Magnetic Resonance Imaging in Patients With Implanted Spinal Cord Stimulation Systems

Spinal cord stimulation (SCS) is an established method of interventional therapy for chronic neuropathic, ischemic, and mixed pain of various origins. The high demand for magnetic resonance imaging (MRI) examinations is a common problem encountered during the follow-up of these patients (1, 2). Current implantable devices are only approved for MRI diagnostics with certain restrictions (3). Older devices are still found in significant numbers and are not usually approved for MRI examinations. Published studies and case reports on MRI examinations in these patients are rare, and there are no publications on products from Abbott Laboratories and their acquired companies (St. Jude Medical, Advanced Neuromodulation Systems [ANS]).

The aim of this retrospective analysis of our treatment data is to assess the safety of MRI examinations, focusing on Abbott Laboratories devices which are the most commonly encountered implantable devices in our region.

Materials and methods

After approval by the responsible ethics committee (State Chamber of Physicians of Thuringia, registration number 64752/ 2019/28), a search was conducted in the database of our hospital information system for MRI requests and coexisting evidence of the presence of a neurostimulator (keywords, ICD or OPS codes, contact with our department) for the period from November 2011 (foundation of our pain center) to March 2019. We reviewed the patient records for all search results and extracted the required data from it.

Having established a medical indication, all MRI examinations were performed with a 1.5 Tesla MRI scanner (Achieva, Philips) using a structured process based on the recommendations of De Andres et al (4). All patients were informed in detail about the lack of approval for the devices and the associated consequences (off-label use, termination criteria, risks). After a complete system check, the MRI mode was activated – if available and possible – otherwise the neurostimulator was programmed to lowest settings (anode and cathode placed on functional contacts, pulse width 50 µs, minimum and maximum current 0.1 mA and 0.3 mA, respectively, magnetic switch set to turn off stimulation only). After the MRI examination, a complete system check was repeated, and the patient was also offered long-term follow-up care. We used a spreadsheet software program for data collection and analysis (Microsoft Excel 365, 64 bit; Microsoft Inc.).

Results

A total of 149 MRI examinations on 86 patients with a median age of 58 years (28–87 years) using an SCS system from Abbott Laboratories were evaluated.

One hundred cases involved implantable pulse generator (IPG) not approved for MRI examinations (Genesis: n = 20, Eon: n = 14, EonC: n = 23, Eon mini: n = 21, Prodigy: n = 19, no IPG: n = 3).

Forty-nine cases involved IPGs approved for MRI examination (Prodigy MRI: n = 28; Proclaim: n = 21). However, the entire system (including leads and, where required, extension cables) was approved for the planned MRI examination in only 15 of 49 cases (30.6%). Of these, 4 of 28 cases involved Prodigy MRI IPGs (approved for MRI of the skull and limbs, except shoulder and hip), and 11 of 21 involved Proclaim IPGs (approved for examination of all anatomical regions). Reasons for lack of approval for Prodigy MRI were defective IPGs (n = 2), unapproved lead types (n = 6), lead position (n = 6), presence of an extension (n = 1), and limitations regarding the anatomical regions examined (n = 9). For Proclaim IPGs, these were lead position (n = 5), no lead (n = 2), lead type (n = 1), and inability to enable MRI mode (n = 2). The examined anatomical regions were distributed as follows: trunk (n = 99), skull (n = 32), upper limbs (n = 4) and lower limbs (n = 14).

We observed a total of twelve complications which were potentially implant-related (8.1%; 95% confidence interval [3.7; 12.4]; 0.14 complications/patient) and two definitely non-implant related adverse events. Details are provided in the Table. We saw complications at follow-up examinations after unremarkable initial examinations as well as complications at the initial examination with unremarkable follow-up examinations. In no single case were the manufacturer’s specifications for technical MRI parameters (coils, specific absorption rate, slew rate, maximum examination time) fully met.

Table. Adverse events during magnetic resonance imaging examinations.

Description	Generator	Lead(s), lead tip position	Examined anatomical region	max. SAR (Watt/kg)	Number of sequences
Serious adverse events (n = 1)
Defective generator	EonC	Octrode, C3	Hand	2.2	10
Adverse events: device heating (n = 3)
Discontinued due to mild heating and panic attack	Eon	Octrode, T10	Skull	0.3	1
Discontinued due to heating along the lead	Genesis	Octrode, T 3	Shoulder	2.0	4
Generator heating	Genesis (discharged)	Lamitrode S-8, T7	TSp/LSp	4.0	8
Adverse events: undesired stimulation (n = 8)
Discontinued due to electrifying sensations	Eon mini	Octrode, T8	LSp	0.5	1
Discontinued due to electrifying sensations	Prodigy MRI	Octrode, T8	LSp	n.s.	0
Discontinued due to electrifying sensations	Eon mini	Lamitrode S-8, T7	TSp/LSp	0.5	1
Discontinued due to electrifying sensations*	Genesis	Lamitrode S-8, T8	Shoulder	3.8	9
Discontinued due to involuntary movements	Eon mini	Octrode, T7 Octrode, T8 (displaced ventrally)	TSp/LSp	0.5	1
Discontinued due to involuntary movements	Genesis	Octrode, C3	Shoulder	0.3	1
Electrifying sensations after contrast agent infusion	Prodigy MRI	Octrode, T8Octrode, T9	Foot	4.0	10
Electrifying sensations after contrast agent infusion	Eon mini	Octrode, C2	CSp	4.0	7
Adverse events unrelated to neurostimulator (n = 2)
Discontinued due to position-related back pain*	Genesis	Lamitrode S-8, T8	Shoulder	3.8	6
Discontinued due to claustrophobia	Proclaim 5 Elite	Octrode, T8	Skull	0.6	2

There was no statistically significant correlation between complication rate and examined anatomical region. There was also no evidence for differences related to the number of examinations performed to date.

* One patient experienced discontinuation in each of two independently performed examinations. All other adverse events occurred in different patients. s

T, C, thoracic, cervical vertebra; TSp, CSp, LSp, thoracic, cervical, lumbar spine; n.s., not specified; SAR, specific absorption rate (measure of the energy absorbed by the body per unit of time, mostly six minutes).

Discussion

Only in 10.1% of the cases were the implanted devices approved for the requested MRI examination. Even with an approved IPG, this was only true in 30.6 % of the cases. The reasons for this were, in particular, restrictions regarding the anatomical regions to be examined and the position of the leads. Compliance with the specified technical parameters proved to be another problem.

Despite the absence of approval, it was still possible to carry out the examinations with an acceptable complication rate. Defective IPGs (n = 1; 0.67%) and device heating (n = 3; 2.0%) are known complications after MRI examinations (3). The three patients all reported gradual heating of their devices which, however, allowed sufficient time to terminate the examination without causing harm to the patient. Adverse sensory or motor stimulation events (n = 8; 5.4%) occurred either when starting examination planning on the device (survey scan) or shortly after administration of the contrast agent (known side effect). Stimulation of peripheral nerves by the MRI environment is a well-known phenomenon that occurs even without the presence of a neurostimulator and depends on the MRI scanner, the anatomical region examined, and the examination protocols (5). In the absence of a comparison group, it remains unclear to what extent these events are related to the neurostimulator. Further limitations of the study are its retrospective, single-center design and the focus on devices from one manufacturer, which limits any general applicability of the data.

In our view, where there is an indication for an MRI examination, it should not be withheld from the patient solely because of the absence of approval for the neurostimulator. We recommend performing the examinations on an awake and cooperative patient, primarily at specialized centers. The indication should be established carefully in each individual case, taking any alternative imaging options into account, and the patient should be informed accordingly. If new pain, paresthesias or involuntary movements develop, the examination should be stopped immediately for safety reasons. Manufacturers are encouraged to develop future devices to meet clinical requirements for MRI diagnostic scanning.