Post-operative imaging in deep brain stimulation: A controversial issue

Abstract

In deep brain stimulation (DBS), post-operative imaging has been used on the one hand to assess complications, such as haemorrhage; and on the other hand, to detect misplaced contacts. The post-operative determination of the accurate location of the final electrode plays a critical role in evaluating the precise area of effective stimulation and for predicting the potential clinical outcome; however, safety remains a priority in postoperative DBS imaging. A plethora of diverse post-operative imaging methods have been applied at different centres. There is neither a consensus on the most efficient post-operative imaging methodology, nor is there any standardisation for the automatic or manual analysis of the images within the different imaging modalities. In this article, we give an overview of currently applied post-operative imaging modalities and discuss the current challenges in post-operative imaging in DBS.

Introduction

In deep brain stimulation (DBS), post-operative imaging (POI) is a crucial aspect to assess complications and to evaluate the exact electrode position.^1–8 It is, as well, a critical aspect for fundamental research in studying structure-function relationships, in order to understand the pathophysiology of the underlying disease and to elucidate further the mechanism of action of DBS. The current literature describes an impressive array of methodologies for post-operative contact localisation. While complications such as haemorrhage, brain shift and venous infarction can be identified by magnetic resonance imaging (MRI) or computed tomography (CT), there is a vivid debate over the most suitable methodology in determining post-operatively the precise contact location.^9–11 Intra- or post-operative MRI,^7,12–16,63 or intra- or post-operative CT imaging,^17–19 are both frequently applied methods. The aim of this article is to provide a brief overview on the adopted post-operative DBS imaging methods and to discuss, within the frame of this article, their advantages, limitations and challenges presented.

Methods

A literature search was carried out on PubMed to identify relevant papers published between 1994 and 2016. Only articles in the English language were selected. We used ‘deep brain stimulation’ and terms such as ‘imaging, post-operative contact localisation, fusion, clinical outcome, surgical issues, brain shift, adverse effects, complications’ as keywords in the search. Studies that employed MRI were further classified into two categories, T1-weighted and T2-weighted applications. The initial search included 176 articles. As the quantity and quality of the provided information in these papers varied to a great extent, a systematic analysis was not possible. Many of the papers reviewed did not provide sufficient imaging details, as imaging was a secondary or tertiary concern in the given manuscript. In order to provide an overview on the most common imaging methods and to discuss their advantages and disadvantages, we selected therefore articles that were representative of a specific DBS post-operative imaging modality and that had detailed their post-operative imaging technique.

CT imaging

CT images (Figure 1) manifested with almost no image distortion nor inhomogeneity, but with electrode-induced artefacts; CT imaging requires a shorter image acquisition time, in comparison to MRI. In addition, it is suitable for certain patients who cannot undergo MR scans due to safety concerns^20–24 and also there is no risk of tissue damage due to heating, as for example in MRI²⁵; however, CT entails ionising radiation and provides less tissue contrast, and usually it is required to be co-registered with the pre-operative MRI,^26,27 because the MRI allows a more detailed visualisation of the target structures.

Figure 1.

One day post-operative axial CT, bilateral STN DBS.

CT: computed tomography, DBS: deep brain stimulation, STN: subthalamic nucleus.

MRI imaging

Although it has been reported that post-operative acquisition of MRI images (Figure 2) on patients with implanted DBS devices may result in severe potential hazards,²⁵ recent studies have demonstrated that this procedure can be performed safely without causing any adverse effects, especially for ≤ 1.5 T field strength²⁴; however, it is important to keep in mind that not all currently available DBS stimulating systems on the market are validated for post-operative MRI (for further information, see MRI safety.com). MRI should not be applied with multiple implanted electrodes and stimulators, or only with reduced SAR values.

Figure 2.

One year post-operative axial T2-weighted MRI.

MRI: magnetic resonance imaging.

By way of example, the Medtronic guidelines (http://manuals.medtronic.com/manuals/mri/region) indicate to limit the displayed average head specific absorption rate (SAR) to 0.1 W/kg or less. The limited SAR strongly influences the image quality and limits the usefulness of MRI significantly, although the use of MRI with higher SAR was reported to occur without complications.¹³ More data on MRI safety in DBS is needed.²⁸

Post-operative MRI can be applied as T1-weighted^6,29,30 and as T2-weighted scans.^31,32 Some centres acquire both the T1- and T2- weighted images.³³ T1-weighted and T2-weighted imaging methods have their own advantages and shortcomings, in terms of their use for a post-operative study of DBS electrode positioning. T1-weighted images are less susceptible to the detrimental effect of geometric distortion,³⁴ whereas T2-weighted images exhibit better contrast between subcortical structures with high iron concentration and the surrounding brain tissues, such as in the sub-thalamic nucleus (STN) region.³⁵ In recent years, susceptibility weighted imaging (SWI) has become an important protocol in DBS imaging for the location of for example, the STN, as it in general visualises structures with high iron content better than T2 images do,³⁶ but for post-operative contact localisation it might be less suitable.

There are several factors that need to be taken into account when choosing the scanning parameters for post-operative MRI acquisition, foremost are: the imaging time, patient safety and image quality. As more MRI scanners with high field strengths (≥ 3 T) are available at clinical centres around the world, it is also necessary to consider the benefits and risks of employing scanners with high field strengths (≥ 3 T) for post-operative imaging and to determine the optimal field strength for this type of application.^37–39 More powerful imaging with detailed anatomic delineation, such as high-resolution three-dimensional (3D) T2*-weighted images, 3.0 T T2*-FLASH2D and 7.0 Tesla are currently being discussed^40–42; however, to the best of our knowledge, not a single study has been published that demonstrates that patients implanted with DBS devices can be safely scanned by 3 T MRI.

Fusion

Images of different modalities provide complementary information, thus the fusion of inter-modality images may enable a more detailed illustration of the target region and allow a more straightforward comparison of the pre- and post-operative conditions. On MRI, the electrode induces a massive non-linear distortion artefact that makes the determination of the precise electrode position not possible.

To ascertain the electrode position, several fusion modalities were employed and reported in previous studies, including co-registering of: the intra- or post-operative CT with the pre-operative MRI,^26,27,43 the post-operative MRI with the pre-operative MRI,^44,45 the post-operative CT with the pre-operative CT,^46,47 or the post-operative MRI with the pre- or intra-operative CT.⁴⁸ Promising applications consist in co-registering the pre-operative, intra-operative and post-operative data.⁴⁹ As generally either an affine or a rigid-body transformation is performed, merely linear errors (and translations) can be corrected. Nonlinear errors, such as for example due to brain shift, cannot be accounted for with these simple transformations. Fusion errors of up to 3 mm have been reported.²⁶

Currently there is no evidence that the post-operative CT with stereotactic frame is better than a post-operative CT without frame.

Lead location

The position of the electrode can be determined in relation to the anatomy of the very same patient, such as the placement relative to the target structure and to the fibre tracts. For example, for subthalamic nucleus (STN) targeting, T2-weighted or SWI MRI images are essential in order to visualise a sufficient level of anatomical detail, as CT and T1-weighted MRI images are not able to visualise all details in the STN region.

Most currently-available DBS electrodes have four contacts with different geometries. Each contact is made of a platinum/iridium alloy. The inter-contact spacing can differ, for example 1.5 mm for the Medtronic lead 3387® and 0.5 mm for the 3389® lead, resulting in a total span of 10.5 mm and 7.5 mm, respectively. The most distal part of the electrode (i.e. the very tip) has a length of 1.5 mm and is made of solely plastic/silicone in the Medtronic electrode (in the St. Jude electrode, the tip is made of metal and also represents the contact). The electrode does not contain a metal wire as the remaining electrode segment. The most critical concern in lead location determination continues to be the electrode-induced distortion artefact on post-operative images. The CT distortion artefact is smaller than that on MRI. On MRI, the artefacts present are less pronounced on T1-weighted than on T2-weighted images; however, with the drawback that T1-weighted images display less tissue contrast (especially concerning the STN). T1-weighted images are better for GPi visualisation, while T2-weighted images are more adapted for STN imaging.⁵⁰

Proper steps should be carried out to analyse and correct the artefact, for more accurate identification and evaluation of the electrode positioning.

Although Pollo et al.^15,51 showed that each contact (Medtronic) induces an ellipsoid-shaped artefact on MRI, allowing to deduce from it the contact position (foremost from the most distal contacts), indentations have not been always observed.³⁴ The distal portion of the lead may appear very irregular in the post-operative image. A point to further consider is that the tip of the electrode is not made of a metal wire, but as mentioned above, of sole plastic/silicone (length of 1.5 mm), leading to totally different Hounsfield units/MRI intensities, compared to the metal part. This is important to consider when applying an electrode model to the artefact.

Furthermore, it should be considered that lead location is not a static phenomenon, because brain shift^10,52–56 and re-shift, and the natural course of neurodegenerative diseases, may modify the lead location with time (Figure 3). This might also be an explanation for the eventual diminishing effect of DBS, upon long-term use. To ascertain the position of the contacts, one single scan in the immediate post-operative setting may consequently not be sufficient, and time-delayed post-operative scans may be required as well; however, one should bear in mind that neither in CT nor in MRI does the centre of the electrode artefact correspond to the exact electrode position,¹⁴ as the artefact configuration will differ from the angle of the implanted electrode with respect to the scanning plane. The determination of the electrode position remains a matter of assumption within a range of 1–2 mm.

Figure 3.

One day post-operative axial CT with air, STN DBS.

CT: computed tomography, DBS: deep brain stimulation, STN: subthalamic nucleus.

Conclusions

Postoperative imaging remains a challenging and controversial issue. Safety concerns with an implanted DBS device is of prime concern, while accurate visualisation of implanted contacts remains equally fundamental in the critical interpretation of clinical outcome. The rapid advances in imaging techniques will continue to have a fundamental influence on the practices in functional neurosurgery.

DTI may provide a promising tool in the future, to align the post-operative electrode position in relation to the intra-cerebral fibre tracks⁵⁷; however, the fibre tracking algorithms need validating. Promising non-invasive methods, such as ultrasound imaging for the localisation of deep brain stimulation electrodes⁵⁸; and intra-operative X-ray,⁵⁹ intra-operative 3D X-ray,⁶⁰ Medtronic O-arm,^61,62 intra-operative MRI^12,63 or CT^17,19 may provide useful complementary information to the current post-operative techniques and allow for a more accurate lead identification with fewer side-effects.

At the current standing of applied imaging technology in DBS, post-operative CT may represent a safe tool to assess for complications and to accurately determine the electrode position.

Precise post-operative contact determination is not solely critical for the correct interpretation of the clinical effects, but also for the understanding of the functionality of the stimulated target(s), and ultimately to elucidate further the mechanism of DBS.

Conflict of interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.