Article Text

Download PDFPDF
Original article
MRI of the Ex-PRESS stainless steel glaucoma drainage device
  1. Leonard K Seibold1,
  2. Ronald A L Rorrer2,
  3. Malik Y Kahook1
  1. 1Department of Ophthalmology, Rocky Mountain Lions Eye Institute, University of Colorado at Denver, Aurora, Colorado, USA
  2. 2Department of Mechanical Engineering, University of Colorado at Denver, Denver, Colorado, USA
  1. Correspondence to Dr Malik Y Kahook, Department of Ophthalmology, Rocky Mountain Lions Eye Institute, University of Colorado at Denver, 1675 Aurora Court, Mail Stop F-731, PO BOX 6510, Aurora, CO 80045, USA; malik.kahook{at}


Aim To evaluate the magnetic properties of the Ex-PRESS stainless steel glaucoma drainage device during MRI.

Design Experimental study.

Methods The Ex-PRESS glaucoma drainage device (316L stainless steel) was examined for magnetic field interactions under standard 1.5, 3.0, and 4.7 T MRI scanning protocols. Testing included measurements of translational and rotational motion of the device induced by static magnetic fields. In addition, the change in the temperature of the device was measured to assess the presence of radiofrequency heating of the stainless steel device.

Main outcome measures Degree of angular deflection, device displacement and rotation, and change in temperature.

Results During induced torque testing, displacement did not occur under 1.5 and 3.0 T conditions, although a significant amount of displacement occurred in the 4.7 T environment. Increasing amounts of angular deflection were demonstrated at all three field strengths. We did not record significant temperature changes during brain MRI sequences at any of the three MRI strengths.

Conclusions The Ex-PRESS glaucoma drainage device, manufactured from grade 316L stainless steel, does move in the presence of high magnetic fields. The clinical significance of this finding and translation to in vivo conditions are not currently known. Further studies are needed to better understand how patients might be affected by the magnetic properties of this device post implantation and how patients should be counselled in regards to safety of MRI in the early and late postoperative period.

  • Ex-PRESS
  • 316L stainless steel
  • MRI
  • magnetic
  • safety
  • glaucoma
  • intraocular pressure
  • imaging
  • diagnostic tests/investigation
  • treatment surgery
  • experimental laboratory

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Glaucoma is a chronic disease characterised by progressive loss of the retinal nerve fibre layer and associated visual field loss.1 2 Intraocular pressure (IOP) is currently the only modifiable glaucoma risk factor.3 The treatment approach to decrease IOP traditionally starts with topical medications followed by laser trabeculoplasty, eventually leading to invasive surgery in those refractory to earlier interventions. Trabeculectomy remains the gold standard surgical intervention when non-invasive techniques have failed, although newer techniques are currently in use and under investigation. While trabeculectomy is very effective at lowering IOP, many intra-operative and postoperative complications can develop, making other options for lowering IOP in a safer manner more desirable.4 5

The Ex-PRESS drainage device (Optonol Ltd, Neve Ilan, Israel), made of stainless steel, has been introduced as an alternative to traditional trabeculectomy. Early implantations of this device took place under conjunctiva without use of a scleral flap and led to many complications.6 7 The device is now implanted under a scleral flap creating a uniform channel between the anterior chamber and sub-Tenon's space. Initial retrospective studies comparing surgical outcomes between traditional trabeculectomy and the Ex-PRESS shunt showed no difference in IOP reductions at 1 year, with a statistically significant decrease in postoperative hypotony and choroidal effusions in the Ex-PRESS group.8 More recent studies demonstrated significantly higher success rates with Ex-PRESS versus trabeculectomy with a lower complication profile.9 These positive findings have led to the increased popularity of the device with thousands of implantations since its introduction.

As the popularity of the Ex-PRESS device has grown, concerns have been raised about the safety of the metallic implant in MRI conditions.10 11 The device is composed of 316L grade stainless steel considered to be austenitic and non-ferromagnetic.12 Until recently, the only reported MRI testing of the Ex-PRESS device's magnetic properties was commissioned by the manufacturer (Optonol Ltd). The executive summary of these studies, done by Frank G Shellock (Shellock R & D Services, Inc., Los Angeles, California, USA) states the device should not present an additional hazard or risk to a patient undergoing an MRI procedure using a static 3 T magnetic field or less. ( However, the manufacturer recommends that patients avoid MRI within 2 weeks post implantation but no reason for this is offered. In a recent report, De Feo and colleagues evaluated MRI brain images taken in patients with the Ex-PRESS device in place. Although no safety concerns arose, they did note imaging artifacts generated by the device potentially compromising diagnostic evaluation.10 Even more recently, Geffen et al found that the Ex-PRESS device does experience some movement under 1.5 and 3.0 T conditions. However, they failed to show any movement in a simulated in vivo setting.11

Previous studies evaluating the safety of middle ear prostheses made of 316L stainless steel and of similar size found that the implants do move in the presence of high magnetic fields.13–15 In addition to movement, metallic implants are also subject to heating during radiofrequency (RF) pulses that, if significant, could potentially lead to destructive tissue burn and necrosis.16 The American Society for Testing and Materials (ASTM) has guidelines for MRI safety requiring testing that includes data on torque, displacement force and RF heating. In this study, we evaluate all three of these elements for the Ex-PRESS device under 1.5, 3.0, and 4.7 T MRI conditions.

Materials and methods

The protocol used to test the Ex-PRESS device was modelled after that used by Williams et al for middle ear implant testing.15 The Ex-PRESS devices were obtained from stock surgical supply having never been used in patient surgery. All implants were initially screened with a handheld magnet (0.2 T) prior to introduction into the MRI magnetic fields. Two different R-50 Ex-PRESS devices were used in all three strengths of the MRI to avoid the possibility of a single defective device. In order to investigate whether the observed results were exclusive to the R-50 model, all three study components were repeated on two different P-50 devices in the 4.7 T MRI.

Magnetic field-induced torque test

The purpose of this test was to evaluate any rotation or movement of the device in a static magnetic field. The device was placed in a plastic Petri dish lined with grid paper (figure 1). The device was then fed into the bore of the magnet starting at 20 cm outside the bore edge and continued inside until the isocentre was reached. Its position was constantly monitored by two observers for movement. The device was carefully removed and position noted for any change. This process was performed with the device oriented both parallel and perpendicular to the magnet bore with each position being repeated twice for both the 1.5 and 3.0 T magnets. Due to the small bore of the 4.7 T magnet, the device was placed on a flat wooden platform lined with grid paper. The position was marked in a similar fashion as before and the device was introduced into the bore of the magnet. Again, the position of the device was constantly monitored by two observers for change in position. A mirror and external light source were used to aid in visualisation of the device. The device was observed for rotational movement as well as displacement.

Figure 1

The Ex-PRESS device with position marked prior to induced torque test.

Magnetic field translational force test

The purpose of this test was to evaluate for the presence and magnitude of translational force acting on the device in a static magnetic field. A similar apparatus to that used by William and colleagues was used for this part of the experiment.15 Each device was suspended from a 7-0 vicryl suture on a wooden dowel. A plastic protractor was placed parallel to the suture and magnetic field so that the angle of deflection from vertical could be measured (figure 2). Again, due to the limited bore size in the 4.7 T machine, an external light source and mirror oriented 45° from the protractor was used to continually visualise the device. The device's position was noted on the protractor a sufficient distance away from the magnet bore and then constantly observed by two researchers as it was introduced into the bore. The maximum angle of deflection was noted during the course of the device all the way to the isocentre and during withdrawal to outside the bore. The force acting upon the device was calculated according to the formula F=mg × tan θ, where F is the force in dyn (cm×g/s2), m is the mass of the device in grams, g is the gravitational acceleration (980 cm/s2) and θ is the angle of deflection.15 A dyne is defined as the force necessary to accelerate a mass of 1 g at a rate of 1 cm/s2 squared. This process was repeated for each device and in each magnet.

Figure 2

Translational force apparatus.

Radiofrequency heating test

The purpose of this test was to evaluate any temperature change experienced during typical brain MRI conditions. The test was conducted by mounting the device onto the MRI patient platform. Due to the small size of the device, a 1 litre plastic container of sterile water was placed adjacent to the device so the MRI technician could locate an object to scan. A non-magnetic, type T, Teflon thermocouple was used for temperature monitoring. The OMEGA HH506RA model digital thermometer (OMEGA Engineering, Inc., Stanford, Connecticut, USA) was used to monitor temperature during the scans; it has a sensitivity of detecting changes as small as 0.1°C. The Ex-PRESS device was affixed to the thermocouple end using a minute amount of cyanoacrylate glue. The temperature was allowed to equilibrate within the bore of the MRI before recording a baseline temperature. The temperature was then recorded every 10 s during each phase of the MRI scan. A post-scan temperature was also recorded and the thermocouple was checked to ensure contact with the device after scanning. A typical brain MRI protocol was used on each scanner including T1, conventional spin echo, T2, flair, and gradient echo sequences. A control set-up was used with the thermocouple only as well as the thermocouple plus cyanoacrylate to ensure no temperature change could be explained by the set-up alone.


The Ex-PRESS device showed no induced torque under the static field of the 1.5 and 3.0 T MRIs. No displacement or rotation was observed with the device oriented parallel or perpendicular to the magnetic field. However, when inserted into the 4.7 T magnetic field, the Ex-PRESS devices were displaced completely to the limits of the Petri dish (>40 mm). Any rotational force could not be calculated given the extreme movement of the device. This same result was obtained with both the R-50 and P-50 models of the implant.

For the translational force test, an increasing level of angular displacement was found with each successive increase in magnetic field strength. Table 1 depicts the degree of displacement under each MRI environment. In addition to the significant amount of displacement in the 4.7 T field, a notable difference was observed between the P-50 and R-50 models of the Ex-PRESS, probably related to the variability in the manufacturing process as well as minor differences in the mass of the different devices. The translational force acting upon each device is also listed in table 1.

Table 1

Movement, force, and temperature change of the Ex-PRESS

The differences recorded between pre- and post-scan temperature was no greater than 0.3°C for any of the MRI machines. The largest temperature change listed in table 1 was observed during the gradient echo sequence at each MRI strength. The minimal temperature change observed did not appear to be dependent on the strength of the static magnetic field. There was no temperature change demonstrated by testing with the probe alone and the probe with cyanoacrylate glue only.


Although 316L stainless steel is viewed as non-magnetic metal, our findings clearly show that the Ex-PRESS device does have magnetic properties.12 Previous studies have demonstrated similar results on implants made of the same 316L grade stainless steel. Initial review by Syms and Petermann of MRI safety for stainless steel middle ear implants revealed some movement occurred in the 1.5 T environment.13 With growing concern for safety and stronger MRI magnet strengths being developed, later studies demonstrated that implants made of 316L stainless steel showed a significant degree of magnetism when subjected to the 3.0 and 4.7 T MRI environment.14 15 However, the same devices failed to show movement or loosening when placed in a simulated in vivo environment of cadaveric bone.15

Optonol, the manufacturer of the Ex-PRESS drainage device has released a statement of MRI safety for the device in a static magnetic field environment of 3.0 T or less (¼1&LNGID¼1&TMID¼84&FID¼675&PID¼0&IID¼0&CMD¼LOGIN). This evaluation, performed by Frank G Shellock (Shellock R & D Services, Inc.), showed no significant heating in the same MRI conditions. However, in the same document, Optonol recommends that no MRI testing be done within 2 weeks of implantation. It is unclear from the report why this recommendation is made as no details of device displacement or movement were included.

We have attempted to cover all three elements from the ASTM guidelines for MRI safety in the design of this study. Our findings partially validate and further extend the work of Geffen and colleagues and parallel those of previous studies on 316L grade stainless steel micro-implants.11 While fairly inert in the 1.5 T environment, the Ex-PRESS drainage device shows increasing angular deflection in the 3.0 T MRI and extreme angular deflection in the 4.7 T environment. A significant amount of induced torque was also displayed under 4.7 T conditions. Despite the demonstrated movement, no significant temperature changes were noted in any of the three MRI environments suggesting no RF heating. This is clinically relevant as the eye is unique in its limited ability for heat dissipation.16Since we found similar results using four different devices (two from each model) taken from our regular surgical stock, we do not think a claim can be made that our findings were artifacts caused by defective devices.

The clinical significance of the findings in this study remains unclear and requires further testing. The demonstrated movement in this study confirms that while the 316L stainless steel Ex-PRESS drainage device may be considered MRI-safe up to 3.0 T, it cannot be considered non-ferromagnetic. A more appropriate description would be weakly ferromagnetic, only reactive in high magnetic fields. Simulated in vivo studies thus far have failed to show macro-positional changes in a static magnetic field up to 3.0 T. Further studies should be carried out to investigate the potential for macro and micro positional changes of the device under actual MRI scanning conditions, especially in the first 2 weeks post implantation prior to maximal healing and development of scar tissue surrounding the device. It is unclear if micro-movements during the early postoperative time could lead to induced inflammation and decrease the success of filtration surgery. It is also unclear how micro-movements of the device might influence nearby tissues such as a thin scleral bed or thin bleb at any point after implantation.


In conclusion, the stainless steel Ex-PRESS glaucoma drainage device does possess ferromagnetic properties. While the free-standing device experienced significant movement in high magnetic field environments, no significant RF heating occurred. Further studies are needed to better understand the safety implications of these findings so that we can better counsel our patients and be more informed when the need for MRI arises post-implantation of the Ex-PRESS glaucoma drainage device.


The author would like to acknowledge Dr Natalie Serkova for her assistance with 4.7 T imaging at the University of Colorado Cancer Center Animal Imaging Core.



  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles

  • At a glance
    Harminder S Dua Arun D Singh