Clinical utility of functional magnetic resonance imaging for brain mapping in epilepsy surgery

doi:10.1016/j.eplepsyres.2009.12.001

Epilepsy Research

Volume 89, Issue 1, March 2010, Pages 126-132

https://doi.org/10.1016/j.eplepsyres.2009.12.001 Get rights and content

Summary

Functional magnetic resonance imaging (fMRI) is commonly used to localize brain function, but its utility in the clinical setting remains unclear. Subdural electrode implantation provides opportunities to correlate the spatial relationship of the blood oxygen level-dependent (BOLD) response to areas defined by extraoperative electrical stimulation mapping (ESM) in patients undergoing staged epilepsy surgery. 4 subjects underwent pre-operative fMRI using the analogous paradigms to those used for ESM to delineate language and motor function. Coregistration of the pre-operative MRI to a post-operative CT and MRI scan was performed in order to assess the spatial relationship between the BOLD response and the location of electrode contacts used for ESM while accounting for brain shift. fMRI was accurate in predicting the location of motor cortex with sensitivity and negative predictive value (NPV) of 1.0. Specificity was .96 with a positive predictive (PPV) value of .8. In all 4 subjects, a laterality index of the fMRI for language was accurate in predicting lateralization measured by Wada testing. While T-scores over regions where ESM-induced language deficits occurred were significantly higher (p < .05, Student's t-test) than those over regions where there was no ESM-induced deficit, sensitivity, specificity and predictive values were poor over a range of threshold criteria. Sensitivity and specificity were improved by excluding sites within 1 cm of the base of the frontal and temporal bone and sites where ESM showed motor function of face. Despite this, sensitivity and specificity were .47 and .76, respectively (T score 2.5, p < .01 corrected FDR) with PPV and NPV of .40 and .77, respectively. Sensitivity for predicting areas within 1 cm of ESM-defined language sites was higher at .82 with an NPV of .94. The results indicate that fMRI is clinically useful for lateralizing language and the localizing motor cortex. fMRI localizes language less accurately, but it may be useful in estimating the region of ESM-induced deficit in areas away from the base of the frontal and temporal bone.

Introduction

fMRI, which measures the blood oxygen level-dependent (BOLD) response (Ogawa et al., 1990) is a noninvasive methodology that is commonly used to localize functional brain areas. In the clinical setting, fMRI is increasingly being used to lateralize language function and to localize motor and language function (Hirsch et al., 2000, Ruge et al., 1999, Binder et al., 1996, Binder et al., 1997, Carpentier et al., 2001, Desmond et al., 1995, Gaillard et al., 2004, Rosenberger et al., 2009). However, the utility of fMRI in guiding surgical decision-making remains unclear. The ability to validate fMRI signal analysis with a more direct invasive method may help resolve this issue. Staged epilepsy surgery, where invasive electrodes are implanted in order to define the ictal onset and functional zones, provides an opportunity to correlate the noninvasive BOLD signal to more invasively obtained methods such as electrical stimulation mapping (ESM) in the extraoperative setting (Carpentier et al., 2001).

A good correlation has been demonstrated between language lateralization assessed using fMRI and the intracarotid amytal (Wada) test (Gaillard et al., 2004, Thesen et al., 2007). Localization of the motor cortex using BOLD signal has been accurate in some studies (Ruge et al., 1999). However, factors such as brain shift may limit their utility in patients undergoing electrode implantation for staged epilepsy surgery (Hill et al., 2000). Prior studies have shown poor sensitivity, specificity and predictive values of the location of the BOLD signal to areas defined by language using intraoperative ESM (Roux et al., 2003). The present study aims to assess the clinical utility of fMRI in the circumstance of extraoperative ESM with additional image processing is performed to account for the brain shift caused by the electrodes. Use of extraoperative ESM allows for more accurate integration and coregistration of intracranial electrodes while accounting for brain shift. In addition, time is permitted to perform reliability testing and to integrate ESM and fMRI data prior to the operative intervention that involves the possibility of resection.

Some studies have examined the effect of varying the threshold for activation for language localization in efforts to reduce Type 1 and Type 2 error, but no standard criteria exist (Loring et al., 2002). This type of analysis is critical for determining the clinical utility of fMRI and may be explored by examining the relationship between ESM and fMRI across a range of statistical values. A further analysis beyond the traditional post hoc analysis would be focused upon determining which ESM-positive sites do not show fMRI activation in order to determine whether certain areas should always be tested via ESM, regardless of fMRI activation. The present study investigates these issues in order to determine the clinical utility of fMRI language mapping in predicting the effects of extraoperative ESM.

Section snippets

Subjects

Four patients were used as subjects for this study. Demographics are shown in Table 1. All patients, after undergoing video electroencephalographic testing, MRI and neuropsychological assessment, were deemed candidates for epilepsy surgery based upon medication-resistant partial complex epilepsy. All subjects had normal MRI with no lesions, dysplasia or masses, and therefore, electrodes were implanted primarily in order to assess seizure localization. When electrodes were near regions of

Motor cortex

67 sites were tested with ESM for motor function of the hand in 2 subjects. ESM at 8 sites elicited motor response of the contralateral hand. For these sites, using a threshold of a T score of 2.5 (p < .01 corrected FDR), sensitivity for hand motor cortex was 1.0 and specificity was .96 for the two subjects where electrode contacts lay over the hand area of motor cortex (Fig. 1). PPV and NPV were 0.8 and 1.0, respectively. In the two remaining subjects an additional 10 sites were found on ESM

Sensitivity and specificity

Comparing the results of fMRI and ESM is complicated by the fact that fMRI involves the observation of normal function while ESM involves interference with function. Therefore, it may be hypothesized that fMRI could show participation of brain areas that are involved in, but not critical to, language function and that these areas overlap with critical sites identified by ESM. If this were the case, then the sensitivity would be high and the specificity would be low. Thus, the BOLD response may

Acknowledgements

Funding support is from the Page and Otto Marx Foundation. Additional support provided by the Feinstein Institute for Medical Research NSLIJ General Clinical Research Center, Grant #M01 RR018535 from the National Center for Research Resources (NCRR).

References (21)

M Jenkinson et al.
Improved optimization for the robust and accurate linear registration and motion correction of brain images
Neuroimage
(2002)
M. Jenkinson et al.
A global optimisation method for robust affine registration of brain images
Med. Image Anal.
(2001)
D.W. Loring et al.
Now you see it now you don’t: statistical and methodological considerations in fMRI
Epilepsy Behav.
(2002)
M.W. Woolrich et al.
Temporal autocorrelation in univariate linear modeling of fMRI data
Neuroimage
(2001)
K.J. Worsley
An improved theortical P value for SPMs based on discrete local maxima
Neuroimage
(2005)
Arora, J., Pugh, K.R., Westerveld, M., Spencer, S., Spencer, D.D., Constable, R.T., in press. Language lateralization...
J.R. Binder et al.
Human brain language areas identified by functional magnetic resonance imaging
J. Neurosci.
(1997)
J.R. Binder et al.
Determination of language dominance using functional MRI: a comparison with the Wada test
Neurology
(1996)
A. Carpentier et al.
Functional MRI of language processing: dependence on input modality and temporal lobe epilepsy
Epilepsia
(2001)
J.E. Desmond et al.
Functional MRI measurement of language lateralization in WADA-tested patients
Brain
(1995)

There are more references available in the full text version of this article.

Cited by (22)

A Novel Functional Magnetic Resonance Imaging Paradigm for the Preoperative Assessment of Auditory Perception in a Musician Undergoing Temporal Lobe Surgery
2018, World Neurosurgery
Citation Excerpt :
Although multiple studies have used fMRI to map the auditory processing pathways,1,11,12 its accuracy compared with the gold standard of extraoperative ESM has never been investigated.13-15 Furthermore, prior studies assessing the accuracy of fMRI used inconsistent methodologies for comparison both between and within studies.16-22 Although fMRI has been demonstrated to have a high positive predictive value for locating motor and language areas compared with ESM, the negative predictive value for fMRI locating functional areas of the brain, which in this case is essential for determining areas safe to surgically excise, is much more variable.16-22
Presurgical evaluation for temporal lobe epilepsy routinely assesses speech and memory lateralization and anatomic localization of the motor and visual areas but not baseline musical processing. This is paramount in a musician. Although validated tools exist to assess musical ability, there are no reported functional magnetic resonance imaging (fMRI) paradigms to assess musical processing. We examined the utility of a novel fMRI paradigm in an 18-year-old left-handed pianist who underwent surgery for a left temporal low-grade ganglioglioma.
Preoperative evaluation consisted of neuropsychological evaluation, T1-weighted and T2-weighted magnetic resonance imaging, and fMRI. Auditory blood oxygen level–dependent fMRI was performed using a dedicated auditory scanning sequence. Three separate auditory investigations were conducted: listening to, humming, and thinking about a musical piece.
All auditory fMRI paradigms activated the primary auditory cortex with varying degrees of auditory lateralization. Thinking about the piece additionally activated the primary visual cortices (bilaterally) and right dorsolateral prefrontal cortex. Humming demonstrated left-sided predominance of auditory cortex activation with activity observed in close proximity to the tumor.
This study demonstrated an fMRI paradigm for evaluating musical processing that could form part of preoperative assessment for patients undergoing temporal lobe surgery for epilepsy.
Combining task-evoked and spontaneous activity to improve pre-operative brain mapping with fMRI
2016, NeuroImage
Citation Excerpt :
Now that our technique is validated with motor mapping, there is good motivation to begin testing it for language and memory. Pre-operative fMRI mapping of these functions tends to be less robust than motor mapping and thus there is greater need for improvement (Hirsch et al., 2000; Kapsalakis et al., 2012; Mehta and Klein, 2010). Further, there is good reason to think that our algorithm will prove useful in these other domains.
Noninvasive localization of brain function is used to understand and treat neurological disease, exemplified by pre-operative fMRI mapping prior to neurosurgical intervention. The principal approach for generating these maps relies on brain responses evoked by a task and, despite known limitations, has dominated clinical practice for over 20 years. Recently, pre-operative fMRI mapping based on correlations in spontaneous brain activity has been demonstrated, however this approach has its own limitations and has not seen widespread clinical use. Here we show that spontaneous and task-based mapping can be performed together using the same pre-operative fMRI data, provide complimentary information relevant for functional localization, and can be combined to improve identification of eloquent motor cortex. Accuracy, sensitivity, and specificity of our approach are quantified through comparison with electrical cortical stimulation mapping in eight patients with intractable epilepsy. Broad applicability and reproducibility of our approach are demonstrated through prospective replication in an independent dataset of six patients from a different center. In both cohorts and every individual patient, we see a significant improvement in signal to noise and mapping accuracy independent of threshold, quantified using receiver operating characteristic curves. Collectively, our results suggest that modifying the processing of fMRI data to incorporate both task-based and spontaneous activity significantly improves functional localization in pre-operative patients. Because this method requires no additional scan time or modification to conventional pre-operative data acquisition protocols it could have widespread utility.
Clinical applications of the functional connectome
2013, NeuroImage
Citation Excerpt :
Clearly, a noninvasive methodology to image cortical function without sampling bias that is sensitive to activity in deeper regions and that does not disrupt function would be desirable. While task-based fMRI has been used extensively to study normal brain function, the clinical utility of this method has yet to be firmly established (Hill et al., 2000; Mehta and Klein, 2010; Roux et al., 2003). A major factor that limits clinical utility involves poor task performance in patients with cognitive impairment and neurological deficits — the very patients who need neurosurgical intervention (Pujol et al., 1998).
Central to the development of clinical applications of functional connectomics for neurology and psychiatry is the discovery and validation of biomarkers. Resting state fMRI (R-fMRI) is emerging as a mainstream approach for imaging-based biomarker identification, detecting variations in the functional connectome that can be attributed to clinical variables (e.g., diagnostic status). Despite growing enthusiasm, many challenges remain. Here, we assess evidence of the readiness of R-fMRI based functional connectomics to lead to clinically meaningful biomarker identification through the lens of the criteria used to evaluate clinical tests (i.e., validity, reliability, sensitivity, specificity, and applicability). We focus on current R-fMRI-based prediction efforts, and survey R-fMRI used for neurosurgical planning. We identify gaps and needs for R-fMRI-based biomarker identification, highlighting the potential of emerging conceptual, analytical and cultural innovations (e.g., the Research Domain Criteria Project (RDoC), open science initiatives, and Big Data) to address them. Additionally, we note the need to expand future efforts beyond identification of biomarkers for disease status alone to include clinical variables related to risk, expected treatment response and prognosis.
Language dominance in children with epilepsy: Concordance of fMRI with intracarotid amytal testing and cortical stimulation
2013, Epilepsy and Behavior
Citation Excerpt :
In particular, language localization by fMRI in patients 5, 14, and 15 was confirmed by ESM, which positively mapped language in left frontal and temporal lobe regions. In a recent study by Mehta and Klein comparing fMRI with ESM, the language fMRI t-score was significantly higher under electrode sites that were associated with a language deficit during ESM testing [34]. In our cohort, ESM failed to identify language in 13 children.
Accurate localization of language function is critical in children undergoing epilepsy surgery. Functional magnetic resonance imaging (fMRI) is a noninvasive mapping method that has begun to replace electrocortical stimulation mapping (ESM) and the intracarotid amytal test (IAT). We used both quantitative and qualitative methods to evaluate the concordance of fMRI with ESM and IAT in 20 children using a panel of language tasks. In no cases did fMRI assessment of language hemisphere dominance identify the opposite hemisphere from assessment by IAT or ESM. Concordance with IAT and ESM was higher using fMRI visual inspection than an fMRI laterality index, which failed to lateralize language in a number of the subjects. We have demonstrated that fMRI has good concordance with more traditional methods of language mapping. When fMRI demonstrates bilateral language activations, however, we continue to recommend confirmatory testing by either IAT or ESM prior to resection in classic language regions.
AveLI: A robust lateralization index in functional magnetic resonance imaging using unbiased threshold-free computation
2012, Journal of Neuroscience Methods
Citation Excerpt :
If a specific LI computational method were able to better estimate the recovery rate from strokes (Naccarato et al., 2006; Calautti et al., 2007, 2010), then that LI method would be optimal for that purpose. If we consider the congruence of LIs to the Wada test as the standard, the LI that best replicates the results of the Wada test is the optimal method (Desmond et al., 1995; Deblaere et al., 2002; Powell and Duncan, 2005; Chlebus et al., 2007; Kamada et al., 2007; Binder et al., 2008; Suarez et al., 2009; Jones et al., 2010; Mehta and Klein, 2010; Binder, 2011). In addition, quick and practical evaluations are essential in clinical situations (Fernández et al., 2001; Deblaere et al., 2002; Jones et al., 2010).
The laterality index (LI) is often applied in functional magnetic resonance imaging (fMRI) studies to determine functional hemispheric lateralization. A difficulty in using conventional LI methods lies in ensuring a legitimate computing procedure with a clear rationale. Another problem with LI is dealing with outliers and noise. We propose a method called AveLI that follows a simple and unbiased computational principle using all voxel t-values within regions of interest (ROIs). This method first computes subordinate LIs (sub-LIs) using each of the task-related positive voxel t-values in the ROIs as the threshold as follows: sub-LI = (Lt − Rt)/(Lt + Rt), where Lt and Rt are the sums of the t-values at and above the threshold in the left and right ROIs, respectively. The AveLI is the average of those sub-LIs and indicates how consistently lateralized the performance of the subject is across the full range of voxel t-value thresholds. Its intrinsic weighting of higher t-value voxels in a data-driven manner helps to reduce noise effects. The resistance against outliers is demonstrated using a simulation. We applied the AveLI as well as other “non-thresholding” and “thresholding” LI methods to two language tasks using participants with right- and left-hand preferences. The AveLI showed a moderate index value among 10 examined indices. The rank orders of the participants did not vary between indices. AveLI provides an index that is not only comprehensible but also highly resistant to outliers and to noise, and it has a high reproducibility between tasks and the ability to categorize functional lateralization.
A practical clinical method to quantify language lateralization in fMRI using whole-brain analysis
2011, NeuroImage
Surgery is often the only effective treatment for intractable epilepsy, but its benefits must be balanced by potential disruption of eloquent cortical functions. Wada test is the standard technique to lateralize language before surgery; however, it is invasive and associated with complications. fMRI provides an attractive noninvasive alternative, which has been previously shown to correlate with Wada results. However this correlation is imperfect since standard fMRI laterality indices are dependent on a particular arbitrary statistical threshold used in the data processing. We report a novel automated, threshold-independent fMRI methodology to assess language lateralization, which we hypothesize provides a robust and unbiased pre-operative assessment. This hemispheric histogram analysis method can accurately interrogate language lateralization, as validated against the Wada test.
Fifty-nine subjects with intractable epilepsy received preoperative evaluation for language lateralization using fMRI. fMRI data then were analyzed using a novel automated threshold-independent method for determining language lateralization. The methodology generated a lateralization score based on hemispheric activation of language areas and a quality index based on multiple factors, including patient motion and signal-to-noise characteristics. Lateralization scores were compared to Wada test results (51 patients), direct cortical stimulation (3 patients), and subdural grid stimulation (5 patients). Data sets were used to generate a probability score for language lateralization for each subject.
The lateralization scores correlated well with the objective measures of language lateralization (r² = 0.46). Cumulative historical data were utilized to prospectively determine probabilities of language lateralization for individual patients. In conclusion, hemispheric language lateralization can be accurately determined using a novel objective and automated methodology that calculates language lateralization in a threshold-independent manner and can be used to determine the probability of language dominance in individual patients.

View all citing articles on Scopus

View full text

Clinical utility of functional magnetic resonance imaging for brain mapping in epilepsy surgery

Summary

Introduction

Section snippets

Subjects

Motor cortex

Sensitivity and specificity

Acknowledgements

Neuroimage

Med. Image Anal.

Epilepsy Behav.

Neuroimage

Neuroimage

Human brain language areas identified by functional magnetic resonance imaging

J. Neurosci.

Determination of language dominance using functional MRI: a comparison with the Wada test

Neurology

Functional MRI of language processing: dependence on input modality and temporal lobe epilepsy

Epilepsia

Functional MRI measurement of language lateralization in WADA-tested patients

Brain