| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Magnetic Resonance Imaging Center, University of Iowa
College of Medicine, 200 Hawkins Dr., Iowa City, IA 52242.
2
Present address: Department of Radiology, Nara Medical University, 840 Shijo,
Kashihara, Nara, 634-8522 Japan.
Received April 25, 2000;
accepted after revision August 10, 2000.
Address correspondence to T. Taoka.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Three hundred consecutive MR imaging examinations were prospectively screened for patients with sulcal hyperintensity on FLAIR images. Nine patients with clinical, CT, or laboratory evidence suggesting abnormal CSF were excluded. The distribution of sulcal hyperintensity on FLAIR images and associated abnormal enhancement were evaluated. The presence of the "dirty CSF" sign (mild increase in CSF signal on unenhanced T1-weighted images or mild decrease on T2-weighted images) in the corresponding hyperintense sulcus was also assessed.
RESULTS. Twenty-six (8.9%) of the 291 patients had sulcal hyperintensity (16 focal, 10 diffuse) associated with 18 masses (6.1%) and eight vascular abnormalities (2.7%). Sulcal hyperintensity was frequently associated with the dirty CSF sign (69.2%) and abnormal contrast enhancement (overall, 96.2%; 88.5%, leptomeningeal; 53.8%, vascular enhancement).
CONCLUSION. Our study shows that sulcal hyperintensity on FLAIR imaging can occur in patients without apparent CSF abnormality. Its frequent association with mass effect, vascular disease, abnormal vascular enhancement, and dirty CSF sign suggests that an increase in blood pool, a small amount of protein leakage, and the "flow-entering" phenomenon of the congested blood may contribute to sulcal hyperintensity on FLAIR images.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our project was initiated after observing sulcal hyperintensity on FLAIR imaging in several patients with acute stroke who had no clinical, CT, or laboratory evidence of subarachnoid hemorrhage. Our purpose was to prospectively investigate whether sulcal hyperintensity can occur in patients without apparent CSF abnormality and to correlate the presence of sulcal hyperintensity with clinical history, underlying pathophysiology, and associated MR findings. In addition, we propose several possible mechanisms to explain sulcal hyperintensity on FLAIR imaging in patients without CSF abnormality.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
On FLAIR images, the pattern of the sulcal hyperintensity was further classified as focal or diffuse. Sulcal hyperintensity limited to a single lobe of the brain was considered focal, whereas more widespread abnormality involving two or more lobes was considered diffuse. In the cerebellum, the focal group was defined as sulcal hyperintensity that did not extend across the great horizontal fissure or across the midline.
The sulci with hyperintensity on FLAIR imaging were also evaluated for the presence of mild hyperintensity on unenhanced T1-weighted images or mild hypointensity on T2-weighted images ("dirty CSF" sign) [14]. The presence and pattern of abnormal enhancement in the corresponding sulcal space with sulcal hyperintensity on FLAIR were also assessed. Patterns of abnormal enhancement included leptomeningeal enhancement, sheetlike enhancement located deep in and along the sulci, and vascular (linear or tubular) enhancement.
MR imaging was performed with either a 1.5-T Signa MR imaging system (General Electric Medical Systems, Milwaukee, WI) or a 1.5-T Vision MR imaging system (Siemens, Erlangen, Germany). Unenhanced MR examinations included FLAIR images (TR/TEeff range, 8000/80-130; inversion time, 2000 msec; echo train length, 8), fast spin-echo T2-weighted images (8000/90-100; echo train length, 8), and spin-echo T1-weighted images (TR range/TE range, 350-500/16-20), with the axial plane; section thickness, 5 mm; matrix size, 256 x 256; field of view, 24 x 24 cm; and a head coil. All FLAIR images were obtained before the administration of contrast medium. Contrast-enhanced T1-weighted images with parameters identical to unenhanced T1-weighted images were obtained after the IV injection of 0.1 mmol/kg of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The pattern of sulcal hyperintensity on FLAIR imaging limited to one lobe (focal pattern) was found in 15 patients and extended to two or more lobes (diffuse pattern) in 11 patients. Although there was no statistical significance, patients with mass effect, 12 (66.7%) of 18 patients, were more likely to have a focal pattern of sulcal hyperintensity, whereas patients with vascular abnormalities, five (62.5%) of eight patients, were more likely to have a diffuse pattern.
Abnormal contrast-enhancement pattern was frequently seen, occurring in 25 (96.1%) of 26 patients with sulcal hyperintensity. Leptomeningeal enhancement in the corresponding area was the most frequent pattern observed, occurring in 23 (88.5%) of 26 patients with sulcal hyperintensity and associated with all cases of vascular abnormality (100%) and 15 (83.3%) of the 18 cases of mass effect groups. Absence of leptomeningeal enhancement occurred in only three (11%) of 26 small mass lesions (meningioma, subcortical hematoma, and metastatic brain tumor). Vascular (arterial or venous) enhancement was seen in 14 (53.8%) of 26 patients.
The dirty CSF sign (mild hyperintensity in the sulcus on T1-weighted images) corresponding to sulcal intensity on FLAIR images was seen in 18 (69.2%) of 26 patients with sulcal hyperintensity on FLAIR images (Fig. 2A,2B,2C,2D,2E). The dirty CSF sign was more prevalent in six (75%) of eight patients in the vascular group, whereas the sign was seen in 12 (66.7%) of 18 patients in the mass effect group.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To investigate possible mechanisms causing sulcal hyperintensity on FLAIR imaging in patients without apparent CSF abnormality, we evaluated several characteristic MR findings that may reflect the underlying pathophysiologies. These included the distribution (focal versus diffuse) of sulcal hyperintensity on FLAIR imaging, association of abnormal vascular enhancement (leptomeningeal and vascular), and corresponding CSF signal changes (dirty CSF sign) on the unenhanced conventional T1- and T2- weighted images. These findings were then correlated with the underlying pathophysiologies (mass effect or vascular disease) of the patients. The distribution pattern of sulcal hyperintensity on FLAIR imaging may reflect the spatial relationship between the lesion and sulcal hyperintensity. The frequent association of an abnormal vascular enhancement pattern may reflect the alteration of underlying hemodynamics associated with various vascular diseases or with breakdown of the blood—brain barrier. The dirty CSF sign in the sulcus may reflect a similar mechanism causing the sulcal hyperintensity on FLAIR imaging.
Our findings show that sulcal hyperintensity on FLAIR imaging caused by vascular disease in patients without apparent CSF abnormality tended to be more diffuse and extensive than sulcal hyperintensity caused by mass effect. In patients with vascular abnormalities, sulcal hyperintensity was always associated with abnormal vascular enhancement and suggested that the alteration of the underlying hemodynamics in the involved sulcal space may play a role in the failure of nulling the normal CSF signal (Figs. 2A,2B,2C,2D,2E,3A,3B,3C,3D,4A,4B,4C,4D). Sulcal hyperintensity on FLAIR imaging in patients with mass effect tended to be limited to those compressed sulcal spaces immediately adjacent to the mass. In this group of patients, sulcal hyperintensity was again frequently associated with focal vascular enhancement particularly associated with large masses. Therefore, sulcal hyperintensity on FLAIR imaging associated with mass effect in patients without apparent CSF abnormality may also be caused by the alteration of regional hemodynamics due to sulcal and vascular compression by the mass itself. Abnormal enhancement suggests that the breakdown of the blood—brain barrier and leakage of the protein (serum) into the local sulcus may exist concurrently. The dirty CSF sign was frequently seen in our patients in the corresponding sulcal spaces with sulcal hyperintensity, but not in the remaining sulcal spaces distant to or on the contralateral side of the brain (Fig. 2A,2B,2C,2D,2E). The dirty CSF sign has been reported in cases of subarachnoid hemorrhage [14], presumably because of the paramagnetic effect from the blood product and high protein concentration in the CSF itself. Although the cause of the shortening of the T1 and T2 relaxation times of the CSF in the sulcal space is unknown, in our patients, it is unlikely that it originates from the CSF alone because no evidence suggests abnormality in the CSF. The mechanism causing changes in the sulcal signal on FLAIR images and unenhanced T1- and T2-weighted imaging (dirty CSF sign) may be similar in patients without CSF abnormality.
Several mechanisms have been proposed to explain the failure of the nulling of CSF signal intensity (sulcal hyperintensity) on FLAIR imaging, mostly involving patients with CSF abnormality [12, 17, 19]. In patients with abnormal CSF, it is generally accepted that the failure to suppress CSF on FLAIR imaging is caused by the shortening of T1 relaxation of the protons in the CSF due to blood products and higher protein concentration in the CSF [12, 15]. In our study, patients with sulcal hyperintensity on FLAIR images had no apparent clinical, laboratory, or CT evidence of abnormal CSF. However, on the basis of the frequent association of abnormal enhancement to sulcal hyperintensity on FLAIR images, local breakdown of the blood—brain barrier may be another cause of sulcal hyperintensity [20, 21]. Local serum leakage may cause local higher protein concentration of CSF, which does not reflect overall CSF findings as in subacute stroke. Local CSF with higher protein concentration may show reduced T1 relaxation time causing hyperintensity on FLAIR imaging. Because some instances show sulcal hyperintensity on FLAIR imaging without evidence of blood—brain barrier breakdown as in venous abnormality or mass effect, local protein leakage may not be the only cause of sulcal hyperintensity on FLAIR imaging.
Another possible mechanism is the "flow-entering" phenomenon of CSF flow because FLAIR imaging applies an inversion pulse. With the flow-entering phenomenon, the protons of the CSF outside the imaging plane that have not experienced inversion pulses may have a higher signal than those protons that remain in the imaging plane on FLAIR images. However, CSF flow effect should differ from section to section, and the signal of the entry section should be the highest. Although we cannot completely exclude the CSF flow-entering phenomenon, this mechanism was not readily appreciated in our study. Although this effect is prevalent in the basal cisterns in which CSF pulsation can be a significant factor, it is uncommon over the convexities of the brain where CSF motion is diminished. In addition, signal changes caused by turbulent CSF flow tended to be more inhomogeneous, whereas by visual examination, sulcal hyperintensity was generally seen diffusely in the sulcal spaces in our patients.
On the other hand, the flow-entering phenomenon of the vessels should be taken into account. The blood from outside the imaging plane will have high signal intensity on FLAIR images. CSF signal changes on FLAIR have been reported in one case of moyamoya disease [19], in which engorged pial vasculature from leptomeningeal anastomoses was the proposed cause of the sulcal hyperintensity as seen in our patients (Figs. 2A,2B,2C,2D,2E,3A,3B,3C,3D,4A,4B,4C,4D). In our study, abnormal enhancement patterns including leptomeningeal and vascular enhancement were frequently associated with hyperintense CSF on FLAIR images. Leptomeningeal enhancement extended into the depths of the sulci and reflected the vascular hemodynamic changes at pial-to-arachnoid (arterial capillary and venous) levels. Although slow flow in leptomeningeal collaterals has been proposed and pathophysiologically correlates with vascular enhancement [22], other factors including arterial, venous, perivascular, and leptomeningeal contrast medium accumulation may also cause intravascular enhancement [23,24,25,26,27]. All cases of hydrocephalus, subdural hematoma, and large mass showed diffuse CSF hyperintensity on FLAIR imaging and diffuse leptomeningeal enhancement. In these cases, intracranial pressure is expected to be elevated and again leads to more extensive venous congestion as compared with those cases of small focal lesions. Our data cannot reveal why the abnormal enhancement pattern has a high association with CSF hyperintensity on FLAIR imaging or how its mechanism contributes to the high signal intensity. However, these findings appear to show, at least in part, that the alteration of the regional vascular dynamics such as congestion, inflammation, or slow flow can play a role in the presence of CSF hyperintensity on FLAIR imaging in patients with normal CSF.
The expansion of the blood pool in the corresponding sulcal space causing failure of the nulling of the CSF signal may be a possible mechanism. The sulcal space consists primarily of the CSF and, to a much lesser extent, the blood vessels consisting of arteries, arteriocapillary bed (leptomeningeal), and venous network. In a normal sulcus, the relatively small volume and oxygenation status of the blood pool in the vascular structure has a minimal effect on the local magnetic field and signal intensity of CSF protons. Therefore, in the healthy patient, the signal intensity of the sulcal space on FLAIR imaging is low because of a successful suppression of the dominant CSF signal in the sulci. However, the blood pool can become more influential in patients with an alteration of the underlying hemodynamics. The blood pool effect can theoretically increase when the relative volume ratio of CSF-blood in a voxel is decreased (as with a decrease in sulcal CSF space or an increase in intravascular volume caused by mass effect or vascular disease). Diffuse sulcal hyperintensity on FLAIR images was seen in our patients with evidence of an increase in venous pressure and blood pool such as in a case of dural sinus occlusion (Fig. 3A,3B,3C,3D) and in another case of pre- and post-treatment of dural arteriovenous fistula (Fig. 4A,4B,4C,4D). The congested venous blood pool increases in volume and contains a higher concentration of deoxyhemoglobin compared with the normal state. This condition can cause shortening of T2* relaxation time due to the paramagnetic effects of deoxyhemoglobin. In our study, we obtained gradient-echo T2*-weighted images in a stroke patient. In this patient, the area corresponding to the sulcal hyperintensity on FLAIR images and congested vasculature on the T2*-weighted images showed mild hypointensity suggesting an increase in local field inhomogeneity (Fig. 2A,2B,2C,2D,2E). These cases again suggest that alteration of hemodynamics causing increased blood pool may play a role in sulcal hyperintensity on FLAIR imaging.
Our study had several limitations. Our preliminary data cannot prove or disprove the mechanism explaining the sulcal hyperintensity on FLAIR imaging in patients without apparent CSF abnormality. In addition, we only prospectively screened 300 MR images to recruit patients for retrospective review. Only those cases with MR images that contained unenhanced and contrast-enhanced T1-weighted and T2-weighted FLAIR images were included. Therefore, incidence and sensitivity of detecting sulcal hyperintensity and its associated pathophysiology and other MR findings were biased. Finally, the proposed mechanisms in this study are speculative and are based indirectly on the MR imaging and clinical findings without definitive proof.
Sulcal hyperintensity on FLAIR is a non-specific finding and occurs frequently in patients without CSF abnormality. Therefore, caution should be taken in interpreting such findings particularly in the emergency treatment of central nervous system diseases that requires information to avoid potential hemorrhagic complications. With the frequent association of vascular disease, mass effect, abnormal vascular enhancement, and dirty CSF sign, we speculate that alteration in vascular hemodynamics with slow flow, increased blood pool, and a small amount of protein leakage may be the cause for the failure in suppression of the CSF signal (sulcal hyperintensity) in patients without apparent CSF abnormality.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K K THA, S TERAE, K KUDO, and K MIYASAKA Differential diagnosis of hyperintense cerebrospinal fluid on fluid-attenuated inversion recovery images of the brain. Part I: pathological conditions Br. J. Radiol., May 1, 2009; 82(977): 426 - 434. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Nakagawa, F. A. Sorond, and A. H. Ropper Ultra-Early Magnetic Resonance Imaging Findings of Eclampsia Arch Neurol, July 1, 2008; 65(7): 974 - 976. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B.E. Hamilton and G.M. Nesbit Delayed CSF Enhancement in Posterior Reversible Encephalopathy Syndrome AJNR Am. J. Neuroradiol., March 1, 2008; 29(3): 456 - 457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.M. Morris and G.M. Miller Increased Signal in the Subarachnoid Space on Fluid-Attenuated Inversion Recovery Imaging Associated with the Clearance Dynamics of Gadolinium Chelate: A Potential Diagnostic Pitfall AJNR Am. J. Neuroradiol., November 1, 2007; 28(10): 1964 - 1967. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. L. Stuckey, T. D. Goh, T. Heffernan, and D. Rowan Hyperintensity in the Subarachnoid Space on FLAIR MRI Am. J. Roentgenol., October 1, 2007; 189(4): 913 - 921. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. V. Maramattom, E. M. Manno, E. F.M. Wijdicks, and E. P. Lindell Gadolinium encephalopathy in a patient with renal failure Neurology, April 12, 2005; 64(7): 1276 - 1278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Galassi, W. Phuttharak, J. R. Hesselink, J. F. Healy, R. B. Dietrich, and S. G. Imbesi Intracranial Meningeal Disease: Comparison of Contrast-Enhanced MR Imaging with Fluid-Attenuated Inversion Recovery and Fat-Suppressed T1-Weighted Sequences AJNR Am. J. Neuroradiol., March 1, 2005; 26(3): 553 - 559. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Frigon, D. W. W. Shaw, S. R. Heckbert, E. Weinberger, and D. S. Jardine Supplemental Oxygen Causes Increased Signal Intensity in Subarachnoid Cerebrospinal Fluid on Brain FLAIR MR Images Obtained in Children during General Anesthesia Radiology, October 1, 2004; 233(1): 51 - 55. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Anzai, M. Ishikawa, D. W. W. Shaw, A. Artru, V. Yarnykh, and K. R. Maravilla Paramagnetic Effect of Supplemental Oxygen on CSF Hyperintensity on Fluid-Attenuated Inversion Recovery MR Images AJNR Am. J. Neuroradiol., February 1, 2004; 25(2): 274 - 279. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Iancu-Gontard, C. Oppenheim, E. Touze, E. Meary, M. Zuber, J.-L. Mas, D. Fredy, and J.-F. Meder Evaluation of Hyperintense Vessels on FLAIR MRI for the Diagnosis of Multiple Intracerebral Arterial Stenoses Stroke, August 1, 2003; 34(8): 1886 - 1891. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Frigon, D. S. Jardine, E. Weinberger, S. R. Heckbert, and D. W. W. Shaw Fraction of Inspired Oxygen in Relation to Cerebrospinal Fluid Hyperintensity on FLAIR MR Imaging of the Brain in Children and Young Adults Undergoing Anesthesia Am. J. Roentgenol., September 1, 2002; 179(3): 791 - 796. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
