Biosensing Contact Lenses and Glaucoma
BY ERIC PAPAS, PHD, MCOPTOM, DIPCL, FAAO
It has been 40 years since the first contact lens-based device for measuring intraocular pressure (IOP) was constructed (Greene and Gilman, 1974). Essentially a strain gauge embedded in a silastic gel lens, the arrangement functioned to measure changes of the meridional angle at the corneo-scleral junction induced by corresponding fluctuations in IOP. The idea was not widely taken up at the time, probably because each lens had to be manufactured from molds taken from the ocular surface and then individually fitted to the eye. Nevertheless, the attempt recognized the potential for contact lenses to function as biosensors, in this case driven by the desire for continuous, noninvasive IOP measurement.
The Ups and Downs of Measuring IOP
It is well known that IOP shows diurnal variation, so assessing the difference between the maximum (acrophase) and the minimum (bathyphase) values, as well as when they occur during the 24-hour cycle period, helps with diagnosis, design of treatment strategies, and monitoring of their efficacy. Historically, the only way to get this information has been to take hourly tonometry readings. This is extremely inconvenient, as patients have to remain close to the clinical site for the entire period. Perhaps even worse is that both patients and clinicians have to be woken for measurement every 60 minutes during the nighttime. Apart from being exhausting, it is unclear what effect such repeated rousing may have on the behavior of IOP itself.
Against this background, the attraction of a continuously worn sensor that provides regular pressure readings while permitting normal activities, including normal sleep patterns, is pretty obvious. It has not been surprising, therefore, to see the contact lens-based IOP sensor concept re-emerge over the last 10 years (Leonardi et al, 2004). The culmination of this activity is the commercial availability of one such device incorporating a relatively sophisticated sensor array into a silicone contact lens carrier, with data being transferred wirelessly to an external recorder (www.sensimed.ch, Figure 1).
Figure 1. The Sensimed Triggerfish lens for continuous IOP tracking.
As in the original concept, the technology is said to respond to curvature changes in the region of the corneo-scleral junction, although the measurement sensors are now orientated circumferentially rather than meridionally (Leonardi et al, 2004; Agnifili et al, 2014). It is claimed that an IOP change of 1 mmHg will produce a 3μm alteration in corneal radius of curvature, which can be easily detected by the sensing elements in the contact lens.
So, how well does the contact lens sensor (CLS) measure IOP? We immediately encounter a problem in trying to answer that question because the output from the CLS is in the form of a voltage, so the measurement units are millivolts (mV) and not mmHg. There currently is no way to convert between these two quantities, meaning that absolute IOP values are not readily obtainable (Agnifili et al, 2014; Mottet et al, 2013). The practical consequence of this is that we have to interpret the CLS data output in relative terms. Thus, after an initial baseline voltage level is established during the setup procedure, subsequent responses manifest as voltage fluctuations around this value. Assuming that these variations reflect real changes in IOP, this is potentially helpful for following changes over time, but the actual maximum and minimum pressure values remain unknown.
With this in mind, placing the device onto a cannulated pig eye showed that the output signal closely followed the peaks and troughs of induced, cyclical IOP changes in the range of 17 mmHg to 29 mmHg (Leonardi et al, 2004). This gave some comfort that the response reflected IOP behavior as suggested by theory. However, more recent work, admittedly on only two live human eyes, has been unable to replicate this relationship, perhaps suggesting that the mechanisms linking IOP and corneal curvature change in real eyes are more complex than originally proposed (Sunaric-Megevand et al, 2014).
Making direct comparisons between the CLS measurements and traditional tonometric IOPs in the same eye is the obvious way to resolve this debate, but the need to repeatedly remove and reapply the lens every time that readings are required makes this practically difficult to accomplish over an extended time period. An alternative approach has thus been to leave the CLS on one eye and take regular tonometry reading of the other (Mottet et al, 2013). Doing this in a group of normal subjects produced only moderate correlations between CLS output and air puff, non-contact tonometry (NCT) readings. Although at first glance this is not encouraging, real differences in IOP may well have existed between the right and left eyes of people in the sample. Furthermore, when NCT was used to measure both eyes of each subject, the interocular correlation was actually not that great either, being about the same magnitude as that between CLS and NCT.
While more work is clearly needed to characterize the relationship between CLS output and true IOP, its value currently appears to be as an adjunct, rather than an alternative, to traditional tonometry. That it is well tolerated and can be safely worn for extended periods (Mansouri et al, 2012; Lorenz et al, 2013) makes it attractive as a means of establishing the 24-hour nyctohemeral rhythm, or IOP profile. Repeated measurement over several months suggests that these cyclical waveforms are quite reproducible in normal eyes (Mottet et al, 2013).
Studies in which CLSs have been applied to individuals who have and do not have glaucoma are also beginning to emerge, with interesting outcomes. For example, comparison of nyctohemeral rhythms in groups of normal individuals and patients who have either primary open-angle glaucoma (POAG) or normal tension glaucoma (NTG) (Agnifili et al, 2014) shows differences that may usefully inform both clinical assessment and treatment practices. For example, the bathyphase for all of these groups occurs during the afternoon, meaning that this is perhaps the worst time to conduct in-office IOP measurement. In fact, further scrutiny of the CLS output suggests that IOP is relatively low throughout all regular office hours, so by conducting tonometry during these times we may be underestimating the stresses to which our patients’ retinal nerve fiber layers are exposed. Unfortunately, acrophase events occur inconveniently in the early hours of the morning. Nevertheless, the CLS output does offer insight into the different behavior of each group. Thus, while POAG patients have a maximum that comes relatively early in the morning, somewhere around 1 o’clock am, those who have NTG peak roughly one hour later. People who do not have glaucomatous disease approach their acrophase later still, typically about 4 o’clock am.
Confirmation of these behaviors in future research will give clinicians the potential to assess patients on an individual basis and tailor treatment plans to each particular need. Tuning drug instillation times so that their absorption and diffusion kinetics respect the nyctohemeral rhythm might be one way to achieve this.
Although questions remain, it seems quite likely that IOP assessment using a CLS is only the first example of a field that will be expanding over the coming years. While the tonometer is safe for now, the technology does appear to offer interesting and attractive performance features that deserve our careful attention. CLS
For references, please visit www.clspectrum.com/references and click on document #230.
Professor Papas is executive director of Research & Development, Brien Holden Vision Institute and Vision Cooperative Research Centre, and professorial visiting fellow, School of Optometry & Vision Science, University of New South Wales, Sydney, Australia. The Brien Holden Vision Institute and Vision Cooperative Research Centre has received research funds from B+L, AMO, and Allergan and has proprietary interest in products from Alcon, CooperVision, and Carl Zeiss. You can reach him at email@example.com.