Advertisement

Telemetric Measurement of Intraocular Pressure via an Implantable Pressure Sensor—12-Month Results from the ARGOS-02 Trial

Open AccessPublished:September 20, 2019DOI:https://doi.org/10.1016/j.ajo.2019.09.011

      Purpose

      The aim of this study was to investigate the safety and performance of the second generation of an implantable intraocular pressure (IOP) sensor in patients with primary open angle glaucoma (POAG).

      Design

      prospective, noncomparative, open-label, multicenter clinical investigation.

      Methods

      In this study, patients with POAG, regularly scheduled for cataract surgery, were implanted with a ring-shaped, sulcus-placed, foldable IOP sensor in a single procedure after intraocular lens implantation. Surgical complications as well as adverse events (AEs) during 12 months of follow-up were recorded. At each follow-up visit, a complete ophthalmic examination, including visual acuity, IOP, slit lamp examination, and dilated funduscopy as well as comparative measurements between Goldmann applanation tonometry and the EYEMATE-IO implant were performed.

      Results

      The EYEMATE-IO implant was successfully implanted in 22 patients with few surgical complications and no unexpected device-related AEs. All ocular AEs resolved quickly under appropriate treatment. Comparative measurements showed good agreement between EYEMATE-IO and Goldmann applanation tonometry (GAT) with an intraclass correlation coefficient (ICC(3,k)) of 0.783 (95% confidence interval [CI]: 0.743, 0.817). EYEMATE-IO measurements were higher than GAT, with a mean difference of 3.2 mm Hg (95% CI: 2.8, 3.5 mm Hg).

      Conclusions

      The EYEMATE-IO sensor was safely implanted in 22 patients and performed reliably until the end of follow-up. This device allows for continual and long-term measurements of IOP.
      Intraocular pressure (IOP) is currently the only modifiable risk factor for the onset and progression of glaucoma. Therefore, accurate and reliable measurement of this parameter is the cornerstone of diagnostic and therapeutic decision making in glaucoma.
      • Weinreb R.N.
      • Aung T.
      • Medeiros F.A.
      The pathophysiology and treatment of glaucoma: a review.
      Currently, best practice patterns in IOP evaluation are defined by infrequent measurements of approximately 3 to 6 times per year. It is known, however, that IOP is a highly dynamic biological parameter with short- and long-term fluctuations.
      • Aptel F.
      • Weinreb R.N.
      • Chiquet C.
      • Mansouri K.
      24-h monitoring devices and nyctohemeral rhythms of intraocular pressure.
      The existence of individual, relatively stable circadian IOP profiles has been proposed by some studies but remains controversial.
      • Hatanaka M.
      • Babic M.
      • Susanna R.
      Reproducibility of the mean, fluctuation, and IOP peak in the diurnal tension curve.
      • Mansouri K.
      • Weinreb R.N.
      • Liu J.H.K.
      Efficacy of a contact lens sensor for monitoring 24-h intraocular pressure related patterns.
      • Mansouri K.
      • Weinreb R.N.
      • Medeiros F.A.
      Is 24-hour intraocular pressure monitoring necessary in glaucoma?.
      It is estimated that up to 80% of peak IOP occur outside of normal office hours, often during the nocturnal period.
      • Mansouri K.
      • Weinreb R.N.
      • Liu J.H.K.
      Efficacy of a contact lens sensor for monitoring 24-h intraocular pressure related patterns.
      ,
      • Arora T.
      • Bali S.J.
      • Arora V.
      • Wadhwani M.
      • Panda A.
      • Dada T.
      Diurnal versus office-hour intraocular pressure fluctuation in primary adult onset glaucoma.
      ,
      • Nuyen B.
      • Mansouri K.
      Detecting IOP fluctuations in glaucoma patients.
      There is some evidence to suggest that IOP variability itself may be an independent risk factor for the development and progression of glaucoma.
      • Mansouri K.
      • Weinreb R.N.
      • Medeiros F.A.
      Is 24-hour intraocular pressure monitoring necessary in glaucoma?.
      However, because of the unavailability of prospective studies using continuous 24-hour IOP monitoring technology, this question remains disputed with contrasting data.
      Most methods for measuring IOP require trained personnel and specialized equipment, which limits IOP monitoring to ophthalmologists' or optometrists' offices. The current gold standard for measuring IOP is Goldmann applanation tonometry (GAT), which requires contact of a measuring probe to the cornea and thus the use of topical anesthetics. GAT has remained the standard despite its many shortcomings mainly because of its ease of use and low cost. One of the major sources of error in current tonometry is dependence on the biomechanical properties of the cornea, including central corneal thickness,
      • Ehlers N.
      • Bramsen T.
      • Sperling S.
      Applanation tonometry and central corneal thickness.
      corneal astigmatism,
      • Holladay J.T.
      • Allison M.E.
      • Prager T.C.
      Goldmann applanation tonometry in patients with regular corneal astigmatism.
      corneal curvature,
      • Francis B.A.
      • Hsieh A.
      • Lai M.-Y.
      • et al.
      Effects of corneal thickness, corneal curvature, and intraocular pressure level on Goldmann applanation tonometry and dynamic contour tonometry.
      and rigidity.
      • Whitacre M.M.
      • Stein R.
      Sources of error with use of Goldmann-type tonometers.
      There can be significant discrepancies between measured and true IOP, as well as between measurements in the same subject by different examiners.
      • Sudesh S.
      • Moseley M.J.
      • Thompson J.R.
      Accuracy of Goldmann tonometry in clinical practice.
      Some commercially available tonometry devices, such as the ocular response analyzer
      • McMonnies C.W.
      Assessing corneal hysteresis using the ocular response analyzer.
      and dynamic contour tonometer, claim to compensate for corneal parameters.
      • Kotecha A.
      • White E.T.
      • Shewry J.M.
      • Garway-Heath D.F.
      The relative effects of corneal thickness and age on Goldmann applanation tonometry and dynamic contour tonometry.
      These are, however, either expensive or difficult to use, limiting their use in everyday clinical practice. Moreover, they do not solve the problem of “undersampling” IOP in the treatment of glaucoma.
      The novel EYEMATE-IO sensor is an implantable device for continual IOP monitoring that is placed within the ciliary sulcus during cataract surgery. An external handheld reader enables on-demand measurements of IOP and wireless power supply of the device. The first-generation device has been shown to be safe and able to obtain IOP measurements in 7 eyes with up to 1 year of follow-up.
      • Koutsonas A.
      • Walter P.
      • Roessler G.
      • Plange N.
      Implantation of a novel telemetric intraocular pressure sensor in patients with glaucoma (ARGOS study): 1-year results.
      The purpose of this open-label single-arm clinical investigation was to evaluate the safety and performance of the second-generation EYEMATE-IO sensor in patients with primary open angle glaucoma (POAG).

      Methods

       Study Design

      The ARGOS-02 trial was designed as a prospective, open-label, single-arm, multicenter observational study, aimed at assessing the safety and performance of the EYEMATE-IO system in patients with POAG. The study adhered to the tenets of the Declaration of Helsinki, received prospective ethics committee/institutional review board approval at each study site, and was registered at clinicaltrials.gov (NCT02434692).
      The primary objectives were to evaluate the safety and tolerability of the EYEMATE-IO sensor by assessing the number and nature of serious adverse device-related events (SADEs) within the first 3 months after implantation. The primary performance endpoint was to evaluate the limits of agreement between measurements with the EYEMATE-IO and GAT up to 3 months after implantation.
      Secondary safety endpoints included the incidence, nature, severity, and seriousness of all observed AES, including ADEs, in the 12 months following implantation. Secondary performance endpoints included the number and nature of device malfunctions, limits of agreement between EYEMATE-IO and GAT measurements, as well as assessment of diurnal IOP profiles obtained via self-guided self-measurements with EYEMATE-IO at home between study visits.
      Safety and performance endpoints were assessed at regular prescheduled visits on days 1, 3, 10, 30, 60, 90, 120, 180, 240, and 360 after surgery. For safety assessment, a complete ophthalmologic exam including best corrected visual acuity, slit lamp biomicroscopy, funduscopy, external eye photography, and tonometry was performed at each visit and any AE or change in medication was recorded. Additionally, perimetry, gonioscopy, specular microscopy, anterior segment optical coherence tomography (OCT), posterior pole OCT, and fundus photography were performed at screening as well as on days 90, 180, and 360 after surgery. Starting on day 30 after surgery, either 2 or 4 series of comparative measurements between GAT and EYEMATE-IO were performed at each visit. GAT was measured before EYEMATE-IO to avoid influencing the subjective GAT readout. Additionally, any device deficiencies were recorded at any visit.

       Description of Device

      The EYEMATE-IO system comprises a ring-shaped foldable sensor device to be implanted into the ciliary sulcus during cataract surgery, an injector device for the facilitation of this implantation procedure, as well as an external handheld reader device (MESOGRAPH) for the activation, power supply, and readout of the implanted sensor.

       Sensor ring

      The EYEMATE-IO pressure sensor, which is intended for permanent implantation in the ciliary sulcus of the eye, bears a microelectromechanical system application-specific integrated circuit (ASIC) that integrates pressure and temperature sensors, identification, and analog-to-digital encoders and a telemetry unit. The ASIC is bonded to a gold microcoil and hermetically encapsulated in a ring of medical-grade silicone rubber material that has been validated through its previous use for intraocular lenses. An initial description of the underlying technology was published by Todani and associates.
      • Todani A.
      • Behlau I.
      • Fava M.A.
      • et al.
      Intraocular pressure measurement by radio wave telemetry.
      The final dimensions of the implant are an inner diameter of 7 mm and an outer diameter of 11.3, 11.7, or 12.1 mm, to be used depending on the size of the ciliary sulcus, a thickness of 0.9 mm at the ASIC, and a thickness of 0.5 mm around the microcoil (see Figure 1). These changes constitute a modification and miniaturization compared to the first-generation device. Further modifications include tapered edges on the outer ring as well as introduction of 4 small pedicles to act as haptics for secure but strainless placement in the ciliary sulcus.
      Figure thumbnail gr1
      Figure 1The left panel shows the first-generation EYEMATE-IO sensor design with a uniform thickness of 0.9 mm and sharp edges, which caused mechanical problems during the first series of implantations.
      • Koutsonas A.
      • Walter P.
      • Roessler G.
      • Plange N.
      Implantation of a novel telemetric intraocular pressure sensor in patients with glaucoma (ARGOS study): 1-year results.
      The second-generation design used in this study is shown in the right panel. It is significantly thinner at 0.5 mm with rounded edges tapered to 0.1 mm and features 4 haptics to prevent unwanted motion in the sulcus.
      When the external reading device (Mesograph) is activated in close proximity to the eye (within approximately 5 cm), an electromagnetic inductive connection is formed between it and the microcoil that provides the ASIC with power and enables the measurement and telemetric data transfer between the sensor and the reader. Each sensor is calibrated individually to be highly accurate between absolute pressures ranging from 800 to 1150 hPa (normal atmospheric pressure) at physiological temperatures (30°C-42°C) before sterilization and packing. Each sensor is checked for plausible readings again immediately before and after implantation.

       Injector device

      The implantation procedure is performed using a multipart injector device that uses a single-use Teflon-based cartridge, holding the partly rolled-up sensor ring. The injector is used to facilitate placement of the sensor ring in the eye through the widened phacoemulsification tunnel to minimize surgical trauma. This is a departure from the original “manual” implantation method used for the previous study. The use of an injector was enabled by the reduction of thickness and rigidity of the sensor ring.

       Mesograph—External reader device

      The external reader device is a hand-held device that provides power to the sensor when activated within 5 cm of the eye. It comprises a power source (2CR5 lithium battery), a coil generating the electromagnetic field to power the sensor via electromagnetic coupling, which also acts as an antenna for the transmission of the signals provided by the sensor. The device also contains a separate pressure sensor measuring external atmospheric pressure (absolute) that is compared to the pressure reading provided by the sensor. The difference between both absolute pressures, expressed in mm Hg, represents the IOP, which is displayed on a LED display of the reader device. The reader can store up to 3000 individual IOP readings, which can be transferred to a computer via a cable connection or wirelessly through a GSM module.

       Patient Population

      Patients with cataract and concomitant POAG, who were regularly scheduled for cataract surgery, were asked to participate in the study based on the following inclusion and exclusion criteria.

       Inclusion criteria

      The inclusion criteria were as follows: mentally competent and willing to provide written informed consent; male or female aged ≥40 years and ≤85 years on the day of screening; diagnosis of POAG as defined by the European Glaucoma Society guidelines
      European Glaucoma Society
      Terminology and Guidelines for Glaucoma (4th edition).
      ; controlled IOP, as determined by treating clinicians; cataract surgery indicated irrespective of study participation; preoperative anterior chamber depth ≥2.0 mm as measured from the corneal endothelium; axis length >22 mm; corneal endothelial cell density ≥2000 cells/mm2.

       Exclusion criteria

      The exclusion criteria were as follows: any type of glaucoma other than POAG; severe visual field loss of –20 dB or worse in both eyes (In cases with different stages of disease, the sensor was to be implanted in the worse eye.); presence of other severe sight-threatening ocular conditions (eg, age-related macular degeneration, retinal detachment, ocular tumors, severe dry eye syndrome); corneal diseases, especially those affecting the corneal endothelium and conditions potentially affecting assessment of visual acuity and/or IOP by GAT; diabetes mellitus; history of connective tissue disease (ie, Marfan syndrome, Ehlers-Danlos syndrome, or Weill-Marchesani syndrome); history or evidence of severe inflammatory eye diseases (ie, uveitis, retinitis, scleritis) in one or both eyes within 6 months prior to EYEMATE-IO implantation; intraocular surgical procedure(s) within 6 months prior to EYEMATE-IO implantation or any surgical procedure with the potential to affect the assessment of IOP by GAT; anterior chamber configuration with high risk of an angle closure glaucoma; subjects with planned ancillary procedures in the study eye at the time of implantation or during the postoperative study period.
      After informed consent was given for both cataract surgery and study participation, patients meeting all inclusion and exclusion criteria were scheduled for surgery at one of the 11 sites participating in the study.

       Description of implantation procedure

      The sensor ring is placed in the ciliary sulcus of the eye during routine cataract surgery. As the size of the implant and the sleeve of the injector tip used for this study required an incision width of approximately 6 mm, most sites chose a sclerocorneal tunnel (temporal approach) to the implantation. After clearing the conjunctiva, the tunnel was pre-formed without penetration of the anterior chamber. Depending on surgeon preference, 1 or 2 paracenteses were then performed using a standard 15° stab incision blade followed by filling of the anterior chamber with an ocular viscoelastic device (OVD) of high viscosity. The phaco tunnel was created using a standard phaco slit knife (2.8-3.2 mm), followed by regular capsulorrhexis, phaco-emulsification, and intracapsular placement of the intraocular lens. The posterior chamber was then widened by use of the same high-viscosity OVD, carefully avoiding too much OVD in the incision quadrant so as to prevent iris prolapse. The pre-formed sclerocorneal tunnel was then widened to fit the sleeve of the injector device, which was either inserted into or docked onto the tunnel opening. The sensor ring was then inserted into the posterior chamber and guided into the sulcus using a flat atraumatic spatula through a paracentesis, carefully avoiding touch of the sensor's ASIC. The OVD was then removed, the pupil constricted medically, the sclerocorneal tunnel sutured, the globe toned to intermediate IOP and the conjunctiva reattached with sutures. Figure 2 shows the injection and unfolding of the sensor ring.
      Figure thumbnail gr2
      Figure 2Implantation of the EYEMATE-IO sensor ring in the ciliary sulcus of the left eye through a temporally placed sclerocorneal tunnel using the injector device. The implantation with an injector is facilitated by the slimmer design of the second-generation sensor compared to the first-generation.

      Results

       Patient characteristics

      A total of 24 patients with POAG and concurrent cataract were initially enrolled into this clinical trial, 22 of whom successfully received the EYEMATE-IO. Of the 2 patients initially enrolled but not implanted, 1 was excluded because of a screening failure. In the second patient, after surgical complications early during cataract surgery, an implantation of the EYEMATE-IO sensor was not attempted. Thus, exclusion of this patient was unrelated to the EYEMATE-IO device. A total of 22 sensors were used for this study, with no device deficiencies occurring during the implantation procedure. In all implanted patients, the sensors remained in the eye to the end of the study.
      At enrollment, patients (8 female and 14 male) were between 55 and 78 years of age (mean 67.8 years, SD 6.8 years) and had been diagnosed with glaucoma for up to 29 years (mean duration 8.74 years, SD 7.94 years). At the time of enrollment, 17 subjects were using at least 1 anti-glaucoma medication. Ten patients had concurrent ocular conditions other than glaucoma and cataract and had undergone other surgical ophthalmic procedures, including laser procedures, prior to enrollment. Two of the patients had had previous glaucoma surgery in the study eye: trabeculectomy and selective laser trabeculoplasty in one patient and canaloplasty in the other patient.

       Safety

       Surgical complications

      Surgical complications were reported in 7 of 23 patients. In 5 of these patients, complications occurred during the implantation of the EYEMATE-IO device. The complications that occurred most often (5 times each) were iris prolapse/floppy iris and pigment dispersion. Both flat anterior chamber and “vis a tergo” (“pressure from behind”) were reported twice. All surgical complications occurred early in the study. After re-evaluation of the surgical procedure and retraining of the surgeons, no further complications occurred during surgery.

       Adverse events

      Four serious ocular AEs were reported for the study after implantation. These included 2 cases of fibrin reaction in the anterior chamber, 1 case of temporary corneal decompensation, and 1 case of intractable IOP increase requiring subsequent glaucoma surgery. This patient underwent trabeculectomy 345 days after initial surgery. Although it cannot be ruled out that pigment dispersion due to iris manipulation during the implantation procedure, which was reported as an adverse event in this case, contributed to increased IOP in this patient, the cause of the increase could not be conclusively determined.
      The 2 cases of severe fibrin reaction (7 and 13 days after initial implantation) lead to precautionary hospitalization because of the novelty of the device and procedure, but resolved quickly and without sequelae under intensive anti-inflammatory treatment. No incidences of hypopyon or uveitis were reported.
      One patient was treated for corneal decompensation 9 days after surgery (treatment period 22 days). Careful review of the implantation process revealed a complicated surgical procedure with potential corneal touch of either the IOL or the sensor implant. In all, 70 nonserious AEs in 18 patients were considered to be at least possibly related to the medical device, and 90 AEs in 21 patients had a potential relationship to the medical procedure at initial assessment. These included increased IOP (22 times in 14 patients), anterior chamber inflammation with increased anterior chamber cells and Tyndall flare (11 times in 9 patients), and mild to moderate pigment dispersion in 8 patients. One case of persistent cystoid macular edema was treated with nonsteroidal anti-inflammatory drugs for 7 months and resolved without sequelae.
      Visual acuity 3 months after surgery was better than at screening owing to removal of the cataract. Best corrected visual acuity increased from 65.0 ± 21.6 ETDRS letters before surgery to 77.5 ± 13.2 letters 3 months after surgery (P = .0045) and remained stable throughout the trial: at follow-up visit 11 (day 360), it was recorded at 79.8 ± 13.5 ETDRS letters (P = .0012).
      The mean endothelial cell density decreased from 2403 ± 198 cells/mm2 at screening to 2177 ± 438 cells/mm2 at the end of follow-up, constituting a loss of 9.4% over the course of 12 months (P = .0502). This higher than expected cell loss mainly stems from one patient who suffered temporary corneal decompensation after corneal touch during surgery, whose endothelial cell density dropped from 2361 to 1599 cells/mm2 before stabilizing. Central corneal thickness remained stable throughout the study. There was mild to moderate corneal swelling and turbidity in the early postoperative phase in half of the patients (11 of 22), with mild to moderate Descemet folds resolving in most cases before follow-up visit 5 (day 30).
      There was a slight decrease in anterior chamber angle from median Shaffer grade 3 at screening to a median Schaffer grade 2 at the end of study with an increase in median pigmentation from mild to moderate over the course of the study. No case of angle closure or anterior synechiae occurred. Mild to moderate iris transillumination defects were seen in 8 patients and mild pigment deposits in the anterior chamber in 6 patients at the end of the study. These were most likely attributable to iris manipulation during the implantation procedure, as they were seen either already during, or immediately after, surgery and stabilized after the initial postoperative healing phase. A typical example is shown in Figure 3, placement of the sensor is shown in Figure 4.
      Figure thumbnail gr3
      Figure 3The left panel shows the correct positioning of the EYEMATE-IO sensor in an eye with medically induced mydriasis 1 week after surgery (patient DE_01_001). In maximal dilation, the inner edge of the sensor is visible. The right panel shows the same patient 3 months after surgery. The sensor ring is only visible through an iridectomy placed during a previous trabeculectomy several years before sensor implantation. There are mild defects to the pigment layer of the iris that were first seen directly after surgery and did not change over the course of the study.
      Figure thumbnail gr4
      Figure 4Typical anterior segment OCT of the sensor placement in the ciliary sulcus of a right eye. Notable in this patient is the relatively large iridectomy and a slight pupil distortion, which were not typical.
      Macular OCT demonstrated a slight increase in mean central retinal thickness (CRT) from 247.1 ± 68.1 μm at screening to 286.4 ±117.4 μm at follow-up visit 7 (day 90; P = .251) and 280.6 ± 112.1 μm at follow-up visit 9 (day 180, P = .288) before returning to 243.6 ± 28.3 μm at the end of the follow-up period (day 360, P = .644). The temporary increase was mainly due to the reported case of cystoid macular edema.
      Overall, the majority of ocular adverse events occurred in the early postoperative phase and could be attributed to the additional manipulation during surgery compared to cataract surgery alone. There were no unexpected AEs, and most AEs resolved without sequelae under appropriate medical treatment.

       Performance

       Measurement concordance

      The EYEMATE-IO IOP measurements showed a good concordance to GAT IOP measurements with an overall Cronbach alpha of 0.882 (95% confidence interval [CI] 0.835, 0.915) and an intraclass correlation coefficient ICC(3,k) of 0.783 (95% CI 0.743, 0.817). Starting 30 days after implantation, either 2 or 4 comparative measurements were performed at each scheduled patient visit. A total of 434 valid comparisons were included in the analysis as shown in Figure 4. On average, EYEMATE-IO measurements were higher than GAT measurements with a mean difference of 3.2 mm Hg (95% CI 2.8, 3.5 mm Hg). The 95% limits of agreement were 10.2 to –3.8 mm Hg (Figure 5).
      Figure thumbnail gr5
      Figure 5Bland-Altman-Plot of all valid comparative measurements (n=434). There is good agreement between the methods in the physiological IOP range with on average higher readings for the EYEMATE-IO device.
      Regression analysis of the comparative measurements indicates that the difference between GAT IOP and EYEMATE IOP was dependent on the respective IOP level with a regression line equation of EYEMATE-IOP = 1.04 × GAT IOP + 2.57, thus showing an increasing difference between the methods at higher IOPs.
      A slight long-term drift in EYEMATE-IO measurements from 22.2 ± 9.2 mm Hg at day 30 to 15.7 ± 3.8 mm Hg at day 360 coincided with a decrease in GAT measurements from 19.5 ± 6.8 mm Hg to 14.1 ± 2.2 mm Hg at the end of the study. The difference between the 2 methods remained relatively stable over the course of the study. Sensor drift for the follow-up period was calculated to be –0.003 ± 0.009 mm Hg/d (corresponding to –1.15 ± 3.22 mm Hg/y) based on patient-individual differences between GAT and EYEMATE–IO measurements over time, indicating that the observed IOP decrease was mostly due to glaucoma treatment changes during the study and only partly influenced by sensor drift. Thus, the EYEMATE-IO performed stably and reliably over the duration of the study.
      Because the EYEMATE-IO can be recalibrated to an external standard (eg, GAT) as needed, sensor drift is not expected to affect the long-term utility of the device.

       User acceptance

      As a primary purpose for the development of the device was to enable self-measurement, the patients were instructed at the beginning of the study to self-measure IOP using the EYEMATE-IO 4 times per day at intervals of their own choosing for the duration of the study. No further instructions, nor any reminders during follow-up were given. This was done to get an approximate measure for participant compliance and user acceptance of the device.
      Device utilization was high, with an average of 7.9 ± 1.4 measurements per patient per day throughout the study (range: 2.2 ± 1.6 to 20.5 ± 4.4 measurements per day) for a total number of 49 063 single IOP measurements between days 30 and 360 (Figure 6). Many patients adopted a regular measurement pattern, typically immediately before and up to 2 hours after instillation of pressure-lowering eye drops, and it was possible to discern relatively stable diurnal IOP patterns during the waking hours of most patients (data not shown).
      Figure thumbnail gr6
      Figure 6Mean number of self-measurements throughout the study. The increase in the number of measurements after day 30 is due to the fact that the majority of the patients were only given their device after follow-up visit 5 (day 30), as mandated by the study protocol. This restriction was introduced to prevent any effect of frequent application of an electromagnetic field during the initial postoperative phase. In the 3 patients who received the device immediately after surgery, no detrimental effect of using the device was seen.
      At the end of the study, patients were asked about their user experience, with the vast majority of patients reporting excellent tolerability of the sensor and ease of use of the hand-held reader device, as well as improved (self-reported) adherence to glaucoma medication and recommendation of the device to other patients. As this survey was done with a nonvalidated, tailor-made questionnaire for internal purposes only, and no other vision-related quality of life questionnaire was employed during this study, these findings are not shown in detail here and need to be verified in a more systematic manner in future studies.

       Device malfunctions

      There were no permanent device malfunctions during the study. In one case, the external reader device had to be exchanged owing to a technical fault. In 2 cases, there was a notable shift in EYEMATE-IO measurements that required the recalibration of the sensor. One of these instances occurred immediately after performing an ultrasound biomicroscopy of the implanted eye. It is currently unknown why this shift occurred, as other sensors had no such issues after ultrasound biomicroscopy examination. The second case of shift occurred after YAG-laser capsulotomy. In both cases, the sensor was successfully recalibrated to within 2 mm Hg of GAT. This was done by simultaneously measuring IOP with GAT and EYEMATE-IO several times and adjusting the reader device output to match GAT measurements. The respective protocol-mandated comparative measurements after the shift were excluded from the concordance analysis. Patient safety was not affected by any of the device malfunctions.

      Discussion

      The current study shows that the EYEMATE-IO sensor is a safe, well-tolerated, and efficacious technology to obtain continuous IOP monitoring in a longitudinal manner in glaucoma patients undergoing cataract surgery. Our results also show that patients seemed to value the provided IOP information as evidenced by their regular daily measurements during a 1-year period, despite the absence of reminders.
      In recent years, a number of devices have been introduced that attempt to address either 24-hour monitoring (ie, a contact lens–based method that measures changes in the corneal curvature thought to represent changes in IOP)
      • Mansouri K.
      • Weinreb R.
      Continuous 24-hour intraocular pressure monitoring for glaucoma—time for a paradigm change.
      or self-measurement (ie, a rebound tonometer that does not require topical anesthetics and is designed to be operated by the patient himself).
      • Dabasia P.L.
      • Lawrenson J.G.
      • Murdoch I.E.
      Evaluation of a new rebound tonometer for self-measurement of intraocular pressure.
      ,
      • Sood V.
      • Ramanathan U.S.
      Self-monitoring of intraocular pressure outside of normal office hours using rebound tonometry: initial clinical experience in patients with normal tension glaucoma.
      Although both devices are significant advances toward a better understanding of IOP dynamics, they still have limitations in terms of usability, data interpretation, and dependence on corneal parameters. The contact lens sensor, for instance, although relatively noninvasive and with quasi-continuous measurements for up to 24 hours, currently cannot be directly calibrated to other tonometry methods and it is, therefore, unclear how its measurements correspond to changes in IOP.
      • Dunbar G.E.
      • Shen B.Y.
      • Aref A.A.
      The Sensimed Triggerfish contact lens sensor: efficacy, safety, and patient perspectives.
      In contrast, the rebound tonometer enables self-measurement but is not capable of automatic measurements over extended periods of time. It is useful, however, for obtaining IOP measurements outside of normal office hours and has been successfully used this way in a number of clinical studies. Its limitations lie in restriction of the measurements to an upright (seated) position and open eyes, as well as the fact that the device is not easy to operate. Only about 74% of all study subjects (with a positive selection bias toward patients interested in using the device) were able to successfully self-measure IOP, while the remaining quarter could not properly handle the device despite extensive training.
      • Dabasia P.L.
      • Lawrenson J.G.
      • Murdoch I.E.
      Evaluation of a new rebound tonometer for self-measurement of intraocular pressure.
      Most important, it does not provide IOP measurements during sleep.
      The EYEMATE-IO is designed to overcome these shortcomings while providing long-term IOP monitoring. Once implanted, patients can obtain IOP measurements on demand, simply by approaching the external reader device to the eye to allow for electromagnetic coupling of the sensor. The measurement itself is not influenced by the position of the reader device, thus eliminating the potential for user error. In our study, all patients successfully performed several self-measurements per day as mandated by the study protocol, and there was a high degree of user acceptance (data not shown). Additionally, it is possible to place an external antenna around the eye (eg, in a sleep mask) and set the reader to automatic mode and measure IOP at predefined intervals over extended periods of time, including nighttime measurements while the patient is asleep.
      The EYEMATE-IO is designed to remain in the eye indefinitely and therefore opens the possibility of repeated assessment of diurnal IOP fluctuations, the influence of different glaucoma medications or physiological parameters like blood pressure, physical activity, or body position over several years. Most important, because the sensor is placed inside the eye and measures absolute pressure, its measurements are theoretically unaffected by external factors that limit all other currently available tonometry methods. Some of these, like corneal thickness and astigmatism can partially be corrected for, but others, like corneal rigidity are difficult to measure in vivo and there is limited knowledge on the physiological range of rigidities and how they affect IOP measurements quantitatively. Liu and Roberts showed in a mathematical model that corneal rigidity may have a much greater impact than central corneal thickness on applanation tonometry.
      • Liu J.
      • Roberts C.J.
      Influence of corneal biomechanical properties on intraocular pressure measurement: quantitative analysis.
      Some tonometry methods are designed to compensate for corneal biomechanics. Some modern noncontact tonometers (ie, the ocular response analyzer [ORA] and more recently the Corvis-ST) measure corneal hysteresis as surrogate for corneal rigidity and should theoretically be more accurate than GAT.
      • McMonnies C.W.
      Assessing corneal hysteresis using the ocular response analyzer.
      Similarly, dynamic contour tonometry (DCT) uses a curved probe and a smaller applanation area to offset some of the corneal biomechanics.
      • Kotecha A.
      • White E.T.
      • Shewry J.M.
      • Garway-Heath D.F.
      The relative effects of corneal thickness and age on Goldmann applanation tonometry and dynamic contour tonometry.
      Notably, both ORA and DCT, when directly compared to GAT, systematically show readings approximately 2 mm Hg higher than GAT IOP levels.
      • Cook J.A.
      • Botello A.P.
      • Elders A.
      • et al.
      Systematic review of the agreement of tonometers with Goldmann applanation tonometry.
      • Ceruti P.
      • Morbio R.
      • Marraffa M.
      • Marchini G.
      Comparison of Goldmann applanation tonometry and dynamic contour tonometry in healthy and glaucomatous eyes.
      • Hager A.
      • Annette H.
      • Loge K.
      • et al.
      Effect of central corneal thickness and corneal hysteresis on tonometry as measured by dynamic contour tonometry, ocular response analyzer, and Goldmann tonometry in glaucomatous eyes.
      This is in line with findings from our study, which also show systematically higher readings of the EYEMATE-IO, which are expected to be largely unaffected by corneal parameters. Moreover, there is an increase in difference between EYEMATE-IO and GAT measurements with higher IOPs. This is also an expected result, as it has been shown, that GAT tends to underestimate IOP, especially at higher IOP levels.
      • Whitacre M.M.
      • Stein R.
      Sources of error with use of Goldmann-type tonometers.
      ,
      • McCafferty S.
      • Levine J.
      • Schwiegerling J.
      • Enikov E.T.
      Goldmann applanation tonometry error relative to true intracameral intraocular pressure in vitro and in vivo.
      Overall, the IOP measurements obtained during this study appear to be very reliable, with good agreement between GAT and EYEMATE-IO. This represents a significant improvement over the first-generation sensor,
      • Koutsonas A.
      • Walter P.
      • Roessler G.
      • Plange N.
      Implantation of a novel telemetric intraocular pressure sensor in patients with glaucoma (ARGOS study): 1-year results.
      in which there were large and sometimes implausible differences between the methods, additionally aggravated by sudden unexpected shifts in EYEMATE-IO readings. Two sudden shifts also occurred in the current study. These were attributed to distinct manipulations of the eye (ie, YAG laser capsulotomy for posterior capsule opacification and ultrasound biomicroscopy), and the sensors were successfully recalibrated to within 2 mm Hg of GAT. The subsequent comparative measurements for these 2 sensors were excluded from analysis.
      The second-generation design also led to improved safety of the sensor and implantation procedure. Although there were 2 cases of sterile hypopyon among the 6 patients in the first study, none were observed in the current study. There was still an increased and prolonged inflammatory response compared with cataract surgery alone. This was likely due to the additional manipulation required to implant the sensor rather than an adverse reaction to the implant itself, as there were no protracted or recurring inflammations observed after the initial postoperative phase had subsided. The injector used in this study still required an incision size of 5.5 to 6 mm, resulting in a longer sclero-corneal tunnel to minimize the risk of induced astigmatism and iris prolapse. This surgical approach is akin to extracapsular cataract extraction and known to increase the amount of inflammation in the eye.
      • Dick H.B.
      • Schwenn O.
      • Krummenauer F.
      • Krist R.
      • Pfeiffer N.
      Inflammation after sclerocorneal versus clear corneal tunnel phacoemulsification.
      A future reduction in incision size is expected to decrease the inflammatory response. However, long-term observations of the study patients are needed to make sure, that no late or long-term inflammation occurs in the sensor eyes, that might be related to continued chemical or mechanical exposure to the implant.
      There was also a marked improvement in the amount of pigment dispersion and transillumination defects induced by the sensor compared with the first-generation design, with only mild to moderate pigment deposits found early after surgery in 6 patients, which remained stable throughout the study.
      One potential complication of any sulcus-fixed implant is uveitis-glaucoma-hyphema syndrome, which is caused by iris-chafing leading to pigment release and erosion of blood vessels.
      • Almond M.C.
      • Wu M.C.
      • Chen P.P.
      Pigment dispersion and chronic intraocular pressure elevation after sulcus placement of 3-piece acrylic intraocular lens.
      • Chang D.F.
      • Masket S.
      • Miller K.M.
      • et al.
      Complications of sulcus placement of single-piece acrylic intraocular lenses: recommendations for backup IOL implantation following posterior capsule rupture.
      • Van Liefferinge T.
      • Van Oye R.
      • Kestelyn P.
      Uveitis-glaucoma-hyphema syndrome: a late complication of posterior chamber lenses.
      No uveitis-glaucoma-hyphema syndrome occurred during the study period or the extended monitoring period since then. Overall, the type, number, and severity of AEs were within the expected range for the novel device and implantation procedure, and all AEs resolved quickly under appropriate treatment. At the end of follow-up, all sensors remained in place and in working condition. Based on these results, the EYEMATE-IO system has received regulatory approval in the European Union (CE mark) for use in patients with open-angle glaucoma.
      Given the potential for late occurrence of uveitis-glaucoma-hyphema or pigment dispersion syndrome due to the still relatively bulky ASIC of the EYEMATE-IO, long-term observation is needed to assess potential late complications. These observations are currently ongoing.
      It is worth noting that because of the large number of study sites involved, half of the patients (11 of 22) underwent the first implantation at their respective site, and only 2 sites performed more than 2 implantations. In these 2 sites, it could be seen that there is a significant learning curve for the implantation procedure. It can, therefore, be argued that the number of implantation-related AEs could potentially have been lower if the surgical learning curve had been overcome. This is one of the limitations of this study. Other limitations include the relatively small and heterogeneous patient population with different disease stages, number and type of medications, as well as previous (glaucoma) surgery. The follow-up period is still relatively short, and further observations are needed to establish long-term safety, performance, and reliability of the sensor, in particular with regard to long-term pigment release. Future improvements to the sensor design, for example, smaller dimensions, a smaller injector sleeve, and potentially the placement of the sensor in the capsular bag, are also desirable. Furthermore, future studies need to establish whether the vastly increased number of IOP measurements will lead to better IOP control and glaucoma therapy.
      In summary, the second-generation EYEMATE-IO sensor was successfully and safely implanted in 22 patients with POAG. It performed reliably and without detrimental device malfunctions in all patients for the entire duration of the study. Future investigations of the long-term stability, safety, and performance are warranted. We have shown that the sensor provides continual and accurate measurement of IOP with the potential to enable individualized glaucoma treatment and preservation of vison-related quality of life.
      All authors have completed and submitted the ICMJE form for disclosure of potential conflicts of interest and the following were reported. None of the authors have any direct financial stake in Implandata Ophthalmic Products GmbH, Germany or any of the materials presented here. J.v.d.B. is currently employed at Implandata Ophthalmic Products GmbH as PhD student within the aforementioned EU-funded postgraduate training program. L.C. and H.T. have received travel compensation from Implandata Ophthalmic Products GmbH for ophthalmologic conferences during which preliminary data from the study were presented. No speakers fees were received by any of the authors. K.M. is currently working as consultant for Implandata Ophthalmic Products GmbH. L.C. has received nonfinancial support for other clinical studies from Allergan GmbH , Germany, Novartis Pharma GmbH , Germany, and Santen GmbH, Germany. K.M. has consulted for Santen, Sensimed, Switzerland, and Implandata and received non-financial support from Topcon , Alcon , Allergan , and OptoVue . HBD has consulted for Avedro Inc, Carl Zeiss Meditec Inc, Johnson & Johnson Vision, Medicem Group BV, Optical Express AG, Polytech-Domilens GmbH, RxSIGHT Inc; received grants from AbbVie Deutschland GmbH & Co KG , Acufocus Inc , Allergan GmbH , Avedro Inc , Bayer Healthcare AG , Germany, Carl Zeiss Meditec AG, Germany , Hoffmann-La Roche Ltd , Johnson and Johnson Vision , Medicem Group BV , Novartis Pharma GmbH , Ophtec GmbH, Germany , OSD Medical GmbH , Germany, PowerVision Inc , Rowiak GmbH, Germany , URSAPHARM GmbH , Germany, and received speakers fees from Johnson & Johnson Vision, Bausch + Lomb Inc, Oculus Inc, Pharm-Allergan GmbH, Théa Pharma GmbH. H.T. has consulted for Allergan GmbH, Novartis Pharma GmbH, and Santen GmbH. M.Wa. and M.We. have nothing to disclose. This study was in part financially supported by the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement no. 675033 . The study has also received support by Implandata Ophthalmic Products GmbH (Implandata), Hannover, Germany, who produced and provided the medical devices under investigation. All authors attest that they meet the current ICMJE requirements to qualify as authors.
      The authors wish to thank Daniela Oehring, MSc, for help with the statistical analysis, as well as the ARGOS study group for providing all clinical data during the analysis period.
      Principal investigators of the ARGOS Study and additional staff at the sites involved in the ARGOS-02 study: Coordinating Investigator: University Eye Clinic Magdeburg, Magdeburg, Germany. Prof. Dr. med. Hagen Thieme. Additional sites: nordBlick Augenklinik-Bellevue GmbH, Kiel, Germany. PD Dr. med. Florian Rüfer. Knappschaftsklinikum Saar GmbH, Krankenhaus Sulzbach, Sulzbach, Germany. Prof. Dr. med. Peter Szurmann. Augenärztliches Augenchirurgische Zentrum (AAZ), Nürnberg, Germany. Dr. med. Wolfram Wehner. Universitätsaugenklinik Tübingen, Tübingen, Germany. Prof. Dr. med. Martin Spitzer. Klinikum am Gesundbrunnen, SLK-Kliniken Heilbronn GmbH, Heilbronn, Germany. Prof. Dr. med. Lutz Hesse. Universitätsmedizin Rostock, Rostock, Germany. Prof. Dr. med. Anselm Jünemann. Universitätsaugenklinik Aachen, Aachen, Germany. Prof. Dr. med. Niklas Plange. Augen-Zentrum-Nordwest, Ahaus, Germany. Dr. med. Stefanie Schmickler. Universitätsaugenklinik Bochum, Bochum, Germany. Prof. Dr. med. Burkhard Dick. Internationale Innovative Ophthalmochirurgie, Breyer–Kaymak–Klabe, Düsseldorf, Germany. Dr. med. Kaymak Hakan.

      References

        • Weinreb R.N.
        • Aung T.
        • Medeiros F.A.
        The pathophysiology and treatment of glaucoma: a review.
        JAMA. 2014; 311: 1901-1911
        • Aptel F.
        • Weinreb R.N.
        • Chiquet C.
        • Mansouri K.
        24-h monitoring devices and nyctohemeral rhythms of intraocular pressure.
        Prog Retin Eye Res. 2016; 55: 108-148
        • Hatanaka M.
        • Babic M.
        • Susanna R.
        Reproducibility of the mean, fluctuation, and IOP peak in the diurnal tension curve.
        J Glaucoma. 2013; 22: 390-392
        • Mansouri K.
        • Weinreb R.N.
        • Liu J.H.K.
        Efficacy of a contact lens sensor for monitoring 24-h intraocular pressure related patterns.
        PLoS One. 2015; 10: e0125530
        • Mansouri K.
        • Weinreb R.N.
        • Medeiros F.A.
        Is 24-hour intraocular pressure monitoring necessary in glaucoma?.
        Semin Ophthalmol. 2013; 28: 157-164
        • Arora T.
        • Bali S.J.
        • Arora V.
        • Wadhwani M.
        • Panda A.
        • Dada T.
        Diurnal versus office-hour intraocular pressure fluctuation in primary adult onset glaucoma.
        J Optom. 2015; 8: 239-243
        • Nuyen B.
        • Mansouri K.
        Detecting IOP fluctuations in glaucoma patients.
        Open Ophthalmol J. 2016; 10: 44-55
        • Ehlers N.
        • Bramsen T.
        • Sperling S.
        Applanation tonometry and central corneal thickness.
        Acta Ophthalmol (Copenh). 1975; 53: 34-43
        • Holladay J.T.
        • Allison M.E.
        • Prager T.C.
        Goldmann applanation tonometry in patients with regular corneal astigmatism.
        Am J Ophthalmol. 1983; 96: 90-93
        • Francis B.A.
        • Hsieh A.
        • Lai M.-Y.
        • et al.
        Effects of corneal thickness, corneal curvature, and intraocular pressure level on Goldmann applanation tonometry and dynamic contour tonometry.
        Ophthalmology. 2007; 114: 20-26
        • Whitacre M.M.
        • Stein R.
        Sources of error with use of Goldmann-type tonometers.
        Surv Ophthalmol. 1993; 38: 1-30
        • Sudesh S.
        • Moseley M.J.
        • Thompson J.R.
        Accuracy of Goldmann tonometry in clinical practice.
        Acta Ophthalmol (Copenh). 1993; 71: 185-188
        • McMonnies C.W.
        Assessing corneal hysteresis using the ocular response analyzer.
        Optom Vis Sci Off Publ Am Acad Optom. 2012; 89: E343-E349
        • Kotecha A.
        • White E.T.
        • Shewry J.M.
        • Garway-Heath D.F.
        The relative effects of corneal thickness and age on Goldmann applanation tonometry and dynamic contour tonometry.
        Br J Ophthalmol. 2005; 89: 1572-1575
        • Koutsonas A.
        • Walter P.
        • Roessler G.
        • Plange N.
        Implantation of a novel telemetric intraocular pressure sensor in patients with glaucoma (ARGOS study): 1-year results.
        Invest Ophthalmol Vis Sci. 2015; 56: 1063-1069
        • Todani A.
        • Behlau I.
        • Fava M.A.
        • et al.
        Intraocular pressure measurement by radio wave telemetry.
        Invest Ophthalmol Vis Sci. 2011; 52: 9573-9580
        • European Glaucoma Society
        Terminology and Guidelines for Glaucoma (4th edition).
        (Published 2014. Accessed December 21, 2016)
        • Mansouri K.
        • Weinreb R.
        Continuous 24-hour intraocular pressure monitoring for glaucoma—time for a paradigm change.
        Swiss Med Wkly. 2012; 142: w13545
        • Dabasia P.L.
        • Lawrenson J.G.
        • Murdoch I.E.
        Evaluation of a new rebound tonometer for self-measurement of intraocular pressure.
        Br J Ophthalmol. 2016; 100: 1139-1143
        • Sood V.
        • Ramanathan U.S.
        Self-monitoring of intraocular pressure outside of normal office hours using rebound tonometry: initial clinical experience in patients with normal tension glaucoma.
        J Glaucoma. 2016; 25: 807-811
        • Dunbar G.E.
        • Shen B.Y.
        • Aref A.A.
        The Sensimed Triggerfish contact lens sensor: efficacy, safety, and patient perspectives.
        Clin Ophthalmol Auckl NZ. 2017; 11: 875-882
        • Liu J.
        • Roberts C.J.
        Influence of corneal biomechanical properties on intraocular pressure measurement: quantitative analysis.
        J Cataract Refract Surg. 2005; 31: 146-155
        • Cook J.A.
        • Botello A.P.
        • Elders A.
        • et al.
        Systematic review of the agreement of tonometers with Goldmann applanation tonometry.
        Ophthalmology. 2012; 119: 1552-1557
        • Ceruti P.
        • Morbio R.
        • Marraffa M.
        • Marchini G.
        Comparison of Goldmann applanation tonometry and dynamic contour tonometry in healthy and glaucomatous eyes.
        Eye. 2008; 23: 262-269
        • Hager A.
        • Annette H.
        • Loge K.
        • et al.
        Effect of central corneal thickness and corneal hysteresis on tonometry as measured by dynamic contour tonometry, ocular response analyzer, and Goldmann tonometry in glaucomatous eyes.
        J Glaucoma. 2008; 17: 361-365
        • McCafferty S.
        • Levine J.
        • Schwiegerling J.
        • Enikov E.T.
        Goldmann applanation tonometry error relative to true intracameral intraocular pressure in vitro and in vivo.
        BMC Ophthalmol. 2017; 17: 215
        • Dick H.B.
        • Schwenn O.
        • Krummenauer F.
        • Krist R.
        • Pfeiffer N.
        Inflammation after sclerocorneal versus clear corneal tunnel phacoemulsification.
        Ophthalmology. 2000; 107: 241-247
        • Almond M.C.
        • Wu M.C.
        • Chen P.P.
        Pigment dispersion and chronic intraocular pressure elevation after sulcus placement of 3-piece acrylic intraocular lens.
        J Cataract Refract Surg. 2009; 35: 2164-2166
        • Chang D.F.
        • Masket S.
        • Miller K.M.
        • et al.
        Complications of sulcus placement of single-piece acrylic intraocular lenses: recommendations for backup IOL implantation following posterior capsule rupture.
        J Cataract Refract Surg. 2009; 35: 1445-1458
        • Van Liefferinge T.
        • Van Oye R.
        • Kestelyn P.
        Uveitis-glaucoma-hyphema syndrome: a late complication of posterior chamber lenses.
        Bull Soc Belge Ophtalmol. 1994; 252 (discussion 66): 61-65