Original Article
181
The Role of 3T Magnetic Resonance Imaging for Targeting the Human Subthalamic Nucleus in Deep Brain Stimulation for Parkinson Disease
1 Department of Neurosurgery, Azienda Ospedaliera Universitaria
Integrata di Verona, Verona, Italy 2 Department of Neuroradiology, Azienda Ospedaliera Universitaria Integrata di Verona, Verona, Italy 3 Department of Neurology, Azienda Ospedaliera Universitaria Integrata di Verona, Verona, Italy
Address for correspondence Michele Longhi, MD, PhD, Department of Neurosurgery, Azienda Ospedaliera Universitaria Integrata di Verona, P.le STEFANI, 1, Verona 37126, Italy (e-mail: [email protected]; longhi. [email protected]).
J Neurol Surg A 2015;76:181–189.
Abstract
Keywords
► deep brain stimulation ► 3-Tesla MRI ► subthalamic nucleus targeting
received September 7, 2012 accepted after revision June 7, 2013 published online March 12, 2015
Background Chronic stimulation of the human subthalamic nucleus (STN) is gradually becoming accepted as a long-term therapeutic option for patients with advanced Parkinson disease (PD). 3Tesla (T) magnetic resonance imaging (MRI) improves contrast resolution in basal ganglia nuclei containing high levels of iron, because of magnetic susceptibility effects that increase significantly as the magnetic field gets higher. This phenomenon can be used for better visualization of the STN and may reduce the time necessary for detailed microrecording (MER) mapping, increasing surgery efficacy and lowering morbidity. Objective The objective of this retrospective study is to analyze a population of 20 deep brain stimulation (DBS) electrode implanted patients with PD divided into two groups in which different targeting methods were used. Methods Mean age was 56 years (range 37 to 69 years). Mean disease duration was 11.6 years. Mean follow-up was 12 months (range 6 to 36 months). Patients were divided into two groups: Group A contained 6 patients who underwent STN targeting using 1T stereotactic (T1w þ T2w) MRI plus STN indirect atlas derived targeting. Group B consisted of 14 patients who underwent STN targeting using 3T nonstereotactic (T2w) MRI fused with 1T T1w stereotactic MRI and STN direct targeting. For statistical analysis, we compared (five different parameters in both (matched) groups: Unified Parkinson’s disease rating scale (UPDRS) score reduction (medication off before surgery against stimulation on/medication off after surgery), postoperative drug reduction, duration of surgery, the “central preoperative track” chosen as final implantation track during surgery, and correspondence between the targeted STN and the intraoperative neurophysiologic data. Results Mean UPDRS III score reduction (medication off/stimulation on versus preoperative medication off) was 69% in Group A and 74% in Group B (p ¼ 0.015, log-rank test) respectively. Postoperatively, antiparkinsonian treatment was reduced by 66% in Group A and 75% in Group B (p ¼ 0.006, log-rank test). The preoperative
© 2015 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0033-1354749. ISSN 2193-6315.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Michele Longhi1 Giuseppe Ricciardi2 Giorgio Tommasi3 Antonio Nicolato1 Roberto Foroni1 Laura Bertolasi3 Alberto Beltramello2 Giuseppe Moretto3 Michele Tinazzi3 Massimo Gerosa1
3T MRI in Deep Brain Stimulation for Parkinson Disease
Longhi et al.
“central” track (which corresponds to ideal STN targeting) proved to be the most clinically effective in 2/12 leads for Group A versus 21/28 for Group B (p < 0.001). Neurophysiologic data confirmed these results; the hypothetical target was confirmed by MER data in 76% of tracks in Group A, and in 75% of tracks in Group B (p < 0.001, univariate and multivariate analysis). Conclusion 3T MRI appears to be a useful tool in STN-DBS preoperative targeting. Neurophysiologic testing remains fundamental to determine lead deepness (and prevent clinical side effects.
Introduction Chronic stimulation (deep brain stimulation [DBS]) of the subthalamic nucleus (STN) is gradually becoming accepted as a long-term therapeutic option for patients with advanced Parkinson disease (PD).1–4 For anatomic targeting, various imaging modalities—such as ventriculography, computed tomography (CT), and magnetic resonance imaging (MRI) (direct targeting)5—and intraoperative electrophysiological testing are combined with information from autopsy-based atlases (indirect targeting).6 These atlases provide the coordinates of the STN and other structures relative to established anatomic landmarks (i.e., primarily the commissures and ventricles). In the case of STN localization, this limitation may be overcome by direct visualization of this small but clearly defined nucleus. The combination of intraoperative microrecording (MER) and macrostimulation of STN borders and 1 to 1.5 Tesla (T) MRI has proven to be the most reliable technique for the most effective targeting of STN.7–11 With 3T MRI scanners, the improved contrast resolution from magnetic susceptibility effects at high magnetic field determined by iron storage in basal ganglia nuclei can provide a better visualization of the target.12 This may reduce the need for detailed MER mapping, increase surgery efficacy, and lower side effects and morbiditydue to lead misplacement. Few have used 3T MRI for direct STN targeting, mainly because distortion of peripheral structures is higher at the periphery of the magnetic field, where stereotactic fiducials are placed. This could reduce accuracy in direct targeting.5,13 In this retrospective study we analyzed a population of 20 patients undergoing STN-DBS-implantation (further divided into two groups depending on the targeting method and the used MRI) and statistically analyzed the clinical results.
Patients/Material and Methods Patient Population Between May 2006 and March 2010, 21 patients with advanced PD underwent STN-DBS in the Neurosurgical Department of Verona University Hospital. Mean age was 56 years (range 37–69 years). Male-to-female ratio was 3:1 (15:5). Mean disease duration was 11.6 years. Mean follow-up was 12 months (range 6–36 months). One patient was excluded
Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
from the study due to major surgical morbidity (large left basal ganglia and temporo-insular intracerebral hematoma) that allowed neither bilateral implantation to finish, nor leads position measures to be completed.
Targeting and Surgical Procedure All patients were enrolled in the study after meticulous neurological and neurophysiological screening. Inclusion criteria followed international guidelines included in Core Assessment Program for Surgical Interventional Therapies (CAPSIT). Patients had to fulfill the following criteria: (1) diagnosis of idiopathic PD, (2) age < 70 years, (3) duration of disease > 5 years, (4) dopa-responsive, (5) Unified Parkinson’s Disease Rating Scale (UPDRS) part III (motor examination) score in medication off > 40/108, (6) Hoenn and Yahr in medication off 3, (7) absence of comorbidities (cardiovascular or pulmonary diseases, coagulopathies), (8) absence of MRI vascular or atrophic lesions (►Table 1). Patients were divided into two groups according to the used targeting methods: Group A had 6 patients) underwent 1T stereotactic MRI alone plus STN indirect targeting using atlas coordinated. Group B had 14 patients who underwent 3T nonstereotactic MRI fused with 1T stereotactic MRI followed by STN direct targeting. In group B, one week before surgery a frameless 3T MRI was performed using an acquisition protocol previously tested on PD patients in a clinical setting using 1.5T scanners.7,14 Patients were left free from antiparkinsonian drugs for more than 12 hours before the 3T MRI scan to minimize movement during imaging acquisition. If necessary a bland sedative (midazolam 0.07–0.08 mg/kg, intramuscular) under anesthesiological monitoring was administered. In the scanner head coil the patient’s head was fixed by placing circumaural headphones with large pads. If necessary, depending on different head volumes, additional soft pads were placed between the skin and the coil. MRI acquisition included the following sequences: (1) a gradient-echo medium resolution scout acquired on three planes, (2) a gradient-echo fast coronal acquisition, (3) two axial T2-weighted spin echo image acquisitions of interleaved 2-mm slices without slice gap, in which the first group passed through the anterior commissure-posterior commissure (AC-PC) line and in the second group was
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
182
3T MRI in Deep Brain Stimulation for Parkinson Disease
Longhi et al.
183
Patient
Sex
Age at disease onset (yrs)
Disease duration (yrs)
Presentation
DBS Date
Age at DBS
1
M
55
15
Rigidity
11/05/2005
69
2
M
46
11
Bradykinesia
08/06/2005
57
3
M
47
11
Bradykinesia
01/07/2006
58
4
M
47
10
Rigidity
12/10/2005
55
5
M
48
10
Rigidity
04/05/2006
56
6
M
48
8
Rigidity, bradykinesia
09/11/2005
56
7
F
45
16
Tremor
11/10/2006
61
8
F
50
16
Rigidity
30/05/2007
66
9
F
37
13
Dystonia
18/09/2007
50
10
F
49
10
Rigidity
23/10/2007
59
11
M
40
10
Tremor
12/12/2007
50
12
F
47
19
Tremor
05/03/2008
66
13
M
54
13
Rigidity
16/04/2008
67
14
M
53
20
Tremor
23/07/2008
70
15
M
55
11
Rigidity
14/05/2008
66
16
M
38
10
Bradykinesia
29/10/2008
48
17
M
53
9
Tremor
19/11/2008
62
18
M
31
17
Rigidity
07/02/2009
37
19
M
44
16
Tremor
24/03/2009
60
20
M
43
9
Rigidity
3/05/2009
58
Abbreviation: DBS, deep brain stimulation.
shifted 1 mm superiorly (parameters as follows: repetition time (TR) 2200 millisecondsec, echo time (TE) 90 millisecondsec, 22 slices, thickness 2 mm, matrix 256 192, field of view (FOV) 260 210 mm, flip angle (FA) 90 degrees, number of acquisitions (NEX) 1; acquisition time 21 minute and 16 second for both the groups of slices). These two series, performed to obtain direct visualization of STN, were added together to obtain a series of 44 2-mm slices, positioned every 1 mm, that covered a volume containing both the whole thalamus and the mesencephalic region. The images obtained were sent to the trajectory planning and brain navigation system (Leksell SurgiPlan, Elekta, Stockholm, Sweden; StealthStation, Medtronic, Minneapolis, Minnesota, United States) and kept for the targeting procedure. The day before surgery, after shaving the patient’s head, a Leksell G-Frame (Elekta) was positioned firmly on the skull in a midline position with its base oriented parallel to the AC-PC plane as described in the literature14 The position of the stereotactic system was checked twice by obtaining spherical craniometric measures using a dedicated tool (Elekta) immediately after the frame positioning and just before starting surgery. No frame misplacements were observed in the present study. Care was taken to fix anterior screws at a maximum of 1.5 cm above the eyebrows, thus avoiding any encumbrance during the surgical procedure (frontal approach).
All patients (Group A and Group B), underwent a 1T stereotactic-guided MRI the day before surgery using a unit equipped with an MRI-compatible Leksell G-frame. Three stereotactic MRI acquisition were performed successively: (1) a T1-weighted gradient-echo scout acquired in the three orthogonal planes, (2) a three-dimensional (3D) T1-weighted stereotactic MRI in which 124 contiguous 1.2-mm axial slices were obtained, (3) a 3D angiographic T1-weighted imaging in which 128 axial images were obtained after injection of 0.1 nmol/kg of gadobutrol (TR 45 millisecondsec; TE 8,7 millisecondsec; NEX 1, FA 30 degrees). In Group A targeting was performed both directly on T1weighted images, using as a landmark the anterior border of the red nucleus on the coronal and axial plane and indirectly via the Schaltenbrand stereotactic atlas.15 In this case, the STN position chosen was 12 mm lateral, 1.5 mm posterior, and 2 to 4 mm inferior to the midcommissural point. In Group B, target position was determined directly on 3T direct axial and coronal T2-weighted MRI and reformatted oblique planes, registered and fused by the navigation software (SurgiPlan, Elekta) with 1T stereotactic MRI. STN position was chosen on axial, coronal, and oblique reformatted planes, taking also the anterior border of the red nucleus as landmark. (►Fig. 1). The day after targeting all patients underwent surgery in awake conditions, under the control of vital parameters and free from antiparkinsonian drugs for more than 12 hours. Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Table 1 Patient population data
3T MRI in Deep Brain Stimulation for Parkinson Disease
Longhi et al.
Fig. 1 Direct targeting in group B. (a) T2-weighted spin echo axial and coronal images acquired with a 3 Tesla magnetic resonance scanner in nonstereotactic conditions and registered with T1 volumetric imaging acquired in stereotactic conditions the day before surgery. Leads trajectory is shown in dotted lines; small circles with a cross indicate the target. For each side the circles are colored in green on three different slices. The central slice indicates the target area. (b) The lead trajectory is followed on the axial, coronal, reconstructed sagittal plane and on oblique planes perpendicular to the path parallel and orthogonal to the stereotactic arch (see legend on the gray bar above each picture).
Neurophysiologic intraoperative microelectrode monitoring (MER) was performed in all cases with a mean of three tracks per side, to outline STN activity and borders. The coordinates of the “central track” were directly determined by the image-based targeting, whereas the others were usually positioned 2 mm laterally and/or posteriorly from the first one. Macrostimulation with a 3389 Medtronic quadripolar definitive lead was employed in all cases, to check lead position and to map possible side effects. The optimal site chosen for implantation of the DBS electrode was determined by the presence of (1) characteristic neuronal activity on the STN on the largest millimetric range, (2) motor improvement, and (3) fewr side effects during stimulation with higher intensities. Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
The final depth of the electrode was checked using intraoperative stereotactically (Cross Air Kits, Elekta) guided X-ray images. Mean surgical time was 7 hours for bilateral DBS implants. After surgery, all patients underwent a stereotactic thinslice CT scan to check the position of the leads and the eventual presence of complications (e.g., hemorrhage). In addition, this study was fused with the preoperative MRI to calculate the mean error of targeting based on the stereotactic coordinates before and after lead. Lead extensions and a Kinetra implantable pulse generator (IPG) pacemaker (Medtronic) were placed under general anesthesia 24 to 48 hours after lead implant. Patients were discharged from the hospital on average after 7 days. Follow-up to test UPDRS III and drug reduction
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
184
were performed at 2 weeks, 1 month, 3 months, 6 months, and 1 year, and then twice yearly.
Statistical Analysis For statistic analysis, we matched the two groups with four different parameters: UPDRS score reduction (medication off before surgery against stimulation on/medication off after surgery), postoperative drug reduction, duration of surgery, “central preoperative track” chosen as the final implantation site during surgery, and correspondence between the targeted STN and the intraoperative neurophysiologic data. Univariate analysis was performed using the log-rank test. Subsequently, to test the significance more accurately in multivariate analysis, an ordered logit model was performed (adopting a 95% confidence interval).
Results Mean UPDRS III score reduction (medication off/stimulation on versus preoperative medication off) was 69% in Group A and 74% in Group B. The difference was statistically significant at univariate analysis (p ¼ 0.005), but not at multivariate analysis (p ¼ 0.012). Postoperatively, antiparkinsonian drug treatment, measured in levodopa equivalent daily dose 15 was reduced by 66% in Group A and 75% in Group B. At univariate and multivariate analysis, the difference was statistically significant (p ¼ 0.006, p ¼ 0.009) (►Figs. 2 and 3). The preoperative “central” track proved to be the most clinically effective in 2/12 leads for Group A versus 21/28 leads for Group B (p < 0.001 in univariate and multivariate analysis). In Group A the best track proved to be the lateral in
Longhi et al.
4/12, the posterior in 4/12 and the medial one in 2/12. This difference was statistically significant both in univariate and multivariate analysis. This data was confirmed also by correspondence with neurophysiology: in Group A 16% of the hypothetical target was confirmed by MER data and 75% was confirmed in Group B (p < 0.001 univariate and multivariate analysis). In addition, for Group B, the mean targeting error was calculated by matching postoperative coordinates (obtained by a stereotactic volumetric CT scan fused with preoperative 3T nonstereotactic MRI) of the best performing lead contacts with those obtained during targeting. On axial plane (x coordinate) the mean error was 0.8 mm, on sagittal plane (y coordinate) 0.95 mm, and coronal plance (electrode deepness, z coordinate) 1.3 mm. The statistical analysis results are summarized in ►Table 2.
Discussion Chronic stimulation of the STN is accepted as a long-term therapeutic option for patients with advanced PD.1–4,11 The first step in this procedure is the anatomical targeting of the STN. Two major approaches are commonly used: • Indirect targeting: Various imaging modalities, such as ventriculography, CT, and MRI, and intraoperative electrophysiological testing are combined with information from time-tested autopsy-based atlases. These atlases provide the coordinates of the STN and other structures relative to the established anatomic landmarks (i.e., primarily the commissures and ventricles).5 This kind of approach represents the gold standard in the case of an “invisible” target (structure not definable on
Fig. 2 Graphs indicating Unified Parkinson’s disease rating scale (UPDRS) score reduction pre- and post–deep brain stimulation (DBS) procedure for each patient. (a) Group A. (b) Group B. Precise UPDRS scores are indicated in annexed table. Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
185
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
3T MRI in Deep Brain Stimulation for Parkinson Disease
3T MRI in Deep Brain Stimulation for Parkinson Disease
Longhi et al.
Fig. 3 Graphs indicating antiparkinsonian drug (mg) dosage pre- and post–deep brain stimulation (DBS) procedure for each patient. (a) Group A. (b) Group B. Precise antiparkinsonian drug (mg) dosage indicated in annexed table.
MRI) such as the zona incerta or ventral intermediate nucleus of the thalamus for essential tremor DBS. • Direct targeting: The STN is one of the nuclei of the basal ganglia, a system that has a characteristic hypointense signal intensity on T2-weighted images. This property has been attributed to the presence of iron.9 A progressive increase in iron concentration is a unique characteristic of parts of the basal ganglia system.6 It has been shown that T2 hypointensity in specific gray matter regions is absent in young children but becomes progressively prominent in adulthood .8 Many authors have demonstrated that those “T2 iron maps” can be correlated with anatomical specimens colored with Prussian blue. For this reason, STN can be clearly defined directly, as a small, hypointense structure, on T2-weighted MRI, immediately anterior to the red nucleus rostral border and just above and slightly lateral to the substantia nigra (SNR).13 The combination of 1/1.5T MRI with intraoperative MER mapping of STN borders has proven to be the most reliable technique for the targeting of STN.17
In several studies, however, the authors describe a discrepancy > 2 mm in anterior-posterior (y coordinate) axis of STN between the position of STN seen on the 1/1.5T MRI and that defined during MER. The latter was mainly related to relatively low resolution of STN on imaging.9 In the present study, the mean discrepancy of the anterior-posterior coordinate was 0.95 mm and the lateral error (x-axis) was 0.80 mm. This discrepancy was maximal during the implantation of the second side: in other words, the second lead implanted in each patient was slightly posterior to the first one, probably because of cerebrospinal fluid (CSF) leakage after the first burr hole in the skull and dura mater opening. On the contrary, in the postoperative CT of two cases, we observed an anterior-posterior brain shift only homolateral to the side of implant. This may have been because these were the only two cases of our population in which, because of technical problems such as hardware malfunction, a two-step surgery was performed. In literature it has been demonstrated that the contralateral shift and posterior shift of the AC–PC line are characteristics of unilateral and bilateral procedures, respectively.
Table 2 Statistical analysis Group A
Group B
p value UNIVARIATE
p value MULTIVARIATE
Mean UPDRS III score reduction (medication off/ stimulation on versus preoperative medication off)
69%
74%
0.004
0.01
Postoperative antiparkinsonian treatment reduction (LED)
66%
75%
0.005
0.009
Preoperative “central” track chosen as most effective
2/12
21/28
p < 0.001
p < 0.001
Mean surgery length
8 hours, 14 minutes
6 hours, 23 minutes
0.023
–
Targeting accuracy (correspondence to neurophysiology)
16%
75%
p < 0.001
p < 0.001
Abbreviations: LED, levodopa equivalent daily dose; UPDRS, Unified Parkinson disease rating scale.
Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
186
Terao et al18 demonstrated that the mechanisms of brain shift depend on negative intracranial pressure, atmospheric pressure, and the site of intracranial and extracranial air communication. The same authors demonstrated that the first monolateral procedure for electrode implantation elicits ipsilateral intracranial air invasion through the cranial window and the resultant pressure discrepancy between the hemicranias facilitates the brain shift in the direction contralateral to the cranial window. During the second procedure, if bilateral surgery is performed on the same day, the pressure discrepancy between bilateral hemicranias is lost and the contralateral shift is reset to the midline. At the same time, the anterior-posterior support by the contralateral hemisphere against gravity is lost because of bilateral air invasion, which facilitates a significant posterior shift due to gravity.18,19 As far as the superior-inferior axis is concerned, our mean error of 1.3 mm is the same as that reported in literature.20 This coordinate is considered to be less important in the targeting procedure, especially because of difficulty, on MRI T2 images, in separating the inferior boundary of STN from the SNR. In our experience, this may be overcome by neurophysiological MER mapping, considering that STN has a completely different neuronal firing pattern than the zona incerta above and SNR below; therefore, its boundaries can be identified. Another study showed that, in approximately 10% of cases, the definition of the STN cannot be fully appreciated with a 1T/1.5T MRI.21 In addition, anatomic postmortem studies have confirmed that STN tends to change its borders, size, and shape during aging. On the contrary, the distance between the anterior and posterior commissure remains constant during lifetime.22 These findings once again confirm the variability of STN borders in relation to the commissures and emphasize the need for direct STN visualization as opposed to the atlasbased coordinates. As described in the literature, the high magnetic field of 3T scanners provides improved contrast resolution secondary to microscopic magnetic susceptibility effects caused by iron storage in most of the basal ganglia, which is highly enhanced in T2-weighted imaging. Gradient echo T2 sequences also have the advantage of providing whole-brain 3D high-resolution imaging.12,16,23,24 Recently, 7T scanners have also been used to better define STN borders.25 By using susceptibility-weighted MRI26 and T2-weighted scans 27 at 7T, superior resolution and contrast were found in the basal ganglia region, dramatically improving delineation of the STN. Unfortunately, these advantages come with an important shortcoming—that is, image deformation due to B0 field inhomogeneities, more evident near the skull base and the skull vault, secondary to macroscopic susceptibility artifacts. Although the head coil available in our scanner could fit the Leksell G-Frame and thus would permit acquisition of both T1 high-resolution anatomic images and T2 high-contrast images, deformation was found to increase exponentially outside the field homogeneity volume. This volume was just slightly bigger than the average patient’s head and was
Longhi et al.
not large enough to include the Leksell frame fiducials, whose geometric position was not adequately represented on the anatomic T1 volumetric acquisition. In this study we used conventional T2 SE imaging, which has the disadvantage of a minimum slice thickness of 2 mm and long acquisition times. However, previous studies performed on PD patients have shown that the presence of a 180degree refocusing pulse in conventional SE sequences, although partially reducing susceptibility artifacts and image deformation, still maintains optimal visibility of iron containing nuclei and specifically of the STN.8,14 Although movement and pulsation artifacts were present in T2 SE images acquired in four of our patients, these were never a serious problem in STN identification and were only found once in both the acquired 2 mm 22 slices sequences. A recent review28 has shown that studies do not offer a clear conclusion about the best targeting protocol. Indirect methods are not patient specific, leading to varying results among different cases. On the other hand, direct targeting on MR imaging suffers from lack of contrast within the subthalamic region, resulting in a poor delineation of the STN. Several studies have investigated new sequences at 3T optimized for STN delineation and reducing macroscopic susceptibility artifacts.24,29,30 Unfortunately, nearly all these studies were performed on healthy volunteers only, and none of them has yet been used in a routine clinical setting. Another solution may come from distortion correction by application of field mapping, but this is a mathematical postprocessing whose clinical safety has not yet been demonstrated. Many surgeons when performing electrode implantation for DBS carry out fusion of MRI with CT to overcome the problem of distortion. However, there is evidence that the precision of 1T systems is comparable to that of CT scanners.31,32 3Tscanners, which have become widely available in first-level centers, can be used to better visualize the target, reducing the need for detailed intraoperative MER mapping.12 In the present study, the group of patients with 3T fusion targeting presented a correspondence between preoperative targeting and a significantly higher MER than 1T direct targeting together with indirect targeting. In addition, the preoperative “central” track, which is the ideal targeting trajectory and site of implantation, was chosen as the most suitable in 21/28 leads. In our experience, however, the role of fusion and 1T MRI is crucial for the definition of the trajectory (to avoid vessels, cerebral ventricles, and the like) because the magnetic distortion of image signal at the periphery of the skull is still not known. This could result in a mismatch in the lead implantation trajectory, especially on the brain surface and at a subcortical level. The error resulting from 3T peripheral field inhomogeneity can, in our experience, be reduced by the fusion of images. 3T imaging represents “the core” of targeting, with direct visualization of STN, whereas 1T stereotactic MRI is used to determine the skull entry points and peripheral trajectory. Many centers have tried to increase the accuracy of the procedure by inserting five microelectrodes parallel to each other covering a 4-mm-diameter circle.21 Although this technique does not seem to significantly increase the risk of Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
187
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
3T MRI in Deep Brain Stimulation for Parkinson Disease
3T MRI in Deep Brain Stimulation for Parkinson Disease hemorrhagic complications, fewer passes through the diseased brain may be potentially beneficial in terms of the patient’s outcome.19 Group A patients were implanted bilaterally in a mean time of 8 hours and 14 minutes, Group B in 6 hours and 23 minutes. Even if this difference is not statistically significant, the time saved in locating the STN during surgery, especially if it is done bilaterally in a single session,33 can avoid some complications, such as infections, associated with long procedures. As described in the literature, a long surgical time and multiple passes of microelectrodes, especially through the caudate nucleus, the thalamus, and other deep brain structures, as well as ventriculography, may be the cause of postoperative delirium and other behavioral changes in patients undergoing stereotactic functional procedures for PD.19,34–36 UPDRS III score and Levodopa equivalent daily dose reduction were evaluated by a neurologist in each patient at the scheduled follow-up, and clear neurologic improvement, mainly uncontrollable tremor and rigidity, was documented mainly in Group B patients (UPDRS III score improvement of 75% against 69%). However, the degree of clinical improvement with stimulation varied from patient to patient, and the small sample size made it impossible to find a statistically significant correlation between our approach and the detailed clinical outcome of DBS. In addition, the learning curve of the neurosurgeon concerning this kind of procedure could be a bias. A longer, larger, randomized double-blind study is required to explore the statistical correlation between highfield 3T MRI targeting and the clinical outcome after DBS.
Longhi et al. 2 Limousin P, Krack P, Pollak P, et al. Electrical stimulation of the
3
4
5
6
7
8
9
10
11
12
Conclusions 3T nonstereotactic T2-w MRI appears to be a useful tool in STN-DBS preoperative targeting. Magnetic field distortion should be reduced by fusion with a stereotactic 1 to 1.5T MRI or CT. High spatial and contrast resolution used to directly visualize and locate the STN shortens the surgery time by significantly increasing target calculation precision, which requires only a single pass and fewer tracks for microelectrodes during MER. Stereotactic fusion seems to reduce errors from magnetic distortion in anterior-posterior and lateral-lateral axis. Neurophysiologic monitoring remains fundamental both to determine lead depth (superior-inferior z axis),to clinically confirm the relief of PD symptoms and to detect stimulationbased side effects.
13
14
15 16
17
18
19
Conflict of Interest None.
20
References 1 Benabid AL, Pollak P, Gross C, et al. Acute and long-term effects of
subthalamic nucleus stimulation in Parkinson’s disease. Stereotact Funct Neurosurg 1994;62(1-4):76–84
Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
21
subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 1998;339(16):1105–1111 Obeso JA, Rodriguez-Oroz MC, Rodriguez M, et al. Pathophysiologic basis of surgery for Parkinson’s disease. Neurology 2000;55(12, Suppl 6):S7–S12 Rodriguez-Oroz MC, Obeso JA, Lang AE, et al. Bilateral deep brain stimulation in Parkinson’s disease: a multicentre study with 4 years follow-up. Brain 2005;128(Pt 10):2240–2249 Zonenshayn M, Rezai AR, Mogilner AY, Beric A, Sterio D, Kelly PJ. Comparison of anatomic and neurophysiological methods for subthalamic nucleus targeting. Neurosurgery 2000;47(2):282– 292, discussion 292–294 Morel A, Magnin M, Jeanmonod D. Multiarchitectonic and stereotactic atlas of the human thalamus. J Comp Neurol 1997;387 (4):588–630 Reck C, Maarouf M, Wojtecki L, et al. Clinical outcome of subthalamic stimulation in Parkinsoń s disease is improved by intraoperative multiple trajectories microelectrode recording. J Neurol Surg A 2012;73:377–386 Dormont D, Ricciardi KG, Tandé D, et al. Is the subthalamic nucleus hypointense on T2-weighted images? A correlation study using MR imaging and stereotactic atlas data. AJNR Am J Neuroradiol 2004;25(9):1516–1523 Sterio D, Zonenshayn M, Mogilner AY, et al. Neurophysiological refinement of subthalamic nucleus targeting. Neurosurgery 2002;50(1):58–67, discussion 67–69 Vesper J, Klostermann F, Stockhammer F, Funk T, Brock M. Results of chronic subthalamic nucleus stimulation for Parkinson’s disease: a 1-year follow-up study. Surg Neurol 2002;57(5):306–311, discussion 311–313 Volkmann J, Sturm V, Weiss P, et al. Bilateral high-frequency stimulation of the internal globus pallidus in advanced Parkinson’s disease. Ann Neurol 1998;44(6):953–961 Slavin KV, Thulborn KR, Wess C, Nersesyan H. Direct visualization of the human subthalamic nucleus with 3T MR imaging. AJNR Am J Neuroradiol 2006;27(1):80–84 Zhu XL, Hamel W, Schrader B, et al. Magnetic resonance imagingbased morphometry and landmark correlation of basal ganglia nuclei. Acta Neurochir (Wien) 2002;144(10):959–969, discussion 968–969 Bejjani BP, Dormont D, Pidoux B, et al. Bilateral subthalamic stimulation for Parkinson’s disease by using three-dimensional stereotactic magnetic resonance imaging and electrophysiological guidance. J Neurosurg 2000;92(4):615–625 Schaltenbrand G, Wahren W. Atlas for Stereotaxy of the Human Brain. Stuttgart, Georg Thieme Verlag, 1977. Volz S, Hattingen E, Preibisch C, Gasser T, Deichmann R. Reduction of susceptibility-induced signal losses in multi-gradient-echo images: application to improved visualization of the subthalamic nucleus. Neuroimage 2009;45(4):1135–1143 Hamani C, Richter EO, Andrade-Souza Y, Hutchison W, Saint-Cyr JA, Lozano AM. Correspondence of microelectrode mapping with magnetic resonance imaging for subthalamic nucleus procedures. Surg Neurol 2005;63(3):249–253, discussion 253 Terao T, Okiyama R, Takahashi H, et al. [Comparison and examination of stereotactic surgical complications in movement disorders]. No Shinkei Geka 2003;31(6):629–636 Terao T, Takahashi H, Yokochi F, Taniguchi M, Okiyama R, Hamada I. Hemorrhagic complication of stereotactic surgery in patients with movement disorders. J Neurosurg 2003;98(6):1241–1246 Toda H, Sawamoto N, Hanakawa T, et al. A novel composite targeting method using high-field magnetic resonance imaging for subthalamic nucleus deep brain stimulation. J Neurosurg 2009;111(4):737–745 Benabid AL, Koudsie A, Benazzouz A, Le Bas JF, Pollak P. Imaging of subthalamic nucleus and ventralis intermedius of the thalamus. Mov Disord 2002;17(Suppl 3):S123–S129
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
188
22 Starr PA, Christine CW, Theodosopoulos PV, et al. Implantation of
23
24
25
26
27
28
29
deep brain stimulators into the subthalamic nucleus: technical approach and magnetic resonance imaging-verified lead locations. J Neurosurg 2002;97(2):370–387 Gringel T, Schulz-Schaeffer W, Elolf E, Frölich A, Dechent P, Helms G. Optimized high-resolution mapping of magnetization transfer (MT) at 3 Tesla for direct visualization of substructures of the human thalamus in clinically feasible measurement time. J Magn Reson Imaging 2009;29(6):1285–1292 Vertinsky AT, Coenen VA, Lang DJ, et al. Localization of the subthalamic nucleus: optimization with susceptibility-weighted phase MR imaging. AJNR Am J Neuroradiol 2009;30(9):1717–1724 Duchin Y, Abosch A, Yacoub E, Sapiro G, Harel N. Feasibility of using ultra-high field (7 T) MRI for clinical surgical targeting. PLoS ONE 2012;7(5):e37328 Abosch A, Yacoub E, Ugurbil K, Harel N. An assessment of current brain targets for deep brain stimulation surgery with susceptibility-weighted imaging at 7 tesla. Neurosurgery 2010;67(6):1745– 1756, discussion 1756 Cho ZH, Oh SH, Kim JM, et al. Direct visualization of Parkinson’s disease by in vivo human brain imaging using 7.0T magnetic resonance imaging. Mov Disord 2011;26(4):713–718 Brunenberg EJ, Platel B, Hofman PA, Ter Haar Romeny BM, VisserVandewalle V. Magnetic resonance imaging techniques for visualization of the subthalamic nucleus. J Neurosurg 2011;115(5):971– 984 Elolf E, Bockermann V, Gringel T, Knauth M, Dechent P, Helms G. Improved visibility of the subthalamic nucleus on high-resolution
30
31
32
33
34
35
36
Longhi et al.
stereotactic MR imaging by added susceptibility (T2) contrast using multiple gradient echoes. AJNR Am J Neuroradiol 2007;28 (6):1093–1094 Kerl HU, Gerigk L, Pechlivanis I, Al-Zghloul M, Groden C, Nölte I. The subthalamic nucleus at 3.0 Tesla: choice of optimal sequence and orientation for deep brain stimulation using a standard installation protocol: clinical article. J Neurosurg 2012;117 (6):1155–1165 Mack A, Wolff R, Scheib S, et al. Analyzing 3-tesla magnetic resonance imaging units for implementation in radiosurgery. J Neurosurg 2005;102(Suppl):158–164 Watanabe Y, Lee CK, Gerbi BJ. Geometrical accuracy of a 3-tesla magnetic resonance imaging unit in Gamma Knife surgery. J Neurosurg 2006;105(Suppl):190–193 Liu X, Rowe J, Nandi D, et al. Localisation of the subthalamic nucleus using Radionics Image Fusion and Stereoplan combined with field potential recording. A technical note. Stereotact Funct Neurosurg 2001;76(2):63–73 Beric A, Kelly PJ, Rezai A, et al. Complications of deep brain stimulation surgery. Stereotact Funct Neurosurg 2001;77(1-4): 73–78 Østergaard K, Sunde N, Dupont E. Effects of bilateral stimulation of the subthalamic nucleus in patients with severe Parkinson’s disease and motor fluctuations. Mov Disord 2002;17(4):693–700 Takeshita S, Kurisu K, Trop L, Arita K, Akimitsu T, Verhoeff NP. Effect of subthalamic stimulation on mood state in Parkinson’s disease: evaluation of previous facts and problems. Neurosurg Rev 2005;28 (3):179–186, discussion 187
Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
189
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
3T MRI in Deep Brain Stimulation for Parkinson Disease
Copyright of Journal of Neurological Surgery. Part A. Central European Neurosurgery is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
181
The Role of 3T Magnetic Resonance Imaging for Targeting the Human Subthalamic Nucleus in Deep Brain Stimulation for Parkinson Disease
1 Department of Neurosurgery, Azienda Ospedaliera Universitaria
Integrata di Verona, Verona, Italy 2 Department of Neuroradiology, Azienda Ospedaliera Universitaria Integrata di Verona, Verona, Italy 3 Department of Neurology, Azienda Ospedaliera Universitaria Integrata di Verona, Verona, Italy
Address for correspondence Michele Longhi, MD, PhD, Department of Neurosurgery, Azienda Ospedaliera Universitaria Integrata di Verona, P.le STEFANI, 1, Verona 37126, Italy (e-mail: [email protected]; longhi. [email protected]).
J Neurol Surg A 2015;76:181–189.
Abstract
Keywords
► deep brain stimulation ► 3-Tesla MRI ► subthalamic nucleus targeting
received September 7, 2012 accepted after revision June 7, 2013 published online March 12, 2015
Background Chronic stimulation of the human subthalamic nucleus (STN) is gradually becoming accepted as a long-term therapeutic option for patients with advanced Parkinson disease (PD). 3Tesla (T) magnetic resonance imaging (MRI) improves contrast resolution in basal ganglia nuclei containing high levels of iron, because of magnetic susceptibility effects that increase significantly as the magnetic field gets higher. This phenomenon can be used for better visualization of the STN and may reduce the time necessary for detailed microrecording (MER) mapping, increasing surgery efficacy and lowering morbidity. Objective The objective of this retrospective study is to analyze a population of 20 deep brain stimulation (DBS) electrode implanted patients with PD divided into two groups in which different targeting methods were used. Methods Mean age was 56 years (range 37 to 69 years). Mean disease duration was 11.6 years. Mean follow-up was 12 months (range 6 to 36 months). Patients were divided into two groups: Group A contained 6 patients who underwent STN targeting using 1T stereotactic (T1w þ T2w) MRI plus STN indirect atlas derived targeting. Group B consisted of 14 patients who underwent STN targeting using 3T nonstereotactic (T2w) MRI fused with 1T T1w stereotactic MRI and STN direct targeting. For statistical analysis, we compared (five different parameters in both (matched) groups: Unified Parkinson’s disease rating scale (UPDRS) score reduction (medication off before surgery against stimulation on/medication off after surgery), postoperative drug reduction, duration of surgery, the “central preoperative track” chosen as final implantation track during surgery, and correspondence between the targeted STN and the intraoperative neurophysiologic data. Results Mean UPDRS III score reduction (medication off/stimulation on versus preoperative medication off) was 69% in Group A and 74% in Group B (p ¼ 0.015, log-rank test) respectively. Postoperatively, antiparkinsonian treatment was reduced by 66% in Group A and 75% in Group B (p ¼ 0.006, log-rank test). The preoperative
© 2015 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0033-1354749. ISSN 2193-6315.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Michele Longhi1 Giuseppe Ricciardi2 Giorgio Tommasi3 Antonio Nicolato1 Roberto Foroni1 Laura Bertolasi3 Alberto Beltramello2 Giuseppe Moretto3 Michele Tinazzi3 Massimo Gerosa1
3T MRI in Deep Brain Stimulation for Parkinson Disease
Longhi et al.
“central” track (which corresponds to ideal STN targeting) proved to be the most clinically effective in 2/12 leads for Group A versus 21/28 for Group B (p < 0.001). Neurophysiologic data confirmed these results; the hypothetical target was confirmed by MER data in 76% of tracks in Group A, and in 75% of tracks in Group B (p < 0.001, univariate and multivariate analysis). Conclusion 3T MRI appears to be a useful tool in STN-DBS preoperative targeting. Neurophysiologic testing remains fundamental to determine lead deepness (and prevent clinical side effects.
Introduction Chronic stimulation (deep brain stimulation [DBS]) of the subthalamic nucleus (STN) is gradually becoming accepted as a long-term therapeutic option for patients with advanced Parkinson disease (PD).1–4 For anatomic targeting, various imaging modalities—such as ventriculography, computed tomography (CT), and magnetic resonance imaging (MRI) (direct targeting)5—and intraoperative electrophysiological testing are combined with information from autopsy-based atlases (indirect targeting).6 These atlases provide the coordinates of the STN and other structures relative to established anatomic landmarks (i.e., primarily the commissures and ventricles). In the case of STN localization, this limitation may be overcome by direct visualization of this small but clearly defined nucleus. The combination of intraoperative microrecording (MER) and macrostimulation of STN borders and 1 to 1.5 Tesla (T) MRI has proven to be the most reliable technique for the most effective targeting of STN.7–11 With 3T MRI scanners, the improved contrast resolution from magnetic susceptibility effects at high magnetic field determined by iron storage in basal ganglia nuclei can provide a better visualization of the target.12 This may reduce the need for detailed MER mapping, increase surgery efficacy, and lower side effects and morbiditydue to lead misplacement. Few have used 3T MRI for direct STN targeting, mainly because distortion of peripheral structures is higher at the periphery of the magnetic field, where stereotactic fiducials are placed. This could reduce accuracy in direct targeting.5,13 In this retrospective study we analyzed a population of 20 patients undergoing STN-DBS-implantation (further divided into two groups depending on the targeting method and the used MRI) and statistically analyzed the clinical results.
Patients/Material and Methods Patient Population Between May 2006 and March 2010, 21 patients with advanced PD underwent STN-DBS in the Neurosurgical Department of Verona University Hospital. Mean age was 56 years (range 37–69 years). Male-to-female ratio was 3:1 (15:5). Mean disease duration was 11.6 years. Mean follow-up was 12 months (range 6–36 months). One patient was excluded
Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
from the study due to major surgical morbidity (large left basal ganglia and temporo-insular intracerebral hematoma) that allowed neither bilateral implantation to finish, nor leads position measures to be completed.
Targeting and Surgical Procedure All patients were enrolled in the study after meticulous neurological and neurophysiological screening. Inclusion criteria followed international guidelines included in Core Assessment Program for Surgical Interventional Therapies (CAPSIT). Patients had to fulfill the following criteria: (1) diagnosis of idiopathic PD, (2) age < 70 years, (3) duration of disease > 5 years, (4) dopa-responsive, (5) Unified Parkinson’s Disease Rating Scale (UPDRS) part III (motor examination) score in medication off > 40/108, (6) Hoenn and Yahr in medication off 3, (7) absence of comorbidities (cardiovascular or pulmonary diseases, coagulopathies), (8) absence of MRI vascular or atrophic lesions (►Table 1). Patients were divided into two groups according to the used targeting methods: Group A had 6 patients) underwent 1T stereotactic MRI alone plus STN indirect targeting using atlas coordinated. Group B had 14 patients who underwent 3T nonstereotactic MRI fused with 1T stereotactic MRI followed by STN direct targeting. In group B, one week before surgery a frameless 3T MRI was performed using an acquisition protocol previously tested on PD patients in a clinical setting using 1.5T scanners.7,14 Patients were left free from antiparkinsonian drugs for more than 12 hours before the 3T MRI scan to minimize movement during imaging acquisition. If necessary a bland sedative (midazolam 0.07–0.08 mg/kg, intramuscular) under anesthesiological monitoring was administered. In the scanner head coil the patient’s head was fixed by placing circumaural headphones with large pads. If necessary, depending on different head volumes, additional soft pads were placed between the skin and the coil. MRI acquisition included the following sequences: (1) a gradient-echo medium resolution scout acquired on three planes, (2) a gradient-echo fast coronal acquisition, (3) two axial T2-weighted spin echo image acquisitions of interleaved 2-mm slices without slice gap, in which the first group passed through the anterior commissure-posterior commissure (AC-PC) line and in the second group was
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
182
3T MRI in Deep Brain Stimulation for Parkinson Disease
Longhi et al.
183
Patient
Sex
Age at disease onset (yrs)
Disease duration (yrs)
Presentation
DBS Date
Age at DBS
1
M
55
15
Rigidity
11/05/2005
69
2
M
46
11
Bradykinesia
08/06/2005
57
3
M
47
11
Bradykinesia
01/07/2006
58
4
M
47
10
Rigidity
12/10/2005
55
5
M
48
10
Rigidity
04/05/2006
56
6
M
48
8
Rigidity, bradykinesia
09/11/2005
56
7
F
45
16
Tremor
11/10/2006
61
8
F
50
16
Rigidity
30/05/2007
66
9
F
37
13
Dystonia
18/09/2007
50
10
F
49
10
Rigidity
23/10/2007
59
11
M
40
10
Tremor
12/12/2007
50
12
F
47
19
Tremor
05/03/2008
66
13
M
54
13
Rigidity
16/04/2008
67
14
M
53
20
Tremor
23/07/2008
70
15
M
55
11
Rigidity
14/05/2008
66
16
M
38
10
Bradykinesia
29/10/2008
48
17
M
53
9
Tremor
19/11/2008
62
18
M
31
17
Rigidity
07/02/2009
37
19
M
44
16
Tremor
24/03/2009
60
20
M
43
9
Rigidity
3/05/2009
58
Abbreviation: DBS, deep brain stimulation.
shifted 1 mm superiorly (parameters as follows: repetition time (TR) 2200 millisecondsec, echo time (TE) 90 millisecondsec, 22 slices, thickness 2 mm, matrix 256 192, field of view (FOV) 260 210 mm, flip angle (FA) 90 degrees, number of acquisitions (NEX) 1; acquisition time 21 minute and 16 second for both the groups of slices). These two series, performed to obtain direct visualization of STN, were added together to obtain a series of 44 2-mm slices, positioned every 1 mm, that covered a volume containing both the whole thalamus and the mesencephalic region. The images obtained were sent to the trajectory planning and brain navigation system (Leksell SurgiPlan, Elekta, Stockholm, Sweden; StealthStation, Medtronic, Minneapolis, Minnesota, United States) and kept for the targeting procedure. The day before surgery, after shaving the patient’s head, a Leksell G-Frame (Elekta) was positioned firmly on the skull in a midline position with its base oriented parallel to the AC-PC plane as described in the literature14 The position of the stereotactic system was checked twice by obtaining spherical craniometric measures using a dedicated tool (Elekta) immediately after the frame positioning and just before starting surgery. No frame misplacements were observed in the present study. Care was taken to fix anterior screws at a maximum of 1.5 cm above the eyebrows, thus avoiding any encumbrance during the surgical procedure (frontal approach).
All patients (Group A and Group B), underwent a 1T stereotactic-guided MRI the day before surgery using a unit equipped with an MRI-compatible Leksell G-frame. Three stereotactic MRI acquisition were performed successively: (1) a T1-weighted gradient-echo scout acquired in the three orthogonal planes, (2) a three-dimensional (3D) T1-weighted stereotactic MRI in which 124 contiguous 1.2-mm axial slices were obtained, (3) a 3D angiographic T1-weighted imaging in which 128 axial images were obtained after injection of 0.1 nmol/kg of gadobutrol (TR 45 millisecondsec; TE 8,7 millisecondsec; NEX 1, FA 30 degrees). In Group A targeting was performed both directly on T1weighted images, using as a landmark the anterior border of the red nucleus on the coronal and axial plane and indirectly via the Schaltenbrand stereotactic atlas.15 In this case, the STN position chosen was 12 mm lateral, 1.5 mm posterior, and 2 to 4 mm inferior to the midcommissural point. In Group B, target position was determined directly on 3T direct axial and coronal T2-weighted MRI and reformatted oblique planes, registered and fused by the navigation software (SurgiPlan, Elekta) with 1T stereotactic MRI. STN position was chosen on axial, coronal, and oblique reformatted planes, taking also the anterior border of the red nucleus as landmark. (►Fig. 1). The day after targeting all patients underwent surgery in awake conditions, under the control of vital parameters and free from antiparkinsonian drugs for more than 12 hours. Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Table 1 Patient population data
3T MRI in Deep Brain Stimulation for Parkinson Disease
Longhi et al.
Fig. 1 Direct targeting in group B. (a) T2-weighted spin echo axial and coronal images acquired with a 3 Tesla magnetic resonance scanner in nonstereotactic conditions and registered with T1 volumetric imaging acquired in stereotactic conditions the day before surgery. Leads trajectory is shown in dotted lines; small circles with a cross indicate the target. For each side the circles are colored in green on three different slices. The central slice indicates the target area. (b) The lead trajectory is followed on the axial, coronal, reconstructed sagittal plane and on oblique planes perpendicular to the path parallel and orthogonal to the stereotactic arch (see legend on the gray bar above each picture).
Neurophysiologic intraoperative microelectrode monitoring (MER) was performed in all cases with a mean of three tracks per side, to outline STN activity and borders. The coordinates of the “central track” were directly determined by the image-based targeting, whereas the others were usually positioned 2 mm laterally and/or posteriorly from the first one. Macrostimulation with a 3389 Medtronic quadripolar definitive lead was employed in all cases, to check lead position and to map possible side effects. The optimal site chosen for implantation of the DBS electrode was determined by the presence of (1) characteristic neuronal activity on the STN on the largest millimetric range, (2) motor improvement, and (3) fewr side effects during stimulation with higher intensities. Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
The final depth of the electrode was checked using intraoperative stereotactically (Cross Air Kits, Elekta) guided X-ray images. Mean surgical time was 7 hours for bilateral DBS implants. After surgery, all patients underwent a stereotactic thinslice CT scan to check the position of the leads and the eventual presence of complications (e.g., hemorrhage). In addition, this study was fused with the preoperative MRI to calculate the mean error of targeting based on the stereotactic coordinates before and after lead. Lead extensions and a Kinetra implantable pulse generator (IPG) pacemaker (Medtronic) were placed under general anesthesia 24 to 48 hours after lead implant. Patients were discharged from the hospital on average after 7 days. Follow-up to test UPDRS III and drug reduction
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
184
were performed at 2 weeks, 1 month, 3 months, 6 months, and 1 year, and then twice yearly.
Statistical Analysis For statistic analysis, we matched the two groups with four different parameters: UPDRS score reduction (medication off before surgery against stimulation on/medication off after surgery), postoperative drug reduction, duration of surgery, “central preoperative track” chosen as the final implantation site during surgery, and correspondence between the targeted STN and the intraoperative neurophysiologic data. Univariate analysis was performed using the log-rank test. Subsequently, to test the significance more accurately in multivariate analysis, an ordered logit model was performed (adopting a 95% confidence interval).
Results Mean UPDRS III score reduction (medication off/stimulation on versus preoperative medication off) was 69% in Group A and 74% in Group B. The difference was statistically significant at univariate analysis (p ¼ 0.005), but not at multivariate analysis (p ¼ 0.012). Postoperatively, antiparkinsonian drug treatment, measured in levodopa equivalent daily dose 15 was reduced by 66% in Group A and 75% in Group B. At univariate and multivariate analysis, the difference was statistically significant (p ¼ 0.006, p ¼ 0.009) (►Figs. 2 and 3). The preoperative “central” track proved to be the most clinically effective in 2/12 leads for Group A versus 21/28 leads for Group B (p < 0.001 in univariate and multivariate analysis). In Group A the best track proved to be the lateral in
Longhi et al.
4/12, the posterior in 4/12 and the medial one in 2/12. This difference was statistically significant both in univariate and multivariate analysis. This data was confirmed also by correspondence with neurophysiology: in Group A 16% of the hypothetical target was confirmed by MER data and 75% was confirmed in Group B (p < 0.001 univariate and multivariate analysis). In addition, for Group B, the mean targeting error was calculated by matching postoperative coordinates (obtained by a stereotactic volumetric CT scan fused with preoperative 3T nonstereotactic MRI) of the best performing lead contacts with those obtained during targeting. On axial plane (x coordinate) the mean error was 0.8 mm, on sagittal plane (y coordinate) 0.95 mm, and coronal plance (electrode deepness, z coordinate) 1.3 mm. The statistical analysis results are summarized in ►Table 2.
Discussion Chronic stimulation of the STN is accepted as a long-term therapeutic option for patients with advanced PD.1–4,11 The first step in this procedure is the anatomical targeting of the STN. Two major approaches are commonly used: • Indirect targeting: Various imaging modalities, such as ventriculography, CT, and MRI, and intraoperative electrophysiological testing are combined with information from time-tested autopsy-based atlases. These atlases provide the coordinates of the STN and other structures relative to the established anatomic landmarks (i.e., primarily the commissures and ventricles).5 This kind of approach represents the gold standard in the case of an “invisible” target (structure not definable on
Fig. 2 Graphs indicating Unified Parkinson’s disease rating scale (UPDRS) score reduction pre- and post–deep brain stimulation (DBS) procedure for each patient. (a) Group A. (b) Group B. Precise UPDRS scores are indicated in annexed table. Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
185
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
3T MRI in Deep Brain Stimulation for Parkinson Disease
3T MRI in Deep Brain Stimulation for Parkinson Disease
Longhi et al.
Fig. 3 Graphs indicating antiparkinsonian drug (mg) dosage pre- and post–deep brain stimulation (DBS) procedure for each patient. (a) Group A. (b) Group B. Precise antiparkinsonian drug (mg) dosage indicated in annexed table.
MRI) such as the zona incerta or ventral intermediate nucleus of the thalamus for essential tremor DBS. • Direct targeting: The STN is one of the nuclei of the basal ganglia, a system that has a characteristic hypointense signal intensity on T2-weighted images. This property has been attributed to the presence of iron.9 A progressive increase in iron concentration is a unique characteristic of parts of the basal ganglia system.6 It has been shown that T2 hypointensity in specific gray matter regions is absent in young children but becomes progressively prominent in adulthood .8 Many authors have demonstrated that those “T2 iron maps” can be correlated with anatomical specimens colored with Prussian blue. For this reason, STN can be clearly defined directly, as a small, hypointense structure, on T2-weighted MRI, immediately anterior to the red nucleus rostral border and just above and slightly lateral to the substantia nigra (SNR).13 The combination of 1/1.5T MRI with intraoperative MER mapping of STN borders has proven to be the most reliable technique for the targeting of STN.17
In several studies, however, the authors describe a discrepancy > 2 mm in anterior-posterior (y coordinate) axis of STN between the position of STN seen on the 1/1.5T MRI and that defined during MER. The latter was mainly related to relatively low resolution of STN on imaging.9 In the present study, the mean discrepancy of the anterior-posterior coordinate was 0.95 mm and the lateral error (x-axis) was 0.80 mm. This discrepancy was maximal during the implantation of the second side: in other words, the second lead implanted in each patient was slightly posterior to the first one, probably because of cerebrospinal fluid (CSF) leakage after the first burr hole in the skull and dura mater opening. On the contrary, in the postoperative CT of two cases, we observed an anterior-posterior brain shift only homolateral to the side of implant. This may have been because these were the only two cases of our population in which, because of technical problems such as hardware malfunction, a two-step surgery was performed. In literature it has been demonstrated that the contralateral shift and posterior shift of the AC–PC line are characteristics of unilateral and bilateral procedures, respectively.
Table 2 Statistical analysis Group A
Group B
p value UNIVARIATE
p value MULTIVARIATE
Mean UPDRS III score reduction (medication off/ stimulation on versus preoperative medication off)
69%
74%
0.004
0.01
Postoperative antiparkinsonian treatment reduction (LED)
66%
75%
0.005
0.009
Preoperative “central” track chosen as most effective
2/12
21/28
p < 0.001
p < 0.001
Mean surgery length
8 hours, 14 minutes
6 hours, 23 minutes
0.023
–
Targeting accuracy (correspondence to neurophysiology)
16%
75%
p < 0.001
p < 0.001
Abbreviations: LED, levodopa equivalent daily dose; UPDRS, Unified Parkinson disease rating scale.
Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
186
Terao et al18 demonstrated that the mechanisms of brain shift depend on negative intracranial pressure, atmospheric pressure, and the site of intracranial and extracranial air communication. The same authors demonstrated that the first monolateral procedure for electrode implantation elicits ipsilateral intracranial air invasion through the cranial window and the resultant pressure discrepancy between the hemicranias facilitates the brain shift in the direction contralateral to the cranial window. During the second procedure, if bilateral surgery is performed on the same day, the pressure discrepancy between bilateral hemicranias is lost and the contralateral shift is reset to the midline. At the same time, the anterior-posterior support by the contralateral hemisphere against gravity is lost because of bilateral air invasion, which facilitates a significant posterior shift due to gravity.18,19 As far as the superior-inferior axis is concerned, our mean error of 1.3 mm is the same as that reported in literature.20 This coordinate is considered to be less important in the targeting procedure, especially because of difficulty, on MRI T2 images, in separating the inferior boundary of STN from the SNR. In our experience, this may be overcome by neurophysiological MER mapping, considering that STN has a completely different neuronal firing pattern than the zona incerta above and SNR below; therefore, its boundaries can be identified. Another study showed that, in approximately 10% of cases, the definition of the STN cannot be fully appreciated with a 1T/1.5T MRI.21 In addition, anatomic postmortem studies have confirmed that STN tends to change its borders, size, and shape during aging. On the contrary, the distance between the anterior and posterior commissure remains constant during lifetime.22 These findings once again confirm the variability of STN borders in relation to the commissures and emphasize the need for direct STN visualization as opposed to the atlasbased coordinates. As described in the literature, the high magnetic field of 3T scanners provides improved contrast resolution secondary to microscopic magnetic susceptibility effects caused by iron storage in most of the basal ganglia, which is highly enhanced in T2-weighted imaging. Gradient echo T2 sequences also have the advantage of providing whole-brain 3D high-resolution imaging.12,16,23,24 Recently, 7T scanners have also been used to better define STN borders.25 By using susceptibility-weighted MRI26 and T2-weighted scans 27 at 7T, superior resolution and contrast were found in the basal ganglia region, dramatically improving delineation of the STN. Unfortunately, these advantages come with an important shortcoming—that is, image deformation due to B0 field inhomogeneities, more evident near the skull base and the skull vault, secondary to macroscopic susceptibility artifacts. Although the head coil available in our scanner could fit the Leksell G-Frame and thus would permit acquisition of both T1 high-resolution anatomic images and T2 high-contrast images, deformation was found to increase exponentially outside the field homogeneity volume. This volume was just slightly bigger than the average patient’s head and was
Longhi et al.
not large enough to include the Leksell frame fiducials, whose geometric position was not adequately represented on the anatomic T1 volumetric acquisition. In this study we used conventional T2 SE imaging, which has the disadvantage of a minimum slice thickness of 2 mm and long acquisition times. However, previous studies performed on PD patients have shown that the presence of a 180degree refocusing pulse in conventional SE sequences, although partially reducing susceptibility artifacts and image deformation, still maintains optimal visibility of iron containing nuclei and specifically of the STN.8,14 Although movement and pulsation artifacts were present in T2 SE images acquired in four of our patients, these were never a serious problem in STN identification and were only found once in both the acquired 2 mm 22 slices sequences. A recent review28 has shown that studies do not offer a clear conclusion about the best targeting protocol. Indirect methods are not patient specific, leading to varying results among different cases. On the other hand, direct targeting on MR imaging suffers from lack of contrast within the subthalamic region, resulting in a poor delineation of the STN. Several studies have investigated new sequences at 3T optimized for STN delineation and reducing macroscopic susceptibility artifacts.24,29,30 Unfortunately, nearly all these studies were performed on healthy volunteers only, and none of them has yet been used in a routine clinical setting. Another solution may come from distortion correction by application of field mapping, but this is a mathematical postprocessing whose clinical safety has not yet been demonstrated. Many surgeons when performing electrode implantation for DBS carry out fusion of MRI with CT to overcome the problem of distortion. However, there is evidence that the precision of 1T systems is comparable to that of CT scanners.31,32 3Tscanners, which have become widely available in first-level centers, can be used to better visualize the target, reducing the need for detailed intraoperative MER mapping.12 In the present study, the group of patients with 3T fusion targeting presented a correspondence between preoperative targeting and a significantly higher MER than 1T direct targeting together with indirect targeting. In addition, the preoperative “central” track, which is the ideal targeting trajectory and site of implantation, was chosen as the most suitable in 21/28 leads. In our experience, however, the role of fusion and 1T MRI is crucial for the definition of the trajectory (to avoid vessels, cerebral ventricles, and the like) because the magnetic distortion of image signal at the periphery of the skull is still not known. This could result in a mismatch in the lead implantation trajectory, especially on the brain surface and at a subcortical level. The error resulting from 3T peripheral field inhomogeneity can, in our experience, be reduced by the fusion of images. 3T imaging represents “the core” of targeting, with direct visualization of STN, whereas 1T stereotactic MRI is used to determine the skull entry points and peripheral trajectory. Many centers have tried to increase the accuracy of the procedure by inserting five microelectrodes parallel to each other covering a 4-mm-diameter circle.21 Although this technique does not seem to significantly increase the risk of Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
187
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
3T MRI in Deep Brain Stimulation for Parkinson Disease
3T MRI in Deep Brain Stimulation for Parkinson Disease hemorrhagic complications, fewer passes through the diseased brain may be potentially beneficial in terms of the patient’s outcome.19 Group A patients were implanted bilaterally in a mean time of 8 hours and 14 minutes, Group B in 6 hours and 23 minutes. Even if this difference is not statistically significant, the time saved in locating the STN during surgery, especially if it is done bilaterally in a single session,33 can avoid some complications, such as infections, associated with long procedures. As described in the literature, a long surgical time and multiple passes of microelectrodes, especially through the caudate nucleus, the thalamus, and other deep brain structures, as well as ventriculography, may be the cause of postoperative delirium and other behavioral changes in patients undergoing stereotactic functional procedures for PD.19,34–36 UPDRS III score and Levodopa equivalent daily dose reduction were evaluated by a neurologist in each patient at the scheduled follow-up, and clear neurologic improvement, mainly uncontrollable tremor and rigidity, was documented mainly in Group B patients (UPDRS III score improvement of 75% against 69%). However, the degree of clinical improvement with stimulation varied from patient to patient, and the small sample size made it impossible to find a statistically significant correlation between our approach and the detailed clinical outcome of DBS. In addition, the learning curve of the neurosurgeon concerning this kind of procedure could be a bias. A longer, larger, randomized double-blind study is required to explore the statistical correlation between highfield 3T MRI targeting and the clinical outcome after DBS.
Longhi et al. 2 Limousin P, Krack P, Pollak P, et al. Electrical stimulation of the
3
4
5
6
7
8
9
10
11
12
Conclusions 3T nonstereotactic T2-w MRI appears to be a useful tool in STN-DBS preoperative targeting. Magnetic field distortion should be reduced by fusion with a stereotactic 1 to 1.5T MRI or CT. High spatial and contrast resolution used to directly visualize and locate the STN shortens the surgery time by significantly increasing target calculation precision, which requires only a single pass and fewer tracks for microelectrodes during MER. Stereotactic fusion seems to reduce errors from magnetic distortion in anterior-posterior and lateral-lateral axis. Neurophysiologic monitoring remains fundamental both to determine lead depth (superior-inferior z axis),to clinically confirm the relief of PD symptoms and to detect stimulationbased side effects.
13
14
15 16
17
18
19
Conflict of Interest None.
20
References 1 Benabid AL, Pollak P, Gross C, et al. Acute and long-term effects of
subthalamic nucleus stimulation in Parkinson’s disease. Stereotact Funct Neurosurg 1994;62(1-4):76–84
Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
21
subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 1998;339(16):1105–1111 Obeso JA, Rodriguez-Oroz MC, Rodriguez M, et al. Pathophysiologic basis of surgery for Parkinson’s disease. Neurology 2000;55(12, Suppl 6):S7–S12 Rodriguez-Oroz MC, Obeso JA, Lang AE, et al. Bilateral deep brain stimulation in Parkinson’s disease: a multicentre study with 4 years follow-up. Brain 2005;128(Pt 10):2240–2249 Zonenshayn M, Rezai AR, Mogilner AY, Beric A, Sterio D, Kelly PJ. Comparison of anatomic and neurophysiological methods for subthalamic nucleus targeting. Neurosurgery 2000;47(2):282– 292, discussion 292–294 Morel A, Magnin M, Jeanmonod D. Multiarchitectonic and stereotactic atlas of the human thalamus. J Comp Neurol 1997;387 (4):588–630 Reck C, Maarouf M, Wojtecki L, et al. Clinical outcome of subthalamic stimulation in Parkinsoń s disease is improved by intraoperative multiple trajectories microelectrode recording. J Neurol Surg A 2012;73:377–386 Dormont D, Ricciardi KG, Tandé D, et al. Is the subthalamic nucleus hypointense on T2-weighted images? A correlation study using MR imaging and stereotactic atlas data. AJNR Am J Neuroradiol 2004;25(9):1516–1523 Sterio D, Zonenshayn M, Mogilner AY, et al. Neurophysiological refinement of subthalamic nucleus targeting. Neurosurgery 2002;50(1):58–67, discussion 67–69 Vesper J, Klostermann F, Stockhammer F, Funk T, Brock M. Results of chronic subthalamic nucleus stimulation for Parkinson’s disease: a 1-year follow-up study. Surg Neurol 2002;57(5):306–311, discussion 311–313 Volkmann J, Sturm V, Weiss P, et al. Bilateral high-frequency stimulation of the internal globus pallidus in advanced Parkinson’s disease. Ann Neurol 1998;44(6):953–961 Slavin KV, Thulborn KR, Wess C, Nersesyan H. Direct visualization of the human subthalamic nucleus with 3T MR imaging. AJNR Am J Neuroradiol 2006;27(1):80–84 Zhu XL, Hamel W, Schrader B, et al. Magnetic resonance imagingbased morphometry and landmark correlation of basal ganglia nuclei. Acta Neurochir (Wien) 2002;144(10):959–969, discussion 968–969 Bejjani BP, Dormont D, Pidoux B, et al. Bilateral subthalamic stimulation for Parkinson’s disease by using three-dimensional stereotactic magnetic resonance imaging and electrophysiological guidance. J Neurosurg 2000;92(4):615–625 Schaltenbrand G, Wahren W. Atlas for Stereotaxy of the Human Brain. Stuttgart, Georg Thieme Verlag, 1977. Volz S, Hattingen E, Preibisch C, Gasser T, Deichmann R. Reduction of susceptibility-induced signal losses in multi-gradient-echo images: application to improved visualization of the subthalamic nucleus. Neuroimage 2009;45(4):1135–1143 Hamani C, Richter EO, Andrade-Souza Y, Hutchison W, Saint-Cyr JA, Lozano AM. Correspondence of microelectrode mapping with magnetic resonance imaging for subthalamic nucleus procedures. Surg Neurol 2005;63(3):249–253, discussion 253 Terao T, Okiyama R, Takahashi H, et al. [Comparison and examination of stereotactic surgical complications in movement disorders]. No Shinkei Geka 2003;31(6):629–636 Terao T, Takahashi H, Yokochi F, Taniguchi M, Okiyama R, Hamada I. Hemorrhagic complication of stereotactic surgery in patients with movement disorders. J Neurosurg 2003;98(6):1241–1246 Toda H, Sawamoto N, Hanakawa T, et al. A novel composite targeting method using high-field magnetic resonance imaging for subthalamic nucleus deep brain stimulation. J Neurosurg 2009;111(4):737–745 Benabid AL, Koudsie A, Benazzouz A, Le Bas JF, Pollak P. Imaging of subthalamic nucleus and ventralis intermedius of the thalamus. Mov Disord 2002;17(Suppl 3):S123–S129
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
188
22 Starr PA, Christine CW, Theodosopoulos PV, et al. Implantation of
23
24
25
26
27
28
29
deep brain stimulators into the subthalamic nucleus: technical approach and magnetic resonance imaging-verified lead locations. J Neurosurg 2002;97(2):370–387 Gringel T, Schulz-Schaeffer W, Elolf E, Frölich A, Dechent P, Helms G. Optimized high-resolution mapping of magnetization transfer (MT) at 3 Tesla for direct visualization of substructures of the human thalamus in clinically feasible measurement time. J Magn Reson Imaging 2009;29(6):1285–1292 Vertinsky AT, Coenen VA, Lang DJ, et al. Localization of the subthalamic nucleus: optimization with susceptibility-weighted phase MR imaging. AJNR Am J Neuroradiol 2009;30(9):1717–1724 Duchin Y, Abosch A, Yacoub E, Sapiro G, Harel N. Feasibility of using ultra-high field (7 T) MRI for clinical surgical targeting. PLoS ONE 2012;7(5):e37328 Abosch A, Yacoub E, Ugurbil K, Harel N. An assessment of current brain targets for deep brain stimulation surgery with susceptibility-weighted imaging at 7 tesla. Neurosurgery 2010;67(6):1745– 1756, discussion 1756 Cho ZH, Oh SH, Kim JM, et al. Direct visualization of Parkinson’s disease by in vivo human brain imaging using 7.0T magnetic resonance imaging. Mov Disord 2011;26(4):713–718 Brunenberg EJ, Platel B, Hofman PA, Ter Haar Romeny BM, VisserVandewalle V. Magnetic resonance imaging techniques for visualization of the subthalamic nucleus. J Neurosurg 2011;115(5):971– 984 Elolf E, Bockermann V, Gringel T, Knauth M, Dechent P, Helms G. Improved visibility of the subthalamic nucleus on high-resolution
30
31
32
33
34
35
36
Longhi et al.
stereotactic MR imaging by added susceptibility (T2) contrast using multiple gradient echoes. AJNR Am J Neuroradiol 2007;28 (6):1093–1094 Kerl HU, Gerigk L, Pechlivanis I, Al-Zghloul M, Groden C, Nölte I. The subthalamic nucleus at 3.0 Tesla: choice of optimal sequence and orientation for deep brain stimulation using a standard installation protocol: clinical article. J Neurosurg 2012;117 (6):1155–1165 Mack A, Wolff R, Scheib S, et al. Analyzing 3-tesla magnetic resonance imaging units for implementation in radiosurgery. J Neurosurg 2005;102(Suppl):158–164 Watanabe Y, Lee CK, Gerbi BJ. Geometrical accuracy of a 3-tesla magnetic resonance imaging unit in Gamma Knife surgery. J Neurosurg 2006;105(Suppl):190–193 Liu X, Rowe J, Nandi D, et al. Localisation of the subthalamic nucleus using Radionics Image Fusion and Stereoplan combined with field potential recording. A technical note. Stereotact Funct Neurosurg 2001;76(2):63–73 Beric A, Kelly PJ, Rezai A, et al. Complications of deep brain stimulation surgery. Stereotact Funct Neurosurg 2001;77(1-4): 73–78 Østergaard K, Sunde N, Dupont E. Effects of bilateral stimulation of the subthalamic nucleus in patients with severe Parkinson’s disease and motor fluctuations. Mov Disord 2002;17(4):693–700 Takeshita S, Kurisu K, Trop L, Arita K, Akimitsu T, Verhoeff NP. Effect of subthalamic stimulation on mood state in Parkinson’s disease: evaluation of previous facts and problems. Neurosurg Rev 2005;28 (3):179–186, discussion 187
Journal of Neurological Surgery—Part A
Vol. 76
No. A3/2015
189
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
3T MRI in Deep Brain Stimulation for Parkinson Disease
Copyright of Journal of Neurological Surgery. Part A. Central European Neurosurgery is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
