MR Safety Update Tuesday 14th November 2017, Park Inn Hotel, York

FINAL PROGRAMME

09:00 – 09:25 Coffee and registration

09.25 – 09.30 Introduction

Chair: Lisa Davenport

09:30 – 10.00 Foetal MRI safety Invited Speaker: Prof Penny Gowland, University of Nottingham

10.00 – 10.15 MHRA Update Invited Speaker: David Grainger, MHRA

10.15 – 10.35 MR Safety Expert Accreditation Invited Speaker: Dr Dan Wilson, Leeds Teaching Hospitals NHS Trust

10.35 – 10:50 The Role of the MR Safety Expert: The Italian Perspective Dr Nicola Pace, University of Trento, Italy

10:50 – 11:20 Coffee and exhibition

Chair: Jennifer Macfarlane

11:20 – 11.35 Challenges of establishing MR safe working practices in a new MR-Linac department Dr Glyn Coutts, The Christie NHS Foundation Trust

11.35 – 11.50 IPEM Recommended MRI Safety Notices Invited Speaker: Dr Cormac McGrath, Belfast Health & Social Care Trust

11.50 – 12.05 Experimental assessment of MRI-induced temperature change and SAR distributions in phantoms James Blackwell, National University of Ireland Galway

12.05 – 12.20 Feasibility of MRI Acoustic Noise Measurements Dr Ian Cavin, NHS Lothian

12.20 – 12.50 Interactive session

12.50 – 13.50 Lunch and exhibition

Chair: John Thornton

13.50 – 14.15 Gadolinium retention Invited Speaker: Dr Nigel Hoggard, Royal Hallamshire Hospital, Sheffield

14.15 – 14.30 T1 Dentate Nucleus Signal Intensity Following Multiple Gadoteric acid (Dotarem®)-Enhanced MRI vs Multiple Non-Contrast MRI Laura Young, University of Dundee

14.30 – 14.50 IPEM working party on Blanket policies for scanning implants Invited Speaker: Dr Jonathan Ashmore, NHS Highland

14.50 – 15.05 Compiling and disseminating generalised MRI implant safety policies for the real world Dr John McLean, NHS Greater Glasgow & Clyde

15.05 – 15.35 Coffee and exhibition

Chair: Stephanie Withey

15.35 – 15.50 Scanwise implant – an approach to addressing the issues of safe MR scanning implants Dr Matthew Clemence, Philips Healthcare

15:50 – 16.05 Neurological MRI protocols for patients with DBS equipment in situ consistent with new B1+rms - limited MR Conditional product label Annie Papadaki, National Hospital for Neurology and Neurosurgery, University College London

16.05 – 16.20 Off-Label Magnetic Resonance Imaging of an InterStim II Sacral Nerve Stimulator Device Andrew Fry, Sheffield Teaching Hospitals NHS Foundation Trust

16.20 – 16.35 Comparison of MRI Screening Policies for Metallic Intra-Orbital Foreign Bodies Theodore Barfoot, King’s College Hospital NHS Foundation Trust

16.35 – 16.50 Performing MRI Scans on Cochlear Implant and Auditory Brainstem Implant Recipients - 13 Year Review Yu-Chuen Tam, Addenbrooke’s Hospital

16.50 – 17.05 Q&A & Prizes

17.05 Close

Organised by IPEM’s Magnetic Resonance Special Interest Group

MR Safety Update Tuesday 14th November 2017, Park Inn Hotel, York

POSTERS

1 MRI-related magnetic field exposures and risk of commuting accidents and abnormal uterine bleeding – a cross-sectional survey among Dutch imaging technicians Anke Huss, Utrecht University, The Netherlands

2 Assessment of Noise Levels Outside the MR Environment: A Preliminary Investigation Rebecca Jennings, ABMU Health Board and Swansea University

3 Implantable medical devices in MRI: the serious risks of using out-of-date MR safety information Dr Aaron McCann, Belfast Health & Social Care Trust

4 IPEM Recommended MRI Safety Notices Dr Cormac McGrath, Belfast Health & Social Care Trust

5 The Role of the MR Safety Expert: The Italian Perspective Dr Nicola Pace, University of Trento, Italy

6 Something to get your heart racing: Scanning of MRI Conditional Active Cardiac Devices Harry Poole, Royal Stoke University Hospital

7 A Review of Auditory Implants and MRI Safety Yu-Chuen Tam, Addenbrooke’s Hospital

8 Audit of MR Safety Queries to an MR Physics Team in a Large Tertiary NHS Trust David Broadbent, Leeds Teaching Hospitals NHS Trust

Organised by IPEM’s Magnetic Resonance Special Interest Group

Foetal MRI Safety Invited Speaker: Prof Penny Gowland, University of Nottingham

This talk will present the range of exposures and risks associate with MRI from the point of view of fetal imaging. It will consider the need to provide a calm environment for the mother and the particular risks associated with exposures to static, RF and gradient fields in a fetus, and will also consider the effects of acoustic noise in the fetus. It will highlight areas where future research is required.

MHRA Update Invited Speaker: David Grainger, MHRA

In my talk intend to cover the following issues; • A review of MRI incidents reported to MHRA. • An update on Gadolinium Update • An update on the new Medical Device Regulations and a review of the key changes.

MR Safety Expert Accreditation Invited Speaker: 1Wilson D 1Medical Physics, Leeds Teaching Hospitals NHS Trust, UK

MR safety is a multi-disciplinary team effort and in the UK the following people may be involved – radiologist, radiographer, MR safety expert (often provided by a medical physicist) and MR unit managers.

The MR Safety Expert provides scientific safety advice and this can range from interpretation of national and international guidelines and legislation, establishment of safety frameworks, site design, measurement of electromagnetic fields and assessment of patients with implants.

The MHRA recommend that all MRI units should have an MR responsible person who is delegated the day-to-day responsibility for MR safety. They also recommend that each MR responsible person should be in full consultation with an MR safety expert (MRSE).

IPEM produced a policy statement in 2013 (2) on “Scientific Safety Advice to Magnetic Resonance Imaging Units that Undertake Human Imaging”. It’s main recommendations are shown in figure 1b above but an important recommendation was “that an MRSE register is established. This will require a mechanism of accreditation...”.

Since 2015 an IPEM working party (with representatives from IPEM, BAMRR, SCoR, MHRA and the BC-ISMRM) has been looking at how to implement this recommendation. This work is not yet finalised but the proposed intial scheme is likely to have two components - a knowledge component and an experience/competency compenent and both components will need to be fulfilled to be accredited as an MR safety expert.

There are current moves to develop an MR safety expert course and exam which would be the most transparent way of showing that the knowledge criteria had been satisfied. The experience / competency requirements are currently being piloted.

It has been explicit since the IPEM policy statement that an MR safety expert is not necessarily limited to medical physicists and therefore is not limited to IPEM members. Anybody who fulfils the knowledge and competency requirements can be an MRSE. However, the role is a scientific and technical role and the expectation is that the level of scientific knowledge will be high. Any examination based upon the knowledge syllabus will also require a reasonably high scientific skill set to pass although the recommended prerequisites and exact scientific level are still to be set.

Final decisions are still to be made on which body will hold the register and administer the accreditation process. Assessors have not yet been identified and a process will need to be in place to ensure that assessors are adequately trained, that applications can be dealt with in a timely fashion and that an appeal process is in place.

The Role of the MR Safety Expert: The Italian Perspective 1Pace N, 2McGrath C 1University of Trento, Trento, Italy 2Belfast Health and Social Care Trust, Belfast, UK

Background. The MR Safety Expert (MRSE) is a crucial figure in an Magnetic Resonance (MR) department. The MRSE’s duties and responsibilities change across different countries. The role of the MRSE in Italy is outlined by national laws1,2,3,4,5 and by several guidelines6 issued by a central body, the National Institute for Insurance against Accidents at Work (Istituto Nazionale Assicurazione Infortuni sul Lavoro ,INAIL). The aim of this work is to identify the key features of the MRSE role in the Italian context. Methods. A thorough review of the Italian regulations and guidelines on MR safety has been conducted. A summary of the main duties and responsibilities of the MRSE are presented, with particular focus on those that appear to differ most from the international context. Results. The first Italian law regulating MR installation and operation dates back to 19851. Since then, modifications have been implemented in 19912, 19933, 19944 and 20165. Since 1994, INAIL has been conducting inspections at MR departments across the country to verify the compliance to the regulations in effect. INAIL also issued several guidelines6 to increase the effectiveness of MR departments in reaching the highest possible level of safety. In these laws and guidelines, the MRSE has a key role: he/she must be formally identified by those responsible for the MR department, and has criminal liabilities for the safety related to technical aspects of the installation and operation of the MRI scanner. The MRSE's duties are: validation of the installation project; verification of the acceptance tests performed by the installing company; inspection of the safety devices installed; identification of rules in the event of an emergency; periodic surveillance of the department and the devices therein installed, with verification of the distribution of the magnetic field within the MR department; communication of technical accidents to the regulatory body. Specifically, safety devices considered by the Italian regulation system are: oxygen sensor; Faraday cage; quench pipe; ventilation system. INAIL guidelines identify the methodologies and periodicity to perform safety devices checks; moreover, behavioural and emergency rules are suggested for enforcement. In regards to the installation phase, INAIL guidelines have been published tackling crucial points in the design and installation of oxygen sensor systems, ventilation systems and quench pipes. During the first years of the activity of a MR system, INAIL inspectors perform a thorough visit of the MR department, in order to check the safety conditions of the department and suggest possible actions for safety optimisation. Oddly enough, the law simply described the MRSE as a professional with proven experience in the field of MR safety and with an adequate CV. No professional register is established, nor official examinations to obtain formal certification. While the MRSE is responsible for technical safety considerations, medical safety considerations remain the explicit responsibility of the MR Medical Director (MRMD). Discussion and Conclusion. In the Italian context, the role of the MRSE has well defined responsibilities and duties. Strong enforcement of the role is given by periodic inspections by the regulatory body. A structured and thorough set of guidelines ensure strong homogeneity and standardisation of safety rules adopted by MR departments across Italy. On the other hand, such a rigid framework makes it less likely to easily and promptly adopt changes when there are updates in MRI technology or other developments. Key references.

1. Ministerial Decree of the 29th of November, 1985 2. Ministerial Decree of the 14th of August 1991 3. Ministerial Decree 3rd of August, 1993 4. Presidential Decree No 542, 8th of August, 1994 5. Legislative Decree No 113,24th of June, 2016 6. INAIL 2017. Sicurezza in risonanza magnetica.

https://www.inail.it/cs/Satellite?c=Page&cid=6443097230584&d=Touch&pagename=Internet%2FPage%2FpaginaFoglia%2Flayout Accessed 17/08/2017.

Challenges of establishing MR safe working practices in a new MR-Linac department 1Coutts G, 2McDaid L, 3Edwards T, 1Jackson S, 1Dubec M 1Christie Medical Physics & Engineering, 2Diagnostic Radiology and 3Radiotherapy, The Christie NHS Foundation Trust, UK.

Background. As one of seven members of the Elekta MR-Linac Consortium the Christie is currently in the commissioning phase of a MR-Linac (MRL) which has the capability of concurrently providing MR imaging and radiotherapy treatment [1]. This presents unique challenges for MR safety in that most of the physicists, engineers, radiographers and clinicians involved have never previously worked in an MR environment. In addition there is a range of testing equipment required for use within the magnet bore as well as tools for servicing the Linac in the Machine Room which, because of the necessity of a bunker for radiation protection, is only accessible through the Treatment Room. A MR safety framework was established in order to meet these needs.

Methods. In accordance with the MRHA guidelines [2] an experienced MR physicist was established as the MR Safety Expert for the MRL and a set of MR Local Rules were drawn up. Entry to the Controlled Area for the MRL, containing the Control Room, Maze, Treatment Room and Machine Room is by swipe card which is only enabled on the Staff ID cards of Authorised Persons. The MR Environment for the MRL consists of the MR Treatment and Machine Rooms. Three levels of MR Authorised Person are in place: non-MR Environment, MR Environment and Supervisor.

All Authorised Persons must have 1) completed a medical screening form which is signed off by an experienced MR radiographer or physicist, 2) attended a training session on safe working within the MR Environment, and 3) have read the MR Local Rules, all with annual updates. Records for all these and MR Authorised Person status have been given entries within the Radiotherapy Department Quality System. Additional training is provided for Supervisors who are able to review screening forms and supervise non-Authorised Persons who need access to the department. However any queries regarding the screening form must be referred to an experienced MR person.

All commissioning equipment must be inspected and labelled as MR Safe, MR Conditional or MR Unsafe (for example equipment cases) by an experienced MR physicist and entered into the MRL equipment log. In addition there must always be at least two MR Authorised Persons present for any work being done in the MR Environment, and a System of Work has been established for taking ferromagnetic tools into the Machine Room.

Results. To date on the Quality System there are 18 MR Authorised Persons (MR Environment), (16 physicists, 1 engineer, 1 radiographer) and 8 MR Authorised Persons (Supervisor) (3 physicists, 2 engineers, 3 radiographers) of whom 3 (2 physicists and 1 radiographer) can additionally sign off screening forms and 2 can assess the MR safety of equipment. There are so far 50 tested items on the MRL Quality Manual. There have been no incidents.

Discussion. When preparing for delivery of the MRL it was recognised within the radiotherapy department that lack of MR safety experience would have to be addressed. In this case it was possible to draw on existing MR expertise within Diagnostic Radiology. Staff compliance with the safety measures put in place has been very high. The MR safety framework and training are being introduced in a staged manner. Preparations are in hand for MR scanning of volunteers, and will then work towards treatment of patients.

Although the MRL setup is very new it is likely that increasingly radiotherapy departments will have ownership of MR facilities and establishing MR safe working practices is of paramount importance.

Conclusion. MR safe working practices have been successfully established for a new MRL department.

Key references. [1] Medicines and Healthcare Products Regulatory Agency (MHRA), Safety Guidelines for Magnetic Resonance Imaging Equipment in Clinical Use, 2014 [2] The Magnetic Resonance Imaging Linac System. Lagendijk JJW., Raaymakers BW., van Vulpen M, Seminars in Radiation Oncology. 2014 7; 24(3):207-209.

IPEM Recommended MRI Safety Notices Invited speaker: 1,2McGrath C, 1,3Curtis S, 1,4Wilson D, 1,5,6Charles-Edwards G, 2,7Orr PA, 2McCann AJ, 8Quest RA, 9Grainger D, 10McGreal R and 11Creak J

1 Institute of Physics and Engineering in Medicine Magnetic Resonance Special Interest Group (IPEM MR-SIG)

2 Northern Ireland Regional Medical Physics Service, Belfast Health & Social Care Trust, Belfast, UK

3 Medical Physics & Bioengineering, University Hospitals Bristol NHS Foundation Trust, Bristol, UK

4 Medical Physics and Engineering, Leeds Teaching Hospitals NHS Trust, Leeds, UK

5 Guy's & St Thomas' NHS Foundation Trust, London, UK

6 King's College London, London, UK

7 Institute of Physics and Engineering in Medicine Radiation Protection Special Interest Group (IPEM RP-SIG)

8 Radiological Sciences Unit, Charing Cross Hospital, Fulham Palace Road, London, UK

9 Medicines & Healthcare products Regulatory Agency, London, UK (MHRA)

10 Health and Safety Executive, Liverpool, UK (HSE)

11 Health and Safety Sign Association, Redhill, Surrey, UK (HSSA)

Background At the 2015 IPEM MR Safety Update, the wide disparity in MRI safety notices, as well as the use of inappropriate safety notices was presented

1. An IPEM Working Party was formed to create a

standardised approved MRI safety notice set, freely available in the IPEM website2. The Working Party

comprised members of the IPEM Magnetic Resonance and Radiation Protection Special Interest Groups (IPEM MR-SIG/RP-SIG) in collaboration with the Health and Safety Executive (HSE), the Health and Safety Sign Association (HSSA) and the Medicines & Healthcare products Regulatory Agency (MHRA). MRI Safety Notices Standardised MRI Safety Notices were produced for the following:

MRI scanner room door

MRI scanner room entrance floor

Entrances to the MR Controlled Access Area

Approaches to the MRI quench pipe exhaust

Regions of magnetic flux density > 0.5mT

Emergency Button and Environmental Control/Display Labelling The Working Party has created a sample set of labels for use in MRI departments to clearly and unambiguously identify emergency buttons, the MRI scanner room oxygen alarm/monitor and the MRI scanner room environmental control/display. Additional Safety Measures Additional safety measures also recommended are an easily removable physical barrier across the MRI scanner room entrance, e.g. a retractable belt stating “AUTHORISED ACCESS ONLY”, and magnetic flux density isocontouring on the MRI scanner room floor to aid the use of ancillary equipment. Conclusions The creation of this standardised MRI safety notice set has met a need in the MRI community, as there has been widespread adoption and very positive feedback from MRI staff and manufacturers’ representatives. Welsh language versions are also available

3. Discussions are being had with EFOMP

regarding other European language versions and with Frank Shellock2,4

regarding American-English versions. Acknowledgements Special thanks to Geoffrey Peckham, Chair of ANSI Z535 Standards Committee and Adrian Knipe, Medical Illustration Department, Belfast Health and Social Care Trust for aiding professional colour specification.

References 1. McGrath, C et al 2015 The good, the bad and the ugly: The need for MRI safety signage standardisation in the UK IPEM Conference Abstracts

2015, pg162 http://www.ipem.ac.uk/Portals/0/Documents/Conferences/2015/2015%20CONFERENCE%20ABSTRACTS.pdf [accessed 17 August 2017]

2. https://www.ipem.ac.uk/ScientificJournalsPublications/MRISafetyNotices.aspx [accessed 17 August 2017] 3. http://www.wales.nhs.uk/sites3/page.cfm?orgid=254&pid=92085 [accessed 17 August 2017] 4. Frank Shellock – private communication

Experimental assessment of MRI-induced temperature change and SAR distributions in phantoms J. Blackwell1, W. Van der Putten2, B. Tuohy2 and N. Colgan1

1 Department of Physics, National University of Ireland Galway, Galway, Ireland 2 Medical Physics and Bioengineering, Galway University Hospital, Galway, Ireland

Background: During an MR procedure, most of the transmitted RF power is transformed into heat within the patients’ tissue resulting from resistive losses, referred to as the specific energy absorption rate(SAR) (2). The EU standardisation has mandated that all scanners must measure SAR in patients and develop system safeguards to ensure that the limits(IEC60602-3-33) are not exceeded. Accurate estimation of SAR is critical in safeguarding patients who may be unconscious/sedated, have implants or are pregnant. Modern MRI systems can easily exceed safe SAR levels (1) requiring the independent verification of manufacturers SAR estimations.

The purpose of this research was to develop a protocol to verify the patient specific SAR. To determine RF power deposition experimentally, a T1 doped MR phantom in a birdcage head coil was used, heated solely by the RF fields produced by the imaging coil.

Methods: As there is a negligible contribution from thermal conduction in our SAR assessment and our phantom is a nonperfused material, physiological changes can be ignored. The SAR at discreet points in the observation plane the can be determined by the following equation (2)

𝑆𝐴𝑅 ≈ 𝐶𝑎𝑔𝑎𝑟∆𝑇

∆𝑡[𝑊\𝑘𝑔]

Where 𝐶𝑎𝑔𝑎𝑟 is 4200 J/Kg.K, ∆𝑇 is the change in temperature and ∆𝑡 is the change in time.

Proton Resonance Frequency Shift (PRF) thermometry was utilised to find the change in temperature.

∆𝑇 =𝜑 − 𝜑0𝛼𝛾𝐵0𝑇𝐸

Where 𝛼 is the temperature co-efficient 0.01ppm/C, 𝛾 is the gyromagnetic ratio (MHz/T), 𝐵0 is the

field strength (T), TE is the echo time (ms) and 𝜑−𝜑0 is the phase shift (degrees).

A 3L volume, T1 doped MR phantom was created by dissolving 60g/L agar, 10g/L NaCl, 1g/L NaCl in distilled water. The solution was autoclaved at 121C for 15min to ensure the solution was homogeneous and to remove air bubbles.

Phase maps were created pre and post heating. These were then used to create a phase difference map and a heat map could then be created using PRF. A FLAIR sequence was used over the whole volume to heat the phantom. The calculated SAR was compared to the scanner readout.

Results.

Discussion. Our results agreed with the manufacturer's settings.

Conclusion. We have developed an open source phantom that can independently verify the temperature rise associated with SAR.

Key references. 1. Crook N et al., Radiography. 2009;15(4):351- 2. Shellock FG. J Magn Reson Imaging. 2000;12(1):30-6.

Scanner Calculation

GE 1.5T Signa Explorer 0.42 W/Kg 0.41 W/Kg

Siemens 1.5T Symphony 1.88 W/Kg 1.90 W/Kg

Philips 3T Achieva 1.52 W/Kg 1.52 W/Kg

Figure 1:Heat Map of Phantom

Feasibility of MRI Acoustic Noise Measurements

1Cavin I, 1 Cluny L, 1Phillips M, 2 Tyler N, 2 Childs J. 1Medical Physics, NHS Lothian, Edinburgh, 2Radiology Physics, Kent & Canterbury Hospital, UK.

Background

Acoustic Noise is a hazard in MRI. The sound levels generated are sufficient to cause transient or permanent hearing impairment. A number of factors influence the sound pressure level (SPL) amplitudes; including gradient slew rate, maximum gradient amplitude as well as the different imaging sequence parameters.

UK [1] and international clinical MRI safety guidelines [2] and UK Health and Safety Legislation [3] state the SPL levels above which hearing protection is required but they do not advise or recommend the measurement of the performance of headphones provided to patients and staff either during acceptance or quality assurance tests. The assumption being that vendor-supplied equipment is compliant with the limits stated in the guidance.

The aim of this investigation was to develop a quick, robust method to measure and verify acoustic noise compliance as part of an acceptance and QA testing program.

Methods

SPL peak A-weighted measurements were recorded using an OptiSLM MRI Conditional SPL meter during an Echo Planar Imaging (EPI) diffusion weighted imaging (DWI) sequence on three 1.5 T Siemens systems, a Siemens 3T Prisma, a GE 1.5T Optima 450 and a Philips 1.5T Ingenia. The sound transducer was affixed to a 170 mm spherical head phantom, with and without the headphones. Positional measurements were taken at 20cm intervals along the patient couch. Staff headphone measurements were taken at four positions in the scanner room, likely positions occupied by parents, carers or MR staff with the transducer enclosed between the MR staff member’s ear and the headphone as well as an unprotected measurement.

Results. Figure 1 below shows the patient headphones SPL values obtained for each MR system.

Discussion.

Four of the 7 vendor-supplied and rated patient headphones failed to provide less than the 85dB limit as well as the rated protection at various z-axis positions. All staff headphones provided sufficient protection below the required 80 dB.

Conclusion.

This works suggests SPL measurements can be included in routine testing procedures and patient headphone compliance cannot be assumed. Replacement headphones did comply. Vendors have been contacted to provide a solution.

Key references.[1] MHRA Safety Guidelines for MRI Equipment in Clinical Use, 2015. [2] Int. Comm. on Non-Ionising Radiation (ICNIRP), Medical MR (MR) Proc: Protection of patients. Heath Physics 97(3):259-261; 2009. [3] Control of Noise at Work Regs. 2005, No. 1643.

Gadolinium retention Invited Speaker: Dr Nigel Hoggard, Royal Hallamshire Hospital, Sheffield

Gadolinium based contrast agents are prescribed over half million times per year in the UK. They have become essential to MR radiological practice. After NSF the UK had reacted by change radiological practice: assessing renal function and/or choice of gadolinium based contrast agent to minimise risk. This has been hugely successful in avoiding any cases of NSF arising from new exposures to gadolinium after the recommendations were adopted. However now there have been increasing concerns about gadolinium accumulation in the brain. The reports, at first demonstrated signal change in the cerebellar dentate nucleus on clinical scans of patients exposed to repeated doses. Following these, were reports demonstrating the presence of gadolinium in the brain at post mortem of patients previously exposed gadolinium based contrast agents. The difference this time round from NSF is that there was evidence of accumulation in patients with normal renal function. What was been relatively lacking though, is direct evidence of harm to patients arising from this accumulation. Nevertheless there has been a regulatory mindset of precautionary action being required. The numbers exposed to repeated doses of gadolinium are large. Unchelated gadolinium is known to be a toxin. Yet evidence is lacking to prove any clinically significant effect in this patient population. What has become increasing apparent over the last two years is that there are multiple reservoirs of gadolinium accumulation and there is no real understanding of the pharmacology of movement between these reservoirs and the circulation and if there is gadolinium exchange between the different sites of accumulation, skin, bone, muscle including the heart, liver and brain. Further investigation into potential effects of accumulated gadolinium are required. As we await the final recommendations from the EU on the marketing of gadolinium based contrast agents in the EU due in the next few weeks, we will be asked to implement changes to practice but the evidence base will be incomplete. In this talk we will run through the regulatory approach to gadolinium based contrast agents, looking at what evidence is available and what likely outcomes for practice might look like.

T1 Dentate Nucleus Signal Intensity Following Multiple Gadoteric acid (Dotarem®)-Enhanced MRI vs Multiple Non-Contrast MRI

L K Young1*, S Matthew1, L McCormick1, J Weir-McCall1,2, S J Gandy1,2, J G Houston1,2 1Division of Molecular and Clinical Medicine, University of Dundee, Dundee 2NHS Tayside, Ninewells Hospital and Medical School, Dundee

Introduction: Recent associations have been found between the repeat administration of some Gadolinium based Contrast Agents (GdCAs) and brain hyper-signal intensities1 indicating the deposition of gadolinium. Linear agents have been shown to have a greater propensity to release free gadolinium than the macrocyclic structure2 and three linear agents have recently been removed from the European market3. There has yet to be, however confirmed clinical consequences of the deposition. The aim of this study was to investigate whether the repeat administrations of gadoteric acid in a local cohort between 2004 and 2016 led to significant increases in signal intensity ratios (SIR) compared to a control group of patients having undergone multiple non-contrast MRI.

Method: Images from patients who have undergone multiple brain MRI were obtained and split into 2 groups; those having solely been administered Dotarem® and those having never had contrast. SIRs were assessed on unenhanced T1-weighted spin echo images for the dentate nucleus (DN) to pons, DN to middle cerebellar peduncle (MCP) and DN to cerebrospinal fluid (CSF). Ratio differences were calculated between the first scan and then following an average [range] of 6.2 [5-12] scans for the Dotarem® group and 5.2 [2-10] scans for the No-contrast group with average time intervals ± standard deviation (days) of 199.6 ± 148.7 vs 420.9 ± 355.4.

Results: Independent sample T-Tests revealed no statistically significant increase in SIR difference for any of the areas assessed. Mean SIR difference ± standard deviation of Dotarem® vs No-contrast: DN/pons: -0.0059 ± 0.038 vs -0.0018 ± 0.0722 (p=0.826), DN/MCP: 0.0016 ± 0.0313 vs 0.0058 ± 0.0402 (p=0.689), DN/CSF: 0.0782 ± 0.4289 vs -0.0352 ± 0.2478 (p=0.185).

Discussion: These results compliment the current evidence base4–6 showing that the macrocyclic agent, gadoteric acid, does not cause brain hyper-signal intensities. Further work must be done, however, to investigate the long-term impact any deposition may have.

References:

1. Kanda, T., Ishii, K., Kawaguchi, H., Kitajima, K. & Takenaka, D. High Signal Intensity in Dentate Nucleus and Globus Pallidus on Unenhanced T1-weighted MR Images: Relationship with Increasing Cumulative Dose of a Gadolinium-based Contrast Material. Radiology 270, 834–841 (2014).

2. Frenzel, T., Lengsfeld, P., Schirmer, H., Huetter, J. & Weinmann, H.-J. Stability of Gadolinium-Based Magnetic Resonance Imaging Contrast Agents in Human Serum at 37 degrees C. Invest. Radiol. 43, 817–828 (2008).

3. European Medicines Agency. EMA’s final opinion confirms restrictions on use of linear gadolinium agents in body scans. (2017).

4. Radbruch, A. et al. Gadolinium Retention in the Dentate Nucleus and Globus Pallidus Is Dependent on the Class of Contrast Agent. Radiology 275, 783–791 (2015).

5. Radbruch, A. et al. High-Signal Intensity in the Dentate Nucleus and Globus Pallidus on Unenhanced T1-Weighted Images. Evaluation of the Macrocyclic Gadolinium-Based Contrast Agent Gadobutrol. Invest. Radiol. 50, 805–10 (2015).

6. Weberling, L. D. et al. Increased Signal Intensity in the Dentate Nucleus on Unenhanced T1-Weighted Images After Gadobenate Dimeglumine Administration. Invest. Radiol. 50, 743–748 (2015).

Figure 2: Graph of the mean signal intensity ratio differences between the last and first MRI scan for the 2 patient groups. Error

bars represent standard deviation.

IPEM working party on Blanket policies for scanning implants Invited Speaker: Dr Jonathan Ashmore, NHS Highland

Background. Most MRI departments apply a risk averse policy for scanning patients with implants following the advice from bodies such as the MHRA: “Users should refer to implanting clinicians and the manufacturers for advice on the MR safety of all implants” and “When in doubt the user should assume the device is MR UNSAFE”. This guidance is typically implemented as an “identify make and model prior to scanning” policy. However, in some cases, due to practical and logistical reasons such policies are potentially not always followed and certain common implants are scanned without full prior assurance of safety. In the work presented here I discuss the results of a national audit highlighting the prevalence of this “blanket scanning” approach. I also present results demonstrating the frequency with which patients present with common implants in a clinical setting. In response to these findings I present the development of generic scanning policies which have been implemented at a number of institutions and I will discuss the work from an IPEM working party to develop guidance on the best approach for generating these policies. Such blanket policies can better facilitate safe and efficient scanning in MRI units where resources can already be overstretched.

Methods. Radiographers from 21 separate clinical departments were surveyed regarding practice for scanning patients with common implants (see figure 1 x-axis for list of implants). When asked about each implant the radiographer provided one of the following responses (for both 1.5T and 3T): (a) Always determine if safe through Identifying make and model (b) Assume it is safe if implanted at your trust (c) Just scan them - always assume it is safe with no need to identify make and model. In a second survey, undertaken at a single institution implant details were (prospectively) recorded for all patient’s entering the scanner (total 596 patients). The risk profile of an implant type was assessed through identifying evidence from (a) mrisafety.com, (b) peer reviewed scientific literature (c) email lists where implants are commonly discussed. These risks were documented and generic implant scanning policies were created based on the risk profiles.

Results. Figure 1 shows the results for survey 1, highlighting that many implants are scanned without make and model at 1.5T, although this practice is less likely at 3T. In survey 2 it was found that 21% of patients presented with an implant of some sort (figure 2). We have created blanket implant policies for a number of implant types where our evidence showed the risk of injury would be extremely low. The IPEM working party intends to report on a methodology to develop and implement such blanket policies, highlighting the risks and benefits of this approach and also provide real example of template policies implemented at our own institutions

Conclusion. Our survey shows that many hospitals appear to have blanket practices for scanning implants without assurances of safety. This is most likely due to the high number of patients who present with implants and the difficulties in finding the implant information from various clinical systems. In response, to facilitate safe and efficient scanning we have created blanket scanning policies for certain implant categories. All other implants retain the “identify make and model” approach.

Figure 1

Figure 2

Compiling and disseminating generalised MRI implant safety policies for the real world 1McLean,JR 1Department of Clinical Physics and Bioengineering, NHS Greater Glasgow & Clyde

Background. In the clinical environment, it is not always possible to establish detailed implant information yet a decision as to whether to proceed with an MR scan in a patient with an implant still needs to be made. Providing generalised implant policies for MRI safety has several benefits, notably: minimising scan rejections, minimising delays from rescheduling appointments which can potentially impact on clinical management and empowering Radiographers and Radiologists with the information and governance framework they need to make decisions.

While there are benefits, providing generalised implant policies is challenging. There are vast numbers of implants in existence, new implants are continually being created and existing implants adapted. Thus, it is necessary that evidence and policies in this area to be kept under regular review. A further challenge is how to disseminate this information to the end users, particularly where those end users are spread across many geographical locations and health boards.

Outlined here is the process that was followed in order to create and disseminate a range of MR implant safety policies for NHS GGC and related health boards.

Methods.

If an implant of a particular type is identified as viable for consideration for a generalised policy, the following processes take place: information gathering, risk assessment, policy creation and dissemination and education. The information gathering phase typically includes a review of the scientific literature, existing implant databases and MR technologist message boards. As well as identifying MR safety information about the implant type in question, it is important to also consider, other, similar implants which may have the potential to be confused with those under consideration. These will form our notable exceptions. Once the information has been considered, a risk assessment will then be done outlying the potential hazard from the implant, how these are best managed, limitations are highlighted, and the overall risk is captured. The policy is then written with the risk assessment as its basis.

One of the challenges faced once the risk assessment and policy was written was how we disseminate the information to staff. Our aim was to have a single instance of the resource such as to minimise any management overhead of the information. To achieve this we obtained a web domain and Word Press site from National Services Scotland (NSS). The benefit of this site was that it was not constrained to any single Health Board Intranet. In order to allay some concerns about liability, being that the website was in the public domain, a disclaimer was written and the implant safety policy section of the website became password protected to restrict access.

Results. Our outcome is that we have over a dozen generalised implant safety policies presented on our MRI physics website: http://www.mriphysics.scot.nhs.uk/. The policies are underpinned by a risk assessment which completes the governance framework. This was rolled out in August 2017.

Discussion. MRI scanners continue to become more prevalent, scanning more patients with the need for more staff, working at times when persons who can provide support and advice may not be available. Supporting those staff to make good decisions in regard to implant safety is challenging both in surmising the literature and disseminating the information. Pursuing generalised policies is not without its risks. However, we aim to minimise the likelihood of adverse outcomes by highlighting notable exceptions, clearly stating what the policies do and don’t cover and by continuing to monitor a range of information sources and our own incident reporting system. The end users themselves also have a key role in feeding back any new implants they come across and in reporting incidents related to implants.

Conclusion. A website has been produced that contains generalised MRI implant safety policies and associated risk assessments. This is accessible to four NHS Scotland health boards and acts as both a governance framework and an information source to aid the safe and timely MR scanning of patients with implants.

Scanwise Implant – an approach to addressing the issues of safe MR scanning with implants Clemence, M Philips Healthcare Clinical Science, Philips Centre, Guildford, GU2 8XG.

With an increasingly aging population and the widening of the role in MRI in both diagnosis and therapy, the need to scan patients with a wide variety of implants has become more prevalent. There are several sources of information now available to the MR operator but adapting protocols and ensuring that the scanner meets the requirements for a given implant remains a challenge.

The documentation and traceability of the decision making process is also often poorly implemented.

There have been recently several initiatives to establish standardisation of the information provided by implant manufacturers. However, differences in the design of each manufacturers systems and new developments in hardware can make it difficult to relate the conditions under which an implant was tested to those when the patient will be scanned. This decision burden can require extensive physicist support and is often executed in a time pressured situation.

Not only can inappropriate scanning lead to poor image quality and diagnostic failure, there are also significant harms to the patient if incorrect decisions are made

In 2009 the medical industry and FDA agreed to develop a framework for a mutually agreed limitation called “Fixed parameter option:Basic” (FPO:B) which has recently been published and Philips participation in this process has directly lead to the development of “ScanWise Implant” to both present system information to the operator to allow good decisions to be made based on available manufacturer information and subsequently limit the scanner performance to the safe operating regime.

Key references IEC 60601-2-33,Edition 3 Amendment 2, Particular Requirements for the basic safety and essential performance of magnetic resonance equipment for medical diagnosis. 2015-06 Shellock FG et. al. MR Labelling Information for implants and Devices: Explanation of Terminology. Radiology 2009;253:26-30

Neurological MRI protocols for patients with DBS equipment in situ consistent with new B1+rms -limited MR Conditional product label A. Papadaki1,5, M.J. Cardoso2, K. Shmueli3, S. J. Wastling1,4, T. Yousry1,5, J. Hyam4, L. Zrinzo4, I.

Davagnanam1,5, J. S. Thornton1,5

1Lysholm Department of Neuroradiology, National Hospital for Neurology and Neurosurgery, 2Translational Imaging Group, Centre for Medical Image Computing (CMIC), Department of

Medical Physics and Biomedical Engineering, University College London, 3 MRI Group,

Department of Medical Physics and Biomedical Engineering, University College London, 4Unit of

Functional Neurosurgery, University College London, 5Department of Brain Repair and

Rehabilitation, University College London

Background. Recently, a new MR-conditional product label was released for deep brain stimulation (DBS) devices limiting B1+rms to ≤2.0μT. Post DBS-implantation MRI is an essential part of the functional neurosurgical workflow at our institution, and patients presenting for adjustment of existing DBS therapy electrodes also require MRI to guide this procedure. Until recently “off-label” sequences were used post-implantation based on our own risk assessment, limited by the use of a transmit-receive head coil1. Additionally, the increased numbers of patients worldwide with in situ DBS systems are likely to require MRI at some stage of their lives, possibly for indications completely independent of their DBS therapy. B1+rms-limited product-label compliant versions of routine brain and spine protocols were created by adjusting sequence parameters while aiming to maintain image quality and we report our experience using these in healthy volunteers and patients with implanted DBS devices.

Methods. New routine brain and spine MRI protocols with B1+rms≤2.0μT at 1.5T in whole-body transmit mode, using standard receive-only head and spine coils were set up and tested in 12 healthy adult volunteers. The signal-to-noise ratio (SNR), CSF vs. brain contrast-to-noise ratio (CNR), grey vs. white matter CNR, image sharpness and artefacts were rated by an experienced neuroradiologist. Quantitative SNR and CNR were obtained based on regions from anatomical labelling using a geodesic information flow approach2. To date, 9 patients with in situ DBS were successfully scanned using these protocols. New specialist post DBS-implantation sequences were also designed including 3D T2 SPACE–FLAIR and quantitative susceptibility mapping (QSM)3 and validated in a series of 6 healthy volunteers. Images were rated by a neuroradiologist.

Results. There was a 28.7%-56.6% increase in total scan time with the B1+rms -limited vs. conventional routine brain and spine protocols. For the B1+rms-limited sequences, head specific absorption rate (SAR) varied between 0.5-1.1 W/kg for the brain protocol, and the whole-body SAR varied between 0.3-0.9 W/kg for the spine; as expected, the B1+rms was approximately the same across subjects. For the majority of subjects the image quality of the B1+rms-limited scans was either equivalent to, or exceeded, that of our routine protocols. SNRs and GM/WM CNRs for the routine brain and B1+RMS-limited sequences were similar. The conspicuity of the subthalamic nucleus (STN), the most frequent DBS target, was higher in the SPACE-FLAIR and QSM images than in original post DBS implantation T2-weighted images.

Conclusion. Our analysis showed the new sequences produced images of similar or marginally better image quality than the conventional protocols used in our centre. Monitoring/limiting B1+rms instead of SAR has an additional advantage of values remaining constant irrespective of patient size, so there is no requirement to adjust sequence parameters for each patient. The new protocols will allow the growing number of patients with in situ DBS systems to undergo neurological MRI examinations.

References. 1.L. Zrinzo, F. Yoshida,M.I. Hariz, J. Thornton, T. Foltynie, T.A. Yousry, P. Limousin Clinical

safety of brain magnetic resonance imaging with implanted deep brain stimulation hardware: large case series and review of the literature, 2011 World Neurosurgery, vol 76 (1-2), pp. 164-72 2.M. J. Cardoso, M. Modat, R. Wolz, A. Melbourne, D. Cash, D. Rueckert, and S. Ourselin, Geodesic

Information Flows: Spatially-Variant Graphs and Their Application to Segmentation and Fusion, 2015 IEEE Transactions on Medical Imaging, vol. 34, no. 9, pp. 1976-1988 3 K. Shmueli J.A. de Zwart, P. van Gelderen, T.Q. Li, S.J. Dodd, Duyn JH. Magnetic susceptibility mapping of

brain tissue in vivo using MRI phase data, 2009 Magn Reson Med.vol. 62(6), pp.1510-22

Acknowledgement: Financial support for this study was provided by Medtronic Inc.

Off-Label Magnetic Resonance Imaging of an InterStim II Sacral Nerve Stimulator Device 1Fry, A., 1Edward, R., 1Kotnis, N., 1Wright, P. 1Medical Physics and Medical Imaging, Royal Hallamshire Hospital, STHFT, Sheffield, UK.

Background: The InterStim II sacral nerve stimulator (SNS) (Medtronic Inc, Minneapolis, MN, USA) is indicated for urinary or bowel control for treatment of urinary retention, overactive bladder symptoms and chronic faecal incontinence in patients where conservative treatment has failed.1 It is not unusual for these patients to have comorbidities which require the use of magnetic resonance imaging (MRI), with patient numbers only likely to increase as demand for MRI grows. In England alone the number of MRIs performed has grown 276% during the 10 years from 2006 to 2016 and is now the equivalent of one MRI scan per 21 of the population per year.2 Presentation: A 53-year-old female was referred for MRI of the left hand for further characterisation of an indeterminate mass following an US examination. The patient had the SNS system (InterStim II 3058 generator; 3093 lead electrode) implanted in 2009 and the battery had been changed 3 months prior to the MRI. While this system is classed as MR Conditional for head scans using a Tx/Rx head coil, other body areas are not recommended by the manufacturer. Investigation: The MHRA have produced guidance documents for off-label use of medical devices and for scanning patients with implants where MRI may be contraindicated3,4. Following MHRA guidelines a risk assessment was conducted by an MR Clinical Scientist. The risk assessment determined the presence and severity of any additional risks and was used to evaluate existing and additional control measures that might be implemented to mitigate risks where possible. The manufacturer’s guidelines stipulates that only a Tx/Rx head coil may be used. With this type of coil RF power is largely deposited in the tissue within the coil, dramatically limiting the potential for heating elsewhere5. Additionally, exposing the device to the RF emissions may induce unwanted stimulations. The risk assessment identified that so as not to vastly increase the risk of heating and unintended stimulation, a Tx/Rx knee coil should be used with the patient positioned prone with arm extended above the head (“superman” position). It was considered that the risk of heating and stimulation was slightly reduced compared to a head scan because the device and leads were further away from the Tx/Rx knee coil with the patient in the “superman” position. There was potential for increased movement artefacts on the images due to patient position, which can be difficult to tolerate. The patient was confident that they could tolerate the position for the duration of the scan and foam padding was used to stabilise the hand within the coil. Outcome. Informed consent was obtained from the patient by the MR Safety Expert (MRSE). The patient’s device was turned off with the patient programmer according to the manufacturer’s instructions and the patient positioned prone on the MR scanner table. The left hand was positioned inside the Tx/Rx knee coil and padded with foam to limit movement. MRI continued uneventfully using a standard hand protocol with all sequences having WB-SAR << 2 Wkg-1, as displayed on the scanner console during scanning and reported in the DICOM header information. Communication between the MR radiographer and the patient was maintained between sequences and the patient reported no discomfort or sensations of heating. Diagnostic images were obtained and the SNS device was successfully reactivated using the patient programmer with the patient reporting normal function of the system following the MRI examination. The MRI examination allowed further characterisation of the mass compared to ultrasound. Axial T1 images appeared to show a non-fatty component to the lesion, making it possible that it was not a simple lipoma. The patient was referred to the sarcoma multidisciplinary team (MDT) for urgent follow up. Conclusion. This case demonstrates that MRI scanning at 1.5 T using a Tx/Rx extremity coil may be carried out safely for this SNS system, following a suitable risk assessment and implementation of risk reduction measures. Institutions wishing to scan patients with SNS devices contraindicated for MRI should follow MHRA advice3,4 and produce a risk assessment considering local hazards and variations in addition to the risks highlighted here. References. (1) Medtronic Inc. 2008. Medtronic InterStim Therapy - Indications Insert. MN, USA: Medtronic Inc. (2) NHS England 2016. Diagnostic imaging dataset annual statistical release 2015/16. (3) Medicines and Healthcare products Regulatory Agency, 2014. Guidance: off-label use of a medical device. London, UK: MHRA. (4) Medicines and Healthcare products Regulatory Agency, 2015. Safety guidelines for magnetic resonance imaging equipment in clinical use. V4.2 (5) Nagy Z, Oliver-Taylor A, Kuehne A, Goluch S, Weiskopf N. Tx/Rx head coil induces less RF transmit-

related heating than body coil in conductive metallic objects outside the active area of the head coil. Front Neurosci. 2017; 11: 15.

Comparison of MRI Screening Policies for Metallic Intra-Orbital Foreign Bodies 1,2,3Barfoot T, 4,5Ashmore J, 2,3Charles-Edwards G 1Medical Engineering and Physics, King’s College Hospital NHS Foundation Trust, 2Medical Physics, Guy’s & St. Thomas’ NHS Foundation Trust, 3Division of Imaging Sciences & Biomedical Engineering, King’s College London, 4Department of Neuroradiology, King’s College Hospital NHS Foundation Trust 5Department of Medical Physics and Bioengineering, Raigmore hospital, NHS Highland.

Background. The majority of metallic intra-orbital foreign bodies (mIOFBs) are ferromagnetic (1)

and have been observed ex vivo to undergo significant displacement when exposed to magnetic fields associated with MRI (2). Nevertheless, retained inorganic IOFBs can be well tolerated (3) with a report suggesting 11% are left in situ after ophthalmic investigation (1). Radiographic screening, in the form of an orbital x-ray, to verify or exclude the presence of mIOFBs is widely used as part of MRI safety screening protocols to avoid displacement of the mIOFB causing injury to the eye. Two alternative schools of thought have emerged with regards to the criteria for radiographic screening of patients with a history of mIOFBs: Policy A: Only patients that sought medical attention, regardless of whether they were informed by a medical practitioner that no mIOFB was present or that it was removed. Policy B: All Patients, unless they were informed by a qualified medical practitioner of the absence or complete removal of the mIOFB. This work investigates the supporting evidence and considers arguments for and against each approach. Methods. A literature review of PubMed was performed to identify reported adverse events related to mIOFBs in the MR environment. Additionally, a review of the occurrence and outcomes of orbital x-rays used for radiographic screening of mIOFBs prior to MR examination was performed across two NHS trusts over a 5 year period.

Results. The literature review revealed one case of unilateral blindness involving a 2x3.5mm iron

fragment on a 0.35T MRI scanner (4) two cases resulting in hyphema (5)(6), the second at 1.5T caused by the patient erroneously believing complete mIOFB removal by medical practitioners, retained mIOFB was 1-2mm in length (6). A fourth case of cataracts resulting from a 1mm mIOFB at 1.5T (7). Across both Trusts 0.053% (253/477,257) of MR examinations involved prior orbital x-ray screening, with 4.35% (11/253) having a positive finding for mIOFBs. A change from policy B to A in one trust, for a period of 10 months to date, resulted in a non-statistically significant change (p>0.05) from 0.056% (108/193,885) to 0.048% (24/50,048) orbital x-rays respectively.

Discussion. Policies A and B have different potential failings. For policy A, retained mIOFBs for

which medical attention was never sought will be missed. One such case was identified in the literature (8). For policy B, any mIOFBs retained after medical attention, which are unknown to the patient, will be missed. Again, an example of this resulting in an adverse incident has been reported (6). Both policies have potential failings in being reliant on accurate information from the patient, although perhaps remembering whether medical attention was sought may be more likely to be remembered correctly than whether the patient was told whether everything had been removed. Nevertheless, the very small number of reported incidents of ocular damage due to mIOFBs following exposure to MRI suggests that most cases of mIOFB that are missed do not lead to eye damage (9). Conclusion. Comparison of these policies highlights different, potentially relevant failings in each. Further data are required to establish whether one approach is more appropriate overall. From the data compared here, neither policy leads to significantly different numbers of patients sent for orbital x-ray. References. (1) Roper-Hall, BJO 38(2):65-99 (1954) (2) Gunenc et al, Doc Opthalmol 81(4):369-78 (1992) (3) Ho et al, OPRS 38(2):232-6 (2004) (4) Kelly et al, AJN &(2):243-5 (1986) (5) Ta et al, AJO 129(4):533-4 (2000) (6) Lawrence et al, MRI 33(3):358-61 (2015) (7) Vote and Simpson, Clin and Exp Opthamol 29(4):262-4 (8) Siedlecki et al, Case Rep Opthalmol Med 2016:3918592 (2016) (9) Siedenwurm et al, AJNR 21:426-433 (2000)

Performing MRI Scans on Cochlear Implant and Auditory Brainstem Implant Recipients - 13 Year Review 1Tam Y, 2Patterson I, 3Neil Donnelly, 3Tysome J, 3Axon P, 4Bance M 1Emmeline Centre for Hearing Implants, Addenbrookes Hospital; 2Department of Radiology, MRIS, Addenbrookes Hospital 3Department of Otolaryngology, Addenbrookes Hospital 4Department of Neurosciences, Cambridge University

Background.

The requirement for scanning of recipients with cochlear implant and their derivatives has always been a requirement. But MRI departments have been reluctant to scan these implants due to the metal casing of the receiver stimulator package. The biggest drawback to safely perform MRI scans is the presence of the internal magnet located in the receiver stimulator.

Outside the MRI scanner, the receiver stimulator magnet is subjected to a linear force pulling it into the MRI scanner. Inside the MRI scanner the magnet is subjected torque forces causing it to dislocate from its retainer in the receiver stimulator. The metal case and the magnet causes image voids and spatial distortion in MRI images of the head.

Methods.

Initially in vitro test were performed on cochlear implants from 3 different manufacturers, Medel Pulsar ci100, Advanced Bionics HiRes90k and Cochlear CI24RE. Qualitative force tests were performed on each of the three implants in the magnetic field of a 1.5T scanner. The implants were attached to MRI phantoms to test the effects of the cochlear implants on quality, with magnets in situ and with magnets removed (where possible). Heating effects on the cochlear implants from standard T1 and T2 head scans were assessed qualitatively.

After undergoing the MRI scan sequences, the implant function was checked by connecting the implants to clinical software and running telemetry checks to ensure the receiver electronics was functional, and any heating effects of the scans did not short out the insulation of the electrode array1.

Results. Over the past 13 years the MRI unit at Addenbrookes Hospital, from Sep 2004 to Aug 2017. Total number of patients scanned 72, youngest 2 year and oldest 86 years. 29 recipients have more than one MRI sessions and 5 recipients have 20 MRI sessions or greater. Total MRI sessions 336 and 508 individual MRI procedures, resulting in a total of 2217 MRI sequences excluding localisers.

Discussion. Complications is 2.7% for CI recipients and 1.8% for ABI recipients, these are caused by dislocation of the receiver stimulator magnet or excessive pressure caused by the head bandage. Most of the complications occurred when scans were performed with the receiver stimulator magnet in situ. Primarily caused by dislocation of the magnet from its retainer or pain caused by torque forces on the magnet whilst in the MRI scanner2.

Conclusion. No implant failures occurred in this 12 year period where cochlear implant recipients received one or more MRI procedure. No problems encountered with demagnetisation of the internal receiver stimulator magnet.

Modifications to scanning procedures with introduction of stiffening inserts and use of local anaesthetic over the receiver stimulator package and keeping sequences to a minimum has reduced complications and aid in the comfort of the patient during the MRI scan sequence.

Key references. 1Tam YC et al, Magnetic Resonance Imaging in Patients with Cochlear Implants and Auditory Brain Stem Implants, Cochlear Implants International, Volume 11, 2010, Pages 48-51 - Issue sup2: 2F. Hassepass, et al, Magnet Dislocation: An Increasing and Serious Complication Following MRI in Patients with Cochlear Implants. Fortschr Röntgenstr 2014; 186(07): 680-685

POSTER: MRI-related magnetic field exposures and risk of commuting accidents and abnormal uterine bleeding – a cross-sectional survey among Dutch imaging technicians 1Huss A, 1Schaap K, 1Kromhout H 1Institute for Risk Assessment Sciences, Utrecht University, the Netherlands

Background: We assessed two different health effects in imaging technicians working with magnetic resonance imaging (MRI). (1) Imaging technicians may experience acute effects such as vertigo or dizziness when being exposed. A previous study had reported an increased risk of accidents in MRI exposed engineers (Bongers et al 2015). We evaluated reports of (near) commuting accidents in imaging technicians. (2) In a second research question, we assessed whether we could confirm previous case reports of menometrorrhagia (prolonged/ excessive uterine bleeding, occurring at irregular and/or frequent intervals; Gobba et al 2011) in MRI workers using intrauterine devices (IUD).

Methods: We invited Dutch imaging technicians registered with their national association, 490 (29%) imaging technicians filled in a questionnaire, of these, 381 (78%) were women. Questions pertained to work practices; (near) commuting accidents when driving or riding a bike, health and lifestyle. Female imaging technicians were asked to fill in questions regarding IUD use and menometrorrhagia.

We used logistic regression to evaluate the association between exposure to MRI-related stray fields (frequency of working with MRI scanners, presence near the scanner/in the scanner room during image acquisition and scanner strength or type), and risk of commuting (near) accidents in the year prior to the survey, adjusted for a range of potential confounders. Logistic regression was also used to evaluate the association with menometrorrhagia.

Results: Our cross-sectional study indicated an increased risk of (near) accidents if imaging technicians had worked with MRI in the year prior to the survey (odds ratio OR 2.13, 95%CI 1.23 - 3.69). Risks appeared to be higher in persons who worked with MRI more often (OR 2.32, 95%CI 1.25-4.31) compared to persons who worked sometimes with MRI (OR 1.91, 95%CI 0.98 - 3.72), and higher in those who had likely experienced higher peak exposures to static and time-varying magnetic fields (OR 2.18, 95%CI 1.06-4.48). The effect was seen on commuting accidents that had occurred on the commute from home to work as well as accidents from work to home or elsewhere.

Regarding menometrorrhagia, compared to unexposed women not using IUDs, the odds ratio (OR) in women with IUDs working with MRI scanners was 2.09 (95% confidence interval (CI) 0.83 - 3.66). Associations were stronger if women working with MRI reported being present during image acquisition (OR 3.43, 95% CI 1.26 - 9.34). Associations with scanner strength or type were not consistent.

Conclusion: Imaging technicians working with MRI scanners may be at an increased risk of commuting (near) accidents. Exposed imaging technicians using IUDs report menometrorrhagia more often than their co-workers without an IUD or non-exposed co-workers with an IUD. For both of these observations, an explanation of the underlying mechanism is not at hand. The results need confirmation and potential risks for other groups (volunteers, patients) should be investigated.

Key references: (1) Bongers et al, Magn Reson Med. 2016;75(5); (2) Gobba et al, Int J Occ Med Env Health 2012;25(1).

POSTER: Assessment of Noise Levels Outside the MR Environment: A Preliminary Investigation 1Jennings R, 1,2Phillips J 1Deparment of Medical Physics and Clinical Engineering, Singleton Hospital, ABMU Health Board, UK. 2Institute of Life Science, Medical School, Swansea University, UK.

Background. A characteristic of gradient field switching in MRI is the production of acoustic noise. It is well established that for most MRI systems the noise is high enough for anyone inside the MR Environment to require the use of ear protection while the scanner is in operation [1]. There are action values and limits in place for occupational exposure to noise [2]. Based on an observation by the MR Safety Expert that the MR Controlled Access Area seemed louder than expected, a preliminary investigation into the noise levels was performed.

Methods. A type 2250-S hand-held analyser (Brüel and Kjær) was calibrated and four areas were assessed during routine head MRI examinations in the MR Controlled Access Area i.e. outside the MR Environment: close to the exam room door, on a desk 3m away from the exam room door, at ear level by the receptionist’s desk and in the control room.

Results. Based on head examinations that were being run (known pulse-sequences and timings)

the summary of results is given in Fig. 1 for the daily personal exposure level (Lep,d) (db(A) scale) based only on the limited measurements taken in the four areas.

Fig. 1: Daily personal exposure based on an 8 hour day in various areas of the MR Controlled Access Area.

Discussion. The lower exposure action level [2] for peak sound pressure level of 135dB has not been exceeded. The daily personal exposure level (Lep,d) is less than 5% below the lower exposure action level [2] of 80dB. Note that this is based on limited observation. Also, the staff members to not stand directly outside the exam room door for extended periods.

Conclusion. No action values have been reached. However, while obtaining measurements it was loud enough in the areas surrounding the MRI exam room (control room excluded) that voices had to be raised in order to have conversations, particularly near the door to the MRI exam room. Only preliminary results have been obtained so far. Further investigation into noise levels should be undertaken based on worst-case scenarios for the loudest sequences e.g. sequences with echo planar imaging (EPI) acquisition. The seal around the door should be investigated to ensure that the maximum attenuation of sound possible is achieved. MR staff involved in the design/refit of MR units should be aware that it is possible for the sound levels to approach the lower exposure action level in the MR Controlled Access Area.

Key references.

[1] Health Protection Agency. Protection of Patients and Volunteers Undergoing MRI Procedures. Documents of the Health Protection Agency Radiation, Chemical and Environmental Hazards August 2008. RCE-7. ISBN 978-0-85951-623-5

[2] The Control of Noise at Work Regulations 2005. Statutory Instrument 2005 No. 1643. The Stationery Office Limited. ISBN 011072984.

By door On desk

Receptionist's area Control room

0

20

40

60

80

Lep

,d (

dB

)

Lower Exposure Action Level

POSTER: Implantable medical devices in MRI: the serious risks of using out-of-date MR safety information McCann AJ, McGrath C Northern Ireland Regional Medical Physics Service, Belfast Health & Social Care Trust, Belfast, UK

Background. Implantable medical devices (IMDs) are an indispensable component of modern medicine, used in the treatment, regulation and monitoring of a wide range of conditions. In patients referred for MRI, the presence of an IMD presents a safety concern since it may interact with the extreme electromagnetic fields associated with the scanner, with the potential for serious injury or death. These safety concerns are handled routinely by all MRI departments via exhaustive screening processes, expert knowledge of the risks and benefits of scans, and the ability to minimize risk during scanning via technical expertise and appropriate provision for emergency scenarios.

International standards have been published1 on the marking of IMDs to indicate their safety with respect to MRI. In line with these standards, extensive guidance is provided by many implant manufacturers, independent testers2,3 and regulatory bodies4,5 on how to safely proceed with scanning patients with IMDs.

Discussion. The authors present specific examples of IMDs whose MR safety guidance has been altered over time. In the majority of instances, such changes have been to diminish the restrictions on scanning (e.g. removal of ‘landmark exclusion zones’ for certain cardiac devices), following supplementary testing of the device. However, there have been instances where MR conditions have been made more restrictive or involved. Recently (May 2017), it was reported that a patient implanted with a Flowonix Prometra-II programmable drug infusion pump may have received a fatal drug overdose during an MRI procedure6. As a direct result, revised MRI Conditional labelling was issued by the manufacturer, containing the additional instruction that all drug be removed from the pump prior to an MRI procedure. Subsequent to the change, it has still been possible to obtain the outdated version of the MR conditions online.

Conclusion. These issues illustrate how critical it is to have the most up-to-date MR safety labelling information for certain types of IMD encountered in patients referred for MRI. For devices whose restrictions have been relaxed, it may enable a scan to be facilitated where previously it would have been denied, or permit the scanner to be run in a manner that improves image acquisition. Of much greater importance, awareness of implants whose levels of restriction have been increased may mean the difference between a safe scan and a serious adverse incident. The authors recommend that a system be established for alerting MRI users of IMDs that have been affected by MRI and their MR conditions altered as a result.

Key references.

1. Standard practice for marking medical devices and other items for safety in the magnetic resonance environment. ASTM International F2503-13, 2013.

2. Shellock FG, http://www.mrisafety.com

3. MR:comp, https://www.mrcomp.com/en

4. Safety guidelines for magnetic resonance imaging equipment in clinical use. MHRA version 4.2, 2015.

5. Kanal et al. ACR guidance document on MR safe practices. Journal of Magnetic Resonance Imaging 37 501-30, 2013.

6. Flowonix Urgent Medical Device Correction: Prometra II Programmable Pump PL-15071-01. 22nd May 2017.

POSTER: Something to get your heart racing: Scanning of MRI Conditional Active Cardiac Devices 1Poole H, 1Prescott S, 1Bentley R,1Butler S 1Medical Physics, Royal Stoke University Hospital, UK.

Background

Developing and maintaining a safe working practice for the scanning of patients with MR conditional cardiac devices.

With the ever increasing availability of MRI Conditional Cardiac devices (ICDs and pacemakers) [1, 2, 3, 4] comes the increasing demand for MRI scans of these patients. Since 2014 the cardiology department at Royal Stoke University Hospital has been implanting MRI Conditional devices to help reduce the number of patients unable to undergo MRI. In anticipation of these patients coming for MRI, a robust working practice had to be developed in order to maintain patient’s safety during the scans.

Methods

Discussions were held as part of a multimodality meeting involving Cardiology, Radiology and Medical Physics. From this a Standard Operating Procedure (SOP) was written. This outlined the requirements and methods needed to safely scan patients with these devices.

The SOP is reviewed 6-monthly for content and an audit on the patients we have scanned is performed annually. The audit has been used to further develop the process and procedures.

Results

Year

Number of … 2015 2016

Referrals with implanted active cardiac devices 35 77

Referrals rejected due to the device being MR Unsafe 27 (77%)

39 (51%)

Referrals with MRI Conditional active cardiac devices 8 (23%)

38 (49%)

Referrals rejected due to conditions not being met e.g. Exclusion Zone 1 1

Scans carried out safely (without AI / AE) 7 36*

Devices confirmed as working during follow up after scan 7 (100%)

36 (100%)

* One patient did not attend their scan

Discussion

The overall number of referrals for patients with pacemakers or ICDs increased from 35 in 2015 to 77 in 2016. Of these referrals, 23% had MRI Conditional devices in 2015 and 49% in 2016 meaning that more patients are able to safely undergo MRI than before. All accepted referrals were carried out safely without Adverse Incidents (AI) or Adverse Events (AE) as a result of having a robust procedure in place.

Conclusion

Performing this type of audit has helped to identify how the service is expanding with the number of referrals doubling between 2015 and 2016. The audit has helped when reviewing and updating the SOP ensuring that the most up-to-date information is available. In order to ensure that this process remains with patient safety as its primary focus, it is vital to establish a strong working relationship between the MRI and Cardiology department.

Key references [1] Biotronik ProMRI Devices (www.biotronik.com) [2] Boston Scientific ImageReady Systems (www.bostonscientific.com) [3] Medtronic SureScan Pacing Devices (www.medtronic.com) [4] St. Jude Medical MRI Ready Devices (www.sjmglobal.com)

POSTER: A Review of Auditory Implants and MRI Safety Tam Y1, Patterson I2, Donnelly N3, Tysome J3, Axon P3, Bance M4

1Emmeline Centre for Hearing Implants, Addenbrookes Hospital 2Department of Radiology, MRIS, Addenbrookes Hospital 3Department of Otolaryngology, Addenbrookes Hospital 4Department of Neurosciences, Cambridge University

Background There are many types of auditory implants in existence; these can be divided into active and passive auditory implants. Active implants include cochlear implants, but less well know are the middle ear implants (MEI) which is becoming more popular. Passive implants include Bone Anchored Hearing Aids (BAHA).

Active Auditory Implants In 2016 there were 14,300 CI recipients in the UK, with an estimated 700 adults and 550 children implanted annually, in total 1800 implants (the majority of children have bilateral implants, and some bilateral adults with conditions such as ushers).

Middle ear implants are a more recent development to provide hearing for patients with conductive hearing losses. They have moving mass transducers attached to or connected to one of the ossicles, with a subcutaneous receiver package. There are around 1000 MEI implanted patients in the UK and around 100 implants performed per year.

Passive Auditory Implants Around 1000 BAHA’s are implanted each year to aid the hearing of patients with conductive hearing losses, most are percutaneous fixtures but there are larger implanted subcutaneous BAHA fixtures. Numbers are in the tens of thousands but there are no accurate numbers.

Methods Most safety data is published on The “List” at 1MRIsafety.com and 2Shellock. The site and the reference book are updated regularly, but it is difficult for the site to maintain up to date information frequently.

Most active implantable devices are tested for MRI compatibility before coming to market. The devices are tested for force effects in the MRI magnetic field and heating of the device during the scan sequences. Majority of tests are performed with 1.5T and 3T magnetic fields.

Results.

Cochlear CI are MRI conditional for 1.5T and 3T (with the magnet removed), except for CI22M (MRI conditional not implanted in UK), to 1.5T and 3T with magnet removed.

Medel All models MRI conditional for 1.5T, latest model (Synchrony) 3T with magnet in situ. Do not scan with 0.5T and 0.7T MRI scanner because the lamour frequencies of these scanners are close to the working frequency of the receiver stimulator.

Advanced Bionics, Latest implants are MRI conditional for 1.5T and 3T with the magnet removed. The older CI and CII models are contra indicated.

Medel MEIs are MRI conditional for the VORP 503, but not the VORP 502. Cochlear Carina MEI is MRI contra indicated.

Oticon BAHA and Cochlear BAHA and BAHA Attract are MRI safe.

Discussion. With some exceptions, most active implantable auditory implants are either MRI conditional or MRI safe. Shellock appear over cautious with its recommendations for BAHA implants. Published safety information can lag behind manufacturer developments.

Conclusion. If the devices are not published on MRISafety.com or Shellock, advice is to contact the manufacturer of the device for further advice.

Key references. 1MRIsafety.com, http://mrisafety.com/TheList_search.asp 2Shellock, Reference Manual for Magnetic Safety, Implants and Devices: 2017

POSTER: Audit of MR Safety Queries to an MR Physics Team in a Large Tertiary NHS Trust 1Broadbent DA, 1Goodall AF, 1Bacon SE, 1Wilson DJ 1Medical Physics and Engineering, Leeds Teaching Hospitals NHS Trust, UK.

Background Safe running of clinical MRI units relies on the availability of MR Safety Expert (MRSE) advice from appropriately qualified personnel [1,2]. This advice is often provided in-house or by an externally contracted MRI physics team. Support required ranges from queries relating to scanning specific patients (or volunteers) to advice on local rules and procedures and on planning or commissioning new units. This work reviews the characteristics of patient/volunteer specific queries received, and advice given, by an MRI physics team in a tertiary NHS trust which supports clinical and research scanners both within the trust and in other NHS, university and private institutions. Advice is given under a BS EN ISO 9001:2008 accredited quality management system.

Methods Patient/volunteer specific queries are logged when received and database entries are updated as information is gathered and advice issued. Each entry in a 12 month period (1/7/16-30/6/17) was reviewed and categorised in terms of: 1: the site raising the query, 2: type of query (non-active implant, implanted active cardiac device (pacemaker/defibrillator), other active implant, foreign body or other), 3: whether advice was required immediately (within an hour), 4: whether full, partial or no details were provided of make/model (implants), 5: expertise required, 6: time spent on the query and 7: advice given (proceed with conditions, proceed with no extra conditions, seek further information (including by x-ray) or do not scan. The total number of queries for the audit period and each of the 8 previous twelve-month periods were also recorded.

Results 203 queries were received from the 22 scanners supported. The scanners (1.5T closed bore except where stated) included 6 internal clinical scanners, 3 research scanners (two 3T closed bore) and 13 external clinical scanners (one open 0.5T system). The number of queries was similar for the preceding three periods, but before that there had been a steady increase from 33 (2008-09) to a peak of 226 (2013-14). Of the 203 queries, 28% required immediate (<1 hour) advice. 58% related to non-active implants, 33% to active implants (17% cardiac, 16% other), and a small number related to either foreign bodies (6%) or non-implant queries (3%, e.g. pregnancy). External clinical scanners raised far fewer queries (3.1 queries per scanner) than internal clinical (19.8) or research (14.7) scanners. For implants, full details were provided in most queries (54%), partial details in a further 15%, and no make or model was provided for the remaining 31%. Most queries (55%) required less than 30 minutes work to answer, and 11% took over an hour (the longest taking three hours). Advice given was to proceed with scanning, subject to specified conditions or restrictions, in the majority of cases (58%), 17% were advised to proceed with no additional limitations, and 9% were advised not to proceed with scanning due to the risks determined. For the remaining 16% of queries the advice given was to seek further information. This might include: seeking further details about the implant, obtaining x-rays to check for presence of metallic components or requiring a clinician to perform a risk vs benefit assessment based on the clinical need for the scan and the risks described by the MR safety expert issuing the advice.

Discussion This audit summarised the nature of safety queries received by an MRI physics group supporting a variety of MRI units. Most were received in advance of the patient attending, although a significant minority (raised after patient arrival) required immediate advice. Across all cases most could proceed, generally with additional scanning restrictions applied. Scans were advised against in only a few cases. Most queries related to passive implant. For the majority of these details were absent or incomplete, so advice was often given assuming a worst-case scenario. Such advice may be based on local procedures or research conducted for the specific query. Where procedures exist, better communication of these to radiographers may avoid some such queries being raised. Similarly many cases did not require MRSE level input (e.g. where implant details were fully known and MRI conditions readily retrievable). Increased handling of such queries by radiographers could reduce demand on the MRI physics team, and also may allow some out-of-office hours cases to proceed without awaiting MRSE availability, but relies on availability of radiographer time.

Conclusion The typical nature of queries received by an MRI physics group at a large tertiary NHS trust is presented, providing data useful for planning the delivery of an MRSE advice service.

References 1. Safety Guidelines for Magnetic Resonance Imaging Equipment in Clinical Use, MHRA, Nov 2014

2. Scientific Safety Advice to Magnetic Resonance Imaging Units that Undertake Human Imaging, IPEM, Oct 2013