Evidence review

Advice

Clinical and technical evidence
Costs and resource consequences
Strengths and limitations of the evidence

Evidence review

Clinical and technical evidence

No studies reporting direct agreement of SENSIMED Triggerfish with a reference standard such as Goldmann Applanation Tonometry (GAT) were identified. No studies with clinical effectiveness outcomes, such as rates of progression to COAG or vision loss, were found.

In preparing this briefing 13 studies were identified which contained relevant outcomes, including studies on the safety and tolerability of the SENSIMED Triggerfish device in healthy people. Of these 13: 9 were open, observational studies (Agnifili et al. 2014; De Smedt et al. 2012; Faschinger and Mossbock 2013; Freiberg et al. 2012; Hervas et al. 2013; Lorenz et al. 2013; Mansouri and Shaarawy 2011; Mansouri et al. 2012; Mottet et al. 2013 [both Faschinger and Mossbock 2013 and Herves et al. 2013 were letters to journal editors]; 3 were open interventional studies in which recording took place before and after IOP-lowering drugs or surgery with no reference diagnostic (Hollo et al. 2014; Pajic et al. 2011; Tojo et al. 2013); and 1 was an observational study, the results of which were presented only in a clinical trial register (NCT01390779). The details and results of these studies are provided in tables 1 to 13.

Safety

Table 14 summarises the safety outcomes as reported by 9 of the 13 studies. Commonly reported adverse events were diffuse conjunctival keratitis, superficial punctate keratitis, blurred vision, sensor pressure mark, corneal abrasion, ocular hyperaemia, discomfort/irritation, conjunctivitis and mild corneal abrasion or erosion.

No serious adverse events were reported. Among the 40 people in the Lorenz et al. (2013) study, 5 severe adverse events were reported: 3 corneal epithelial defects, 1 incidence of severe pain, and 1 conjunctival erythema. The study did not report the total number of adverse events. Mansouri et al. (2012) described 2 people who had 5 adverse events of conjunctival hyperaemia (5 of 149 [3%] events), which were classified as severe. All adverse events in this study resolved in the 24 hours after the device was removed.

Patient comfort and tolerability

According to the 5 of the 13 studies which reported these outcomes, the SENSIMED Triggerfish was generally well tolerated.

Technical results

One study reported fair to good reproducibility in pairs of SENSIMED Triggerfish readings 1 week apart (Mansouri et al. 2012). However, a letter to the editor did not report this reproducibility (Faschinger and Mossbock 2013). There is uncertainty about the SENSIMED Triggerfish's ability to accurately and reproducibly measure IOP-related patterns over 24 hours, and this is compounded by there being no studies which directly measure its performance against a reference technique such as GAT. There are, however, validity outcomes in the included studies which give an indication of the device's recording capabilities (reported in table 14).

Table 1 Summary of the observational study: Agnifili et al. (2014)

Study component	Description
Objectives/hypotheses	To examine the circadian IOP-related patterns in healthy people and in people with POAG and NTG using the SENSIMED Triggerfish.
Study design	Open, uncontrolled, observational study.
Setting	Patients were consecutively enrolled and measured with the SENSIMED Triggerfish. Patients were enrolled from an ophthalmology clinic in Italy (no dates provided).
Inclusion/exclusion criteria	Healthy eyes showed a BCVA ≥20/40, refractive error <4 dioptres and mean IOP <18 mm Hg, CCT ranging from 530 micrometres to 560 micrometres, the absence of signs of glaucomatous optic neuropathy and normal visual-field test. Inclusion criteria for POAG and NTG eyes were a BCVA ≥20/40, refractive error ≤4 dioptres (spherical equivalent) and CCT ranging from 530 micrometres to 560 micrometres. The initial diagnosis of POAG needed an indication in the patient chart of an untreated IOP of 24‑37 mm Hg, the absence of secondary glaucoma and the presence of glaucomatous optic neuropathy. The diagnosis of NTG needed untreated mean diurnal IOP ≤21 mm Hg from diagnosis to enrolment as indicated by the patient chart, IOP asymmetry <4 mm Hg and open iridocorneal angle.
Primary outcomes	24-hour SENSIMED Triggerfish pattern (secondary outcomes were sub-period patterns, peaks and prolonged peaks).
Statistical methods	Growth modelling to address within-person and between-person variability simultaneously.
Participants	10 healthy patients, 10 patients with POAG and 10 with NTG.
Results	See table 14.
Conclusions	Adverse events were reported during lens wear. No changes in visual acuity were found after sensor removal and no patients requested removal of the sensor. Mean SENSIMED Triggerfish patterns were significantly different between healthy, POAG, and NTG groups.
Abbreviations: BCVA, best corrected visual acuity; CCT, central corneal thickness; IOP, intraocular pressure; NTG, normal tension glaucoma; POAG, primary open angle glaucoma.

Table 2 Summary of the observational study: De Smedt et al. (2012)

Study component	Description
Objectives/hypotheses	To evaluate the tolerability, comfort, and reliability of the signal transmission of the SENSIMED Triggerfish used for 24-hour IOP-related fluctuation recording in humans.
Study design	Open, uncontrolled, observational study (healthy volunteers).
Setting	Healthy volunteers were recruited in to the study and fitted with SENSIMED Triggerfish (8.7 mm curvature radius prototype). Patients were seen by the investigators at 5 minutes, 30 minutes, 4 hours, 12 hours and 24 hours after fitting of the SENSIMED Triggerfish. Full eye examinations were done before and after SENSIMED Triggerfish sensor placement. Dates and setting were not reported.
Inclusion/exclusion criteria	Inclusion: aged>21 years, not pregnant, not suffering from glaucoma/ocular surface disorder, no ocular surgery in past 3 months.
Primary outcomes	No primary outcome reported. Other outcomes included BCVA, subjective wearing comfort (scored between 0 and 10; 0=intolerable; 10=perfect), position and surface wetting ability of the sensor and its mobility on blinking and adverse events.
Statistical methods	Wilcoxon signed rank test for analysis of the differences in parameters over time.
Participants	10 healthy volunteers.
Results	See table 14.
Conclusions	Subject-reported comfort score was acceptable. Visual acuity reduced during and immediately after SENSIMED Triggerfish wear and mild and moderate adverse events were reported. Three patients needed antibiotic treatment for corneal abrasion.
Abbreviations: BCVA, best corrected visual acuity; IOP, intraocular pressure.

Table 3 Summary of the observational study: Faschinger and Mossbock (2013) (letter to editor)

Study component	Description
Objectives/hypotheses	Not specified
Study design	Open, semi-comparative, observational study (healthy volunteers). Letter to editor.
Setting	Healthy people were monitored with the SENSIMED Triggerfish in 1 eye and simultaneously measured with applanation tonometry 5 times in the other eye (total duration approximately 2 hours). Head and body positions were changed during this period to artificially induce IOP changes. The procedure was repeated 2–8 weeks later in all 5 patients.
Inclusion/exclusion criteria	Young people with healthy eyes (no further information provided).
Primary outcomes	Not reported.
Statistical methods	None reported.
Participants	5 people with healthy eyes.
Results	See table 14.
Conclusions	Applanation tonometry and the SENSIMED Triggerfish did not produce similar curves during change in body and head positions. Applanation curves appeared reproducible but those from the SENSIMED Triggerfish did not.
Abbreviations: IOP, intraocular pressure.

Table 4 Summary of the observational study: Freiberg et al. (2012)

Study component	Description
Objectives/hypotheses	Investigate the effect of overnight wear of the SENSIMED Triggerfish on CCT.
Study design	Open, uncontrolled, observational study
Setting	Patients were monitored with the SENSIMED Triggerfish in hospital during their sleep period (no dates reported). The device was randomly installed in 10 right and 10 left eyes. Comprehensive eye assessment including GAT measurement was done before the device was inserted and after it was removed. One eye was monitored and the opposite eye (without sensor) was used as control.
Inclusion/exclusion criteria	Inclusion criteria: patients over 18 years of age diagnosed with either ocular hypertension or glaucoma. Adults scheduled to be hospitalised as part of their standard glaucoma care. Anti-glaucomatous drug treatment, if any, had to be stable since at least 4 weeks prior to study entry. Exclusion criteria: contact lens wear within the last 2 years, ocular surface status preventing contact lens fitting or ocular surgery within the last 3 months.
Primary outcomes	Effect of overnight wear of the SENSIMED Triggerfish on CCT. Other outcomes included effect of on peripheral corneal thickness, CCR and safety.
Statistical methods	Corneal thickness was assessed by computing Spearman correlation coefficients. Paired t-tests or Wilcoxon signed rank tests were used as appropriate.
Participants	20 patients with glaucoma (15 included in primary data, all 20 evaluated for safety)
Results	See table 14.
Conclusions	Adverse events were reported including severe ocular hyperaemia and bacterial conjunctivitis. No changes in visual acuity were reported.
Abbreviations: CCT, central corneal thickness; CCR, central corneal radius; GAT, Goldmann Applanation Tonometry.

Table 5 Summary of the observational study: Hervas Ontiveros et al. (2013) (letter to editor)

Study component	Description
Objectives/hypotheses	No objectives reported
Study design	Open, uncontrolled, observational study. Letter to editor.
Setting	Patients had 24-hour recording with the SENSIMED Triggerfish in hospital in Spain (6 patients completed the recording). GAT measurements were taken at the start, middle and end of this 24-hour period. One patient had GAT measurements taken at 3-hour intervals.
Inclusion/exclusion criteria	Aged 18–85 years with similar open angle glaucoma diagnosed in both eyes and topical antihypertensive treatment stable for at least 4 weeks. Included patients had not had any type of eye surgery in the past 6 months.
Primary outcomes	No primary outcomes defined. Other outcomes were tolerability and serious adverse events.
Statistical methods	None reported.
Participants	8 patients with open angle glaucoma.
Results	See table 14.
Conclusions	Device tolerability issues were reported and no severe adverse events were reported.
Abbreviations: GAT, Goldmann Applanation Tonometry.

Table 6 Summary of the interventional study: Hollo et al. (2014)

Study component	Description
Objectives/hypotheses	To evaluate 24-hour continuous recording of IOP-related patterns with the SENSIMED Triggerfish to detect prostaglandin-induced IOP reduction.
Study design	Interventional study.
Setting	Between October 2011 and February 2012, patients with ocular hypertension and POAG were recruited in Hungary. All glaucoma medications were stopped for 6 weeks prior to study IOP‑related measurements. One eye from each patient had 24‑hour measurements 3 times, 4 days apart (twice with the SENSIMED Triggerfish and once with GAT). GAT measurements were taken at 6 time points with the patient in a sitting position during hospitalisation. SENSIMED Triggerfish curves were obtained in an outpatient setting. Patients then had IOP-lowering therapy (travoprost) for 3 months, after which 24‑hour monitoring was done with GAT and the SENSIMED Triggerfish (separated by 4 days). Eye examination was done before and after sensor placement.
Inclusion/exclusion criteria	Inclusion criteria: all patients were aged 18 years and over. Ocular hypertension was defined as untreated IOP consistently >21 mm Hg, open anterior chamber angle, reproducible normal visual field, and no glaucomatous optic nerve head and retinal nerve fibre layer damage. POAG was defined as typical glaucomatous optic nerve head and retinal nerve fibre layer damage, reproducible glaucomatous visual field deterioration with automated threshold perimetry, untreated IOP>21 mm Hg, open anterior chamber angle, and no sign of any secondary glaucoma (further inclusion criteria are provided in the full text). Exclusion criteria: any uncontrolled systemic disease, known intolerance to contact lens material and topical prostaglandin analogue medication, IOP>40 mm Hg during the washout period, and participation in any clinical drug trial 30 days before the enrolment.
Primary outcomes	No primary outcome described. Other outcomes included safety, effect of IOP-lowering medication on the SENSIMED Triggerfish and GAT curves, comparison of GAT values before and after sensor fitting with first and last 50 minutes of SENSIMED Triggerfish measurements, and effect of change in sitting position.
Statistical methods	Pearson correlation for relationship between GAT IOP values measured immediately before and after SENSIMED Triggerfish curves, and the difference between the first and last 50-minute periods of the SENSIMED Triggerfish curves. Friedman test to investigate treatment-related change in the mean and SD values of the complete SENSIMED Triggerfish curves. Paired t-test was used to compare mean GAT measurements before and after IOP-lowering drugs.
Participants	9 patients (4 with ocular hypertension; 5 with POAG).
Results	See table 14.
Conclusions	IOP-lowering medication changed the shape of GAT curves but not SENSIMED Triggerfish curves. No correlation was seen in the difference between the first and last SENSIMED Triggerfish measurements compared to the first and last GAT measurements. The authors suggest that these results show limited agreement between methods.
Abbreviations: GAT, Goldmann Applanation Tonometry; IOP, intraocular pressure; POAG, primary open angle glaucoma; SD, standard deviation.

Table 7 Summary of the observational study: Lorenz et al. (2013)

Study component	Description
Objectives/hypotheses	To investigate tolerability and safety of the SENSIMED Triggerfish for 24-hour intraocular pressure recording in healthy people and age-matched glaucoma patients.
Study design	Open, uncontrolled, observational study.
Setting	Healthy people and patients with glaucoma had 24-hour recording with the SENSIMED Triggerfish (8.7 mm curvature radius only) as part of this study in Germany (no dates provided). Comprehensive eye assessment including GAT measurement was done before the device was fitted and after it was removed. Discomfort levels were scored. Patients were asked to apply their topical anti-glaucoma medication as usual.
Inclusion/exclusion criteria	Inclusion criteria: male or female, 18–80 years old; healthy person or treated glaucoma patient; cylinder refraction of ≤±2 dioptres in the study eye; visual acuity of 20/80 or better in the study eye. Exclusion criteria: people wearing contact lenses in the last 2 years; people with contraindications for wearing contact lenses; history of refractive surgery; history of intraocular surgery in the last 3 months; severe dry eye syndrome; keratoconus or other corneal abnormalities; conjunctival or intraocular inflammation; pregnancy and lactation; simultaneous participation in other clinical trials.
Primary outcomes	Primary end point was the level of patient-reported discomfort in the study eye at 24 hours using a VAS (0=no discomfort; 100=very severe discomfort). Secondary end points included the following parameters which were measured as change from baseline for the study eye: BCVA, pachymetry, corneal epithelial staining, conjunctival erythema and corneal topography.
Statistical methods	VAS scores for healthy people and glaucoma patients were compared using independent groups' Wilcoxon rank-sum test.
Participants	20 healthy people and 20 age-matched patients with glaucoma.
Results	See table 14.
Conclusions	Patients found the SENSIMED Triggerfish device tolerable and reasonably comfortable. Mild and severe adverse events were reported, including corneal epithelial defects, pain, and conjunctival erythema, all of which resolved within 2 days. Ocular parameters such as conjunctival oedema and erythema were significantly changed following sensor wear.
Abbreviations: BCVA, best corrected visual acuity; GAT, Goldmann Applanation Tonometry; IOP, intraocular pressure; VAS, visual analogue scale.

Table 8 Summary of the observational study: Mansouri and Shaarawy (2011)

Study component	Description
Objectives/hypotheses	None reported.
Study design	Open, uncontrolled, observational study.
Setting	Consecutive patients with progressive open angle glaucoma had 24-hour ambulatory IOP-related change recording with the SENSIMED Triggerfish in Switzerland (no dates reported). Patients were asked to grade their average level of ocular comfort.
Inclusion/exclusion criteria	Patients with progressive open angle glaucoma despite medical treatment and controlled IOPs during office hours.
Primary outcomes	No primary outcome defined; other outcomes were safety and comfort.
Statistical methods	None reported.
Participants	15 patients with glaucoma (12 with POAG and 3 with pseudoexfoliative glaucoma).
Results	See table 14.
Conclusions	Results suggested reasonable tolerability and comfort. Minor complications were reported, but no serious adverse events.
Abbreviations: IOP, intraocular pressure.

Table 9 Summary of the observational study: Mansouri et al. (2012)

Study component	Description
Objectives/hypotheses	To examine the safety, tolerability, and reproducibility of the SENSIMED Triggerfish device.
Study design	Open, uncontrolled, observational study.
Setting	Patients with suspected glaucoma and established glaucoma received 2, 24-hour recording sessions with the SENSIMED Triggerfish with a 1-week interval. Behaviour and sleep were not controlled. Patients had an eye examination before and after 24-hour recording.
Inclusion/exclusion criteria	Suspected glaucoma was defined as those people with eyes with abnormal-appearing optic discs without repeatable abnormal standard automated perimetry results. Patients suspected of having glaucoma also included those with eyes with an IOP>22 mm Hg but with healthy-appearing optic discs and without repeatable abnormal standard automated perimetry results. Established glaucoma was defined as people having at least 2 consecutive, reliable and repeatable standard automated perimetry examinations with either a pattern standard deviation outside the 95% confidence limits or a glaucoma hemifield test result outside the 99% confidence limits. Inclusion criteria: aged 18–80 years, best-corrected visual acuity of 20/80 or greater in the study eye, spherical refraction between −5 and 3 dioptres, cylinder correction of 2 dioptres or less, and open angles on gonioscopy (further inclusion criteria reported in full text). Exclusion criteria: previous glaucoma surgery or any intraocular surgery 3 months before study inclusion, known intolerance to silicone, contraindications for contact lens wear, severe dry eye disease, keratoconus, or other corneal abnormality.
Primary outcomes	Primary safety outcomes were adverse events (defined as any change from baseline in relevant ocular parameters, as assessed by the investigator or the patient) and subjective comfort level measured on a VAS (0=no discomfort; 100=very severe discomfort). Secondary outcomes were reproducibility of the SENSIMED Triggerfish patterns on 2 separate readings.
Statistical methods	Reproducibility of signal patterns was assessed using Pearson correlations.
Participants	40 patients (21 with suspected glaucoma and 19 with established glaucoma).
Results	See table 14.
Conclusions	Adverse events were reported in the majority of patients; most were classified as mild, and 2 were severe. All adverse events resolved within 24 hours and no serious adverse events were reported. Tolerability scores were encouraging. Correlation between repeated SENSIMED Triggerfish curves was fair to good.
Abbreviations: IOP, intraocular pressure; VAS, visual analogue scale.

Table 10 Summary of the observational study: Mottet et al. (2013)

Study component	Description
Objectives/hypotheses	To evaluate 24-hour IOP rhythm reproducibility during repeated continuous 24-hour IOP monitoring with NCT and the SENSIMED Triggerfish in healthy people.
Study design	Observational, uncontrolled, open study.
Setting	Volunteers at a referral centre of chronobiology in France had 4 24-hour IOP monitoring sessions over 6 months. The IOP‑related patterns of the first eye were continuously recorded using the SENSIMED Triggerfish and the IOP of the opposite eye was measured hourly using NCT. Two sessions with NCT measurements in 1 eye and SENSIMED Triggerfish measurements in the opposite eye, 1 session with SENSIMED Triggerfish measurements in only 1 eye, and 1 session with NCT measurements in both eyes were performed.
Inclusion/exclusion criteria	Inclusion criteria: people free of sleep disturbance, endocrine illness or ocular disease (spherical equivalent between −1 and +1 dioptre), with regular lifestyle habits and a habitual total sleep time of approximately 8 hours. Exclusion criteria: people who worked shifts, had taken a transmeridian flight less than 2 months before the beginning of the study, had any medical treatment, or smoked.
Primary outcomes	A nonlinear regression analysis was used to model the 24‑hour IOP curve. Comparison of curve characteristics was done and agreement evaluated .
Statistical methods	Model was fitted to the SENSIMED Triggerfish data. ICC was used to measure reproducibility. A Bland-Altman plot was used to compare different methods.
Participants	12 young healthy adults.
Results	See table 14.
Conclusions	The SENSIMED Triggerfish produced a reproducible curve in healthy people but the relative variation in electrical signal cannot be used to estimate IOP measurements in mm Hg.
Abbreviations: ICC, intraclass correlation; IOP, intraocular pressure; NCT, non-contact tonometry.

Table 11 Summary of the interventional study: Pajic et al. (2011)

Study component	Description
Objectives/hypotheses	To perform 24-hour recording in 5 patients with NTG in the presence and absence of anti-glaucomatous treatment.
Study design	Interventional study.
Setting	Patients who were untreated or with IOP‑lowering drugs stopped for a minimum of 6 weeks and were recorded with the SENSIMED Triggerfish for 24 hours. Patients then had IOP‑lowering medication for at least 6 weeks. SENSIMED Triggerfish recording was then done again. Recording was conducted in an ambulatory setting in Switzerland (no dates reported). GAT IOP measurements were taken before and after SENSIMED Triggerfish recording. Patients were instructed to use their IOP-lowering medication as usual during SENSIMED Triggerfish recording.
Inclusion/exclusion criteria	NTG was defined as IOP without treatment less than 21 mm Hg on diurnal testing, gonioscopic open anterior chamber angle, typical glaucomatous optic disc damage with glaucomatous cupping and loss of neuroretinal rim, visual field defect compatible with the glaucomatous cupping and progressive damage.
Primary outcomes	No primary outcome was defined. Other outcomes included description of SENSIMED Triggerfish output with and without IOP-lowering drugs, relationship between GAT values before and after SENSIMED Triggerfish recording and the respective initial and final SENSIMED Triggerfish output.
Statistical methods	Pearson correlation was used to compare patient patterns over 2 sessions.
Participants	5 patients diagnosed with NTG
Results	See table 14.
Conclusions	SENSIMED Triggerfish profiles were repeatable and highly individual. The effect of IOP-lowering medication was not reflected in all SENSIMED Triggerfish patterns, and there was no relationship between GAT IOP values and SENSIMED Triggerfish output (no analysis reported).
Abbreviations: GAT, Goldmann Applanation Tonometry; IOP, intraocular pressure; NTG, normal tension glaucoma.

Table 12 Summary of the trial record: NCT01390779, Efficacy of 24-hour IOP recording with the SENSIMED Triggerfish

Study component	Description
Objectives/hypotheses	Assess the safety and effectiveness of the SENSIMED Triggerfish device in continuous recording of relative IOP‑related patterns.
Study design	Open, uncontrolled, observational study.
Setting	Patients were recruited between July 2011 and May 2012 and had 24-hour recording with the SENSIMED Triggerfish in 1 randomly selected eye, while in a sleep laboratory. An eye examination was done at screening and before and after the device recording.
Inclusion/exclusion criteria	Inclusion criteria: diagnosis of POAG, including normal tension glaucoma, or healthy people, including people with ocular hypertension for whom no evidence or suspicion of structural or functional glaucomatous damage exists, no anti-glaucomatous drug treatment or washed-out for 4 weeks, IOP symmetry of ±3 mm Hg between fellow eyes, aged 18‑80 years. Exclusion criteria: patients who had eye surgery in the last 3 months, corneal or conjunctival abnormality hindering contact lens adaptation, wear of full frame metallic glasses during SENSIMED Triggerfish recording, severe dry eye, secondary forms of open angle glaucoma, allergy to corneal anaesthetic.
Primary outcomes	The SENSIMED Triggerfish device's capacity to detect changes in IOP-related patterns from wake to sleep, and its ability to detect ocular pulse frequency concurrent to heart rate.
Statistical methods	Not reported.
Participants	33 patients with or without glaucoma (32 completed).
Results	See table 14.
Conclusions	These preliminary results in a trial record showed that the SENSIMED Triggerfish is associated with adverse events but none was serious.
Abbreviations: IOP, intraocular pressure; POAG, primary open angle glaucoma.

Table 13 Summary of the interventional study: Tojo et al. (2013)

Study component	Description
Objectives/hypotheses	To examine the effects of SLT on IOP-related patterns in patients with normal tension glaucoma using the SENSIMED Triggerfish.
Study design	Interventional study.
Setting	Patients were recruited between April 2012 and November 2012 in a hospital in Japan. All patients had SLT, and had IOP measurement and 24-hour recording with the SENSIMED Triggerfish before and after treatment.
Inclusion/exclusion criteria	Diagnosis of NTG was made if all of the following criteria were satisfied: presence of glaucomatous optic disc neuropathy with corresponding visual field defects; a threshold examination of Swedish interactive thresholding algorithm 30‑2 program showing 'outside normal limits' in a glaucoma hemifield test and a cluster of 3 contiguous points on the pattern deviation plot depressed at p<5% level; an open angle by gonioscopy; and a baseline IOP<21mm Hg. Inclusion criteria: NTG for over 20 years, BCVA ≥0.3, spherical equivalent <−6 dioptres, and no clinical problems in the ocular surface for wearing contact lens for 24 hours. Exclusion criteria: pseudo exfoliation syndrome, neovascular glaucoma, steroid glaucoma, primary angle closure glaucoma, history of ocular trauma, history of retinal diseases or uveitis, history of vitrectomy, and history of laser trabeculoplasty or other glaucoma surgeries.
Primary outcomes	No primary outcome was defined. Other outcomes included range of IOP-related patterns measured by the SENSIMED Triggerfish before and after SLT (including nocturnal and diurnal measurements), complications and corneal changes.
Statistical methods	The range of IOP‑related patterns, which was defined as the difference between the maximum and minimum value during the course of 24 hours, was calculated from the data of IOP‑related patterns. Wilcoxon signed-rank test was used for statistical analyses.
Participants	10 people with NTG.
Results	See table 14.
Conclusions	The range of IOP-related patterns measured by the SENSIMED Triggerfish over 24 hours showed no statistically significant difference before and after SLT. The range of nocturnal fluctuations was significantly different before and after SLT. Minor complications were reported which resolved quickly and without treatment
Abbreviations: BCVA, best corrected visual acuity; IOP, intraocular pressure; NTG, normal tension glaucoma; SLT, selective laser trabeculoplasty.

Table 14 Summary of results from non-randomised studies

Study	Agnifili et al. 2014
Design	Open, uncontrolled, observational study, n=30
Safety	4 people (13.3%; 3 with glaucoma and 1 without) reported blurred vision; 30 patients (100%) reported diffuse conjunctival hyperaemia; 12 study eyes (40%) developed superficial punctate keratitis and 2 eyes (6.7%) showed a conjunctival pressure mark from the contact lens sensor edge. No significant changes in BCVA were found after sensor removal.
Technical information relating to measurement capabilities	Differences in SENSIMED Triggerfish trends between patients with and without glaucoma were seen (p<0.05). Other outcomes were reported but not relevant.
Study	De Smedt et al. 2012
Design	Open, uncontrolled, observational study, n=10
Patient tolerability/ acceptability	Comfort scores did not change significantly from 8.35/10 at 5 minutes after fitting to 7.5/10 at 24 hours after fitting (it is not clear whether the sensor was in place at the 24-hour time point). The minimum detectable difference was 1.84 (p=0.16).
Safety	9 patients (90%) had localised fluorescein-positive staining and 1 (10%) had generalised staining after device removal; 3 (30%) had a corneal epithelial micro-defect/abrasion for which ofloxacin antibiotics were prescribed (resolved quickly). Contact lens impression mark was seen in 8 patients (80%). BCVA significantly reduced while wearing the SENSIMED Triggerfish (1.07 to 0.85; p=0.008). This reduction did not continue beyond 48 hours after removal of the device. Lubrication of the sensor lens surface was good in 9 eyes (90%) at baseline, but 1 eye showed hydrophobic spots. There was no significant change in this parameter after 30 minutes or 4, 12 and 24 hours. Lens mobility reduced during the 24 hours in which the SENSIMED Triggerfish was in place; during the final visit no spontaneous mobility was present in 9 eyes (90%) and in 1 eye (10%) it was difficult. At 24 hours the push-up test was difficult in all eyes. Decentration of the sensor lens in the vertical axis was seen in 7 patients; the authors report that this may indicate a too tight fit. At the time of publication only 1 lens size was available (8.7 mm curvature radius). In 2 patients disconnection between cable and antenna was seen.
Technical information relating to measurement capabilities	Other outcomes reported but not relevant.
Study	Faschinger and Mossbock 2013 (letter to editor)
Design	Open, semi-comparative, observational study, n=5.
Technical information relating to measurement capabilities	Applanation tonometry (method not specified) showed an increasing slope when patients were positioned to artificially increase IOP (supine or with head and thorax down). No agreement in curve profiles were seen between applanation and the SENSIMED Triggerfish, and no statistics were reported. Repeated applanation curves looked similar on the publication graph, whereas repeated SENSIMED Triggerfish curves looked less similar.
Study	Freiberg et al. 2012
Design	Open, uncontrolled, observational study, n=20
Safety	4 adverse events were reported (20%). 1 patient reported moderate eye irritation and blurred vision; another had severe ocular hyperaemia and was diagnosed with a bacterial conjunctivitis. Issues not defined as adverse events were superficial punctate keratitis in 10 eyes (50%), conjunctiva contact lens sensor edge pressure mark in 2 eyes (10%), and diffuse conjunctiva hyperaemia in all study eyes (100%). No significant changes in BCVA were reported (no data reported).
Technical information relating to measurement capabilities	Mean ultrasound CCT changed from 523 to 537 micrometres (p=0.015) in the eyes wearing the SENSIMED Triggerfish device, and from 518 to 522 micrometres in the opposite eyes (p=0.206).
Study	Hervas Ontiveros et al. 2013
Design	Open, uncontrolled, observational study, n=8.
Diagnostic accuracy/efficacy	Authors report that 1 patient was surgically intervened following high IOP measurements during the trial, although it is unclear whether SENSIMED Triggerfish or GAT measurements provided this information.
Patient tolerability/ acceptability	2 patients (25%) did not complete SENSIMED Triggerfish recording due to device intolerance (both reported discomfort such as a foreign body feeling and moderate conjunctival hyperaemia).
Serious adverse events	None.
Technical information relating to measurement capabilities	Other outcomes reported but not relevant.
Study	Hollo et al. 2014
Design	Interventional study, n=9.
Safety	100% of patients had mild transient conjunctival hyperaemia which resolved within 48 hours after SENSIMED Triggerfish recording; 1 patient (11%) had moderate discomfort and corneal epithelial defects which healed spontaneously in 1 day; 1 patient (11%) had mild corneal epithelium erosion with no discomfort which also healed spontaneously within 1 day; 1 patient (11%) had acute conjunctivitis treated with topical levofloxacin medication which healed in 4 days.
Technical information relating to measurement capabilities	Comparison of the means of 3 SENSIMED Triggerfish curves (2 untreated baseline curves and 1 curve under treatment with IOP‑lowering medication) showed no difference (152.94, 142.35, and 132.98 au; p=0.273). Differences in mean GAT measurements over 24 hours before and after treatment were significantly different (22.91±5.11 mm Hg after wash-out of medication and 8.24±2.49 mm Hg at 3 months of IOP-lowering treatment (p<0.001). This suggests that IOP-lowering medication did not change the shape of 24-hour SENSIMED Triggerfish curves, in contrast to GAT measurements. Mean differences in GAT IOP values from immediately before SENSIMED Triggerfish fitting and immediately after its removal did not differ significantly from zero, either when the untreated baseline values (mean difference: −0.722 mm Hg) or the values measured under treatment (mean difference: 0.111 mm Hg) were compared (p=0.083 and 0.884 respectively). In contrast, when the mean value of the first 50‑minute period of the SENSIMED Triggerfish curve was subtracted from the mean of the last 50‑minute values, the difference was significant for both the untreated and the treated periods (mean difference: 233.56 and 203.34 au; p=0.001 and <0.001 respectively). No correlation was seen between the difference of the first and last 50‑minute periods of the SENSIMED Triggerfish curves (mean values) and the corresponding difference of the first and last GAT IOP values either at baseline (Pearson correlation, r=-0.223; p=0.546) or under treatment (r=0.320; p=0.402).
Study	Lorenz et al. 2013
Design	Open, uncontrolled, observational study, n=40.
Patient tolerability/ acceptability	95% of patients with glaucoma (19/20) agreed to use the device again. Mean discomfort measured on a VAS was 21.82 (range 7–67, median 13.5) in the healthy group and 26.8 (range 0–71, median 22) in patients with glaucoma. Mean VAS for all patients was 24.3. 38 patients (95%) completed the study; 1 patient without glaucoma did not continue with monitoring due to improper device fitting (the device was too big and did not stay in place); 1 patient with glaucoma discontinued the study due to severe foreign body sensation and pain. The device was removed after 40 minutes and deterioration of corneal epithelial staining was detected. Discomfort had resolved after 1 hour. Retrospective analysis of the device by the manufacturer showed an incorrect encapsulation of the microelectronic components.
Safety	A total of 32 patients had an adverse event (80%) which was probably or definitely related to the study device (table below). All adverse events resolved within 2 days. Any adverse event: 80% of all participants (75% of healthy participants; 85% of glaucoma patients). Corneal epithelium defect: 58% of all participants (45% of healthy participants; 70% of glaucoma patients). Visual acuity reduced: 20% of all participants (20% of healthy participants; 20% of glaucoma patients). Foreign body sensation: 15% of all participants (10% of healthy participants; 20% of glaucoma patients). Conjunctival hyperemia: 15% of all participants (10% of healthy participants; 20% of glaucoma patients). Eye irritation: 8% of all participants (0% of healthy participants; 15% of glaucoma patients). Lacrimation increased: 5% of all participants (5% of healthy participants; 5% of glaucoma patients). Eye pain: 5% of all participants (0% of healthy participants; 10% of glaucoma patients). Conjunctivitis bacterial: 5% of all participants (10% of healthy participants; 0% of glaucoma patients). Conjunctival haemorrhage: 3% of all participants (0% of healthy participants; 5% of glaucoma patients). Eyelid oedema: 3% of all participants (5% of healthy participants; 0% of glaucoma patients). Sicca syndrome: 3% of all participants (0% of healthy participants; 5% of glaucoma patients). Abnormal sensation in eye: 3% of all participants (5% healthy participants; 0% glaucoma patients). Visual impairment: 3% of all participants (0% of healthy participants; 5% of glaucoma patients). Discomfort: 3% of all participants (0% of healthy participants; 5% of glaucoma patients). Administration site reaction (injury): 3% of all participants (0% of healthy participants; 5% of glaucoma patients). The following study parameters showed statistically significant changes when compared before and after fitting of the SENSIMED Triggerfish device: conjunctival oedema, (mean change=0.18; p=0.0016); conjunctival erythema (mean change=1.05; p<0.001); epithelial defects (mean change=1.08; p<0.001); lid oedema (mean change=0.16; p=0.063).
Serious adverse events	No serious adverse events were reported; 11% of adverse events were deemed severe: corneal epithelial defects (n=3), sharp pain in the eye (n=1) and conjunctival erythema (n=1). All adverse events resolved within 2 days.
Study	Mansouri and Shaarawy 2011
Design	Open, uncontrolled, observational study, n=15.
Diagnostic accuracy/efficacy	Following the findings of 24-hour recording with the SENSIMED Triggerfish, therapy was changed in 11 patients (73%).
Patient tolerability/ acceptability	13 patients (87%) completed 24-hour SENSIMED Triggerfish recording. Of the 2 who did not, 1 had pre-existing severe dry eye disease and discontinued IOP-related recording after 13 hours due to device intolerance; the device malfunctioned in the other. Average patient score for comfort was 7 out of 10.
Safety	5 minor complications were reported: 1 case of corneal erosion in a patient with severe dry eye disease and 4 cases of superficial punctate keratitis. All complications resolved after 24 hours.
Serious adverse events	None (0%)
Study	Mansouri et al. 2012
Design	Open, uncontrolled, observational study, n=40.
Patient tolerability/ acceptability	Mean comfort score for all patients was 27.2 out of 100 (SD 18.5) in the first recording session with SENSIMED Triggerfish and 23.8 (SD 18.7) in the second session a week later. Poor tolerability was reported in 4 patients in the first session and 3 patients in the second session.
Safety	151 adverse events were reported, of which 149 (in 38 patients) were device-related (142 were classified as mild). There were 77 in the first session and 72 in the second; 2 patients had moderate adverse events (1 superficial punctate keratitis and 1 blurred vision); 33 (82%) had blurred vision, 32 (80%) had hyperaemia of the bulbar alpebral conjunctiva, and 6 (15%) had superficial punctate keratitis. All adverse events resolved within 24 hours and there was none in the opposite eye. 15% of patients developed corneal staining. Technical issues which prevented complete data recording occurred in 4 cases: battery insufficiency (2), disconnection of device (1) and unknown (1).
Serious adverse events	No serious adverse events were reported during the study. Two patients exhibited conjunctival hyperaemia that was qualified as severe (5 of 149 events) but these severe adverse events were resolved within 24 hours.
Technical information relating to measurement capabilities	Correlation of two 24-hour recording sessions was r=0.59 (defined as a fair to good correlation).
Study	Mottet et al. 2013
Design	Open, uncontrolled, observational study, n=12.
Technical information relating to measurement capabilities	ICCs were measured to compare SENSIMED Triggerfish measurements across repeated visits. ICCs were significant (indicating fair to good agreement according to the authors) for 9 of 25 readings for the device. Results show that the SENSIMED Triggerfish cannot estimate the absolute value of IOP in mm Hg when using pre- and post-SENSIMED Triggerfish GAT measurements or when using pre- and post-SENSIMED Triggerfish NCT measurements. It is not possible to use the relative change in SENSIMED Triggerfish signal multiplied by the IOP measured before the SENSIMED Triggerfish fitting to estimate the absolute IOP at each point of the 24-hour session. Regarding the pre-session and post-session measurements with GAT and NCT, the agreement on the change in IOP between the SENSIMED Triggerfish and NCT or GAT was also poor. Bland‑Altman analyses suggest that SENSIMED Triggerfish overestimates the large IOP changes compared with NCT.
Study	Pajic et al. 2011
Design	Interventional study, n=5.
Technical information relating to measurement capabilities	In 3 patients absolute device output values were generally lower with IOP-lowering treatment than without, and for 2 patients SENSIMED Triggerfish output was lower in the control session (without IOP-lowering treatment) than in the treated session (the publication text contains an error relating to this point; the figures are assumed to be correct). There was no relationship between the reference GAT IOP values before and after device recording and the respective initial and final device output (no statistical testing done).
Study	Trial record (NCT01390779): Efficacy of 24-hour IOP recording with the SENSIMED Triggerfish
Design	Open, uncontrolled, interventional study, n=33.
Safety	Adverse events were recorded in 15 of 33 patients: 8 (24%) ocular hyperaemia, 7 (21%) punctate keratitis, 5 (15%) eye pressure mark, 1 (3%) blurred vision, 1 (3%) eyelid oedema and 1 (3%) corneal disorder.
Serious adverse events	0 (0%)
Study	Tojo et al. 2013
Design	Interventional study, n=10.
Safety	Minor complications were reported in some cases (number not reported), such as conjunctivitis, slight hyperaemia, or peripheral corneal oedema in the eyes following SENSIMED Triggerfish wear for 24 hours. These complications resolved within a few days (exact duration not reported) and without treatment. Visual acuity did not change before or after SLT.
Serious adverse events	No serious complications were observed.
Technical information relating to measurement capabilities	Mean IOP measured by tonometry at 1 month was significantly reduced following SLT (p=0.002). This was not reflected in the SENSIMED Triggerfish results (measured post-operatively at 1‑2 months); there was no significant change in 24-hour IOP changes before and after SLT (p=0.77). Nocturnal changes were significantly different before and after SLT (p=0.014).
Abbreviations: BCVA, best corrected visual acuity; GAT, Goldmann Applanation Tonometry; ICC, intraclass correlation; IOP, intraocular pressure; n, number of patients; NCT, non-contact tonometry; NTG, normal tension glaucoma; POAG, primary open angle glaucoma; SD, standard deviation; SLT, selective laser trabeculoplasty; VAS, visual analogue scale.

Nine newly-completed or ongoing trials on the SENSIMED Triggerfish device were identified:

Study ID 12426: Continuous Recording of Short Time Fluctuations in Intraocular Pressure using the Sensimed Triggerfish Sensor (Observational). This trial has completed recruitment (in follow-up) and has not been published.
NCT01972997: A Single-center, Randomized, Double-blinded, Prospective Study to Assess the Changes in the 24-hour IOP (Intraocular Pressure) Pattern in Relation to SENSIMED Triggerfish Sensor Sizes in Healthy Subjects. This trial has completed but has not been published.
NCT01906138: A Prospective, Open Label Study Assessing the 24-hour Intraocular Pressure Pattern Monitored by SENSIMED Triggerfish in Primary Angle Closure and Primary Angle Closure Glaucoma Patients, Before and After Laser Peripheral Iridotomy. This trial has completed but has not been published.
NCT01906502: A Prospective, Open Label Study to Assess the 24-hour IOP Pattern Recorded With SENSIMED Triggerfish in a Healthy Population. This trial has completed but has not been published.
NCT01767753: A Prospective, Observational, Open Label Study to Assess the 24-hour IOP Fluctuation Pattern Recorded With SENSIMED Triggerfish in Patients With Primary Open-angle Glaucoma, Before and After Selective Laser Trabeculoplasty. This trial has completed but has not been published.
NCT01467453: Clinical Application of Sensimed Triggerfish Sensor (TS) With Wireless Signal Transmission for Continuous Intraocular Pressure Measurement. The trial record indicated that this study is currently recruiting participants (estimated completion date December 2012).
NCT01507584: The Effects of the Water Drinking Test on Intraocular Pressure of Glaucoma Patients Undergoing 24 Hour Continuous Monitoring With the SENSIMED Triggerfish. The trial record indicated that this study is currently recruiting participants (estimated completion date Sept 2012).
NCT01912599: Pilot Study on Ambulatory Intraocular Pressure and Blood Pressure Monitoring in Glaucoma. This trial has completed but has not been published.
NCT02030886: A Single-center, Open Label, Prospective Study Assessing the 24-hour IOP Patterns Using SENSIMED Triggerfish in Ocular Hypertensive Patients Newly Converted to Glaucomatous Disease Versus Stable Ocular Hypertensive Patients. The trial record indicated that this study is currently recruiting participants (estimated completion date October 2014).

Costs and resource consequences

No published evidence relating to the cost or resource consequences of using the SENSIMED Triggerfish in the NHS was identified for this briefing.

In order to provide 24-hour recording with the SENSIMED Triggerfish device, an NHS eye clinic would need to purchase a SENSIMED Triggerfish sensor, antenna, sleeve and booklet for every patient. In cases where patients are monitored before and after an intervention, such as medication or surgery, each patient would require another sensor, antenna, sleeve, and booklet. The clinic would also need to buy at least 1 each of the reusable data recorder, cable, battery charger and Bluetooth stick. In addition, it would need a computer with the SENSIMED Triggerfish software installed.

If another SENSIMED Triggerfish lens needs to be used because of an ill-fitting lens, patient discomfort, or after a lens is dropped or damaged during fitting, its cost would be doubled.

It is assumed that an NHS eye clinic would have the necessary equipment to perform eye examinations before and after placement of the SENSIMED Triggerfish sensor.

Each patient would require at least 2 sessions with an appropriately trained ophthalmologist or optometrist for the fitting and removal of the SENSIMED Triggerfish sensor. An eye examination including visual acuity, central corneal radius and thickness measurement, slit lamp examination, and IOP tonometry should be done during the SENSIMED Triggerfish fitting session. Slit lamp examination and tonometry should be done after the device is removed, as well as analysis of the SENSIMED Triggerfish output. There is no information on the time taken to carry out these sessions. There is also uncertainty as to how existing ophthalmology services would manage this additional staff need.

Strengths and limitations of the evidence

The key limitation of the available evidence was that no studies investigated the impact of the SENSIMED Triggerfish device on clinical outcomes, such as progression to glaucoma or vision deterioration. There was no evidence that having a better understanding of IOP‑related patterns using the SENSIMED Triggerfish over 24 hours influenced the need for or success of any subsequent treatment. None of the selected studies show agreement between the SENSIMED Triggerfish and a reference standard, such as GAT (although correlations were analysed in some cases), and therefore there was no evidence that the SENSIMED Triggerfish provided more informative results than conventional techniques.

An important barrier to interpretation of the study findings on the SENSIMED Triggerfish in its current form is that it records dimensional changes related to IOP patterns using a millivolt scale, which cannot be referenced or calibrated to a standard measure of IOP in mm Hg. Also, the patterns that the SENSIMED Triggerfish produces are qualitative, and the values that would indicate inadequate management of ocular hypertension have not yet been defined.

All studies included in this briefing had small sample sizes (the largest was 40 patients) and most did not report whether patients were consecutively recruited, therefore introducing the risk of selection bias. Despite being of interest from a safety point of view, inclusion of healthy adults may reduce the generalisability of results to people with suspected or established glaucoma.

Given the nature of the device, blinded studies would be inappropriate. There is a risk of reporting bias in open-label studies published to date, specifically in outcomes such as patient-reported adverse events and tolerability. Many studies provided neither clear definitions of what constitutes an adverse event or complication, nor descriptions of tolerability measures.

Two letters to editors (Faschinger and Mossbock 2013; Hervas et al. 2013) and 1 trial register entry (NCT01390779) were included in the evidence for this briefing. These publications have not been peer-reviewed and contained limited information on their methodology. As such, the results should be interpreted with caution.

Seven publications reported either financial support from or conflicts of interest from the authors. This may be a further source of bias.