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


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Okuda, T.
Right arrow Articles by Takahashi, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Okuda, T.
Right arrow Articles by Takahashi, M.
(Radiology. 1999;212:793-798.)
© RSNA, 1999

Neuroradiology

Brain Lesions: When Should Fluid-attenuated Inversion-Recovery Sequences Be Used in MR Evaluation?1

Tomoko Okuda, MD, Yukunori Korogi, MD, Yoshinori Shigematsu, MD, Takeshi Sugahara, MD, Toshinori Hirai, MD, Ichiro Ikushima, MD, Luxia Liang, MD and Mutsumasa Takahashi, MD

1 From the Department of Radiology, Kumamoto University School of Medicine, 1-1-1 Honjo, Kumamoto 860, Japan. From the 1997 RSNA scientific assembly. Received February 13, 1998; revision requested April 15; final revision received December 16; accepted March 16, 1999. Address reprint requests to T.O.


   Abstract
TOP
 Abstract
Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
PURPOSE: To compare qualitatively and quantitatively the contrast of brain lesions detected with fluid-attenuated inversion-recovery (FLAIR) and intermediate-weighted sequences at magnetic resonance (MR) imaging.

MATERIALS AND METHODS: In this prospective study, 47 patients suspected of having a brain lesion underwent MR imaging with FLAIR, intermediate-weighted, and T2-weighted sequences. Qualitative assessment was performed of lesion conspicuity, detection, overall image artifact, and additional clinical information. Contrast and contrast-to-noise ratio (CNR) were calculated between lesions and the normal brain or cerebrospinal fluid (CSF).

RESULTS: FLAIR images were equal to intermediate-weighted images for overall lesion conspicuity and detection but were associated more often with image artifacts. Lesion-to-background contrast was significantly higher on FLAIR than on intermediate-weighted images. FLAIR images failed to demonstrate multiple sclerosis (MS) plaques located in the basal ganglia and brain stem.

CONCLUSION: Although FLAIR images provided additional information in some cases, they did not have distinct advantages over intermediate-weighted images. When cases of MS are evaluated, intermediate-weighted images are preferable to FLAIR images. Except in cases of MS, either FLAIR or intermediate-weighted sequences should be added to T2-weighted sequences at MR imaging.

Index terms: Basal ganglia, MR, 14.121411, 14.121413, 15.121416 • Brain, diseases, 14.30, 14.76, 14.871, 15.30, 15.76, 15.871 • Brain, ischemia, 14.76, 15.76 • Brain, neoplasms, 14.30, 15.30 • Brain stem, MR, 15.121411, 15.121413, 15.121416 • Magnetic resonance (MR), pulse sequences, 18.121411, 18.121413, 18.121416 • Sclerosis, multiple, 18.871


   Introduction
TOP
 Abstract
 Introduction
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Fluid-attenuated inversion-recovery (FLAIR) sequences have been developed that null the signals from cerebrospinal fluid (CSF) and yield heavily T2-weighted images. The usefulness of FLAIR sequences has been reported in central nervous system diseases, including infarction, multiple sclerosis (MS), subarachnoid hemorrhage, head injury, and others (1–14). Most of these reports confirm the superiority of FLAIR over spin-echo (SE) magnetic resonance (MR) imaging, and some reports suggest the possibility of replacing SE with FLAIR technique (1,2), although other investigators have recently pointed out disadvantages of FLAIR sequences (3,4,15–17).

At present, FLAIR images are usually obtained in addition to T1- and T2-weighted images, since they are expected to give additional useful information. There is some similarity between intermediate-weighted and FLAIR imaging in regard to suppression of the signal of CSF. However, the number of examinations with intermediate-weighted sequences has tended to decrease with the development of the fast SE sequence. In the previously mentioned reports about the usefulness of FLAIR images, however, they were compared with T2-weighted, short inversion time inversion-recovery, or STIR, or gadolinium-enhanced T1-weighted images (1–17). Only a few of these articles reported a direct comparison between FLAIR and intermediate-weighted sequences.

The purpose of this study was to compare qualitatively and quantitatively the image contrast of brain lesions detected on FLAIR images and intermediate-weighted images and to document which is more suitable as an additional sequence in routine MR examinations.


   MATERIALS AND METHODS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
In this prospective study, 47 patients (23 male and 24 female; age range, 0–76 years; mean age, 37.6 years) suspected of having brain lesions on the basis of clinical history or neurologic findings underwent MR imaging at 1.5 T (Magnetom-Vision; Siemens Medical Systems, Erlangen, Germany) with FLAIR, intermediate-weighted, and T2-weighted pulse sequences.

All images were obtained in the axial plane (5-mm section thickness with 1-mm section gap). In all patients, T2-weighted imaging was performed with the following sequences and parameters: T2-weighted fast SE (repetition time msec/effective echo time msec = 4,500/96, one signal acquired, 19 sections, echo train length of seven) or T2-weighted SE (2,400/90, one signal acquired, 19 sections). In all patients, intermediate-weighted fast SE imaging was performed with a single-echo sequence (2,400/15, two signals acquired, 19 sections, echo train length of seven). In all patients, FLAIR imaging (8,000/119/2,000 [inversion time msec]) was performed with two signals acquired, acquisition time of 4 minutes 24 seconds for 15 sections, echo train length of seven, 180–224 x 256 matrix, and field of view of 150–220 mm. In all patients, nonenhanced T1-weighted images (670/14) were obtained with one signal acquired. As proof of existence of a lesion on the MR images, two neuroradiologists (T.O., Y.K.) carefully reviewed the nonenhanced T1-weighted, T2-weighted, and gadolinium-enhanced T1-weighted images (if available) by consensus.

FLAIR images were compared with intermediate-weighted images qualitatively and quantitatively. Qualitative evaluations were performed independently by two neuroradiologists (T.S., Y.S.) with knowledge of the patient's clinical diagnosis. FLAIR and intermediate-weighted images were presented at the same time as were the T2-weighted images. Four qualitative evaluations were performed: the value of additional clinical information, image artifacts, lesion conspicuity, and lesion detection.

The value of additional clinical information was defined on the basis of the following grading system: +2, FLAIR or intermediate-weighted images contributed to clinical diagnosis or treatment; +1, FLAIR or intermediate-weighted images contributed to lesion detection or visualization; 0, FLAIR or intermediate-weighted images added no information to T2-weighted images; -1, FLAIR or intermediate-weighted images caused misdiagnosis. The statistical significance of differences with this criterion was determined with the paired t test.

The following three-point grading system was used to evaluate the criteria of overall image artifacts, lesion conspicuity (defined as ease of lesion visualization), and lesion detection (defined as the presence or absence of lesions): +1, FLAIR images were superior to intermediate-weighted images; 0, FLAIR and intermediate-weighted images were equal; -1, FLAIR images were inferior to intermediate-weighted images. All studies were evaluated for overall image artifact, and studies with pathologic conditions were evaluated for lesion conspicuity and lesion detection. The statistical significance of differences with these three qualitative criteria was determined with the signed rank test.

FLAIR and intermediate-weighted images were also compared quantitatively by one neuroradiologist (T.O.). The quantitative criteria were lesion-to-background contrast and contrast-to-noise ratio (CNR) and lesion-to-CSF contrast and CNR. "Background" was defined as the normal brain parenchyma adjacent to the lesion. For each patient, contrast and CNR were measured and compared for a maximum of two representative lesions, which were chosen randomly from the group of lesions visualized equally well with both sequences. Mean lesion signal intensity (SI) was measured in regions of interest placed within similar uniform areas of the lesion. Mean background SI was measured in areas immediately surrounding the lesion. The SDs of noise were measured on the image along the phase-encoding direction in space outside the head. With these measurements, contrast and CNR for each lesion were calculated as follows: contrast = (SIlesion - SIB or CSF)/SIB or CSF and CNR = (SIlesion - SIB or CSF)/SD SInoise, where B is background. Paired t tests were performed to evaluate the statistical significance of differences with these quantitative data.


   RESULTS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
No lesions were seen on the MR images of seven of the 47 patients. Two different types of lesions were observed in nine of the 47 patients. Therefore, a total of 49 different lesions or groups of lesions were identified and evaluated. The final clinical diagnoses included MS (n = 12), ischemia or infarction (n = 9), brain tumor (n = 12), or miscellaneous other conditions (three, tuberous sclerosis; three, white matter lesions; three, basal ganglia lesions of uncertain pathogenesis; two, edema; one, parenchymal hemorrhage; one, subdural fluid collection; one, acquired immunodeficiency syndrome, or AIDS-related lesions; one, dilated Virchow-Robin space; and one, laminar necrosis).

The 12 brain tumors were extraaxial (five, arachnoid cyst or other cysts; two, epidermoid; one, meningioma; one, chondroma) or intraaxial (one, lymphoma; two, glioma). Diagnoses of the brain tumors were proved histopathologically except for the arachnoid cyst and other cysts.

In cases without verification by means of biopsy, diagnoses were based on clinical history, clinical presentation, and findings on nonenhanced T1- and T2- weighted images and gadolinium-enhanced T1-weighted images (if available). MS, ischemia or infarction, and tumor were evaluated as separate entities. Because of the small number of cases, the remaining 16 lesions were evaluated together as miscellaneous lesions.

Results of the qualitative comparison of FLAIR and intermediate-weighted images are summarized in Tables 1–4. Additional clinical information based on FLAIR and intermediate-weighted images was evaluated in 56 studies (49 with pathologic conditions and seven without lesions). Radiologists 1 and 2, respectively, evaluated as useful (grades +1 or +2) 16 (29%) and 18 (32%) FLAIR studies and 12 (21%) and 17 (30%) intermediate-weighted studies. There were no significant differences between FLAIR and intermediate-weighted images (Table 1). For overall image artifact, radiologists 1 and 2 evaluated FLAIR studies as inferior to intermediate-weighted studies in 33 (70%) and 25 (53%) of 47 patients, respectively. FLAIR images received an overall score of -0.72 ± 0.81 (mean ± SD) (radiologist 1) and -0.32 ± 0.90 (radiologist 2), and they were rated as significantly inferior to intermediate-weighted images (P < .001, P < .05, respectively) (Table 2). Artifacts tended to occur in areas of prominent CSF pulsatility, such as inferiorly located sections and areas near the foramen of Monro.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Additional Clinical Information: FLAIR versus Intermediate-weighted Images
 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Overall Assessment of Artifacts: FLAIR versus Intermediate-weighted Images
 

View this table:
[in this window]
[in a new window]
  		TABLE 3. Lesion Conspicuity and Lesion Detection: FLAIR versus Intermediate-weighted Images
 

View this table:
[in this window]
[in a new window]
  		TABLE 4. Lesion Conspicuity and Lesion Detection in the 12 Patients with MS
 
For lesion conspicuity and detection, FLAIR studies were rated as roughly equivalent to intermediate-weighted studies. MS plaques, however, were demonstrated less clearly on FLAIR than on intermediate-weighted images (Table 3). When the locations of the MS plaques were separated into supra- and infra-tentorial regions, both radiologists graded FLAIR studies as inferior to intermediate-weighted studies for infratentorial lesions (Table 4). In particular, lesion contrast was insufficient at the brain stem (Fig 1). For supratentorial lesions, both radiologists graded FLAIR studies as roughly equivalent to intermediate-weighted studies, although grades for conspicuity and detection of basal ganglia lesions were lower for FLAIR than for intermediate-weighted studies. In one case of MS, a basal ganglia lesion had low SI on FLAIR images and was difficult to identify, but it had high SI on intermediate- and T2-weighted images (Fig 2). Lesions of the cerebral white matter appeared to be more extensive on FLAIR images.



View larger version (167K):
[in this window]
[in a new window]
  		Figure 1a. Axial MR images in a 53-year-old man with MS. (a) T2-weighted image (4,500/96). (b) Intermediate-weighted image (2,400/15). (c) FLAIR image (8,000/119/2,000). MS plaque in the pons (arrow), which is clearly depicted in a and b, is difficult to identify in c because of insufficient contrast.

 


View larger version (162K):
[in this window]
[in a new window]
  		Figure 1b. Axial MR images in a 53-year-old man with MS. (a) T2-weighted image (4,500/96). (b) Intermediate-weighted image (2,400/15). (c) FLAIR image (8,000/119/2,000). MS plaque in the pons (arrow), which is clearly depicted in a and b, is difficult to identify in c because of insufficient contrast.

 


View larger version (178K):
[in this window]
[in a new window]
  		Figure 1c. Axial MR images in a 53-year-old man with MS. (a) T2-weighted image (4,500/96). (b) Intermediate-weighted image (2,400/15). (c) FLAIR image (8,000/119/2,000). MS plaque in the pons (arrow), which is clearly depicted in a and b, is difficult to identify in c because of insufficient contrast.

 


View larger version (147K):
[in this window]
[in a new window]
  		Figure 2a. (a-f) Axial MR images in a 20-year-old woman with MS. (a, d) T2-weighted images (4,500/96). (b, e) Intermediate-weighted images (2,400/15). (c, f) FLAIR images (8,000/119/2,000). (a-c) MS plaques in the basal ganglia (long arrows) were slightly hyperintense in a and were clearly depicted in b but have insufficient contrast in c. However, another MS plaque (short arrow) in the frontal lobe lateral to the Sylvian fissure is clearly seen in c but not in a or b. (d-f) Images were obtained at a level slightly inferior to that in a-c. MS plaque in the left basal ganglia (open arrow) has high SI in d and e but low SI in f. Another MS plaque in the midbrain (arrowhead) is seen more clearly in e than in f.

 


View larger version (158K):
[in this window]
[in a new window]
  		Figure 2b. (a-f) Axial MR images in a 20-year-old woman with MS. (a, d) T2-weighted images (4,500/96). (b, e) Intermediate-weighted images (2,400/15). (c, f) FLAIR images (8,000/119/2,000). (a-c) MS plaques in the basal ganglia (long arrows) were slightly hyperintense in a and were clearly depicted in b but have insufficient contrast in c. However, another MS plaque (short arrow) in the frontal lobe lateral to the Sylvian fissure is clearly seen in c but not in a or b. (d-f) Images were obtained at a level slightly inferior to that in a-c. MS plaque in the left basal ganglia (open arrow) has high SI in d and e but low SI in f. Another MS plaque in the midbrain (arrowhead) is seen more clearly in e than in f.

 


View larger version (167K):
[in this window]
[in a new window]
  		Figure 2c. (a-f) Axial MR images in a 20-year-old woman with MS. (a, d) T2-weighted images (4,500/96). (b, e) Intermediate-weighted images (2,400/15). (c, f) FLAIR images (8,000/119/2,000). (a-c) MS plaques in the basal ganglia (long arrows) were slightly hyperintense in a and were clearly depicted in b but have insufficient contrast in c. However, another MS plaque (short arrow) in the frontal lobe lateral to the Sylvian fissure is clearly seen in c but not in a or b. (d-f) Images were obtained at a level slightly inferior to that in a-c. MS plaque in the left basal ganglia (open arrow) has high SI in d and e but low SI in f. Another MS plaque in the midbrain (arrowhead) is seen more clearly in e than in f.

 


View larger version (161K):
[in this window]
[in a new window]
  		Figure 2d. (a-f) Axial MR images in a 20-year-old woman with MS. (a, d) T2-weighted images (4,500/96). (b, e) Intermediate-weighted images (2,400/15). (c, f) FLAIR images (8,000/119/2,000). (a-c) MS plaques in the basal ganglia (long arrows) were slightly hyperintense in a and were clearly depicted in b but have insufficient contrast in c. However, another MS plaque (short arrow) in the frontal lobe lateral to the Sylvian fissure is clearly seen in c but not in a or b. (d-f) Images were obtained at a level slightly inferior to that in a-c. MS plaque in the left basal ganglia (open arrow) has high SI in d and e but low SI in f. Another MS plaque in the midbrain (arrowhead) is seen more clearly in e than in f.

 


View larger version (166K):
[in this window]
[in a new window]
  		Figure 2e. (a-f) Axial MR images in a 20-year-old woman with MS. (a, d) T2-weighted images (4,500/96). (b, e) Intermediate-weighted images (2,400/15). (c, f) FLAIR images (8,000/119/2,000). (a-c) MS plaques in the basal ganglia (long arrows) were slightly hyperintense in a and were clearly depicted in b but have insufficient contrast in c. However, another MS plaque (short arrow) in the frontal lobe lateral to the Sylvian fissure is clearly seen in c but not in a or b. (d-f) Images were obtained at a level slightly inferior to that in a-c. MS plaque in the left basal ganglia (open arrow) has high SI in d and e but low SI in f. Another MS plaque in the midbrain (arrowhead) is seen more clearly in e than in f.

 


View larger version (172K):
[in this window]
[in a new window]
  		Figure 2f. (a-f) Axial MR images in a 20-year-old woman with MS. (a, d) T2-weighted images (4,500/96). (b, e) Intermediate-weighted images (2,400/15). (c, f) FLAIR images (8,000/119/2,000). (a-c) MS plaques in the basal ganglia (long arrows) were slightly hyperintense in a and were clearly depicted in b but have insufficient contrast in c. However, another MS plaque (short arrow) in the frontal lobe lateral to the Sylvian fissure is clearly seen in c but not in a or b. (d-f) Images were obtained at a level slightly inferior to that in a-c. MS plaque in the left basal ganglia (open arrow) has high SI in d and e but low SI in f. Another MS plaque in the midbrain (arrowhead) is seen more clearly in e than in f.

 
In most of the cases of ischemia or infarction, conspicuity was better on FLAIR than on intermediate-weighted studies. In one case of infarction, the lesion of the middle cerebellar peduncle had low SI on FLAIR images. For extraaxial tumors, there were no differences in conspicuity or detection between the FLAIR and intermediate-weighted studies, whereas FLAIR studies were graded as slightly superior to intermediate-weighted studies for the intraaxial tumors. In cases of tuberous sclerosis, some cortical tuber lesions not detected on the FLAIR images were detected on the intermediate-weighted images (Fig 3).



View larger version (159K):
[in this window]
[in a new window]
  		Figure 3a. Axial MR images of cortical tubers in a 5-month-old infant with tuberous sclerosis. (a) T2-weighted image (4,500/96). (b) Intermediate-weighted image (2,400/15). (c) FLAIR image (8,000/119/2,000). Multiple cortical tubers (arrows in a and b) are depicted most clearly in a, followed by b. Most of these lesions are not depicted in c, but the subcortical lesion (arrowhead in c) in the right parietal area is clearly depicted in c as cystic.

 


View larger version (170K):
[in this window]
[in a new window]
  		Figure 3b. Axial MR images of cortical tubers in a 5-month-old infant with tuberous sclerosis. (a) T2-weighted image (4,500/96). (b) Intermediate-weighted image (2,400/15). (c) FLAIR image (8,000/119/2,000). Multiple cortical tubers (arrows in a and b) are depicted most clearly in a, followed by b. Most of these lesions are not depicted in c, but the subcortical lesion (arrowhead in c) in the right parietal area is clearly depicted in c as cystic.

 


View larger version (168K):
[in this window]
[in a new window]
  		Figure 3c. Axial MR images of cortical tubers in a 5-month-old infant with tuberous sclerosis. (a) T2-weighted image (4,500/96). (b) Intermediate-weighted image (2,400/15). (c) FLAIR image (8,000/119/2,000). Multiple cortical tubers (arrows in a and b) are depicted most clearly in a, followed by b. Most of these lesions are not depicted in c, but the subcortical lesion (arrowhead in c) in the right parietal area is clearly depicted in c as cystic.

 
Results of the quantitative evaluation are shown in Table 5. Lesion-to-CSF and lesion-to-background contrast were significantly superior on FLAIR compared with intermediate-weighted studies. Lesion-to-background CNR was significantly superior on intermediate-weighted studies, but differences in lesion-to-CSF CNR were not significant.


View this table:
[in this window]
[in a new window]
  		TABLE 5. Lesion-to-Background and Lesion-to-CSF Contrast and CNR: FLAIR versus Intermediate-weighted Images
 

   DISCUSSION
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
FLAIR imaging is a relatively new technique with a long inversion time and a long echo time that can null the signals from CSF and produce heavily T2-weighted studies. This type of sequence is particularly useful in the detection of subtle changes at the periphery of the hemispheres and in the periventricular region close to CSF. Although advantages with intermediate-weighted sequences are similar to those with FLAIR sequences in regard to suppression of the signal of CSF, intermediate-weighted studies have insufficient contrast owing to a short echo time, and the number of intermediate-weighted examinations has decreased with the development of the fast SE sequence. The usefulness of FLAIR sequences has been evaluated in diseases of the central nervous system such as infarction, MS, subarachnoid hemorrhage, head injuries, and others (1–14). The authors of most of these reports have concluded that FLAIR sequences are superior to intermediate- and T2-weighted sequences, and some suggest a possibility of replacing the SE technique with FLAIR (1,2). However, a few investigators have recently pointed out disadvantages of FLAIR sequences (3,4,15–17). It is now necessary to evaluate which pulse sequence is more useful for the imaging of brain abnormalities.

Since a long acquisition time is needed with conventional FLAIR sequences, a faster FLAIR sequence is now routinely used with multiple echoes per repetition time that is based on rapid acquisition with relaxation enhancement, or RARE, and fast SE sequences (1–7). Although the current technique represents an improvement over previous methods, another problem with FLAIR sequences remains. Incomplete nulling of CSF signals due to CSF inflow effects has been reported to produce imaging artifacts (1,3,5). These artifacts occurred in areas of prominent CSF pulsatility, such as inferiorly located sections and those containing foramina of the CSF ventricular system. In this study, we also encountered similar CSF flow-related artifacts on FLAIR images, which contributed to failure to detect lesions located in the brain stem.

In addition to the appearance of CSF flow-related artifacts, poor lesion contrast may be another drawback with FLAIR sequences. For lesion conspicuity and detection of MS plaques, FLAIR studies received a lower grade and were judged inferior to intermediate-weighted studies, particularly in the posterior fossa and basal ganglia. Flow-related artifacts may not be the sole contributing factor to poor lesion conspicuity and detection. Although the cause is not clear, there are some possible explanations for the poor contrast on FLAIR images. One possible explanation is that the MS plaques in the brain stem or basal ganglia may have different T2 characteristics (15). Hittmair et al (3) reported limitations of FLAIR sequences in patients with MS of the spinal cord. With only slightly prolonged T2, characteristic of inactive MS plaques, the signals of the normal cord and pathologic lesions have already decreased at a long echo time. In these cases, use of shorter echo times can produce a higher contrast between the normal cord and pathologic lesions. Since the fast FLAIR sequence consists of an inversion pulse followed by a heavily T2-weighted fast SE sequence, the long echo time and echo train length in fast FLAIR studies may explain the poor plaque contrast (3,15). In addition, the inversion pulse in FLAIR sequences introduces considerable T1 weighting, which acts antagonistically to the T2 contrast (3). The brain stem and basal ganglia have a structure different from that of white matter. In these regions, MS plaques may have relaxation times similar to those of adjacent brain parenchyma.

In chronic infarctions, some lesions with high SI on T2-weighted images are depicted with low SI or isointensity on FLAIR images and are difficult to identify (2,4). It is presumed that these foci may be liquefied or gliotic and, therefore, their SI matches that of CSF and is nulled with FLAIR sequences. This opens the possibility that FLAIR images can help differentiate acute from chronic infarctions on the basis of their SI. However, it should be noted that FLAIR studies sometimes could not help identification of lesions with low SI or isointensity. In our study, MS plaques depicted with low SI were also observed on FLAIR images, and they were deemed chronic lesions.

There are several limitations in this study. First, many different kinds of disease were evaluated as one group. For example, miscellaneous lesions included nine kinds of diseases, and the number of these lesions was insufficient to enable definitive conclusions. Also, we did not differentiate acute MS plaques or infarctions from chronic lesions. Evaluation of a large number of each type of lesion may be more effective to choose the optimum sequence. Furthermore, use of an optimum pulse sequence suitable for each disease may be necessary. Second, two different T2-weighted sequences were used. Since examinations with each sequence included a variety of lesions, comparison of the two sequences was impossible. Third, the presence of CSF flow-related artifacts on FLAIR images, which contributed to failure to demonstrate lesions located in the brain stem, may be directly related to the pulse sequence design and its implementation. CSF signals would be completely nulled if the pulse sequence were further optimized. Fourth, the subjective part of the study was somewhat biased. The reviewers could identify which was the FLAIR image, they evaluated the images side by side rather than separately and independently, and the grading criteria were subjective. We included seven patients who had no lesions and evaluated additional clinical information and image artifacts on their studies. The subjective nature of the criteria regarding additional information may also be a source of potential bias.

In some cases, the FLAIR studies offered useful clinical information as an adjunct to T2-weighted images, particularly in the detection of subtle changes in the region contiguous to CSF. The FLAIR studies also helped differentiate cystic lesions on the basis of their SI, whereas T2-weighted images depicted almost all lesions as hyperintense areas (2,4–7). Intermediate-weighted studies were equivalent to FLAIR studies, however, in their ability to provide useful clinical information. FLAIR studies had some pitfalls, such as CSF flow-related artifacts, insufficient lesion contrast in the basal ganglia and posterior fossa (particularly MS plaques), and the inability to clearly depict cystic lesions.

In summary, FLAIR sequences have advantages and disadvantages in the demonstration of central nervous system diseases and cannot replace T2-weighted sequences. It is important to recognize that lesions in the basal ganglia or brain stem may be missed with FLAIR sequences. Intermediate-weighted images should be obtained in patients with MS, but the FLAIR sequence may be omitted from the MR evaluation. Except in cases of MS, FLAIR images provided additional information compared with T2-weighted images and offered higher lesion contrast than did intermediate-weighted images, and we recommend routine use of FLAIR sequences in addition to T2-weighted sequences. Intermediate-weighted images can substitute for FLAIR images.


   Footnotes
 
Abbreviations: CNR = contrast-to-noise ratio CSF = cerebrospinal fluid FLAIR = fluid-attenuated inversion recovery MS = multiple sclerosis SE = spin echo SI = signal intensity

Author contributions: Guarantor of integrity of entire study, Y.K.; study concepts and design, T.O., Y.K.; definition of intellectual content, T.O., Y.K.; literature research, T.O.; clinical studies, L.L., I.I., T.H., T.S., Y.S.; data acquisition and analysis, T.O.; statistical analysis, T.O.; manuscript preparation, T.O.; manuscript editing and review, Y.K., M.T.


   References
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Hashemi RH, Bradley WG, Jr, Chen DY, et al. Suspected multiple sclerosis: MR imaging with a thin-section fast FLAIR pulse sequence. Radiology 1995; 196:505-510.[Abstract/Free Full Text]
  2. Alexander JA, Sheppard S, Davis PC, Salverda P. Adult cerebrovascular disease: role of modified rapid fluid-attenuated inversion recovery sequences. AJNR 1996; 17:1507-1513.[Abstract]
  3. Hittmair K, Mallek R, Prayer D, Schindler EG, Kollegger H. Spinal cord lesions in patients with multiple sclerosis: comparison of MR pulse sequences. AJNR 1996; 17:1555-1565.[Abstract]
  4. Shiono T, Teshima Y, Iwasaki N. Advantages and disadvantages of fast fluid-attenuated inversion recovery sequences in the evaluation of brain infarction. Nippon Igaku Hoshasen Gakkai Zasshi 1996; 56:625-628.[Medline]
  5. Rydberg JN, Hammond CA, Grimm MS, et al. Initial clinical experience in MR imaging of the brain with a fast fluid-attenuated inversion-recovery pulse sequence. Radiology 1994; 193:173-180.[Abstract/Free Full Text]
  6. Hajnal JV, Bryant DJ, Kasuboski L, et al. Use of fluid attenuated inversion recovery (FLAIR) pulse sequence in MRI of the brain. J Comput Assist Tomogr 1992; 16:841-844.[Medline]
  7. Coene BD, Hajnal JV, Gatehouse P, et al. MR of the brain using fluid-attenuated inversion recovery (FLAIR) pulse sequence. AJNR 1992; 13:1555-1564.[Abstract]
  8. Coene BD, Hajnal JV, Pennock JM, Bydder GM. MRI of the brain stem using fluid-attenuated inversion recovery pulse sequence. Neuroradiology 1993; 35:327-331.[Medline]
  9. Noguchi K, Ogawa T, Inugami A, et al. Acute subarachnoid hemorrhage: MR imaging with fluid attenuated inversion recovery pulse sequences. Radiology 1995; 196:773-777.[Abstract/Free Full Text]
  10. Bergin PS, Fish DR, Shorvon SD, Oatridge A, deSouza NM, Bydder GM. Magnetic resonance imaging in partial epilepsy: additional abnormalities shown with the fluid attenuated inversion recovery (FLAIR) pulse sequence. J Neurol Neurosurg Psychiatry 1995; 58:439-443.[Abstract/Free Full Text]
  11. Murata T, Itoh S, Koshino Y, et al. Serial cerebral MRI with FLAIR sequences in acute carbon monoxide poisoning. J Comput Assist Tomogr 1995; 19:631-634.[Medline]
  12. Baratti C, Barkhof F, Hoogenraad F, Valk J. Partially saturated fluid attenuated inversion recovery (FLAIR) sequence in multiple sclerosis with fully relaxed FLAIR and conventional spin-echo. Magn Reson Imaging 1995; 13:513-521.[Medline]
  13. Ashikaga R. Clinical utility of MR FLAIR imaging for head injury. Nippon Igaku Hoshasen Gakkai Zasshi 1996; 56:1045-1049.[Medline]
  14. Tsuchiya K, Inaoka S, Mizutani Y, Hachiya J. Fast fluid-attenuated inversion-recovery MR of intracranial infections. AJNR 1997; 18:909-913.[Abstract]
  15. Bastianello S, Bozzao A, Paolillo A, et al. Fast spin-echo and fast fluid-attenuated inversion-recovery versus conventional spin-echo sequences for MR quantification of multiple sclerosis lesions. AJNR 1997; 18:699-704.[Abstract/Free Full Text]
  16. Yousry TA, Filippi M, Becker C, Horsfield MA, Voltz R. Comparison of MR pulse sequences in the detection of multiple sclerosis lesions. AJNR 1997; 18:959-963.[Abstract]
  17. Keiper MD, Grossman RI, Brunson JC, Schnall MD. The low sensitivity of fluid-attenuated inversion-recovery MR in the detection of multiple sclerosis of the spinal cord. AJNR 1997; 18:1035-1039.[Abstract]



This article has been cited by other articles:


Home page
 Br. J. Radiol.Home page
A Saleh, F Wenserski, M Cohnen, G Furst, E Godehardt, and U Modder
Exclusion of brain lesions: is MR contrast medium required after a negative fluid-attenuated inversion recovery sequence?
Br. J. Radiol., March 1, 2004; 77(915): 183 - 188.
[Abstract] [Full Text] [PDF]

Home page
 StrokeHome page
A. J. Bastos Leite, E. C.W. van Straaten, P. Scheltens, G. Lycklama, and F. Barkhof
Thalamic Lesions in Vascular Dementia: Low Sensitivity of Fluid-Attenuated Inversion Recovery (FLAIR) Imaging
Stroke, February 1, 2004; 35(2): 415 - 419.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
D. F. Kallmes, F. K. Hui, and J. P. Mugler III
Suppression of Cerebrospinal Fluid and Blood Flow Artifacts in FLAIR MR Imaging with a Single-Slab Three-dimensional Pulse Sequence: Initial Experience
Radiology, October 1, 2001; 221(1): 251 - 255.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
M. V. Okujava, F. G. Woermann, T. Hirai, and Y. Korogi
Evaluation of the Human Limbic Lobe Drs Hirai and Korogi respond:
Radiology, August 1, 2001; 220(2): 555 - 556.
[Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
K. Toyoda, M. Ida, and K. Fukuda
Fluid-attenuated Inversion Recovery Intraarterial Signal: An Early Sign of Hyperacute Cerebral Ischemia
AJNR Am. J. Neuroradiol., June 1, 2001; 22(6): 1021 - 1029.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
A. H. Herlihy, J. V. Hajnal, W. L. Curati, N. Virji, A. Oatridge, B. K. Puri, and G. M. Bydder
Reduction of CSF and Blood Flow Artifacts on FLAIR Images of the Brain with k-Space Reordered by Inversion Time at each Slice Position (KRISP)
AJNR Am. J. Neuroradiol., May 1, 2001; 22(5): 896 - 904.
[Abstract] [Full Text]

Home page
 Am. J. Roentgenol.Home page
E. H. Herskovits, R. Itoh, and E. R. Melhem
Accuracy for Detection of Simulated Lesions: Comparison of Fluid-Attenuated Inversion-Recovery, Proton Density-Weighted, and T2-Weighted Synthetic Brain MR Imaging
Am. J. Roentgenol., May 1, 2001; 176(5): 1313 - 1318.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Okuda, T.
Right arrow Articles by Takahashi, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Okuda, T.
Right arrow Articles by Takahashi, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE