The role of intraoperative MRI in resective epilepsy surgery for peri-eloquent cortex cortical dysplasias and heterotopias in pediatric patients

Free access

OBJECTIVE

Previous studies have demonstrated that an important factor in seizure freedom following surgery for lesional epilepsy in the peri-eloquent cortex is completeness of resection. However, aggressive resection of epileptic tissue localized to this region must be balanced with the competing objective of retaining postoperative neurological functioning. The objective of this study was to investigate the role of intraoperative MRI (iMRI) as a complement to existing epilepsy protocol techniques and to compare rates of seizure freedom and neurological deficit in pediatric patients undergoing resection of perieloquent lesions.

METHODS

The authors retrospectively reviewed the medical records of pediatric patients who underwent resection of focal cortical dysplasia (FCD) or heterotopia localized to eloquent cortex regions at the Children's National Health System between March 2005 and August 2015. Patients were grouped into two categories depending on whether they underwent conventional resection (n = 18) or iMRI-assisted resection (n = 11). Patient records were reviewed for factors including demographics, length of hospitalization, postoperative seizure freedom, postoperative neurological deficit, and need for reoperation. Postsurgical seizure outcome was assessed at the last postoperative follow-up evaluation using the Engel Epilepsy Surgery Outcome Scale.

RESULTS

At the time of the last postoperative follow-up examination, 9 (82%) of the 11 patients in the iMRI resection group were seizure free (Engel Class I), compared with 7 (39%) of the 18 patients in the control resection group (p = 0.05). Ten (91%) of the 11 patients in the iMRI cohort achieved gross-total resection (GTR), compared with 8 (44%) of 18 patients in the conventional resection cohort (p = 0.02). One patient in the iMRI-assisted resection group underwent successful reoperation at a later date for residual dysplasia, compared with 7 patients in the conventional resection cohort (with 2/7 achieving complete resection). Four (36%) of the patients in the iMRI cohort developed postoperative neurological deficits, compared with 15 patients (83%) in the conventional resection cohort (p = 0.02).

CONCLUSIONS

These results suggest that in comparison with a conventional surgical protocol and technique for resection of epileptic lesions in peri-eloquent cortex, the incorporation of iMRI led to elevated rates of GTR and postoperative seizure freedom. Furthermore, this study suggests that iMRI-assisted surgeries are associated with a reduction in neurological deficits due to intraoperative damage of eloquent cortex.

ABBREVIATIONSCNHS = Children's National Health System; DTI = diffusion tensor imaging; DWI = diffusion-weighted imaging; ECoG = electrocorticography; FCD = focal cortical dysplasia; GTR = gross-total resection; iMRI = intraoperative MRI; MEP = motor evoked potential; SEM = standard error of the mean; SSEP = somatosensory evoked potential.

OBJECTIVE

Previous studies have demonstrated that an important factor in seizure freedom following surgery for lesional epilepsy in the peri-eloquent cortex is completeness of resection. However, aggressive resection of epileptic tissue localized to this region must be balanced with the competing objective of retaining postoperative neurological functioning. The objective of this study was to investigate the role of intraoperative MRI (iMRI) as a complement to existing epilepsy protocol techniques and to compare rates of seizure freedom and neurological deficit in pediatric patients undergoing resection of perieloquent lesions.

METHODS

The authors retrospectively reviewed the medical records of pediatric patients who underwent resection of focal cortical dysplasia (FCD) or heterotopia localized to eloquent cortex regions at the Children's National Health System between March 2005 and August 2015. Patients were grouped into two categories depending on whether they underwent conventional resection (n = 18) or iMRI-assisted resection (n = 11). Patient records were reviewed for factors including demographics, length of hospitalization, postoperative seizure freedom, postoperative neurological deficit, and need for reoperation. Postsurgical seizure outcome was assessed at the last postoperative follow-up evaluation using the Engel Epilepsy Surgery Outcome Scale.

RESULTS

At the time of the last postoperative follow-up examination, 9 (82%) of the 11 patients in the iMRI resection group were seizure free (Engel Class I), compared with 7 (39%) of the 18 patients in the control resection group (p = 0.05). Ten (91%) of the 11 patients in the iMRI cohort achieved gross-total resection (GTR), compared with 8 (44%) of 18 patients in the conventional resection cohort (p = 0.02). One patient in the iMRI-assisted resection group underwent successful reoperation at a later date for residual dysplasia, compared with 7 patients in the conventional resection cohort (with 2/7 achieving complete resection). Four (36%) of the patients in the iMRI cohort developed postoperative neurological deficits, compared with 15 patients (83%) in the conventional resection cohort (p = 0.02).

CONCLUSIONS

These results suggest that in comparison with a conventional surgical protocol and technique for resection of epileptic lesions in peri-eloquent cortex, the incorporation of iMRI led to elevated rates of GTR and postoperative seizure freedom. Furthermore, this study suggests that iMRI-assisted surgeries are associated with a reduction in neurological deficits due to intraoperative damage of eloquent cortex.

Resection of focal cortical dysplasia (FCD) and heterotopia localized adjacent to eloquent cortex presents a significant challenge of competing objectives. A balance must be attained between achieving the ultimate goal of complete seizure freedom while minimizing postoperative neurological deficits.

Previous studies have demonstrated that an important factor in achieving postoperative seizure freedom for patients presenting with intractable epilepsy secondary to a lesion adjacent to eloquent cortex is gross-total resection (GTR).9,28 Traditionally, aggressive resections in the peri-eloquent regions have been tempered by high rates of neurological deficit, and seizure-free rates in pediatric patients following surgery range from 43% to 64%.1,2,8,24 Intraoperative identification of the extent of resection with compensation for brain shift is a critical factor in peri-eloquent resections, which has been shown to be aided by intraoperative MRI (iMRI).20,22 iMRI has previously been shown to aid surgeries of eloquent cortex gliomas, leading to increased rates of complete resection and fewer postoperative deficits.13,19,30,31 Furthermore, iMRI has been successfully integrated into pediatric cases and has been shown to aid in intraoperative evaluation of the extent of resection in epilepsy surgery.3,4,21,32 Hence, we sought to adapt this technology in resective epilepsy surgery for epileptogenic lesions located adjacent to eloquent cortex. The current study is a comparative analysis of postoperative seizure and neurological outcomes in patients undergoing resection of FCD and heterotopias adjacent to eloquent cortex using iMRI-assisted versus conventional resective techniques.

Methods

Patient Population

We retrospectively reviewed the medical records and MR images of 29 pediatric patients undergoing surgery at Children's National Health System (CNHS) for pharma-coresistant epilepsy secondary to FCD localized adjacent to eloquent cortex (motor, sensory, visual, and speech) who met inclusion criteria. Resections between March 2005 and April 2013 (n = 18) were performed via a conventional microsurgical technique with the aid of neuronavigation (Stealth Station navigation system, Medtronic-Sofamor Danek), neurophysiological monitoring, and intraoperative electrocorticography (ECoG), while surgeries conducted between January 2014 and August 2015 (n = 11) were performed with the additional aid of 1.5-T iMRI to the previously used assistive neurotechnologies. Upon approval from the institutional review board, patient records were assessed for demographic data, new postoperative deficits, completion of resection, postoperative seizure outcome, length of surgery, length of hospitalization, and need for reoperation. For conventional resection patients undergoing repeat resection of residual dysplasia, duration of follow-up was extended until the time of repeat surgery.

Eloquent Cortex

Eloquent cortex is defined as areas of brain that are responsible for highly critical and complex functions that the brain cannot easily compensate for, including areas of brain that subserve motor, sensory, language, or visual function. Eloquent cortex was defined by 4 parameters. All patients presented with structural lesions on preoperative MRI indicating proximity to eloquent regions. Functional MRI (fMRI) with motor and language paradigms was performed as indicated in patients who were old enough to cooperate with this investigation. Patients with perirolandic lesions underwent intraoperative brain mapping with electrophysiological phase-reversal assessment to identify the central sulcus. Additionally, 2 patients in the conventional resection cohort underwent extraoperative brain mapping through grids that had been placed intraoperatively.

Surgical Procedure

We previously described a conventional microsurgical technique for resection of FCD at our institution.23 In the iMRI cohort, ECoG was performed in 9 of the 11 resections but was not used in patients with deep lesions that posed a significant risk of vascular injury. Intraoperative neurophysiological monitoring of somatosensory evoked potentials (SSEPs) and motor evoked potentials (MEPs) was used in all cases. Standard microsurgical resection was conducted until the surgeon believed that a complete resection had been performed. The patient was then prepared and transferred to the dedicated 1.5-T iMRI suite (Greenline Achieva Nova Dual, Philips Medical Systems). The intraoperative sequences taken in the iMRI cohort included 1) a T1-weighted 3D fast field echo (TE 3.2 msec, TR 6.7 msec, matrix 240 × 240, FOV 300 mm, slice thickness 1.6 mm, slab 34 cm) reformatted into orthogonal planes; 2) T2-weighted turbo spin-echo (TE 100 msec, TR 4019 msec, matrix 296 × 225, FOV 280 mm, slice thickness 3 mm); and 3) a diffusion-weighted imaging (DWI) sequence (TE 103 msec, TR 5096 msec, matrix 152 × 106, FOV 230 mm, slice thickness 4 mm). Images were independently viewed and analyzed by the attending neurosurgeon and neuroradiologist and discussed jointly. In cases of residual dysplasia, the patient was returned to surgery for further reexploration at the same operative session.

Duration of Operative Event

Duration of the operative event was defined as the recorded time of entry into the operating room suite until exit. Hence, in addition to surgical time, this encompassed time for anesthesia, placement of peripheral lines, preparing the patient, acquisition of all MR images, and anesthesia recovery. These extrasurgical activities accounted for 2–3 hours of the duration on average. We believed it was relevant to include these other (extraoperative) components because they are a reflection of the complexity of coordination of care and are therefore indirect costs in undertaking this type of procedure.

Neurological and Seizure Outcomes

Neurological deficits were retrospectively assessed at discharge, the first postoperative visit, and at the last postoperative visit. Postoperative neurological function was compared with the preoperative baseline level and evaluated for alterations. Transient deficit was defined as complete recovery to preoperative baseline by the last postoperative follow-up evaluation. Permanent deficit was defined as persistence of the deficit at the last postoperative follow-up. Mild deficits were defined as nondisabling hemiparesis or hemianesthesia. Severe deficits were defined as disabling hemiparesis or hemianesthesia. Postsurgical seizure outcome was assessed utilizing the Engel Epilepsy Surgery Outcome Scale10 at the last postoperative follow-up visit.

Statistical Analysis

Clinical and demographic statistics were recorded as means ± standard errors of the mean (SEM), and duration of follow-up was reported as median. Statistical analysis of continuous variables between the independent cohorts was performed via the nonparametric Mann-Whitney U-test. The Fischer exact test was used to assess categorical variables between groups. All tests performed were 2-tailed, and a p value < 0.05 was considered statistically significant.

Results

Patient Population

Between March 2005 and August 2015, 29 consecutive patients underwent resection of FCD and heterotopias localized adjacent to eloquent cortex at CNHS. Eighteen patients (13 males and 5 females) underwent resection utilizing conventional epilepsy protocol and techniques, and 11 patients (3 males and 8 females) underwent iMRI-assisted resection. There were no statistically significant differences between the two groups for the variables of mean age at surgery, mean duration of seizures, or mean age at seizure onset (Table 1). The median duration of follow-up was 5.72 months (range 0.46–10.61 months) for the iMRI-assisted cohort and 17.75 months (range 0.13–71.85 months) for the conventional cohort. Patients in the iMRI cohort had a shorter mean length of stay than those in the conventional cohort (3.8 ± 0.4 days vs 4.9 ± 0.5 days, respectively), although this difference was not statistically significant (p = 0.08).

TABLE 1.

Comparative analysis between cohorts*

VariableiMRI Incorporated Resection GroupConventional Epilepsy Protocol Resection Groupp Value
Age at surgery (yrs)9.1 ± 2.06.1 ± 1.10.2
Age at seizure onset (yrs)3.49 ± 1.02.7 ± 9.10.5
Duration of seizures (mos)59.5 ± 20.041.4 ± 11.30.4
Duration of follow-up (mos)5.02 ± 1.223.3 ± 5.00.005
Duration of operative event (mins)409.5 ± 44.9289.6 ± 13.90.004
Postop length of stay (days)3.8 ± 0.44.9 ± 0.50.08

All data given as mean ± SEM.

Demographics and surgical data are illustrated in Table 2 for patients in the conventional resection cohort, and in Table 3 for patients in the iMRI-assisted resection cohort. Pathohistological analysis of patients undergoing iMRI-assisted resection revealed 2 Type I FCDs, 5 Type IIA FCDs, 2 Type IIB FCDs, and 2 FCDs of undetermined histology. In the conventional resection cohort there were 6 Type I FCDs, 3 Type IIA FCDs, 7 Type IIB FCDs, and 2 FCDs of undetermined histology.

TABLE 2.

Descriptive characteristics for patients undergoing conventional peri-eloquent resections

Case No.Age (yrs), SexHistopathology (FCD type)Lesion LocalizationOperative Duration (mins)Follow-up (mos)Engel ClassPrimary Postop Morbidity
10.2, MIIBRt frontal21313.0IVLt hemiparesis
22.5, MIIBLt frontal2702.5IIRt hemiparesis
311, FNSLt temporal37761.9IRt superior quadranopsia
40.3, MIIARt parietal35522.5IIILt hemiparesis
51.5, MIIBRt parietal31671.9ILt hemiparesis
66, FNSRt parietal2180.6IIILt hemiparesis
74.5, MIIALt frontal26312.0IIIRt hemiparesis
810, MIIBRt frontal28050.2IILt hemiparesis
92.5, FIIBLt temporal23924.0IExpressive language deficit
104, MIRt occipital2880.1IExophoria
113, MIIALt frontotemporal4314.8IIIRt hemiparesis
128, MILt temporal3277.1IIIRt hemiparesis
1313, MILt frontal3289.8IIINo
1415, FILt parietal30341.3IReceptive language deficit
150.2, MIIBRt parietal23433.4IILt hemiparesis
167.5, FILt frontal transmantle heterotopia23410.7IRt hemiparesis
176.5, MIRt frontal29427.6INo
1813.5, MIIBRt frontoparietal24225.5IINo

NS = not specified.

TABLE 3.

Descriptive characteristics for patients undergoing iMRI-assisted peri-eloquent resections

Case No.Age (yrs), SexHistopathology (FCD type)Lesion LocationOperative Duration (mins)Return to Surgery After iMRIFollow-Up (mos)Engel ClassPostop Morbidity
117, FILt temporal261No0.5INo
26, FIIBLt temporoparietal382Yes10.6INo
31, MNSRt parietal382No0.8INo
43, FIRt parietal482Yes9.1ILt hemiparesis
513, FIIARt parietal371No9.8IAltered lt lower-extremity sensation
610, FIIALt parietal293No7.3IRt upper-extremity monoparesis
77, FIIARt insular361No6.9IIINo
82, MNSLt parietooccipital transmantle heterotopia798Yes5.7II*No
98, FIIALt parietal514Yes1.8IAltered rt lower-extremity sensation
109, MIIBLt frontal352No1.3INo
1122, FIIALt frontal309No1.5INo

Patient underwent repeat iMRI-assisted surgery 1 month postoperatively, achieving GTR and complete seizure freedom (Engel Class I).

Intraoperative MRI Surgeries

As expected, iMRI-resections were statistically longer in duration than conventional resections (409.5 ± 44.9 minutes vs 289.6 ± 13.9 minutes, respectively; p = 0.004). Mean total acquisition time per case was 14.7 ± 2.4 minutes. In addition to acquisition time, preparation and transport for iMRI accounted for approximately 45–60 minutes. Four of the patients (36%) returned to the operating suite in the same operative session for further exploration of the resection cavity following the iMRI indication of residual dysplasia, and further resection was performed in 2 (18%) of these patients (Cases 4 and 9). Further intraoperative exploration confirmed complete resection in 1 patient (Case 2), and further resection was halted in 1 patient (Case 8) due to reduction in evoked potentials.

Resective and Postoperative Seizure Outcomes

The iMRI cohort achieved a greater rate of GTR of the MRI-indicated dysplastic lesion than the conventional resection group (p = 0.02). Intraoperative 1.5-T or postoperative 3.0-T MRI confirmed complete resection in 10 (91%) of the 11 patients undergoing iMRI-assisted FCD resection, whereas postoperative 1.5- or 3.0-T MRI confirmed complete resection in 8 (44%) of the 18 patients in the conventional resection cohort. One patient in the iMRI cohort underwent a further resection at a later date to achieve a GTR, while 7 patients in the conventional FCD resection cohort underwent repeat resections at a later date and 2 of 7 achieved GTRs.

At the time of the last postoperative follow-up evaluation, 9 (82%) of the 11 patients in the iMRI-assisted group achieved complete seizure freedom (Engel Class I) compared with 7 (39%) of the 18 patients in the conventional resection group (p = 0.05; Fig. 1). The median duration to seizure recurrence in patients from the conventional resection group (n = 11) was 2.23 months postoperatively (ranging from postoperative Day 1 to 28.5 months postoperatively). One patient (Case 8) in the iMRI cohort underwent further reoperation at a later date, achieving complete seizure freedom postoperatively. Seven patients in the conventional resection cohort underwent further surgery at a later date with 2 achieving complete seizure freedom.

FIG. 1.
FIG. 1.

Pie charts illustrating postoperative seizure freedom at the last postoperative follow-up evaluation in the iMRI-assisted resection cohort (upper) and the conventional resection cohort (lower).

Neurological Outcome

Patients undergoing conventional FCD resection were more likely to develop postoperative neurological deficits than patients undergoing iMRI-assisted resections (p = 0.02; Fig. 2). Four (36%) of the patients in the iMRI cohort developed transient neurological deficits (resolved by the last postoperative visit) following surgery. Two patients (Cases 4 and 6) developed mild motor deficits, and 2 patients (Cases 5 and 9) reported transient alteration of sensation in the lower extremities. Fifteen (83%) of the patients in the conventional resection group developed transient (n = 3) or more permanent (n = 12) deficits following surgery. Nine patients developed permanent motor deficit following surgery; 8 patients developed mild unresolved weakness, and 1 patient developed more severe hemiparesis that persisted despite rehabilitation. There were 2 cases of transient hemiparesis. Of the remaining patients who developed postoperative deficits, there were 2 cases of permanent visual field deficits, 1 case of transient language deficit, and 1 case of permanent language deficit.

FIG. 2.
FIG. 2.

Postoperative neurological deficits in the iMRI-assisted and conventional resection cohorts.

Illustrative Case

A 20-month-old boy (Case 8) presented with intractable epilepsy secondary to left parietooccipital transmantle heterotopia extending into the ventricle and abutting the thalamus, immediately adjacent to the posterior limb of the internal capsule. Additionally, the patient presented with a large occipital meningeal cyst (Fig. 3). Following discussion at the multidisciplinary epilepsy meeting, it was decided to proceed with resection of the heterotopia and to perform a left temporal lobectomy. BrainLAB neuronavigation uploaded with preoperative MRI was used to localize the initial site of resection, and ECoG was performed indicating discharge in the left parietooccipital junction adjacent to the anterior margin of the occipital cyst, along the temporal lobe including the left superior temporal gyrus. To avoid the risk of hemiparesis, an approach from a posterior-inferior aspect was undertaken, resecting the dysplasia from its inferior margin medially to the choroidal fissure. Next, resection of the left posterior parietal gyri was carried through to the midline. In the depths of the cortical dissection, heterotopia was identified. It should be noted that at this point SSEPs and MEPs remained stable. However, as the resection continued anteriorly, there was a reduction in evoked potential. At this point, the resection was stopped and the patient prepared for iMRI. Both T1 spoiled gradient recalled echo and T2-weighted sequences confirmed residual dysplasia, and the patient was returned to surgery for further resection of the anterolateral margins (Fig. 4). Again, attempted resection was accompanied by decreased MEPs and the procedure was stopped. Postoperatively, the patient saw a reduction in seizures from 9/day to 2/day; however, a decision was made to repeat surgery 1 month later to resect the residual dysplasia. Resection of the deeper aspect of the residual heterotopia was conducted until the atrium of the left lateral ventricle was encountered, then continued laterally and inferiorly. Throughout this point, there were no changes in the MEPs. A complete disconnection of the occipital lobe was then undertaken. Intraoperative T1 spoiled gradient recalled echo, T2-weighted, and DWI images confirmed a complete resection (Fig. 5). Postoperative follow-up at 5 months indicated the patient had achieved complete seizure freedom (Engel Class I).

FIG. 3.
FIG. 3.

Case 8. Preoperative MR images. A: Axial T2-weighted image showing left-sided heterotopia (outline) just behind the left central sulcus. B: Axial T2-weighted image showing the deeper aspects of the left transmantle heterotopia just behind the left internal capsule and left thalamus. C: Sagittal T1-weighted image showing left transmantle heterotopia (note a congenital meningeal cyst posterior to the heterotopia).

FIG. 4.
FIG. 4.

Case 8. Axial intraoperative T2-weighted MR image at the first surgery showing residual heterotopia (outlined) adjacent to the left internal capsule. Surgery was discontinued at this stage due to changes in MEPs.

FIG. 5.
FIG. 5.

Case 8. Left: Axial intraoperative T2-weighted MR image at the second surgery showing complete resection of the residual heterotopia. Right: Sagittal intraoperative T2-weighted MR image at the second surgery showing complete resection of the residual heterotopia.

Discussion

Resection of FCD localized adjacent to eloquent cortex presents a distinct challenge. A balance must be achieved between achieving the ultimate goal of complete seizure freedom with preservation of important neurological function. We present evidence for the integration of iMRI into preexisting epilepsy protocol techniques, allowing for improved results on both sides of the spectrum: a more complete resection leading to greater rates of seizure freedom, along with decreased rates of resection-induced new neurological deficits.

Intraoperative MRI-assisted surgeries resulted in nearly 50% greater rate of GTR than conventional surgeries (p = 0.02). Furthermore, patients in the iMRI cohort achieved a higher ratio of complete seizure freedom (Engel Class I) than patients in the conventional resection cohort (p = 0.05). Using iMRI in conjunction with functional MRI and diffusion tensor imaging (DTI), Sommer et al. previously reported achieving GTR in all patients and 72% complete seizure freedom in resection of lesional epilepsy in close proximity to eloquent cortex.32 Of note, their cohort consisted of 7 (28%) of 25 patients with intractable epilepsy secondary to FCD (1 Type IIA and 6 Type IIB), with the remainder of the cases secondary to mixed etiology including posttraumatic glial scars and cavernous hemangiomas. Importantly, surgical outcomes in our study are consistent with those reported by Sommer et al. and considerably improved over conventional resection both in our current study and in the literature.23 Prior studies of resection of lesional epilepsy adjacent to regions of eloquent cortex in pediatric patients have reported complete seizure freedom (Engel Class I) rates of 43%–64%.1,2,8,24

Patients in the conventional resection cohort developed postoperative neurological deficits at a higher rate than patients undergoing iMRI-assisted resection (p = 0.02). Again, our reported rates of 36% transient and no permanent new neurological deficits in the iMRI cohort are consistent with those of Sommer et al., who reported 20% transient and 12% permanent neurological deficits in their cohort of iMRI-resected extratemporal epilepsy adjacent to eloquent cortex.32 Importantly, this represents a reduction in deficits over conventional resections in our study, and in previous studies that have reported postoperative neurological deficits ranging from 50% to 83%.1–3,25 Hence, our results demonstrate that iMRI-assisted resections are less prone to incurring neurological deficits. This may be due, in part, to the surgeon undertaking a more conservative extirpation with the knowledge that iMRI may provide a “second chance” to reevaluate the surgical course of action; thus in cases of FCD adjacent to eloquent cortex this minimizes the risk of damage to functional regions and enhances postoperative quality of life.

The most significant factor leading to postoperative seizure freedom in patients with intractable epilepsy secondary to FCD is GTR of the MRI-indicated lesion.6,11,12,14,16,27 We found that iMRI allowed for the intraoperative ability to examine the extent of resection once the surgeon believed a complete surgery had been accomplished. In cases of MR-indicated residual dysplasia, iMRI became especially critical in enabling the surgeon to account for brain shift (which may distort accurate depiction of eloquent anatomical landmarks on preoperative images) and update neuronavigation for a reevaluation of the resection cavity. In our present study, iMRI impacted the course of surgery in 4 cases, with repeat surgery precluded in 3 of those cases. Conversely, more than one-third of the patients in the conventional resection cohort underwent reoperations for residual dysplasia. In preventing unnecessary reoperations, iMRI acts as a quality control mechanism that both reduces the risk associated with further surgery (a factor that becomes even more substantial in eloquent cortex resections) and has profound implications from a health economics standpoint.

A critique of iMRI utilization has been that it may lead to more aggressive resections of MR-indicated lesions, thus increasing the risk of postoperative morbidity in eloquent cortex surgeries.29 As Case 8 illustrates, iMRI seamlessly integrates with intraoperative neurophysiological monitoring (a functional “security margin” that is often more precise than anatomical indication of eloquent cortex alone), thus reducing the risk associated with overly aggressive resection of intraoperative MR-indicated residual lesions. Similar findings have been reported in resection of gliomas localized to eloquent cortex where iMRI in combination with intraoperative neurophysiological monitoring allows for an optimal resection within functional boundaries.30 Equally important may be avoiding essential white matter tracts, which has been aided by the advent of DTI. Yet, a known intraoperative challenge of functional MRI with DTI utilization is the effect of brain shift. To counter this, iMRI has been successfully integrated with these imaging techniques in lesional epilepsy surgery.7,32 In addition to traditional intraoperative CT and MRI, further technologies such as advanced intraoperative ultrasonography and intraoperative elasticity imaging may better aid in real-time identification of dysplastic lesions and in more accurate depiction of margins.5,18 Furthermore, intraoperative ultrasonography has been shown to successfully integrate with functional MRI and DTI to provide an intraoperative update to neuronavigation accounting for brain shift.26

While we found iMRI to be a powerful addition to preexisting epilepsy protocol techniques, its resolution is limited to approximately 1 mm.17 In eloquent cortex resections where lesions less than 5 mm from functional tissue are associated with increased risk of neurological deficit, iMRI resolution may potentially be limited.15 Although our study represents one of the larger investigations of lesional resection adjacent to eloquent cortex, due to the relatively small number of cases involving FCD localized to this region, our study is limited by sample size (n = 29). Furthermore, while we are among the first to illustrate the capacity of iMRI to aid in FCD eloquent cortex resections, the median follow-up duration for iMRI-assisted resections (5.72 months) is relatively short compared with conventional resections. Of note, however, we found that for the 11 patients who suffered seizure recurrence following conventional FCD resection, the median duration to postoperative seizure recurrence was 2.23 months. Finally, although microsurgery was used in all cases of our study, our results may be confounded by advanced experience over time and in the utilization (or lack thereof) of other neurosurgical assistive technologies. Additionally, differences in surgeons performing the conventional and iMRI-assisted surgeries may present a limitation of operator dependency bias. Furthermore, differences in radiological interpretation from 1.5-T versus 3.0-T MRI may have skewed our results. The current study may benefit from a multiinstitutional study with a larger cohort of patients and longer follow-up duration.

Conclusions

The results of this study suggest that in comparison with a conventional surgical protocol and technique for resection of FCD localized adjacent to eloquent cortex, the incorporation of iMRI aids in achieving greater rates of GTR and complete seizure freedom. Additionally, iMRI-assisted surgeries are associated with reduced rates of neurological deficit in the functional cortex. When paired with neurophysiological monitoring, iMRI may provide additional information to guide intraoperative decision making in regions of eloquent cortex.

References

  • 1

    Behdad ALimbrick DD JrBertrand MESmyth MD: Epilepsy surgery in children with seizures arising from the rolandic cortex. Epilepsia 50:145014612009

    • Search Google Scholar
    • Export Citation
  • 2

    Benifla MSala F JrJane JOtsubo HOchi ADrake J: Neurosurgical management of intractable rolandic epilepsy in children: role of resection in eloquent cortex. Clinical article. J Neurosurg Pediatr 4:1992162009

    • Search Google Scholar
    • Export Citation
  • 3

    Buchfelder MFahlbusch RGanslandt OStefan HNimsky C: Use of intraoperative magnetic resonance imaging in tailored temporal lobe surgeries for epilepsy. Epilepsia 43:8648732002

    • Search Google Scholar
    • Export Citation
  • 4

    Buchfelder MGanslandt OFahlbusch RNimsky C: Intraoperative magnetic resonance imaging in epilepsy surgery. J Magn Reson Imaging 12:5475552000

    • Search Google Scholar
    • Export Citation
  • 5

    Chan HWPressler RUff CGunny RSt Piers KCross H: A novel technique of detecting MRI-negative lesion in focal symptomatic epilepsy: intraoperative ShearWave elastography. Epilepsia 55:e30e332014

    • Search Google Scholar
    • Export Citation
  • 6

    Cohen-Gadol AAOzduman KBronen RAKim JHSpencer DD: Long-term outcome after epilepsy surgery for focal cortical dysplasia. J Neurosurg 101:55652004

    • Search Google Scholar
    • Export Citation
  • 7

    Cui ZQLing ZPSong HFHu SSun GCChen XL: Combining pyramidal tract mapping, microscopic-based neuronavigation, and intraoperative magnetic resonance imaging improves outcome of epilepsy foci resection in the sensorimotor cortex. Turk Neurosurg 24:5385452014

    • Search Google Scholar
    • Export Citation
  • 8

    de Oliveira RSSantos MVTerra VCSakamoto ACMachado HR: Tailored resections for intractable rolandic cortex epilepsy in children: a single-center experience with 48 consecutive cases. Childs Nerv Syst 27:7797852011

    • Search Google Scholar
    • Export Citation
  • 9

    Delev DSend KWagner Jvon Lehe MOrmond DRSchramm J: Epilepsy surgery of the rolandic and immediate perirolandic cortex: surgical outcome and prognostic factors. Epilepsia 55:158515932014

    • Search Google Scholar
    • Export Citation
  • 10

    Engel JVan Ness PCRasmussen TBOjemann LMOutcome with respect to epileptic seizures. Engel J: Surgical Treatment of the Epilepsies ed 2New YorkRaven Press1993. 609621

    • Search Google Scholar
    • Export Citation
  • 11

    Fauser SBast TAltenmüller DMSchulte-Mönting JStrobl KSteinhoff BJ: Factors influencing surgical outcome in patients with focal cortical dysplasia. J Neurol Neurosurg Psychiatry 79:1031052008

    • Search Google Scholar
    • Export Citation
  • 12

    Hader WJMackay MOtsubo HChitoku SWeiss SBecker L: Cortical dysplastic lesions in children with intractable epilepsy: role of complete resection. J Neurosurg 100:2 Suppl Pediatrics1101172004

    • Search Google Scholar
    • Export Citation
  • 13

    Hatiboglu MAWeinberg JSSuki DTummala SRao GSawaya R: Utilization of intraoperative motor mapping in glioma surgery with high-field intraoperative magnetic resonance imaging. Stereotact Funct Neurosurg 88:3453522010

    • Search Google Scholar
    • Export Citation
  • 14

    Kim DWLee SKChu KPark KILee SYLee CH: Predictors of surgical outcome and pathologic considerations in focal cortical dysplasia. Neurology 72:2112162009

    • Search Google Scholar
    • Export Citation
  • 15

    Krishnan RRaabe AHattingen ESzelényi AYahya HHermann E: Functional magnetic resonance imaging-integrated neuronavigation: Correlation between lesion-to-motor cortex distance and outcome. Neurosurgery 55:9049152004

    • Search Google Scholar
    • Export Citation
  • 16

    Krsek PMaton BJayakar PDean PKorman BRey G: Incomplete resection of focal cortical dysplasia is the main predictor of poor postsurgical outcome. Neurology 72:2172232009

    • Search Google Scholar
    • Export Citation
  • 17

    Lüsebrink FWollrab ASpeck O: Cortical thickness determination of the human brain using high resolution 3T and 7T MRI data. Neuroimage 70:1221312013

    • Search Google Scholar
    • Export Citation
  • 18

    Miller DKnake SBauer SKrakow KPagenstecher ASure U: Intraoperative ultrasound to define focal cortical dysplasia in epilepsy surgery. Epilepsia 49:1561582008

    • Search Google Scholar
    • Export Citation
  • 19

    Nimsky CGanslandt OBuchfelder MFahlbusch R: Intraoperative visualization for resection of gliomas: the role of functional neuronavigation and intraoperative 1.5 T MRI. Neurol Res 28:4824872006

    • Search Google Scholar
    • Export Citation
  • 20

    Nimsky CGanslandt OCerny SHastreiter PGreiner GFahlbusch R: Quantification of, visualization of, and compensation for brain shift using intraoperative magnetic resonance imaging. Neurosurgery 47:107010802000

    • Search Google Scholar
    • Export Citation
  • 21

    Nimsky CGanslandt OGralla JBuchfelder MFahlbusch R: Intraoperative low-field magnetic resonance imaging in pediatric neurosurgery. Pediatr Neurosurg 38:83892003

    • Search Google Scholar
    • Export Citation
  • 22

    Nimsky CGanslandt OHastreiter PFahlbusch R: Intraoperative compensation for brain shift. Surg Neurol 56:3573652001

  • 23

    Oluigbo COWang JWhitehead MTMagge SMyseros JSYaun A: The influence of lesion volume, perilesion resection volume, and completeness of resection on seizure outcome after resective epilepsy surgery for cortical dysplasia in children. J Neurosurg Pediatr 15:6446502015

    • Search Google Scholar
    • Export Citation
  • 24

    Otsubo HChitoku SOchi AJay VRutka JTSmith ML: Malignant rolandicsylvian epilepsy in children: diagnosis, treatment, and outcomes. Neurology 57:5905962001

    • Search Google Scholar
    • Export Citation
  • 25

    Pondal-Sordo MDiosy DTéllez-Zenteno JFGirvin JPWiebe S: Epilepsy surgery involving the sensory-motor cortex. Brain 129:330733142006

    • Search Google Scholar
    • Export Citation
  • 26

    Rasmussen IA JrLindseth FRygh OMBerntsen EMSelbekk TXu J: Functional neuronavigation combined with intraoperative 3D ultrasound: initial experiences during surgical resections close to eloquent brain areas and future directions in automatic brain shift compensation of preoperative data. Acta Neurochir (Wien) 149:3653782007

    • Search Google Scholar
    • Export Citation
  • 27

    Rowland NCEnglot DJCage TASughrue MEBarbaro NMChang EF: A meta-analysis of predictors of seizure freedom in the surgical management of focal cortical dysplasia. J Neurosurg 116:103510412012

    • Search Google Scholar
    • Export Citation
  • 28

    Sarkis RAJehi LEBingaman WENajm IM: Surgical outcome following resection of rolandic focal cortical dysplasia. Epilepsy Res 90:2402472010

    • Search Google Scholar
    • Export Citation
  • 29

    Senft CGlioma surgery: intraoperative low field magnetic resonance imaging. Hayat MA: Tumors of the Central Nervous System. Gliomas: Glioblastoma DordrechtSpringer2011. 2:180188

    • Search Google Scholar
    • Export Citation
  • 30

    Senft CForster MTBink AMittelbronn MFranz KSeifert V: Optimizing the extent of resection in eloquently located gliomas by combining intraoperative MRI guidance with intraoperative neurophysiological monitoring. J Neurooncol 109:81902012

    • Search Google Scholar
    • Export Citation
  • 31

    Senft CFranz KBlasel SOszvald ARathert JSeifert V: Influence of iMRI-guidance on the extent of resection and survival of patients with glioblastoma multiforme. Technol Cancer Res Treat 9:3393462010

    • Search Google Scholar
    • Export Citation
  • 32

    Sommer BGrummich PCoras RKasper BSBlumcke IHamer HM: Integration of functional neuronavigation and intraoperative MRI in surgery for drug-resistant extratemporal epilepsy close to eloquent brain areas. Neurosurg Focus 34:4E42013

    • Search Google Scholar
    • Export Citation

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Oluigbo, Sacino, Gaillard. Acquisition of data: Oluigbo, Sacino. Analysis and interpretation of data: Oluigbo, Sacino. Drafting the article: Oluigbo, Sacino. Critically revising the article: Oluigbo, Sacino. Reviewed submitted version of manuscript: all authors. Statistical analysis: Oluigbo, Sacino. Study supervision: Oluigbo.

If the inline PDF is not rendering correctly, you can download the PDF file here.

Article Information

INCLUDE WHEN CITING DOI: 10.3171/2016.1.FOCUS15538.

Correspondence Chima O. Oluigbo, Departments of Neurosurgery, Children's National Medical Center, 111 Michigan Ave. NW, Washington, DC 20010. email: coluigbo@cnmc.org.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Pie charts illustrating postoperative seizure freedom at the last postoperative follow-up evaluation in the iMRI-assisted resection cohort (upper) and the conventional resection cohort (lower).

  • View in gallery

    Postoperative neurological deficits in the iMRI-assisted and conventional resection cohorts.

  • View in gallery

    Case 8. Preoperative MR images. A: Axial T2-weighted image showing left-sided heterotopia (outline) just behind the left central sulcus. B: Axial T2-weighted image showing the deeper aspects of the left transmantle heterotopia just behind the left internal capsule and left thalamus. C: Sagittal T1-weighted image showing left transmantle heterotopia (note a congenital meningeal cyst posterior to the heterotopia).

  • View in gallery

    Case 8. Axial intraoperative T2-weighted MR image at the first surgery showing residual heterotopia (outlined) adjacent to the left internal capsule. Surgery was discontinued at this stage due to changes in MEPs.

  • View in gallery

    Case 8. Left: Axial intraoperative T2-weighted MR image at the second surgery showing complete resection of the residual heterotopia. Right: Sagittal intraoperative T2-weighted MR image at the second surgery showing complete resection of the residual heterotopia.

References

  • 1

    Behdad ALimbrick DD JrBertrand MESmyth MD: Epilepsy surgery in children with seizures arising from the rolandic cortex. Epilepsia 50:145014612009

    • Search Google Scholar
    • Export Citation
  • 2

    Benifla MSala F JrJane JOtsubo HOchi ADrake J: Neurosurgical management of intractable rolandic epilepsy in children: role of resection in eloquent cortex. Clinical article. J Neurosurg Pediatr 4:1992162009

    • Search Google Scholar
    • Export Citation
  • 3

    Buchfelder MFahlbusch RGanslandt OStefan HNimsky C: Use of intraoperative magnetic resonance imaging in tailored temporal lobe surgeries for epilepsy. Epilepsia 43:8648732002

    • Search Google Scholar
    • Export Citation
  • 4

    Buchfelder MGanslandt OFahlbusch RNimsky C: Intraoperative magnetic resonance imaging in epilepsy surgery. J Magn Reson Imaging 12:5475552000

    • Search Google Scholar
    • Export Citation
  • 5

    Chan HWPressler RUff CGunny RSt Piers KCross H: A novel technique of detecting MRI-negative lesion in focal symptomatic epilepsy: intraoperative ShearWave elastography. Epilepsia 55:e30e332014

    • Search Google Scholar
    • Export Citation
  • 6

    Cohen-Gadol AAOzduman KBronen RAKim JHSpencer DD: Long-term outcome after epilepsy surgery for focal cortical dysplasia. J Neurosurg 101:55652004

    • Search Google Scholar
    • Export Citation
  • 7

    Cui ZQLing ZPSong HFHu SSun GCChen XL: Combining pyramidal tract mapping, microscopic-based neuronavigation, and intraoperative magnetic resonance imaging improves outcome of epilepsy foci resection in the sensorimotor cortex. Turk Neurosurg 24:5385452014

    • Search Google Scholar
    • Export Citation
  • 8

    de Oliveira RSSantos MVTerra VCSakamoto ACMachado HR: Tailored resections for intractable rolandic cortex epilepsy in children: a single-center experience with 48 consecutive cases. Childs Nerv Syst 27:7797852011

    • Search Google Scholar
    • Export Citation
  • 9

    Delev DSend KWagner Jvon Lehe MOrmond DRSchramm J: Epilepsy surgery of the rolandic and immediate perirolandic cortex: surgical outcome and prognostic factors. Epilepsia 55:158515932014

    • Search Google Scholar
    • Export Citation
  • 10

    Engel JVan Ness PCRasmussen TBOjemann LMOutcome with respect to epileptic seizures. Engel J: Surgical Treatment of the Epilepsies ed 2New YorkRaven Press1993. 609621

    • Search Google Scholar
    • Export Citation
  • 11

    Fauser SBast TAltenmüller DMSchulte-Mönting JStrobl KSteinhoff BJ: Factors influencing surgical outcome in patients with focal cortical dysplasia. J Neurol Neurosurg Psychiatry 79:1031052008

    • Search Google Scholar
    • Export Citation
  • 12

    Hader WJMackay MOtsubo HChitoku SWeiss SBecker L: Cortical dysplastic lesions in children with intractable epilepsy: role of complete resection. J Neurosurg 100:2 Suppl Pediatrics1101172004

    • Search Google Scholar
    • Export Citation
  • 13

    Hatiboglu MAWeinberg JSSuki DTummala SRao GSawaya R: Utilization of intraoperative motor mapping in glioma surgery with high-field intraoperative magnetic resonance imaging. Stereotact Funct Neurosurg 88:3453522010

    • Search Google Scholar
    • Export Citation
  • 14

    Kim DWLee SKChu KPark KILee SYLee CH: Predictors of surgical outcome and pathologic considerations in focal cortical dysplasia. Neurology 72:2112162009

    • Search Google Scholar
    • Export Citation
  • 15

    Krishnan RRaabe AHattingen ESzelényi AYahya HHermann E: Functional magnetic resonance imaging-integrated neuronavigation: Correlation between lesion-to-motor cortex distance and outcome. Neurosurgery 55:9049152004

    • Search Google Scholar
    • Export Citation
  • 16

    Krsek PMaton BJayakar PDean PKorman BRey G: Incomplete resection of focal cortical dysplasia is the main predictor of poor postsurgical outcome. Neurology 72:2172232009

    • Search Google Scholar
    • Export Citation
  • 17

    Lüsebrink FWollrab ASpeck O: Cortical thickness determination of the human brain using high resolution 3T and 7T MRI data. Neuroimage 70:1221312013

    • Search Google Scholar
    • Export Citation
  • 18

    Miller DKnake SBauer SKrakow KPagenstecher ASure U: Intraoperative ultrasound to define focal cortical dysplasia in epilepsy surgery. Epilepsia 49:1561582008

    • Search Google Scholar
    • Export Citation
  • 19

    Nimsky CGanslandt OBuchfelder MFahlbusch R: Intraoperative visualization for resection of gliomas: the role of functional neuronavigation and intraoperative 1.5 T MRI. Neurol Res 28:4824872006

    • Search Google Scholar
    • Export Citation
  • 20

    Nimsky CGanslandt OCerny SHastreiter PGreiner GFahlbusch R: Quantification of, visualization of, and compensation for brain shift using intraoperative magnetic resonance imaging. Neurosurgery 47:107010802000

    • Search Google Scholar
    • Export Citation
  • 21

    Nimsky CGanslandt OGralla JBuchfelder MFahlbusch R: Intraoperative low-field magnetic resonance imaging in pediatric neurosurgery. Pediatr Neurosurg 38:83892003

    • Search Google Scholar
    • Export Citation
  • 22

    Nimsky CGanslandt OHastreiter PFahlbusch R: Intraoperative compensation for brain shift. Surg Neurol 56:3573652001

  • 23

    Oluigbo COWang JWhitehead MTMagge SMyseros JSYaun A: The influence of lesion volume, perilesion resection volume, and completeness of resection on seizure outcome after resective epilepsy surgery for cortical dysplasia in children. J Neurosurg Pediatr 15:6446502015

    • Search Google Scholar
    • Export Citation
  • 24

    Otsubo HChitoku SOchi AJay VRutka JTSmith ML: Malignant rolandicsylvian epilepsy in children: diagnosis, treatment, and outcomes. Neurology 57:5905962001

    • Search Google Scholar
    • Export Citation
  • 25

    Pondal-Sordo MDiosy DTéllez-Zenteno JFGirvin JPWiebe S: Epilepsy surgery involving the sensory-motor cortex. Brain 129:330733142006

    • Search Google Scholar
    • Export Citation
  • 26

    Rasmussen IA JrLindseth FRygh OMBerntsen EMSelbekk TXu J: Functional neuronavigation combined with intraoperative 3D ultrasound: initial experiences during surgical resections close to eloquent brain areas and future directions in automatic brain shift compensation of preoperative data. Acta Neurochir (Wien) 149:3653782007

    • Search Google Scholar
    • Export Citation
  • 27

    Rowland NCEnglot DJCage TASughrue MEBarbaro NMChang EF: A meta-analysis of predictors of seizure freedom in the surgical management of focal cortical dysplasia. J Neurosurg 116:103510412012

    • Search Google Scholar
    • Export Citation
  • 28

    Sarkis RAJehi LEBingaman WENajm IM: Surgical outcome following resection of rolandic focal cortical dysplasia. Epilepsy Res 90:2402472010

    • Search Google Scholar
    • Export Citation
  • 29

    Senft CGlioma surgery: intraoperative low field magnetic resonance imaging. Hayat MA: Tumors of the Central Nervous System. Gliomas: Glioblastoma DordrechtSpringer2011. 2:180188

    • Search Google Scholar
    • Export Citation
  • 30

    Senft CForster MTBink AMittelbronn MFranz KSeifert V: Optimizing the extent of resection in eloquently located gliomas by combining intraoperative MRI guidance with intraoperative neurophysiological monitoring. J Neurooncol 109:81902012

    • Search Google Scholar
    • Export Citation
  • 31

    Senft CFranz KBlasel SOszvald ARathert JSeifert V: Influence of iMRI-guidance on the extent of resection and survival of patients with glioblastoma multiforme. Technol Cancer Res Treat 9:3393462010

    • Search Google Scholar
    • Export Citation
  • 32

    Sommer BGrummich PCoras RKasper BSBlumcke IHamer HM: Integration of functional neuronavigation and intraoperative MRI in surgery for drug-resistant extratemporal epilepsy close to eloquent brain areas. Neurosurg Focus 34:4E42013

    • Search Google Scholar
    • Export Citation

TrendMD

Cited By

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 328 314 24
PDF Downloads 221 199 8
EPUB Downloads 0 0 0

PubMed

Google Scholar