Article Text
PDF
Statistics from Altmetric.com
Introduction
Both 7-Tesla (7T) and 3-Tesla (3T) MRI sequences can be used for identification of the subthalamic nucleus (STN) in deep brain stimulation (DBS) for Parkinson’s disease.1
On our 7T T2 sequence the dorsal STN-border is clearly shown but the ventral STN border cannot be distinguished sufficiently from the substantia nigra (SN).1 The 3T Fast Grey Matter Acquisition T1 Inversion Recovery (FGATIR) and other 3T sequences offer a clear contrast between the ventral border of the STN and the SN2
Several publications have found excellent correlation between the MR-visualised STN and microelectrode recordings (MER). Here, we present a novel combination of 7T T2 (dorsal border) and 3T FGATIR MRI (ventral border) which also accomplishes this aim.
Methods
Patients
Data were collected from all patients undergoing DBS surgery at our institution between November 2019 and October 2020 in whom 7T T2 and 3T FGATIR MRI were available in addition to the standard 3T T1 and T2 sequences used for DBS planning.
All patients gave consent for the extra 7T T2 MRI and 3T FGATIR sequence acquisition.
Image acquisition and surgical procedure
The extra MRI sequences were: (1) 7T 3D T2-weighted and (2) 3T 3D FGATIR. Stereotactic registration and electrode localisation were done using intraoperative CT. See online supplemental methods for imaging parameters and surgical procedure.3
Supplemental material
STN borders determined by MER
MERs were started 6 mm above the calculated target and advanced with 0.5 mm steps until the electrophysiological STN signal ended. The central MER trajectories were used in the analyses. Electrophysiological activity of the STN were recognised as a typical broadening of background signal with tonic and irregular discharge patterns and occasional bursts. The ventral border was marked by the disappearance of this typical pattern. The dorsal and ventral electrophysiological STN borders were expressed in millimetre distance to the calculated target
STN borders determined by 7T T2 and 3T FGATIR MRI
STN border identification on 7T T2 and 3T FGATIR was done postoperatively for this analysis using the Brainlab Elements software (Brainlab AG, Munich, Germany). The trajectory of the microelectrode used for the analysis was placed as depicted by the intraoperative cone-beam CT. Border determination on the MRI sequences was done using trajectory-view, axial and coronal orientated 7T T2 and 3T FGATIR. Distances were expressed in millimetres relative to the target depth. For dorsal border identification, the trajectory-view image was advanced (regularly moving backward and forward) in 0.5 mm steps until the hypointense STN depiction was noted (figure 1A) on the 7T T2. For identification of the ventral border, 7T T2 was used for acquiring an indication of STN ending (figure 1C). Subsequently definite border determination was done using the 3T FGATIR MRI (in 0.5 mm steps) until the hyperintense signal of the SN (compared with hypointense STN) became visible (figure 1D). The FGATIR sequence was used for identification of the border with the SN only.
Overview of the mesencephalic area showing STN and substantia nigra (SN) on 7 T T2 and 3T FGATIR MR and corresponding MER. The first row of four panels (vertical orientation, A1.1–D1.1) shows an in-line trajectory view at various depths. The second row (A1.2–D1.2) shows a corresponding coronal view. The third row (A1.3–D1.3) shows the same coronal imaging with red contouring indicating either STN or SN borders. (A) show a 7 Tesla T2 MRI with the electrode (yellow dot) located on the dorsal STN border, (B) show a 7 Tesla T2 MRI with the electrode on target level, (C) show a 7 Tesla T2 MRI with the electrode on the ventral STN border and (D) show a 3 Tesla FGATIR MRI again with the electrode on the ventral STN border. E1.1 represents the corresponding MER trajectory, expressed in millimetre relative to the target depth. Negative distances dorsal to the target depth and positive distances ventral to the target depth were used. During MER, 0.5 mm steps were used until electrophysiological STN signal appeared or disappeared (a typical broadening of background) noise with tonic and irregular discharge pattern and occasional burst or a complete disappearance of this typical pattern and the appearance of more regular tonic activity of smaller amplitudes (SN signal) to determine the dorsal and ventral border, respectively. Typical electrophysiological STN signal starts (representing the dorsal border) at −4.0 mm and ends at +2.0 mm (representing the ventral border). The dorsal border on 7T T2 was identified at −4.0 mm, the ventral border on +2.0 mm using a combination of 7T T2 and 3T FGATIR. In current example MER and MRI borders exactly correspond. For dorsal border identification, the trajectory was advanced (regularly moving backward and forward) in 0.5 mm steps until hypointense STN depiction was noted on the 7T T2 (A). For ventral border identification 7T T2 was used for acquiring an indication of STN ending (C). Subsequently definite localisation was done using the 3T FGATIR MRI as the trajectory was advanced in 0.5 mm steps until the hyperintense signal of the SN was noted (D). Note the difference in SN depiction on 7T T2 (hypointense, not distinguishable from STN, C1.3) and 3T FGATIR (hyperintense, clearly distinguishable from STN, border indicated by red horizontal line, D1.3). Stereotactic Brainlab (Brainlab AG) software was used for visualisation. FGATIR, Fast Grey Matter Acquisition T1 Inversion Recovery; MER, microelectrode recordings; STN, subthalamic nucleus.
Data analysis
The differences in border positions as defined by MER and the MRI sequences are presented as mean±SD. Spearman’s rank correlation was used for the correspondence between STN border depiction by MRI and MER.
Results
A total of 64 tracks were evaluated in 32 patients. There were 17 male patients, average age at surgery was 62±8 years.
In all but one microelectrode trajectory typical STN activity was measured. This single trajectory was excluded from border comparison. It was located outside MRI STN for its entire course due to a 2 mm medial deviation.
Identification of the dorsal STN border was well feasible on 7T T2 and ventral border on 3T FGATIR (figure 1). The average distance from MRI to MER was 0.3 mm (range –1.0 mm to +1.0 mm) and 0.2 mm (range –1.0 mm to +1.0 mm) for dorsal and ventral borders, respectively. The dorsal STN border on MRI coincided with MER in 50% of cases and was located more dorsal in 32% of the trajectories. The ventral border coincided with MER in 61% and was located more dorsal in 24% of the trajectories. Average length of the STN was 6.0±1.3 mm for MER and 6.0±1.2 mm for MRI. The length of the STN corresponded exactly on MRI and MER in 40% of the trajectories.
The Spearman correlation test used the distance of the dorsal and ventral borders relative to the target on MRI and MER (online supplemental figure 1). Significant correlations were found between the dorsal border identification on 7T T2 MRI and MER (r=0.89, p=<0.01) and between the ventral border identification by 3T FGATIR and MER (r=0.93, p<0.01).
Supplemental material
Discussion
We showed that in DBS the dorsal and ventral borders of the STN can be clearly defined using 7T T2 and 3T FGATIR MRI. The distance of these borders to those determined by MER averaged 0.3 mm, with a maximum of 1.0 mm. These differences are small enough to be considered not clinically relevant. This combination of sequences for STN identification has not been reported before.
The consistency in electrophysiological delineation of the STN and that based on imaging is of particular importance when considering replacing MER by solely (single track) MRI-based DBS.
The combination of 7T T2 and 3T FGATIR MRI provides sufficient representation of the expected electrophysiological STN activity. However, these sequences do not depict the exact location of motor segment within the STN. In the future, this may be facilitated by implementing connectivity derived STN segmentation using 7T MRI.
Limitations
The study was not set up in a prospective manner. Second, applicability for other groups would possibly increase when 3T T2 was also included for dorsal STN identification. Third, we did not perform a comparison to an atlas or machine-learning based approach.4 5
Conclusion
We showed that in DBS the dorsal and ventral borders of the (electrophysiological) STN can be clearly defined using 7T T2 and 3T FGATIR MRI.
Ethics statements
Patient consent for publication
Ethics approval
The Medical Ethics Committee of our institution waived formal protocol approval under the Dutch Medical Research Involving Human Subjects Act.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors NR, MB, PvdM and RS have overseen the main writing of this article and implementing of the MRI sequences. WvdZ and MdW have overseen MRI techniques and details. JD and WVP have overseen the electrophysiological matters. RMADB has overseen the neurological theory.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.