A systematic review of magnetic resonance imaging in patients with an implanted vagus nerve stimulation system

Purpose Vagus nerve stimulation (VNS) is an effective adjunctive treatment for drug-resistant epilepsy (DRE) and difficult-to-treat depression (DTD). More than 125.000 patients have been implanted with VNS Therapy® System (LivaNova PLC) since initial approval. Patients with DRE often require magnetic resonance imaging (MRI) of the brain during the course of their disease. VNS Therapy System devices are labeled to allow MRI under certain conditions; however, there are no published comprehensive articles about the real-world experience using MRI in patients with implanted VNS devices. Methods A systematic review in accordance with PRISMA statement was performed using PubMed database. Full-length articles reporting MRI (1.5 T or 3 T scanner) of patients with implanted VNS for DRE or DTD and published since 2000 were included. The primary endpoint was a positive outcome that was defined as a technically uneventful MRI scan performed in accordance with the VNS Therapy System manufacturer guidelines and completed according to the researchers’ planned scanning protocol without harm to the patient. Results Twenty-six articles were eligible with 25 articles referring to the VNS Therapy System, and 216 patients were included in the analysis. No serious adverse events or serious device-related adverse events were reported. MRI scan was prematurely terminated in one patient due to a panic attack. Conclusion This systematic review indicates that cranial MRI of patients with an implanted VNS Therapy System can be completed satisfactorily and is tolerable and safe using 1.5 T and 3 T MRI scanners when performed in adherence to the VNS manufacturer’s guidelines. Supplementary Information The online version contains supplementary material available at 10.1007/s00234-021-02705-y.


Introduction
Epilepsy and depression are prevalent neurological and psychiatric diseases that are often associated with a pharmaco-refractory course, high long-term morbidity, and decreased quality of life [1].
In children and adults, epilepsy is associated with a drugresistant course in more than 30% of the patients [2]. Patients with DRE [3] are at risk of physical and psychological comorbidities as well as psychosocial problems in addition to their seizure burden. Among the broad spectrum of comorbidities requiring continuous and comprehensive treatment as well as long-term interdisciplinary care, disturbances of cognition, behavior, and communication as well as psychiatric illness are common in children and adults with DRE [4][5][6]. A variety of brain anomalies, genetic mutations, socio-economic implications, anticonvulsive polytherapy, and inter-ictal epileptiform activity may additionally modify and aggravate the complex course of DRE of various etiologies [3,4,6].
In a multi-center trial of 406 patients with epilepsy and primary generalized tonic-clonic seizures and/or focal to bilateral tonic-clonic seizures, 59.6% of patients had experienced at least one seizure-related accidental injury in the last 12 months with the most common being head injuries (35.%) [7]. A quarter of these patients suffered at least one serious injury, and it has been reported that patients with epilepsy are 2.2-4.8 times more likely to die by some type of accident than the standard population [8].
Seizure-associated accidents and injuries further impair quality of life in patients with DRE and may pose an indication for magnetic resonance imaging (MRI). MRI is commonly performed for epilepsy-related injuries or status epilepticus which are major contributors to poor quality of life and mortality. Patients with DRE may also undergo repetitive MRI of the brain (cMRI) for comprehensive pre-surgical evaluation using advanced techniques of MR scanners [9]. Due to the progressive nature of the disease, changes in clinical symptoms may require performing repeated cMRI, direct intracerebral recordings (e.g., stereotactic electroencephalogram [EEG]) [10], functional MRI [11,12], or novel minimally invasive MRI-guided laser therapies may be needed after VNS implantation [13].
Among adults with depression, more than 25% experience treatment-resistant depression (TRD) that encompasses considerable morbidity and adverse effects on quality of life. TRD comprises failure to respond to two or more antidepressants used at an appropriate dose for an adequate time frame [14]. Since 2001 in the European Union and 2005 in the USA, VNS therapy has also been approved for adjunctive treatment of patients with chronic and recurrent TRD. In addition to standard-of-care treatments with pharmacotherapy, psychotherapy, and electroconvulsive therapy, adjunctive VNS Therapy System has been shown to reduce depressive symptoms, improve quality-of-life, and prevent relapse in patients with TRD [15][16][17]. Major depressive disorder (MDD) is one of the most diagnosed mental disorders in most first world countries, including Europe, China, and the USA. Structural and functional brain alterations are common in patients with MDD. During the last three decades, MRI has played a critical role in deciphering the pathogenesis of this disorder [18].
Since 1994 in Europe and 1997 in the USA, VNS Therapy System has been an approved and well-accepted adjunctive treatment of patients with drug-resistant epilepsy (DRE). More than 125,000 patients have been implanted with VNS Therapy System (data on file; LivaNova PLC).
VNS Therapy Systems are labeled to allow MRI under certain conditions [19] (Supplementary material); however, no comprehensive real-world experience on the use of MRI in patients with implanted VNS systems is available in the published literature. We conducted this systematic review to analyze information on use and safety of MRI scans in patients with an implanted VNS Therapy System for DRE or TRD.

Search strategy and article selection
A literature review (systematic review) was conducted till June 2020 using the PubMed database (https:// pubmed.ncbi.nlm.nih.gov/advanced/). A search syntax strategy was devised using keywords: ("MRI" OR "fMRI" OR "magnetic resonance") AND ("VNS" OR "Vagus nerve stimulator" OR "Vagal nerve stimulator" OR "Vagus nerve stimulation"). In the query box of the PubMed Advanced Search Builder, the following search query was used: ( This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) checklist [20].

Data extraction
Each article was searched for clear evidence that patients underwent an MRI procedure with a 1.5 T or 3 T MRI scanner with an implanted VNS Therapy System for DRE or TRD and labeled for conditional MRI scanning. The primary endpoint was a positive outcome that was defined as a technically uneventful MRI scan performed in accordance with the VNS device manufacturer guidelines and completed according to the researchers' planned scanning protocol without harm to the patient.
Each article was searched for any of the following scanning related adverse events: & Device-related adverse event: Any VNS Therapy Systemrelated non-serious or serious adverse event reported to occur during MRI scanning performed in accordance with manufacturer guideline. & VNS device malfunction: Reported VNS Therapy System device malfunction during or after MRI scanning performed in accordance with manufacturer guidelines. This could include erratic stimulation or generator malfunction, destruction, or necessary re-programing as confirmed by "system diagnostics" after scanning.

Included studies
The search strategy yielded 156 publications (Fig. 1). Among these publications, 87 were excluded for not reporting on patients with VNS undergoing MRI scans and not being relevant to for the present analysis (e.g., animal studies; t-VNS; patient receives MRI before VNS implantation; not original research (review); not English; "VNS" has other meaning). A full review of the remaining 69 publications led to 43 publications being excluded for reporting only baseline MRI scans in the context of a pre-surgical evaluation, MRI scans after VNS Therapy System explanation, same patients reported in other articles (duplicates), or other reason (e.g., only positron emission tomography [PET] scans performed). Finally, 26 articles were included in this analysis and included data from 216 patients ( Table 1). Not all studied specify the field strength of the MRI scanners used, nevertheless 77 patients were reported getting 1.5 T, and 58 patients received 3 T scans. Some studies described in these articles performed multiple scans on the same patient. The duration of the performed scans (exposure time) was only mentioned in a few articles for fMRI sequences and could not be meaningfully evaluated (Table 2). In 23 of the 26 articles, only cranial MRI (cMRI) was performed. Two studies report on patients receiving spinal MRI and one study includes patients receiving MRI of the extremities. Of the eligible articles, 21 studies evaluated patients with DRE and 5 studies evaluated patients with TRD. Nineteen studies addressed either a technical or clinical research question, two studies addressed MRI in the context of medical emergencies (trauma and febrile infection-related epilepsy syndrome [FIRES]), one study was conducted in the context of a diagnostic workup for bradycardia of unknown origin, and one article reviewed all patients with an implanted VNS receiving MRI at their center for any reason. The remaining 3 studies were published in the last 2 years and addressed MRI scans in the context of laser interstitial thermal therapy (LITT) for DRE.
Of the 26 eligible articles, 25 studies were in patients implanted with VNS Therapy System models from LivaNova PLC, London, UK (before Feb. 2015, Cyberonics, Inc.) and one article presented a case study of a patient implanted with a VNS system from PINS Medical (Beijing, China), which from a design point of view is comparable with the VNS Therapy System. MRI procedures included MRI-guided laser interstitial thermal therapy (LITT) in three patients (Table 1).

Adverse events
None of the included articles reported a serious adverse event or a device-related adverse event. In one patient with TRD, scanning was interrupted due to a panic attack [11], and this was described as an event of mild intensity.
The article by de Jonge et al. reported one patient, in whom high lead impedance was detected post MRI scanning [21].   Multi-slice single-shot gradient echo EPI-fMRI;  However, because "system diagnostics" prior to MR scanning in this patient were not performed according to manufacturer guidelines, no temporal relationship between scanning and lead impedance could be established. In a separate article [22], two patients were reported not to tolerate MRI scanning after previous successful MRI scans; the authors did not specify further details. Three articles reported one or more generators failing to start stimulation when in the magnetic field of the scanner when MRIs were performed contrary to instructions for use in the physician's manual [11,22,23,28]. As an example, the design of the study by Nahas and colleagues was based upon achieving uninterrupted VNS at programmed device settings during functional magnetic resonance imaging (fMRI). To DTI: parallel imaging mode with an acceleration factor of 2. A single shot spin echo planar imaging sequence was used, with 5 images obtained without diffusion weighting and 33 isotropically distributed diffusion gradient directions. The b value in the diffusion-weighted images was 1000 s/mm 2 . The TE was 90.3 ms, and the TR was 14,000 ms, but may have varied up to 14,800 ms in some patients. Images were zero filled to a matrix size of 128 × 128, yielding an image resolution of 0.9 × 0.9 × 3 mm 3 . From the diffusion-weighted images, maps of fractional anisotropy, mean diffusivity, and V1 images (the main vector of the diffusion tensor) were calculated using FSL software. achieve this, Model 101 VNS Therapy Systems were programmed to continue during MRI. Instructions for use call for programming current output to 0 mA before MRI (Physician's Manual, 2002). During the study, all Model 101 VNS Therapy Systems performed as designed. This included resetting of stimulation parameters to factory programmed settings when exposed to magnetic and radiofrequency (RF) fields generated during MRI, or movement of the pulse generator's reed switch when exposed to a static magnetic field to interrupt the programmed Model 101 duty cycle. While this caused inability for the investigators to maintain uninterrupted VNS during fMRI and resulted in a number of scans having to be excluded from the final analysis, no stimulation-related adverse events during fMRI and no unintentional resetting of VNS parameters to a higher output by fMRI were reported [11]. There were no reports of adverse events in children with VNS Therapy System during MRI examinations. In three case reports, the unintentional execution of a full-body MRI in patients implanted with an VNS Therapy System was described, with no reported clinical consequences for the patients [21,24,25].
Taken together in this analysis, one non-serious adverse event was reported among all reported patients (0.4%) and was unrelated to the VNS system implanted. No serious adverse events were reported. No VNS Therapy System malfunction was reported when MRI scanning was performed according to instructions for VNS Therapy System use in the physician's manual.

Discussion
Clinical concerns during MRI scanning of patients with an implanted VNS Therapy Systems are typically focused based on three phenomena during MRI: heating, force, and malfunction. These could be caused by the interaction of the VNS Therapy System with the static fields, gradient fields, and the RF fields generated by the MRI scanner. The static field can exert torque and force on ferromagnetic objects, which could result in displacements; RF and gradient fields could each result in excessive heating, and all three of these fields could either separately or in combination theoretically cause generator vibration and malfunction. Clinically, these would be manifested by the presence of pain or loss of therapy. In addition to these hazards, there is also the possibility of unintended stimulation caused by either the RF or gradient fields.
Laboratory testing focusing on the functional aspects of the VNS Therapy System indicated that MRI procedures performed at 1.5 T and 3 T produced no significant alterations in the operation of the generators (Livanova, data on file). These findings coincide with results described by Shellock and coworkers for the VNS Therapy System studied under in vitro conditions in 1.5 T and 3 T scans [19]. However, Shellock and coworkers identified potentially unsafe conditions with regards to MRI-related heating, therefore MRI scans at 1.5 T and 3 T in patients implanted with VNS Therapy System should only be performed respecting the clearly defined "exclusion zones" in the manufacturers' guidelines [19,28].
None of the here reviewed literature on MR imaging studies in patients implanted with the VNS Therapy System did not report any symptoms or signs of pain, local discomfort, or loss of therapy. However, most are brain studies and have scanned fMRI, DTI, and 3D T1 (MP-RAGE or FSPGR), of which all are low SAR sequences. Two studies involving MRI of the spine reporting 4 patient scans in total. Especially, the work of Wilfong [24] where 3 patients underwent MR spine imaging must be assessed critically, as this is an abstract presumably concerns a conference contribution and is not widely available.
Our analysis here has demonstrated that cranial magnetic resonance imaging for soft tissue visualization can be performed safely under appropriate conditions in patients implanted with a VNS Therapy System. Such "MRI conditional" use of MRI means that a VNS Therapy System poses no known hazard in a specified magnetic resonance environment if specific conditions that are described in the physician's manual are met (Supplemental materials).
MRI scans of the head, neck, and extremities are possible currently with 1.5 T and 3 T MRI scanners. Initial studies have demonstrated that 7 Tesla MRI may improve lesion detection in epilepsy patients [26,27,29,30]. Comprehensive technical assessment will be needed in order to evaluate the safety of MRI scanning at this higher magnetic field strength in patients implanted with a VNS Therapy System. VNS has been established as an effective, safe, and welltolerated adjunctive therapeutic option in patients with DRE [31] and TRD [32]. This present systematic review on realworld use of MRI in patients with DRE and TRD has revealed no serious adverse events, device-related adverse events, or VNS Therapy System malfunction when MRI scanning is performed in accordance with VNS Therapy System instructions for use in the physician's manual (LivaNova PLC, 2019; Supplemental materials).
With the first approval for VNS therapy (1994 EU and 1997 USA, Cyberonics Inc.), MRI scans were only allowed using local transmit/receive (T/R) coils. T/R coils are commonly used only for extremity imaging on modern MRI scanners. Brain MRI scan protocols nowadays often use parallel imaging and cannot be scanned unmodified using a T/R head coil. Since 2017, the MRI guidelines for VNS therapy were expanded and permit the usage of a transmit body coil, together with a receive-only local coil, respecting the International Electrotechnical Commission (IEC 60601-2-33) SAR limits (2 and 3.2 W/kg for whole body and head scans, respectively) under normal mode operation (Group A, Supplemental materials) [28]. According to these guideline extensions, 28 out of 216 patients may have been scanned with transmit body coil.

Limitations
This systematic review must be interpreted with caution as it is retrospective and evaluated a heterogenous mix of patient populations, VNS systems, and MRI techniques and methods. Most of the articles summarized experience that was based upon a small number of patients and had a length of longterm follow-up that differed across studies. These limitations notwithstanding, no safety signal emerged from this systematic literature review of real-world MRI use in patients implanted with a VNS Therapy System when MRI is performed according to instructions for MRI system use (LivaNova PLC, 2019).

Conclusion
This systematic review indicates that cranial MRI of patients with an implanted VNS Therapy System can be completed satisfactorily, is tolerable, and safe using 1.5 T and 3 T scanners, when the manufacturer guidelines are followed for MRI scanning. Acknowledgements Open Access funding enabled and organized by Projekt DEAL.
Author's contributions All authors contributed to the conception and design of the review. Literature search and analysis were performed by SF, MD, and RT. SF and RT wrote the manuscript. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Declarations
Conflicts of interest SF and MD are employee of LivaNova PLC Deutschland GmbH-a fully owned subsidiary of LivaNova PLC. MD holds stock options in LivaNova PLC. RT and AN have no conflicts of interest to report.
Ethical standards This work is in compliance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Informed consent The submission is a review and does not require informed consent.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.