Magnetic resonance imaging of focal cortical dysplasia: Comparison of 3D and 2D fluid attenuated inversion recovery sequences at 3 T

Epilepsy Research 116 (2015) 8–14

Contents lists available at www.sciencedirect.com

Epilepsy Research journal homepage: www.elsevier.com/locate/epilepsyres

Magnetic resonance imaging of focal cortical dysplasia: Comparison of 3D and 2D fluid attenuated inversion recovery sequences at 3 T Henriette J. Tschampa a,∗ , Horst Urbach b , Michael Malter c , Rainer Surges c , Susanne Greschus a , Jürgen Gieseke d a

Department of Radiology/Neuroradiology, University of Bonn, Bonn, Germany Department of Neuroradiology, University of Freiburg, Freiburg im Breisgau, Germany c Department of Epileptology, University of Bonn, Bonn, Germany d Philips Medical Systems, Hamburg, Germany b

a r t i c l e

i n f o

Article history: Received 16 April 2015 Received in revised form 25 June 2015 Accepted 5 July 2015 Available online 6 July 2015 Keywords: High field MRI 3D FLAIR Refocusing flip angle Focal cortical dysplasia Epilepsy

a b s t r a c t Purpose: Focal cortical dysplasia (FCD) is a frequent finding in drug resistant epilepsy. The aim of our study was to evaluate an isotropic high-resolution 3-dimensional Fluid-attenuated inversion recovery sequence (3D FLAIR) at 3 T in comparison to standard 2D FLAIR in the diagnosis of FCD. Materials and methods: In a prospective study, 19 epilepsy patients with the MR diagnosis of FCD were examined with a sagittal 3D FLAIR sequence with modulated refocusing flip angle (slice thickness 1.10 mm) and a 2D FLAIR in the coronal (thk. 3 mm) and axial planes (thk. 2 mm). Manually placed regions of interest were used for quantitative analysis. Qualitative image analysis was performed by two neuroradiologists in consensus. Results: Contrast between gray and white matter (p ≤ 0.02), the lesion (p ≤ 0.031) or hyperintense extension to the ventricle (p ≤ 0.021) and white matter was significantly higher in 2D than in 3D FLAIR sequences. In the visual analysis there was no difference between 2D and 3D sequences. Conclusion: Conventional 2D FLAIR sequences yield a higher image contrast compared to the employed 3D FLAIR sequence in patients with FCDs. Potential advantages of 3D imaging using surface rendering or automated techniques for lesion detection have to be further elucidated. © 2015 Elsevier B.V. All rights reserved.

Introduction Focal cortical dysplasia type II (FCD II), is a frequent cause of focal medically intractable epilepsy in surgery series (Bien et al., 2013; Fauser et al., 2009; Frater et al., 2000). FCD belongs to the complex of malformations of cortical development (Barkovich et al., 2005), characterized pathologically by the presence of dysmorphic neurons in type IIa lesions and additionally by balloon cells in type IIb (Blumcke et al., 2011; Palmini et al., 2004; Palmini and Luders, 2002). The identification of a FCD on MR imaging enables the patient to take a surgical option, which yields rate of

Abbreviations: FCD, focal cortical dysplasia; FLAIR, fluid attenuated inversion recovery; MRI, magnetic resonance imaging; FAS, flip angle sweep. ∗ Corresponding author. Tel.: +49 228 287 1 9639; fax: +49 228 287 1 5598. E-mail addresses: [email protected] (H.J. Tschampa), [email protected] (H. Urbach), [email protected] (M. Malter), [email protected] (R. Surges), [email protected] (S. Greschus), [email protected] (J. Gieseke). http://dx.doi.org/10.1016/j.eplepsyres.2015.07.004 0920-1211/© 2015 Elsevier B.V. All rights reserved.

seizure-freedom between 58 and 80% (Bien et al., 2009; Urbach et al., 2002; Wagner et al., 2011). MRI in FCD shows focal areas of increased cortical thickness, blurring of the gray/white matter junction and increased signal from the surface to the ventricle on T2-weighted, proton density weighted or FLAIR sequences (Bronen et al., 1997; Lee et al., 1998). These changes are more likely to occur in balloon cell-containing lesions (Lerner et al., 2009; Urbach et al., 2002). Despite the use of high resolution MR imaging protocols, approximately 25–35% of the FCDs are unrecognized (Krsek et al., 2008; Lerner et al., 2009). Recently, advances in MR with dedicated surface coils (Knake et al., 2005) or post-processing techniques with automated lesion detection (Focke et al., 2008; Huppertz et al., 2008; Wagner et al., 2011) have improved the detection rate of FCDs. Imaging at higher field strengths and the use of new imaging sequences such as diffusion tensor imaging (Widjaja et al., 2007) and positron emission tomography studies (Salamon et al., 2008) further increased yield. However, identifying a focal lesion and determining the exact extension of the diseased brain remains a challenge. Additionally, the imaging time for the diagnostic workup of an epilepsy patient is long, generally more than 30 min.

H.J. Tschampa et al. / Epilepsy Research 116 (2015) 8–14

While up to now, the imaging protocol in our institution includes 2D standard FLAIR sequences in 3 planes (axial, coronal, sagittal) with a total scan time of approx. 12 min., recently, a new 3D FLAIR sequence with a variable refocusing flip angle (in the following text: flip angle sweep = FAS) was available at 3 T (Busse et al., 2008; Hennig et al., 2003). The 3D FLAIR sequence has isotropic voxel, a high in-plane resolution and offers the possibility of multiplanar reformatting in a short scan time of less than 5 min. In our subsequent study, we evaluated this new 3D FLAIR sequence in the diagnostic workup of patients with FCD, regarding image quality, general image contrast and lesion detectability rate. Materials and methods Volunteers For sequence optimization we performed a pre-study with six volunteers. All volunteers had no history of neurological disease and gave informed consent. Patients In our prospective study, we included 19 epilepsy patients with the MR diagnosis of FCD as detected on routine brain imaging. All were in-patients in the Department of Epileptology and had a detailed clinical examination including video-EEG telemetry for seizure recording using scalp EEG (10–20 system) and in some cases additional intracranial electrodes. The study protocol was approved by the ethics committee of the university hospital and patients or their guardians gave written informed consent for inclusion in the study. MRI All imaging was performed at a 3 T Scanner (Achieva 3.0 T or Ingenia 3.0 T; Philips Healthcare; The Netherlands), using a transmit body coil and an 8 channel receive only SENSE head coil. For sequence optimization, the whole-brain 3D FLAIR sequence with isotropic spatial resolution and fast spin echo acquisition, based on the variable refocusing flip angle technique (sweep technique Busse et al., 2008; Chagla et al., 2008; Gieseke et al., 2004; Morakkabati-Spitz et al., 2006) was carried out first in a volunteer study. Starting from a refocusing flip angle of 180◦ , a continuous flip angle reduction was performed at the beginning of the echo train leading to a “pseudo-steady-state” like condition with a flip angle (FA) between 18◦ and 30◦ , which was maintained until the end of the echo train. Quantitative measurements were conducted using manually placed region of interests (ROIs) in 10 brain regions (white and gray matter in the frontal, temporal, parietal and occipital lobe and the cingulate gyrus, bilaterally). The mean region of interest was 4 mm2 for gray matter and 9 mm2 for white matter ROIs. ROIs were oval or round in shape, the ROIs were placed in the same brain regions in the different scans, three ROI measurements were performed per anatomical site, mean values of the ROI measurements were then used. Parallel imaging was used in the 3D sequence, therefore the standard deviation of signal intensity in air outside the object could not be used to determine noise. As the signal is not homogenous, standard deviation of the signal could not be used for the estimation of contrast-to-noise ratio (CNR) as performed e.g. by Stehling et al. (2007). Therefore, the perceived relative contrast between different tissues was calculated to compare the sequences as described by Morakkabati-Spitz et al. (2006) using the formula (mean signal A − mean signal B)/(mean signal A + mean signal B), where A represents the signal intensity of tissue A (e.g. gray matter) and B represents the signal intensity of tissue B (e. g. white matter of the corresponding lobe). Sixteen patients

9

were finally scanned at FA 22◦ . Two patients were scanned with FA 18◦ and one patient with FA 25◦ during the process of flip angle optimization. Our optimized 3D FLAIR sequence delivers the following parameters: Field of view (FOV) 250 mm, Rectangular field of view (RFOV) 100%, Scan% 100; No. of slices 327; acquisition matrix 228 × 228; TR/TE 4800/302 ms; TI 1600 ms; Echo train length 203; variable refocusing angle with 22◦ ; spatial resolution, 1.09 × 1.10 × 1.10 mm; reconstructed resolution 0.53 × 0.53 × 0.55 mm; TSE-Factor: 168; SENSE factor 2.5 in phase encoding direction; fat suppression with SPAIR to reduce infolding and ghosting artefacts from the fat signal of the scalp when high SENSE factors are used (Lutterbey et al., 2007); two signal averages; acquisition time 4:43 min. The parameters for the standard coronal 2D FLAIR imaging were the following: FOV 230 mm; RFOV 80%; Scan% 70.6; No. of slices 40; TR/TE 12000/140 ms; TI 2850 ms; spatial resolution 0.90 × 1.02 × 3 mm; reconstructed resolution 0.48 × 0.48 × 3 mm; TSE Factor 36; SENSE factor 1; 1 signal average; acquisition time 4:48 min. Standard two dimensional axial FLAIR imaging had the following parameters: FOV 256 mm; RFOV 80%; acquisition matrix 256 × 200; TR/TE 12000/140 ms; TI 2850 ms, acquired spatial resolution 1.00 × 1.02 × 2 mm; reconstructed resolution 0.5 × 0.5 × 2 mm; TSE-Factor: 40; SENSE factor 1; 1 signal average; acquisition time 3:36 min. The 3D FLAIR sequence was acquired in the sagittal plane. For comparison with the 2D sequences, we reformatted the 3D sequence in the axial and coronal planes in slices with 2 mm and 3 mm thickness matching the angulations and slice thicknesses of the 2D sequences. To quantitatively compare the 2D and 3D sequences, we used the perceived relative contrast between different tissues as described in the pre-study (Morakkabati-Spitz et al., 2006). Manually drawn ROIs were placed in similar locations in the 2D and the reformatted 3D sequences: in the gray and white matter, the lesion and the hyperintense extension to the ventricle. Three ROI measurements were performed in each region and mean values were calculated. Subsequently, the following contrasts were calculated: contrast between gray and white matter, contrast between the lesion and the white matter, and the contrast between the hyperintense extension to the ventricle and the white matter using the above mentioned formula. Statistical analysis was performed using SPSS Statistics 21. Student’s t-test for paired samples was used to compare the coronal and the axial scans in 2D versus 3D according to the calculated contrasts. p-Values <0.05 were considered significant. Qualitative image analysis was performed by two experienced neuroradiologists in consensus. The following characteristics of the lesions were evaluated on a semi-quantitative scale ranging from 0 = no; 1 = slight; 2 = moderate to 3 = severe: (A) focal cortical thickening; (B) blurring of the gray and white matter interface; (C) increased subcortical signal; (D) increased signal from the surface to the ventricle. Statistical significance (p < 0.05) was tested with the nonparametric paired sample Wilcoxon signed-rank test in case of visual scoring results. Additionally, the size of the lesion was recorded.

Results Patients Patients had a history of seizures in the median time for 8.7 years (range: 2 months to 68 years. Median age was 18.7 years (range 21 months to 74 years). Ten patients were female, nine male. Clinical characteristics of the patients are summarized in Table 1.

10

H.J. Tschampa et al. / Epilepsy Research 116 (2015) 8–14

Table 1 Patients characteristics including clinical data, MRI, histology, and clinical outcome. No.

Age (years)/gender

Duration of seizure disorder (years)

Seizure type1

Location of FCD2

Histology

Outcome

1

17/m

14

SGTCS

FL

Seizure-free under medication

2

6/m

1

FR

3

11/f

3

FR

FCD IIa

Seizure-free

4

40/f

22

3 SPS/week, 1 CPS/week SPS, CPS, variable with several months of complete seizure-freedom 3 SPS/week, 6 CPS/day; 6 GM/year

No op.3 (seizure-free under medication) FCD IIb

PL

FCD IIb

5

2/m

0,2

1–2 CPS/day

No op. (TSC with multiple FCDs)

6 7

9/f 31/f

8 17

1 SPS/month 1 CPS/day

TSC; multiple FCDs, Analysis for lesion T–O R PL OL

One seizure in 12 months not taking medication Unknown

8 9

21/m 17/m

7 1,6

10

51/f

14

3 CPS/day Variable from several/month to several months seizure-freedom 6 CPS/day

11

45/m

21

3 CPS/week

12

5/f

4

13

33/m

28

14

60/f

52

15

19/f

9

16

32/f

15

17

7/m

2

Multiple SPS and CPS/night, sGTCS 5 CPS/week, 1 GM/year 2 SPS/week, 2 CPS/week, 2 GM/year 2 CPS/day, 1 GM/year 5 CPS/day, 1 GM/year 30 SPS/day

18

13/f

1

8 SPS/day

FR

19

74/m

68

1–3 SPS/week

FL

1 2 3

Seizure-free

FCD IIb No op. (patient refused) FCD IIa No op. (patient refused)

Unknown Unknown

No op. (three FCDs)

Unknown

Unknown

FL

No op. (lesion in Broca area) FCD IIb

Seizure-free

T–O R

FCD IIb

Seizure-free

FR

No op. (patient refused)

Unknown

PR

FCD IIa

Seizure-free

FL

FCD IIb

Unknown

FL

No op. (patient refused) No op. (patient refused) No op. (patient refused)

Unknown

TL PR

multiple FCDs; analysis for lesion P L FL

Seizure-free Unknown

Unknown Unknown

SPS = simple partial seizure; CPS = complex partial seizure; GM = grand mal; sGTCS = secondary generalized tonic–clonic seizure. F = frontal, P = parietal; O = occipital; T = temporal; T–O = temporo–occipital; L = left; R = right. op. = operation.

Nine patients underwent epilepsy surgery: histology showed FCD type IIa in three patients (Fig. 2) and FCD type IIb (with balloon cells) in six patients, according to the classification of Palmini and Lüders (Palmini et al., 2004; Palmini and Lüders, 2002).

MRI The FCDs were located in the frontal lobe (10 patients), the temporal lobe (1 patient), the temporo–occipital cortex (2 patients), the parietal lobe (5 patients) and the occipital lobe (1 patient). In 10 patients a high signal band to the ventricle could be identified on coronal and axial scans both. Two patients (patients no. 5 and 10) had multiple FCDs; in patient no. 5 tuberous sclerosis complex was genetically confirmed, while patient 10 had no cutanous stigmata of tuberous sclerosis. All measurements were performed in both patients only in one lesion, which had typical MR aspect of FCD.

None of the patients had hippocampal sclerosis or other lesions typically causing epilepsy. Quantitative image analysis A hyperintense band to the ventricle thick enough to allow ROI measurements was identified in 10 patients on the axial and coronal scans, both. The other measurements were performed in all 19 patients. Comparing 2D versus 3D sequences, we found significantly higher image contrast in 2D sequences: between gray and white matter (p < 0.02), between the lesion and the white matter (p < 0.031), and between the hyperintense extension to the ventricle and the white matter (p < 0.021), (Table 3). Qualitative image analysis In visual rating there was no significant difference between the axial or coronal 2D versus 3D sequences (p > 0.08), neither was

H.J. Tschampa et al. / Epilepsy Research 116 (2015) 8–14

11

Table 2 Qualitative image analysis. Patient

Lesion size1

Sequence

Focal cortical thickening

Blurring of the gray-white matter junction

Marked hyperintensity of the subcortical white matter

Hyperintense extension to the ventricle

1

Medium

2

Medium

3

Small

4

Medium

5

Small

6

Large

7

Small

8

Small

9

Small

10

Large

11

Small

12

Large

13

Small

14

Small

15

Large

16

Small

17

Medium

18

Small

19

Small

2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D 2D 3D

12 1 1 1 1 1 0 0 0 0 1 1 1 1 1 0 1 1 1 0 1 0 1 1 2 2 2 2 2 2 1 1 2 2 1 1 0 0

02 0 0 0 0 0 0 0 0 0 1 1 1 1 2 1 1 2 1 0 0 0 1 1 1 1 2 1 1 2 1 2 2 2 1 1 0 0

22 3 2 3 0 0 2 1 2 1 1 1 2 2 3 3 1 1 3 3 3 3 3 3 0 0 2 3 0 0 0 0 0 0 1 1 2 2

02 0 0 0 0 0 0 0 2 2 1 1 0 0 3 3 0 0 3 3 3 3 3 3 0 0 2 3 0 0 0 0 13 13 13 13 2 2

1 2 3

Lesion size is classified as: small (1 sulcus or 1 gyrus ± adjacent sulci), medium (2 adjacent gyri), large (> = 3 adjacent gyri). 0 = no abnormality; 1 = mild; 2 = moderate; 3 = severe. Hyperintense extension to the ventricle in this patient better discernable on coronal images.

the evaluated size of the lesion influenced by imaging technique (Table 2, Figs. 1 and 2).

Discussion Focal cortical dysplasia is rare disease, which is increasingly detected as a cause of chronic medically intractable epilepsy (Bien et al., 2013). FLAIR imaging has proven to be advantageous in the diagnostic workup of epilepsy patients (Focke et al., 2008, 2009; Knake et al., 2005; Saini et al., 2010; Urbach et al., 2002). Still, using 2D sequences, the necessity to perform scans in three orthogonal planes, affords a long scan time (approx. 12 min. in our institution). We present a new isotropic high resolution 3D FLAIR sequence at 3 T with a scan time of less than five min. using the FAS technique (Busse et al., 2008; Chagla et al., 2008; Gieseke et al., 2004; Hennig et al., 2003; Hennig and Scheffler, 2001), and tested whether this 3D FLAIR sequence would be able to replace the 2D FLAIR sequences and thus save scan time. Overall, the conventional visual image analysis did not reveal significant differences between the 3D and 2D sequences, including the detectability, general image contrast and lesion size (Table 2, Figs. 1 and 2). However, ROI-based image analysis showed significantly higher contrasts in the 2D FLAIR sequences between the gray and the white matter, the lesion and the white matter and the hyperintense signal extending to the ventricle in relation to

the white matter (Table 3). The results were similar concerning the employed axial and coronal 2D sequences. The 3D FLAIR was intended to match image contrast of standard 2D sequence with the preconditions of a short scan time (less than 5 min.) and a high and isotropic spatial resolution of about 1 mm. Differences in image contrast between the employed 2D and 3D sequences can be explained by the design of the 3D sequence, using the FAS technique. In the FAS technique the “effective TE” (traditionally defined as the time at which the center of k-space is sampled) no longer corresponds to the expected amount of spin-echo T2 contrast due to the mixing of stimulated and spin echoes (Busse et al., 2008). Therefore, the echo time in the 3D technique is considerably longer. Specific absorption rate (SAR) limits a scan time reduction. To drop down the scan time and SAR of the 3D sequence, parallel imaging (SENSE) (Lutterbey et al., 2007) is applied. SENSE decreases the number of refocusing pulses and thus reduces SAR. This permits a shorter effective echo time by narrowing the intervals between the refocusing pulses. Using parallel imaging alone in the 3D technique, the scan time cannot be shortened to the desired precondition. Therefore, FAS is additionally employed. With this technique, starting from a refocusing flip angle of 180◦ , an exponential flip angle reduction is performed leading to a “pseudo-steady-state” like condition, which is maintained until the end of the echo train. This leads to a drastic SAR reduction, which minimizes echo spacing of the refocusing pulses and thus cuts down scan time. With the combination of SENSE and FAS,

12

H.J. Tschampa et al. / Epilepsy Research 116 (2015) 8–14

Fig. 1. 17y/o male patient (patient no. 1) with a left frontal FCD: 2D FLAIR (A and C) in axial (A) and coronal (C) plane and corresponding reformats of the 3D FLAIR ((B) in axial and (D) in the coronal plane) show focal cortical thickening and marked subcortical hyperintensity (arrow).

both preconditions were achieved, yielding a 2D-like image contrast on visual inspection. Also the inversion time delay could have an additional effect on the contrast although the cerebral spinal fluid is well suppressed. Saini et al. (2010) reported on a higher detection rate of cortical dysplasias in dedicated MR exams including a 3D FLAIR sequence at 1.5 T in comparison to their standard MRI. One cause of this difference could be the slice thickness, as in the work of Saini and colleagues, the 2D FLAIR sequences had a slice thickness of 5 mm, while the 3D sequence had a thickness of 0.9 mm. With thinner

slices partial volume effect is reduced, therefore subtle lesions were possibly better detected with the 3D sequence in their study. In our institution however, the standard sequences are thin-sliced (2 mm axial and 3 mm coronal scans), so that there was no advantage of 3D imaging. Two other possible explanations for a higher contrast between tissues at 1.5 T as reported by Saini and colleagues, but not at 3 T are: first, relaxation times are field strength dependent (Wansapura et al., 1999), which can lead to different T1- and T2tissue contrasts at 1.5 T as compared to 3 T. Second, the design of the FAS technique used in our study depends on the desired scan

Table 3 Quantitative image analysis. Comparison

Sequences

N

Mean contrast (contrast A − contrast B)/(contrast A + contrast B)

Standard deviation

Gray matter vs. white matter

Axial 2D Axial 3D Coronal 2D Coronal 3D Axial 2D Axial 3D Coronal 2D Coronal 3D Axial 2D Axial 3D Coronal 2D Coronal 3D

19 19 19 19 19 19 19 19 10 10 10 10

0.1551 0.1266 0.1359 0.1137 0.2182 0.1859 0.1855 0.1590 0.2223 0.1767 0.1754 0.1413

0.03874 0.04134 0.04272 0.04894 0.07759 0.09064 0.06138 0.07706 0.10016 0.09291 0.08977 0.08480

Lesion vs. white matter

Hyperintense extension to the ventricle vs. white matter

1

p < 0.05 was considered significant and marked with asterisk.

p1 (Students’ paired t-test) 0.006* 0.020* 0.026* 0.031* 0.021* 0.015*

H.J. Tschampa et al. / Epilepsy Research 116 (2015) 8–14

13

Fig. 2. 19 y/o female patient (patient no. 15) with pharmaco-resistant epilepsy and a pathologically confirmed right parietal FCD IIa. The 2D FLAIR in the axial (A) and coronal (C) planes as well as the reformatted 3D FLAIR in the axial (B) and coronal (D) planes show focal cortical thickening and a subtle subcortical hyperintensity as marked by arrows.

time, relaxation times and spatial resolution and thus modulates tissue contrast (Busse et al., 2008; Hennig et al., 2003; Hennig and Scheffler, 2001). Our results are in accordance with Wieshmann et al. (1998) who reported a higher contrast between gray and white matter, higher contrast for the myelinated tracts and a higher rate of conspicuity for the detection of hippocampal sclerosis in 2D FLAIR as compared to 3D FLAIR at 1.5 T. Apart from epilepsy studies (Wieshmann et al., 1998), the 3D FLAIR sequence at 3 T is mainly used for the anatomy and pathology of the inner ear (Naganawa et al., 2004; Sugiura et al., 2006) and the subarachnoid space (Mills et al., 2007), as with the excellent water suppression in 3D FLAIR, cerebrospinal fluid does not cause artefacts. Other groups reported 3D FLAIR imaging at 3 T in multiple sclerosis patients (Bink et al., 2006) and various brain pathologies (Kakeda et al., 2012). We speculate that a potential of the 3D FLAIR sequence would be the use in automated techniques for the detection of FCDs in voxelbased analyses, where 2D-FLAIR sequences have been proven to be helpful (Focke et al., 2008, 2009; Wagner et al., 2011). An additional advantage of 3D imaging is the possibility to visualize the cortex using 3D surface rendering techniques. These may aid to determine the exact anatomic extent of a lesion e. g. in a pre-surgical setting.

Conclusion The employed 3D FLAIR with the FAS technique yields lower image contrasts as compared to 2D FLAIR sequences in the depiction of FCDs. Thus, the 3D imaging might not replace conventional 2D FLAIR scans yet. However, 3D has two major advantages: (1) better slice profile definition and smaller slice thickness for 3D and therefore less partial voluming and (2) more accurate geometry depiction and isotropic voxel size. We currently use the 3D sequence in clinical routine as a third (sagittal) plane and for the determination of the exact anatomical extent of the lesion. Potential advantages of 3D imaging such as surface rendering or the use in automated techniques for lesion detection have to be further elucidated. References Barkovich, A.J., Kuzniecky, R.I., Jackson, G.D., Guerrini, R., Dobyns, W.B., 2005. A developmental and genetic classification for malformations of cortical development. Neurology 65, 1873–1887. Bien, C.G., Raabe, A.L., Schramm, J., Becker, A., Urbach, H., Elger, C.E., 2013. Trends in presurgical evaluation and surgical treatment of epilepsy at one centre from 1988–2009. J. Neurol. Neurosurg. Psychiatry 84, 54–61. Bien, C.G., Szinay, M., Wagner, J., Clusmann, H., Becker, A.J., Urbach, H., 2009. Characteristics and surgical outcomes of patients with refractory magnetic resonance imaging-negative epilepsies. Arch. Neurol. 66, 1491–1499.

14

H.J. Tschampa et al. / Epilepsy Research 116 (2015) 8–14

Bink, A., Schmitt, M., Gaa, J., Mugler III, J.P., Lanfermann, H., Zanella, F.E., 2006. Detection of lesions in multiple sclerosis by 2D FLAIR and single-slab 3D FLAIR sequences at 3.0 T: initial results. Eur. Radiol. 16, 1104–1110. Blumcke, I., Thom, M., Aronica, E., Armstrong, D.D., Vinters, H.V., Palmini, A., Jacques, T.S., Avanzini, G., Barkovich, A.J., Battaglia, G., Becker, A., Cepeda, C., Cendes, F., Colombo, N., Crino, P., Cross, J.H., Delalande, O., Dubeau, F., Duncan, J., Guerrini, R., Kahane, P., Mathern, G., Najm, I., Ozkara, C., Raybaud, C., Represa, A., Roper, S.N., Salamon, N., Schulze-Bonhage, A., Tassi, L., Vezzani, A., Spreafico, R., 2011. The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 52, 158–174. Bronen, R.A., Vives, K.P., Kim, J.H., Fulbright, R.K., Spencer, S.S., Spencer, D.D., 1997. Focal cortical dysplasia of Taylor, balloon cell subtype: MR differentiation from low-grade tumors. AJNR Am. J. Neuroradiol. 18, 1141–1151. Busse, R.F., Brau, A.C., Vu, A., Michelich, C.R., Bayram, E., Kijowski, R., Reeder, S.B., Rowley, H.A., 2008. Effects of refocusing flip angle modulation and view ordering in 3D fast spin echo. Magn. Reson. Med. 60, 640–649. Chagla, G.H., Busse, R.F., Sydnor, R., Rowley, H.A., Turski, P.A., 2008. Three-dimensional fluid attenuated inversion recovery imaging with isotropic resolution and nonselective adiabatic inversion provides improved three-dimensional visualization and cerebrospinal fluid suppression compared to two-dimensional flair at 3 tesla. Invest. Radiol. 43, 547–551. Fauser, S., Sisodiya, S.M., Martinian, L., Thom, M., Gumbinger, C., Huppertz, H.J., Hader, C., Strobl, K., Steinhoff, B.J., Prinz, M., Zentner, J., Schulze-Bonhage, A., 2009. Multi-focal occurrence of cortical dysplasia in epilepsy patients. Brain 132, 2079–2090. Focke, N.K., Bonelli, S.B., Yogarajah, M., Scott, C., Symms, M.R., Duncan, J.S., 2009. Automated normalized FLAIR imaging in MRI-negative patients with refractory focal epilepsy. Epilepsia 50, 1484–1490. Focke, N.K., Symms, M.R., Burdett, J.L., Duncan, J.S., 2008. Voxel-based analysis of whole brain FLAIR at 3T detects focal cortical dysplasia. Epilepsia 49, 786–793. Frater, J.L., Prayson, R.A., Morris III, H.H., Bingaman, W.E., 2000. Surgical pathologic findings of extratemporal-based intractable epilepsy: a study of 133 consecutive resections. Arch. Pathol. Lab. Med. 124, 545–549. Gieseke, J., Wattjes, M.P., Lutterbey, G., von Falkenhausen, M., Blömer, R., Gür, O., Manka, C., Schild, H., Kuhl, C., 2004. Ultra fast T2-weighted TSE sequences using flip angle sweep with half-Fourier and SENSE at 3 T. Neuroradiology 46 (Suppl. 1), 122 (Ref Type: Abstract). Hennig, J., Scheffler, K., 2001. Hyperechoes. Magn. Reson. Med. 46, 6–12. Hennig, J., Weigel, M., Scheffler, K., 2003. Multiecho sequences with variable refocusing flip angles: optimization of signal behavior using smooth transitions between pseudo steady states (TRAPS). Magn. Reson. Med. 49, 527–535. Huppertz, H.J., Wellmer, J., Staack, A.M., Altenmuller, D.M., Urbach, H., Kroll, J., 2008. Voxel-based 3D MRI analysis helps to detect subtle forms of subcortical band heterotopia. Epilepsia 49, 772–785. Kakeda, S., Korogi, Y., Hiai, Y., Ohnari, N., Sato, T., Hirai, T., 2012. Pitfalls of 3D FLAIR brain imaging: a prospective comparison with 2D FLAIR. Acad. Radiol. 19, 1225–1232. Knake, S., Triantafyllou, C., Wald, L.L., Wiggins, G., Kirk, G.P., Larsson, P.G., Stufflebeam, S.M., Foley, M.T., Shiraishi, H., Dale, A.M., Halgren, E., Grant, P.E., 2005. 3T phased array MRI improves the presurgical evaluation in focal epilepsies: a prospective study. Neurology 65, 1026–1031. Krsek, P., Maton, B., Korman, B., Pacheco-Jacome, E., Jayakar, P., Dunoyer, C., Rey, G., Morrison, G., Ragheb, J., Vinters, H.V., Resnick, T., Duchowny, M., 2008. Different features of histopathological subtypes of pediatric focal cortical dysplasia. Ann. Neurol. 63, 758–769. Lee, B.C., Schmidt, R.E., Hatfield, G.A., Bourgeois, B., Park, T.S., 1998. MRI of focal cortical dysplasia. Neuroradiology 40, 675–683.

Lerner, J.T., Salamon, N., Hauptman, J.S., Velasco, T.R., Hemb, M., Wu, J.Y., Sankar, R., Donald, S.W., Engel Jr., J., Fried, I., Cepeda, C., Andre, V.M., Levine, M.S., Miyata, H., Yong, W.H., Vinters, H.V., Mathern, G.W., 2009. Assessment and surgical outcomes for mild type I and severe type II cortical dysplasia: a critical review and the UCLA experience. Epilepsia 50, 1310–1335. Lutterbey, G., Wattjes, M.P., Kandyba, J., Harzheim, M., Falkenhausen, M.V., Morakkabati, N., Schild, H., Gieseke, J., 2007. Clinical evaluation of a speed optimized T2 weighted fast spin echo sequence at 3.0 T using variable flip angle refocusing, half-Fourier acquisition and parallel imaging. Br. J. Radiol. 80, 668–673. Mills, R.J., Young, C.A., Smith, E.T., 2007. 3D MRI in multiple sclerosis: a study of three sequences at 3 T. Br. J. Radiol. 80, 307–320. Morakkabati-Spitz, N., Gieseke, J., Kuhl, C., Lutterbey, G., von Falkenhausen, M., Traber, F., Park-Simon, T.W., Zivanovic, O., Schild, H.H., 2006. MRI of the pelvis at 3 T: very high spatial resolution with sensitivity encoding and flip-angle sweep technique in clinically acceptable scan time. Eur. Radiol. 16, 634–641. Naganawa, S., Koshikawa, T., Nakamura, T., Kawai, H., Fukatsu, H., Ishigaki, T., Komada, T., Maruyama, K., Takizawa, O., 2004. Comparison of flow artifacts between 2D-FLAIR and 3D-FLAIR sequences at 3 T. Eur. Radiol. 14, 1901–1908. Palmini, A., Lüders, H.O., 2002. Classification issues in malformations caused by abnormalities of cortical development. Neurosurg. Clin. N. Am. 13, 1–16 (vii). Palmini, A., Najm, I., Avanzini, G., Babb, T., Guerrini, R., Foldvary-Schaefer, N., Jackson, G., Luders, H.O., Prayson, R., Spreafico, R., Vinters, H.V., 2004. Terminology and classification of the cortical dysplasias. Neurology 62, S2–S8. Saini, J., Singh, A., Kesavadas, C., Thomas, B., Rathore, C., Bahuleyan, B., Radhakrishnan, A., Radhakrishnan, K., 2010. Role of three-dimensional fluid-attenuated inversion recovery (3D FLAIR) and proton density magnetic resonance imaging for the detection and evaluation of lesion extent of focal cortical dysplasia in patients with refractory epilepsy. Acta Radiol. 51, 218–225. Salamon, N., Kung, J., Shaw, S.J., Koo, J., Koh, S., Wu, J.Y., Lerner, J.T., Sankar, R., Shields, W.D., Engel Jr., J., Fried, I., Miyata, H., Yong, W.H., Vinters, H.V., Mathern, G.W., 2008. FDG-PET/MRI coregistration improves detection of cortical dysplasia in patients with epilepsy. Neurology 71, 1594–1601. Stehling, C., Vieth, V., Bachmann, R., Nassenstein, I., Kugel, H., Kooijman, H., Heindel, W., Fischbach, R., 2007. High-resolution magnetic resonance imaging of the temporomandibular joint: image quality at 1.5 and 3.0 Tesla in volunteers. Invest. Radiol. 42, 428–434. Sugiura, M., Naganawa, S., Teranishi, M., Nakashima, T., 2006. Three-dimensional fluid-attenuated inversion recovery magnetic resonance imaging findings in patients with sudden sensorineural hearing loss. Laryngoscope 116, 1451–1454. Urbach, H., Scheffler, B., Heinrichsmeier, T., von, O.J., Kral, T., Wellmer, J., Schramm, J., Wiestler, O.D., Blumcke, I., 2002. Focal cortical dysplasia of Taylor’s balloon cell type: a clinicopathological entity with characteristic neuroimaging and histopathological features, and favorable postsurgical outcome. Epilepsia 43, 33–40. Wagner, J., Weber, B., Urbach, H., Elger, C.E., Huppertz, H.J., 2011. Morphometric MRI analysis improves detection of focal cortical dysplasia type II. Brain 134, 2844–2854. Wansapura, J.P., Holland, S.K., Dunn, R.S., Ball Jr., W.S., 1999. NMR relaxation times in the human brain at 3.0 tesla. J. Magn. Reson. Imaging 9, 531–538. Widjaja, E., Blaser, S., Miller, E., Kassner, A., Shannon, P., Chuang, S.H., Snead III, O.C., Raybaud, C.R., 2007. Evaluation of subcortical white matter and deep white matter tracts in malformations of cortical development. Epilepsia 48, 1460–1469. Wieshmann, U.C., Barker, G.J., Symms, M.R., Bartlett, P.A., Stevens, J.M., Shorvon, S.D., 1998. Fast fluid-attenuated inversion-recovery imaging: first experience with a 3D version in epilepsy. Neuroradiology 40, 483–489.