INTRODUCTION

Temporal lobe epilepsy (TLE) is one of the most common adult forms of epilepsy. Despite medical therapy, seizures can persist in <35% of patients with partial epilepsy¹. The diagnosis of TLE is primarily clinical; it depends on obtaining a detailed history both from the patient and from an eyewitness. Ictal and interictal EEG is usually performed to confirm the clinical diagnosis of temporal lobe epilepsy but it may be misleading in localization of the primary site of seizure onset, because it has inherently low spatial resolution and it’s dependent on primarily cortical surface effects².

Diffusion Weighted imaging (DWI) is a modality of magnetic resonance imaging (MRI) that allows the evaluation of differences in the extracellular space through variations of molecular water mobility at a microscopic level and has been used to identify pathologic changes in human TLE³.

The aim of this study is to evaluate the possible role of interictal DWI in lateralizing mesial temporal lobe epilepsy in patients with normal conventional MR scans and in understanding the underlying pathology.

SUBJECTS AND METHODS

This study is a case-control study that included thirty right-handed Egyptian epileptic patients (8 males and 22 females) with an age range of 11-45 years and a mean of 27.3±9.7 years. Patients were recruited from the epilepsy and neurology outpatient clinic of Kasr El-Aini hospital in the period between June 2008 to march 2009. The control group included 10 right handed healthy age matched subjects (5 males and 5 females) with age range of 15-39 years and mean of 26±7.6 years.

Inclusion criteria:

All patients were diagnosed to have TLE as defined in 1989 by the International League against Epilepsy (ILAE)⁴ as a condition characterized by recurrent unprovoked seizures originating from the mesial temporal lobe.

Exclusion criteria:

·          Mesial temporal lobe sclerosis evident on conventional MR imaging.

·          Other types of epilepsy (e.g. frontal lobe epilepsy, myoclonic).

·          Abnormal MRI findings (e.g.: tumor, post infection, vascular malformations).

Methods:

1.        Clinical evaluation: Detailed history of epilepsy and clinical examination according to epilepsy sheet, Department of Neurology, Cairo university hospitals.

2.        Electroencephalography (EEG): An awake interictal EEG under standard conditions was done for all patients at the neurophysiology unit of Kasr El-Aini hospital, using the Nikon Kohden 14 channel EEG machine. Electrodes were placed according to the 10-20 international system of electrode placement using mono and bipolar montages. Hyperventilation for 3 minutes and photic stimulation were done for all patients to provoke any existing abnormality. The EEG tracing were analyzed carefully as regards frequency, amplitude and the background activity, as well as the presence of any abnormalities. The abnormalities were described as focal, generalized or focal with secondary generalization. EEG was positive in 15 patients (50%).

3.        Neuropsychological testing: Rey-Osterrieth complex figure test (RCFT) (5) was performed for all TLE patients. The purpose of the test was to compare figural and spatial memory in patients with left and right temporal lobe epilepsy to examine the association between figural and spatial component of RCFT, temporal lobe laterality, and hippocampal ADC values. The figure is broken down into 18 scorable elements: 0.5-2.0 points are awarded for each element depending on accuracy, distortion, and location of its reproduction, so the highest possible score on each figure is 36. Patients were classified according to Rey-Osterrieth complex figure test seizure lateralization into: right TLE (n=14) and left TLE (n=16) patients.

4.                Magnetic resonance imaging (MRI):

a.      Conventional MRI: Non-enhanced MRI of the brain was performed for all patients included in this study. The MRI scans were performed at the Neuroradiology unit, Department of Radiologym Kasr El-Aini University hospitals using a 1.5T Phillips Intera® scanner. Brain MR Images were obtained in both the control subjects and patients. Imaging protocol consisted of : Sagittal 5 mm-thick T1-weighted images (TR470/TE12), Coronal 1.5 mm (no intersection gap) 3D T1- weighted gradient echo images (TR1800/TE4.4, flip angle 8º) through the entire brain, Coronal 2 mm-thick fluid attenuated inversion recovery, (FLAIR) images (TR8600/TE108/TI2400), Coronal 3 mm-thick T2-weighted images (TR6860/TE125) angled perpendicular to the long axis of the hippocampi.  

b.      Diffusion weighted imaging: DWI images were obtained angled along the axis of the hippocampi, with b values of 500 and 1000 s/mm² and ADC maps were generated. On the axial ADC maps each hippocampus was divided into arbitrary four segments and manually placed circular regions of interest (ROIs) of similar size (35-55) pixels corresponding to the head (segment A), body (segments B and C), and tail (segment D) of the hippocampus avoiding the adjacent CSF-containing spaces (Figure 1). Mean and standard deviation (SD) were obtained in each ROI (for both left and right hippocampi). An asymmetry index (AI) was used to ascertain the degree of asymmetry between the values obtained in the right and the left hippocampi.

Figure 1. ADC DWI MRI image of the brain of a female with TLE showing placement of ROIs in both hippocampi (A=head, B= body 1, C= body 2 and D= tail).

Statistical Analysis

Data was summarized using mean and standard deviation (SD) for quantitative variables and number and percent for qualitative values. Comparisons between groups were done using chi- square test, Fisher's exact test for qualitative values while independent sample t-test (student t-test) was used for normally distributive qualitative variables, while non-parametrical Mann-Whitney test and Wilcoxon signed ranks test were used for quantitative variables which are not normally distributive. Analysis of Variance (ANOVA) was used to compare variables among more than two groups. Correlations were done to test for linear relations between qualitative and ordinal variables. A p-value < 0.05 were considered as statistically significant. All calculations were performed using an IBM–compatible personal computer using the SPSS version 15, STATISTICA and EXCEL programs of analysis.

RESULTS

The clinical characteristics of TLE patients included in the study are shown in Table (1). Comparison of ipsilateral ADC (b500) of patients with right TLE (n=14) to controls showed statistically higher ADC in head (p=0.03), body 1 (p=0.04), body 2 (p=0.02), tail (p=0.02), mean (p=0.03) and AI (0.005). Comparison of ipsilateral ADC (b500) of patients with left TLE (n=16) to controls showed statistically higher ADC in head (p=0.03), body 1 (p=0.04), body 2 (p=0.009), tail (p=0.01) mean (p=0.01) and AI (p=0.009).

Comparison of ipsilateral ADC (b1000) of patients with right TLE (n=14) to controls showed only a significantly higher AI (p=0.005) but no significant difference in ADC in all hippocampal regions and mean (p>0.05 respectively). Comparison of ipsilateral ADC (b1000) of patients with left TLE to controls showed only a significantly higher AI (p=0.002) but no significant difference in ADC in all hippocampal regions and mean (p>0.05 respectively).

Comparison of ipsilateral to contralateral ADC (b500) values in patients with right TLE showed significantly higher ipsilateral values in head (p=0.037), body 2 (p=0.009), tail (p=0.005) and mean (p=0.022) but not in body 1 (p>0.05). Comparison of ipsilateral to contralateral ADC (b1000) values in patients with right TLE showed significantly higher ipsilateral values in head (p=0.028), body 2 (p=0.013) and mean (p=0.005) but not in body1 or tail (p>0.05 respectively) (Table 2).

Comparison of ipsilateral to contralateral ADC (b1000) values in patients with left TLE showed significantly higher ipsilateral values in body 2 (p=0.036) and tail (p=0.027) but not in head, body1 nor mean (p>0.05 respectively). Comparison of ipsilateral to contralateral ADC (b1000) values in patients with left TLE showed significantly higher ipsilateral values in body1 (p=0.007), body 2 (0.004), tail (p=0.001) and mean (p=0.031) but not in head (p>0.05) (Table 2).

We found that the maximum differences in the ADC values were detected in the body 2 part of the hippocampus (p=0.018).

The delayed memory phase of RCFT was significantly associated with higher ADC values in the left hippocampus (0.000) while impairment of copy phase had significantly higher ADCs values in the right hippocampus (0.001). Patients with higher ADC in the left hippocampus had more severe degree of impairment of RCFT (p=0.017).

There were no significant differences in ADC values between male and female TLE patients (P>0.05).

A statistically significant negative correlation was found between the mean difference of ADCs values of both hippocampi and the total score of RCFT (p=0.001, r=-0.594) (Figure 2). A statistically significant positive correlation was found between the mean difference of ADCs values of both sides and the degree of impairment in RCFT results (p=0.001, R=0.629).

Compared to RCFT laterality and the clinical seizure semiology, true positive test results for diffusion MRI in both b500 and b1000 and EEG were: Thirty patients in diffusion MRI with b value 500 (100%). Twenty one patients in diffusion MRI with b value 1000 (70%) and Ten patients in EEG (33.3%).

Table1. Characteristics of study population.

SD standard deviation, TLE temporal lobe epilepsy

Table 2. Comparison between ipsilateral and contralateral hippocampal ADCs indices in both left and right TLE patients.

ADC apparent diffusion coefficient, LT-TLE left temporal lobe epilepsy, RT-TLE right temporal lobe epilepsy

* Significant at p<0.05 ** Significant at p<0.01.

Figure 2. Correlation between the mean difference of ADCs values

of both hippocampi and the total score of RCFT.

DISCUSSION

The main purpose of this study was to examine the lateralizing ability of the interictal ADC data for the hippocampus in mesial TLE patients with normal conventional MR scans. We hypothesized that DWI analyses performed interictally in TLE, reflects the characteristic structural abnormalities found in the hippocampus and therefore could be used as a noninvasive mean of diagnosis in the assessment of TLE patients.

Our results showed that the ipsilateral hippocampus had significantly abnormal ADC b500 values compared to the contralateral side and compared to that of the controls.

Similarly, Hugg et al.⁶ found that ADC values ipsilateral to the epileptic focus were elevated with obvious significant difference from the controls. Kantarci et al.⁷ found that among 40 patients with complex partial seizures 32 patients (80%) had elevated ADC values in the ipsilateral side of the focus. Similar results were shown by Londono et al.⁸. Lui et al.⁹ rather showed a bilateral increase in hippocampal ADC values of 20 presurgical TLE patients compared to the control group and hypothesized  that these findings may reflect the underlying microstructural abnormalities including neuronal loss, gliosis and increase in the amount of intracellular fluid. Wang et al.¹⁰ explained increase the hippocampal diffusivity by a decrease in the hippocampal neuronal density, gliosis or loss of neurons results in a relative increase of the extracellular compartment, leading to increased diffusion of water in sclerotic hippocampi, resulting in elevation of ADC values.

We found that the maximum differences in the ADC values were detected in the body 2 part of the hippocampus. This is confirmed by the results of Kim et al.¹¹, who showed that segmental MR measurements of the body of the hippocampus are as accurate as measurements of the whole hippocampus for lateralizing temporal lobe epilepsy before surgery. Also Bronen et al.¹² showed that the most frequently affected region was the hippocampal body and the lateralizing sensitivity was: hippocampal head in 29 patients (51%), hippocampal body in 50 patients (88%), and hippocampal tail in 35 patients (61%). On the other hand Hugg et al.⁶ found that the ADC elevation was apparent throughout the hippocampal body.

These results may suggest that the pathological changes that occur in the hippocampus in TLE patients either starts in the body or occurs mainly in this part. Segmental measurements are less time consuming and require less experience to perform, they may be considered the procedure of choice.

Our findings demonstrate that increased apparent diffusion coefficient measurements in the hippocampal structures correlate well with impaired total score of RCFT. Lui et al.⁹ found a significant correlation between the ADC measurements in the left hippocampus of TLE patients and RCFT score but found only a trend towards statistical significance in the correlation between right hippocampal ADC and RCFT scores.

Furthermore we found that patients with greater diffusion abnormalities in the left hippocampus have decreased delayed memory function and patients with greater diffusion abnormalities in the right hippocampus have visuospatial dysfunction. McConley et al.¹³ showed positive correlations between left Hippocampal volume and RCFT delayed recall scores but they failed to find correlation between right Hippocampal volumes and visuospatial part of the test. This may point towards the ability of DWI to quantitate or differentiate the pathology in the hippocampus of TLE patients.

The sensitivity of ADCs b=500 index to lateralize the abnormal hippocampus compared to RCFT results was 100% compared to b=1000 index (70%) and EEG (33.3%) suggesting that DWI using ADC at b500 gives more accurate results in lateralizing hippocampal side of focus in mesial TLE patients but to the best of our knowledge we didn't find previous results to support our suggestion.

In the present study, the interictal EEG was positive in 15 patients (50%). Adachi et al.¹⁴ showed similar results where EEG localized the site of the focus 43 from 83 TLE patients (51.8%). Murro et al.¹⁵ showed that combined EEG and MRI were localizing in 40% of TLE patients while EEG alone was localizing in 13%. In general about 50% of people with temporal lobe epilepsy may have normal results on their first EEG and about 10-40% of people with temporal lobe epilepsy will have normal EEG results even after having several EEG tests done¹⁶. This reflects the limitation of EEG in localizing the true epileptic focus in mesial TLE.

In conclusion interictal diffusion MRI showed a considerable diagnostic and lateralizing value in the assessment of patients with TLE having normal conventional MR scans and may be helpful in understanding underlying disease pathology.

[Disclosure: Authors report no conflict of interest]

REFERENCES

1.      Wehner T, Lapresto E, Tkach J, Liu P, Bingaman W, Prayson RA, et al. The value of interictal diffusion-weighted imaging in lateralizing temporal lobe epilepsy. Neurology. 2007;68(2):122-7.

2.      Devous MD, Leroy RF, Homan RW. Single photon emission computed tomography in epilepsy. Semin Nucl Med. 1990;20(4):325-41.

3.      Chabolla DR. Characteristics of the epilepsies. Mayo Clin Proc. 2002;77(9):981-90.

4.      Proposal for revised classification of epilepsies and epileptic syndromes. Commission on Classification and Terminology of the International League Against Epilepsy. Epilepsia. 1989;30(4):389-99.

5.      Taylor EM. Psychological appraisal of children with cerebral defects. Cambridge, Mass.,: Harvard Univ. Press; 1959. xvii, 499 p. p.

6.      Hugg JW, Butterworth EJ, Kuzniecky RI. Diffusion mapping applied to mesial temporal lobe epilepsy: preliminary observations. Neurology. 1999;53(1):173-6.

7.      Kantarci K, Shin C, Britton JW, So EL, Cascino GD, Jack CR. Comparative diagnostic utility of 1H MRS and DWI in evaluation of temporal lobe epilepsy. Neurology. 2002;58(12):1745-53.

8.      Londoño A, Castillo M, Lee YZ, Smith JK. Apparent diffusion coefficient measurements in the hippocampi in patients with temporal lobe seizures. AJNR Am J Neuroradiol. 2003;24(8):1582-6.

9.      Lui YW, Nusbaum AO, Barr WB, Johnson G, Babb JS, Orbach D, et al. Correlation of apparent diffusion coefficient with neuropsychological testing in temporal lobe epilepsy. AJNR Am J Neuroradiol. 2005; 26(7): 1832-9.

10.    Wang R, Li SY, Chen M, Zhou C. Diagnostic value of interictal diffusion-weighted imaging in evaluation of intractable temporal lobe epilepsy. Chin Med Sci J. 2008; 23(2): 68-72.

11.    Kim JH, Tien RD, Felsberg GJ, Osumi AK, Lee N. MR measurements of the hippocampus for lateralization of temporal lobe epilepsy: value of measurements of the body vs the whole structure. AJR Am J Roentgenol. 1994; 163(6): 1453-7.

12.    Bronen RA, Fulbright RK, Kim JH, Spencer SS, Spencer DD, al-Rodhan NR. Regional distribution of MR findings in hippocampal sclerosis. AJNR Am J Neuroradiol. 1995; 16(6): 1193-200.

13.    McConley R, Martin R, Palmer CA, Kuzniecky R, Knowlton R, Faught E. Rey Osterrieth complex figure test spatial and figural scoring: relations to seizure focus and hippocampal pathology in patients with temporal lobe epilepsy. Epilepsy Behav. 2008; 13(1): 174-7.

14.    Adachi N, Koutroumanidis M, Elwes RD, Polkey CE, Binnie CD, Reynolds EH, et al. Interictal 18FDG PET findings in temporal lobe epilepsy with déjà vu. J Neuropsychiatry Clin Neurosci. 1999; 11(3): 380-6.

15.    Murro AM, Park YD, King DW, Gallagher BB, Smith JR, Yaghmai F, et al. Seizure localization in temporal lobe epilepsy: a comparison of scalp-sphenoidal EEG and volumetric MRI. Neurology. 1993; 43(12): 2531-3.

16.    Bazil CW MM, Pedley T. Epilepsy. In: Rowland LP, Kneass W, editors. Merritt's Neurology. 11^th ed. Philadelphia: Lippincot Williams & Wilkins; 2005.