Multiple MRIs demagnetized an internal BAHA magnet – An enriching case for the everyday practice

Abstract

We present a Baha patient with demagnetization after repeated magnetic resonance imagings (MRIs) and demonstrate a novel way of monitoring magnet adherence. A 53-year-old man with a Baha Attract System reported reduced holding force of the external audio processor after nine MRIs (eight at 1.5 T and one at 3 T). Subsequently, the original external magnet was replaced by a stronger magnet without benefit. After three additional MRIs at 1.5 T, a total loss of magnetic adherence was reported. During revision surgery replacing the internal magnet, we used three different methods to document the magnetic field of the old and new internal magnet; a gaussmeter, a flux detector film and in-situ determination of the minimum adhesive magnetic force. Measurements performed with the gaussmeter served as the gold standard and were confirmed by the other two methods. The application of MR imaging at 3 T in patients with a Baha system is off-label use and initiates magnetic-implant damage, while repeated MRI at 1.5 T scans seems to be harmless. To document potential demagnetization, a flux detector film can be used in daily practice due to its simplicity of application and broad availability.

Keywords:

• Baha
• bone-anchored hearing aid
• demagnetization
• electromagnetic field
• MRI

Introduction

The Baha^® Attract System is a bone-conductive ­hearing aid, which transmits the sound by bone-conduction to the cochlea, bypassing the external and middle ear. Baha or osteointegrated hearing aids have proven to be beneficial for patient with conductive, mixed or in some cases unilateral hearing loss. The Cochlear™ Baha^® Attract System is based on transcutaneous bone conduction using two magnets. The internal magnet (BIM400) is attached to the osteointegrated screw, which is implanted in the temporal bone underneath the skin. The external removable magnet is coupled to a sound processor with the vibrating transducer. Generally, there are six external magnet strengths available, starting at no. 1 being the weakest magnet and no. 6 being the strongest. The choice of magnet strengths at first fittings depends on various factors such as skin thickness, hairiness, age, skull anatomy, patient’s need and the surgeon’s preference.

Besides sound transmission across the skin, the outer magnet function is to hold the processor at the intendant position. The intact skin is an advantage of this system, in comparison to skin-penetrating Baha-systems (e.g. Baha^® Connect). The former reduces local infections and skin irritations and improves patient compliance since there is no need for daily cleaning of the penetration site. However, the thickness of the scalp needs to be adequate (3–6 mm) to avoid e potential skin damage due to the magnetic strength [1]. However, increased thickness may decrease sound transmission [2–4].

A disadvantage of the Baha^® Attract System compared to the Baha^® Connect system is the magnetic resonance imaging (MRI) conditional application.

Besides significant limitations of the image quality due to the implanted magnet, the strong static field of the MR scanner might have a permanent detrimental influence on the internal magnet. According to the manufacturer’s instructions, the Baha^® Attract System is labeled MRI conditional for use within 1.5 T MR environment but not at 3 T. Baha-systems with abutment or fixture (e.g. Baha^® Connect) are conditional for both 1.5 T and 3 T MRI [4].

Patient information

We present a case of a 53-year-old man with conductive hearing loss after a radical cavity procedure in 2004. In 2015, a Baha^® Attract System was successfully implanted.

Between January 2016 and December 2017, the patient underwent eight MR scans at 1.5 T. The ninth MR scan was accidentally performed for orthopedic reasons with a static magnet strength of 3 T, which is off-label use. The examination was immediately stopped as the patient experienced painful local heat. Subsequently, the patient noticed increased occurrences of audio processor detachments during head or body movements. Nonetheless, he continued to wear the processor. During the following months, another three MR scans at 1.5 T were performed without complications, the first of them in the opposite head position. In December 2018, the patient presented to our outpatient clinic complaining of the inability to attach the sound processor in any situation to the internal magnet. Therefore, the original external magnet (no. 5) was replaced by a stronger magnet (no. 6). However, the external sound processor still did not attach sufficiently. As a consequence, the patient underwent revision surgery replacing the internal magnet. Postoperatively, the new internal magnet showed sufficient strength to hold the external processor in place.

Intervention

We used qualitative and quantitative methods to document the magnetic field of the old defective and the new intact BAHA internal magnet (BIM400). Furthermore, the magnetic fields of the old and new externally worn magnets (nos. 1–6) were quantified to provide an estimate of clinically relevant differences. The BIM 400 is a diametral disc magnet with a central hole that is magnetized parallel to one specific diameter such that an intact magnet results in having two opposite magnetic induction (B₊, B_–) on the hemicycles oriented perpendicular on this specific diameter of the magnet (Tesla: T = V s m²). The direction of magnetization is indicated and labelled on its surface. We measured the magnetic inductance B of both hemicycles of the magnets with a gaussmeter. Beside the absolute value, we also calculated a normalized relative asymmetry index (AI) as it is used in different fields of research with the following formula [5–7]: $$\mathrm{AI}=\frac{{|B}_{+}\left|\text{-}\mathrm{ }{|B}_{-}\right|}{\left|B_{+}\right|+\mathrm{ }{|B}_{-}|}$$

First, we measured the magnetic field strength at the poles of the BIM400 magnet using a gaussmeter (GM05, Gaussmeter, Hirst, Tesla House, Falmouth, UK). The magnetic field of the new intact and sterile BIM400 was measured in the implant package. After surgery, the old defective explanted BIM400 magnet was put in the same implant package and the magnetic field measurement was repeated. The second BIM400 assessment was achieved in situ by determining the tangential gravitational force, which provoked a release of the strongest external magnet (i.e. no. 6) before and after revision surgery using a water-filled bag. A plastic bag was attached to the external magnet and filled with water. One milliliter of water was assumed to weigh 1 g. The bag was slowly and continuously filled with water until the magnet detached. Subsequently, the amount of water was measured and converted to Newton (N) to illustrate gravitational force and to compare the values with experimentally measured retention forces [8]. Finally, we used a flux detector, a magnetic film (M-08 supermagnete webcraft, Uster, Switzerland) providing a simple qualitative visual instrument to check the magnet in situ. Following revision surgery, we applied the additional tangential force to the new magnet by filling a plastic bag with water in milliliter steps using a syringe as described above.

Results

Figure 1 shows the magnetic field of the intact, defective and external magnets (nos. 1–6). The magnetic field of the two poles of the intact magnet showed absolute values of 131 mT and 127 mT, which is in the same level of magnitude as the weakest external magnet (i.e. no. 1) with 122 mT and 124 mT. The absolute values of the poles of defective magnet were significantly lower with absolute values of 94 mT and 25 mT. The larger deterioration corresponds approximately to the difference between the weakest external magnet (no. 1, 122 and 124 mT) and the strongest one (no. 6, 231 and 238 mT). The demagnetization of the BIM400 resulted in a strong asymmetry between its poles (58%), compared to the asymmetry of the intact BIM400 (2%) and the external magnets (1% in average). The defective internal magnet was still strong enough to hold an external magnet no. 6 in place. With an additional tangential gravitational force of 0.14 N, the external magnet fell off in the preoperative condition. After revision surgery, the additional load increased to 0.44 N.

Figure 1. Magnetic induction of different magnets: internal defective (left), internal intact (center) and external magnets nos. 1–6 (right). For each magnet, both poles and the percentage asymmetry (AI) are depicted. mT: milliTesla.

Figure 2 shows a qualitative visualization of the diametrically magnetized magnet in situ using a flux detector film, before (Figure 2(A)) and after revision surgery (Figure 2(B)). Figure 2(A) shows the asymmetry with lower magnetization of the upper magnetic field prior to revision surgery, whereas Figure 2(B) presents a symmetric situation 1 year after revision surgery. Figure 2(C) illustrates the visual comparison of the intact and the defective BIM400 magnet with external magnets nos. 1–6 (weakest to strongest).

Figure 2. Magnetic field visualization with a flux detector film (M-08 supermagnete webcraft, Uster, Switzerland). (A) The condition of the diametral magnet before revision surgery is asymmetric with lower magnetization of the upper magnetic field. (B) One year after revision surgery shows symmetric magnetization. (C) The visual comparison with external magnets (1–6), normal and explanted BIM400 magnet shows asymmetry of the explanted diametral magnet.

Discussion

In this report, we describe a rare case of a patient who experienced a demagnetization of his Baha^® Attract System (BIM400 magnet) after multiple MR scans. The initial reduction of magnet strength was observed after having accidentally performed an MRI at 3 T (off-label). However, following another three MRIs (at 1.5 T), the patient complained again of further insufficient magnet strength. Because there were no changes in skin thickness, hairiness or skull anatomy, the question arose whether gradual loss of magnetization occurred due to cumulative MRIs or if there was a sudden demagnetization because of the single MR scan at 3 T. MR scanners use a combination of electromagnetic fields and a potential concern is demagnetization of the implanted magnet. Other research groups reported a significant reduction in the strength of the internal magnet when placed into an MRI at 1.5 T [9].

Depending on the orientation of the magnet relative to the magnetic field in the scanner, demagnetization is possible [10]. Furthermore, it is known that magnetic field interactions depend on geometric orientation of poles and it has been reported, that one single MRI scan can cause demagnetization if the angle of the magnet relative to the main field is above 80° whereas exposing the implant’s diametral magnet to the MRI field with a relative angle less than 80° has a lesser or no effect in terms of demagnetization of the implant. Multiple exposures lead to a cumulative demagnetization, but if the magnet is aligned antiparallel the primary weakening of the magnet can be considered as sufficient to cause maximal cancellation [11,12].

There are different techniques to measure and visualize magnetic field strength ranging from classical to high-tech [13]. We measured the magnetic fields of the defective and the intact BAHA internal magnet by using three methods. A gaussmeter served as the gold standard as it is reliable and widely used in the technical setting but not generally available in the clinical setting. It was complemented by the use of a flux detector film, which is broadly available and capable of qualitatively visualize magnetic field strength. However, it is not a quantitative measuring technique. In a clinical setting, the flux detector film is most useful to check whether a magnetic failure is present or if an anatomical issue is triggering magnetic detachment. Prior to surgically explanting a magnet, a primary check with the flux detector film to assess the magnet is reasonable. Lastly, measurements were performed by assessing the retention force of the magnet in situ and in vivo. A water-filled bag was used to imitate translational force and torque as similarly portrayed by the American Society for Testing and Materials (ASTM International) and International Organization for Standardization (ISO/TS 10974) in an in vitro setting. In our case, we modified this technique to be used in vivo, which so far has not been described in the literature. The retention force prior to the exchange of the internal magnet was below the range of an optimal retention force of 0.23–0.4 N [8]. After the revision surgery, the retention force with the same external magnet (no. 6) surpassed the optimal retention force without any skin irritations. All three types of measurements indicated equal tendencies and correlation in magnetic field strength within the new and the defective magnet.

One argument for a single time event is given by the company’s manual, which approves the implant as MR conditional at 1.5 T but does not provide access to MRI at 3 T. Neither the information in the operating manual has changed nor a clinical report of an adverse event using MRI at 3 T was found by the authors. While BAHA abutment fixtures resist to magnetic forces up to 9.4 T without adverse effects, an implanted magnet function seems to be damaged by using MRI at more than 1.5 T [14]. Implanted devices containing ferromagnetic substances cause significant artifact limiting the image quality. In terms of artifact, the magnetic field of the MR scanner is competing with the magnetic field of the internal magnet. A higher static magnetic field of the MRI with implanted BAHAs are challenging and there are various techniques to minimize implant related artifacts, e.g. head position [15].

For best results, three things should be considered. First, maximizing image quality by reducing artifacts. Both 1.5 T and 3 T MRIs are prone to the same sort of artefacts among other RF-induced eddy current artifacts [16,17]. The latter can lead to increased susceptibility artifacts, particularly around metallic implants, due to the stronger fields present in 3 T MRI. With an understanding of the artifact mechanism, methods for correcting the RF-induced eddy current can and must be applied [18]. The artifacts surrounding the implant intensify with increase in the degree of angulation from the direction of the main magnetic field of the scanner and can be reduced by change of implant positioning or head rotation, respectively [15,19]. Likewise, artifacts are highly dependent on the technical scanning sequence [20,21]. Second, the device should not be damaged and third, the patient should not be harmed.

Compared to other devices, such as cochlear implants, several MRI contraindications have been adapted in the manual after gaining clinical experience over time [22,23]. Also, the use of other contradicted devices such as cauterization or transcranial electric stimulation does not necessarily produce an implant failure [24].

An argument for the gradual loss of magnetization following MRI at 3 T is the patient history. The patient did not observe any deterioration of the holding force after several images at 1.5 T prior to the MRI at 3 T. He noticed an initial reduction of the holding force after one incomplete MRI at 3 T from which it can be derived that the magnet was not aligned antiparallel because otherwise demagnetization had to be more pronounced. He continued wearing the device. Following additional three MR scans at 1.5 T, the holding force reduced even more such that he presented at our outpatient clinic. Several factors such as thickness of the skin, change of body weight, changed hairstyle may have had an effect by increasing the gap between the magnets. However, we did not observe such changes, in particular as the magnet was exchanged without the need of additional surgical reduction of the skin thickness.

The permanent demagnetization only happens when a certain magnitude (remanence coercivity) of the external magnetic field in the opposite direction is applied. For total demagnetization, if not aligned antiparallel, this process needs to be repeated but once the coercivity is reduced, less strong magnetic fields can demagnetize. According to the patient history, the coercivity value of a normative BIM400 magnet seems to be reached for external magnetic field between 1.5 T and 3 T; otherwise, their decremental effects should have been observed already after eight MRIs at 1.5 T. After one single MRI at 3 T, however, the coercivity value seems to have decreased below 1.5 T, which likely lead to a cumulative effect of any further MRIs. In addition, much stronger eddy currents leading to heat dissipation on the implant surface surrounding tissue unbearable for the patient could be an additional source of demagnetization at least at the magnet’s surface [25].

Conclusions

We recommend to strictly comply with the manufacturer’s guidelines regarding the use of MRI in Baha patients. In this particular case, MR imaging at 3 T should have been avoided because implant demagnetization seemed to have been initiated, and additional MRIs at 1.5 T likely have further demagnetized the magnet.

If an MRI at 3 T is clinically necessary in Baha patients, the manufacturer should be consulted and the health care workers as well as the patients should be made aware of the potential detrimental effect on the magnet. The use of a flux detector showed a correlation and equal tendencies in loss of magnetic field strength as the other two methods, which were more complex in its application. From this particular case, we conclude that in order to document the effects of demagnetization, a flux detector film could be applied as a first-line method of measurement if magnet damage is suspected following MR imaging, due to its simple application, broad availability and reliability.

Informed consent

The project conforms to the Code of Ethics of the World Medical Association in its current version (Declaration of Helsinki). We consulted the Ethikkomission Nordwest- und Zentralschweiz (EKNZ). According to HFG Art. 3, our case report does not require ethical approval as the patient’s personal details were kept anonymous. The authors certify that they have obtained all appropriate patient consent.

Acknowledgements

The authors thank Dr. Flurin Honegger for his input in the physical review.

Disclosure statement

The authors report no conflict of interest.

Additional information

Funding

The external magnets were sponsored by Cochlear AG (Basel, Switzerland).