") {=0A= e.style.wordBreak =3D value;=0A= }=0A= else if (name =3D=3D "word-spacing") {=0A= e.style.wordSpacing =3D value;=0A= }=0A= else if (name =3D=3D "word-wrap") {=0A= e.style.wordWrap =3D value;=0A= }=0A= else if (name =3D=3D "writing-mode") {=0A= e.style.writingMode =3D value;=0A= }=0A= else if (name =3D=3D "z-index") {=0A= e.style.zIndex =3D value;=0A= }=0A= else if (name =3D=3D "zoom") {=0A= e.style.zoom =3D value;=0A= }=0A= }=0A= catch (e) {=0A= /* ignore error on attempt to set e.style.[property] */=0A= }=0A= }=0A= }=0A= }=0A= ------=_NextPart_000_0000_01C8A919.B3CD9390 Content-Type: application/octet-stream Content-Transfer-Encoding: quoted-printable Content-Location: http://www.ajnr.org/javascript/entrez/callback.js /************************************************************************= *****=0A= * javascript/entrez/callback.js=0A= *=0A= * Entrez Linking callback to populate content box.=0A= *=0A= * Copyright 2006 Board of Trustees of the Leland Stanford Junior = University.=0A= = *************************************************************************= ***/=0A= =0A= /*=0A= * Execute callback to fill content box with Entrez Linking information.=0A= */=0A= function entrez_callback(pmid, callback_url) {=0A= /*=0A= * MSIE 5.5 and below have issues with the JavaScript=0A= * used for Entrez Linking. For now we have to disable=0A= * the callback until we can track down a proper fix=0A= * (or everybody sanely upgrades to version 6 or 7!).=0A= */=0A= if (navigator) {=0A= var appname =3D navigator.appName;=0A= if (appname =3D=3D "Microsoft Internet Explorer") {=0A= var userAgent =3D navigator["userAgent"];=0A= var s =3D "MSIE ";=0A= var n =3D -1; =0A= if ((n =3D userAgent.indexOf(s)) !=3D -1) {=0A= var v =3D parseFloat(userAgent.substring(n+s.length));=0A= if (v < 6) {=0A= return;=0A= }=0A= }=0A= }=0A= }=0A= =0A= /*=0A= * Acquire table row element to update, initiate callback=0A= * to update table with Entrez Links.=0A= */=0A= var tr =3D document.getElementById('entrez_callback_'+pmid);=0A= if (!tr) {=0A= return;=0A= }=0A= var req =3D new XMLHttpRequest();=0A= if (!req) {=0A= return;=0A= }=0A= req.onreadystatechange =3D function() {=0A= if (req.readyState =3D=3D 4 && (req.status =3D=3D 200 || req.status = =3D=3D 304)) {=0A= var src =3D req.responseXML.documentElement;=0A= var dst =3D document.createDocumentFragment();=0A= for (var i =3D 0; i < src.childNodes.length; i++) {=0A= copy_xml_to_html(src.childNodes[i], dst);=0A= }=0A= var tbl =3D tr.parentNode;=0A= tbl.replaceChild(dst, tr);=0A= }=0A= }=0A= req.open('GET', callback_url, true);=0A= req.send(null);=0A= }=0A= ------=_NextPart_000_0000_01C8A919.B3CD9390--