Highlights
- •
Corneal topography measurements are affected by in the insertion of fluorescein sodium (NaFl) ocular dye.
- •
A single (not double) dose of NaFl resulted in increased reliability and consistency in corneal topography measurements.
- •
Tear film surface regularity changes significantly with NaFl, although this is not clinically significant in healthy corneas.
Abstract
Purpose
During image acquisition, certain topographers require the addition of sodium fluorescein (NaFl) dye to the tear film. This study investigates the effect of NaFl dye on corneal topography and tear surface quality.
Method
The E300 corneal topographer (Medmont International Pty Ltd., Victoria, Australia) was used to measure ocular surface topography and quality of 57 eyes of 57 healthy individuals without dry eye symptoms, age 35.1 ± 15.2 years (mean ± standard deviation) ranging between 19 and 65 years. The mean of three simulated keratometry values, a variety of corneal shape descriptors, and Tear Film Surface Quality (TFSQ) were measured under three different conditions; without NaFl (baseline), with the addition of a single dose NaFl, and using a double dose of NaFl.
Results
Compared to baseline, the Inferior-Superior (IS) index decreased significantly after a single dose (P = 0.034) or double dose of NaFl (P = 0.030). The corneal surface was significantly more regular without NaFl (P = 0.003) or one insertion of NaFl (P = 0.024) when compared to two doses of NaFl. There was no association with age, or dry eye signs or symptoms on the variance observed in any of the indices between baseline, intervention I, and intervention II (P > 0.05). Agreement between corneal surface indices reduced following the addition of NaFl.
Conclusion
In comparison to measurements taken without an ocular dye, one dose of NaFl resulted in increased reliability and consistency in corneal topography measurements using the E300 topographer, but 2 doses decreased reliability and consistency. Practitioners ought to be aware that tear film surface regularity and inferior-superior corneal power changed significantly following the addition of NaFl in those with healthy corneas. Its effect in diseased corneas is unknown.
1
Introduction
Most common corneal surface devices in an optometric practice measure the curvature of the cornea, a baseline measurement in soft contact lens fittings, over a relatively small central area. These include keratometers (2–4 mm) and small- or large-cone Placido disc videokeratographs (6−8 mm) [ ]. Using Placido disc videokeratography, attempts have been made to extend the corneal coverage to approximately 11 mm using extrapolation techniques [ , ]. Following the recent increase in fittings of large diameter gas permeable contact lenses, such as contact lenses for orthokeratology (>11 mm) and (semi-/ mini-) scleral (13 to 24+ mm) lenses, devices are needed to visualize and measure large geometric areas of the cornea and sclera.
Other approaches to corneal topography, including Optical Coherence Tomographers (OCT) and Scheimpflug-based imaging devices are able to generate a three-dimensional profile of the anterior segment and provide topography information about the anterior as well as posterior cornea including corneal thickness. These imaging techniques are able to measure a larger surface of the anterior eye up to 16 mm [ ]. Scheimpflug camera systems have shown good agreement for anterior corneal geometry compared to Placido-based videokeratography [ ], although this is significantly reduced for the posterior cornea [ ]. Similar results have been reported for OCT devices when compared to Placido-based videokeratography [ ], producing suboptimal peripheral output predominantly associated with refractive and elevation data. It is possible that three-dimensional information obtained using a radial scan mode plays a role, causing oversampling in the central region while undersampling in the outer region, and misalignment of the OCT system caused by the patient or operator proving difficulties with acquiring perfectly centered radial scans [ ].
Devices based on Fourier transform profilometry have been developed to measure the corneo-scleral region up to 16.5 mm [ , ]. Examples of this are the Eye Surface Profiler (ESP, Eaglet-Eye, The Netherlands) and sMap3D (Precision Ocular Metrology, LLC., Cedar Crest, New Mexico, United States). These devices directly measure the elevation of both the anterior and posterior cornea via time domain or light-based analysis, while converting elevation data into anterior and posterior curvatures (in diopters) as well as corneal thickness. To obtain a measurement, the profilometer uses the phase information of the projected images, which only exist by the mirror function of the tear film [ , ]. To obtain reflected light from the scleral epithelium, a significant amount of sodium fluorescein (NaFl) dye is required prior to image capture. NaFl dye is highly water soluble and is used as a diagnostic dye to detect the tear film stability as well as damage of the epithelial cells of the anterior surface of the eye. When artificially increasing tear volume by introducing an ocular dye like NaFl, it is important to understand the effect of this volume increase on the regularity of the corneal topography and quality of resulting images. This cross-sectional study aims to evaluate the effect of various doses of NaFl dye on the image quality of corneal topography when using a Placido disc videokeratograph. Adding NaFl was expected to increase the volume, smooth the tear film and improve optical regularity, and possibly increase reliability and consistency of the corneal curvature measurements. A particular interest was to investigate if topography and tear film quality measurements are affected by multiple doses of NaFl. To do this, variations in the ocular surface topography following instillation of different amounts of NaFl were observed, predominantly in participants with low tear film quality and/or signs of dry eyes.
2
Methods
The research conducted in this study complied with the requirements of the Declaration of Helsinki (2008) and the research protocol and documentation received approval from the School of Health Sciences Research Ethics Committee at City, University of London (United Kingdom) and University of Applied Sciences Utrecht (UAS Utrecht; the Netherlands). Written consent was obtained prior to participation. The study included 57 healthy participants from the Healthcare department at the UAS Utrecht Eyecare Clinic between February and June 2017. Health was determined by general and ocular health questionnaires and anterior eye examination using slit lamp biomicroscopy. Exclusion criteria included a history of ocular surgery, anterior eye trauma or corneal/ corneal-scleral disease resulting in reduced visual acuity. Volunteers were excluded if they were pregnant, diagnosed with amblyopia, rigid contact lenses wearers, or presented with any other corneal abnormalities including suspect keratoconus. Participants were either neophytes or were asked to discontinue soft contact lens wear for at least 48 h prior to the assessment.
All measurements were taken on both eyes during a single visit ( Table 1 ), including a 20-minute break between the baseline measurements and interventions I and II. To rule out observer bias, all measurements were obtained by the same experienced investigator (JM). At the baseline visit, participants underwent a clinical anterior eye examination and symptom assessment. The prevalence of ocular surface disease was determined using the Ocular Surface Disease Index (OSDI) questionnaire culturally translated into Dutch (Oogoppervlak Beperkingen Vragenlijst, Alcon, 1995). OSDI results were considered normal (0–12 points), mild (13–22 points), moderate (23–32 points) or severe (33–100 points) diseased [ ]. A subjective refraction was performed to determine the degree of ametropia and visual acuity. Snellen visual acuity was measured with full correction. A standard optometric slit lamp examination of the anterior eye including (palpebral) eyelids, lid margins, conjunctiva, limbus and cornea was performed and evaluated using CCLRU grading scales (0–4 grades) in 0.5 increments [ ].
| Baseline | Intervention I | Intervention II |
|---|---|---|
| 20 min after baseline | 20 min after intervention I | |
| Health questionnaire | Insertion of one application of NaFl | Insertion of two applications of NaFl |
| Subjective refraction including visual acuity | Corneal topography and surface assessment | Corneal topography and surface assessment |
| Slit lamp examination of anterior eye | ||
| Corneal topography and surface assessment (Medmont topographer) | ||
| Tear Break Up Time (TBUT) [ ] |
2.1
Corneal topography and surface assessment
Corneal topography and surface assessment were determined using the E300 corneal topographer (Medmont International Pty Ltd., Victoria, Australia. Software: Medmont Studio 6.0). Three high-quality measurements were obtained whereby the instrument‘s software automatically calculates the geometric shape of the measured area and provides information about the regularity of the surface using indices including Simulated keratometry values, Inferior Superior index (IS), Surface Asymmetry Index (SAI), Surface Regularity Index (SRI) and the Tear Film Surface Quality (TFSQ; see Table 2 ) [ , ]. For all indices, the average of three measurements were used for analysis.
| Abbreviation | Description | |
|---|---|---|
| IS | Inferior Superior Index (Diopters) | The difference between the average inferior and superior power in the eye is called the IS value. Measured over a 4−5 mm area depending on the position of the eyelids. |
| SAI | Surface Asymmetry Index (Diopters) | Calculated from the centrally weighted summation of differences in corneal power between corresponding points at one of the 128 equidistant chords, 180 degrees apart on the eyes surface. A regular cornea shows a SAI value of <1.0. |
| SRI | Surface Regularity Index (Diopters) | Description of the corneal shape in the central 4.5 mm zone. The power of each point is compared with contiguous points. The calculation is based on the determination of the most frequently occurring dioptric power and the comparative analysis of dioptric powers of adjacent points in 256 hemi-meridians in the 10 central rings [ ]. A cornea with an SRI index of <0.8D is considered regular. |
| TFSQ | Tear Film Surface Quality | Tear film quality indicated by the average organisation of the Placido disc image reflections over the entire cornea. A local TFSQ value of 0.30 or higher correspondents with a visual tear break up [ ]. |
2.2
Tear Break Up Time (TBUT)
To measure TBUT, a sterile BIO-GLO 100 fluorescein sodium 1 mg ophthalmic (HUB Pharmaceuticals, LLC Rancho Cucamonga CA, USA) was moistened with 1 drop of non-preserved saline from a minim (Oté Pharma, Uden, The Netherlands), and was gently shaken once after moistening to remove excess fluorescein solution from the strip. After application on the superior-temporal conjunctiva, the TBUT was observed with a SL-9900D LED slit lamp biomicroscope (CSO Srl, Firenze, Italy) with a cobalt blue filter using a Wratten no 12 (yellow) filter after two blinks. Time in seconds was recorded when the first dry spot was observed after blink using a full width beam at 10x magnification. The mean of three TBUT measurements was calculated for each eye with a time period of at least one minute between measurements to improve measurement accuracy [ ].
2.3
Intervention I and II
After 20 min, corneal topography measurements were repeated after the insertion of NaFl as described above. During the first intervention, one lubricated strip of BIO-GLO 100 was used, while during the second intervention (20 min apart) this was immediately followed by a second strip of NaFl.
2.4
Statistical analysis
Statistical analyses were calculated using SPSS statistical package version 25 (SPSS Inc., Chicago, IL, USA). Mean spherical equivalent and corneal topography indices showed strong positive correlations between both eyes ( p < 0.0005); therefore, only right eyes were included for analysis to alleviate any inter ocular dependency issues and statistical bias due to the mirror-image relations [ ]. At baseline, age-related differences were calculated between mean values (Mann-Whitney test) and proportions (one-sample t -test between percents). Following violation of the assumptions of normality (Kolmogorov-Smirnov tests) for indices SAI, SRI, and TFSQ, data was transformed on a logarithmic scale to achieve normality and for statistical analyses, whereas raw (sample) data is presented as summary statistics (mean ± SD, 95 % confidence intervals CI, etc) [ ]. A one-way repeated measures ANOVA including Least Significant Difference post hoc tests determined the significance between measurements under different conditions (baseline, intervention I and II), while mixed between-within AVOVA tests were used to explore the effect of covariates such as age and dry eye. Intra class correlation (ICC) estimates and their 95 % confident intervals were calculated based on a mean-rating (k = 3), absolute-agreement, 2-way mixed-effects model. Coefficient of Repeatability (CoR) for each of the parameters measured at baseline and each intervention were calculated as 1.96 x S w (within-subject standard deviation). Agreement between the different interventions was calculated using the mean differences and 95 % limits of agreement (LoA). Statistical significance was accepted at the 95 % CI (p < 0.05). The participants were grouped by age, and two-way ANOVA power statistics revealed that a sample size of 39, 19 subjects per group, was needed to detect a standardized difference between the groups using a partial eta squared of 0.033 and 80 % power at 5% significance level [ ]. This calculation was based on an estimated mean of 3 repeated flat keratometry readings of 7.87 mm with group SDs of 0.29 mm, based on data collected from the first 15 subjects.
3
Results
Demographic and dry eye characteristics of the participants are summarized in Table 3 . A total of 57 participants (34 females, 23 males), age range between 19 and 65 years, were divided in two age groups: group A < 40 years (n = 34) and group B ≥ 40 years (n = 23). All participants presented with healthy corneas, without significant ocular surface, eyelid diseases or corneal staining. Both groups were well-matched for gender, OSDI scores, mean spherical equivalent (MSE), and average TBUT measured at baseline.
| All subjects | <40 years | ≥40 years | p | |
|---|---|---|---|---|
| n = 57 | n = 34 | n = 23 | ||
| Gender (male: female) | 23 : 34 | 14 : 20 | 9 : 14 | 0.88 |
| Age (years) | 35.1 ± 15.2 | 23.8 ± 4.5 | 51.7 ± 8.4 | <0.0005 |
| [31–39] | [22–25] | [48–55] | ||
| OSDI score | 11.9 ± 9.9 | 12.0 ± 9.3 | 11.6 ± 10.9 | 0.83 |
| [9.2–14.5] | [8.8–15.3] | [6.9–16.3] | ||
| % normal OSDI score (<13) | 68 % | 62 % | 78 % | 0.21 |
| MSE (Diopters) | −0.82 ± 2.38 | −1.11 ± 2.17 | −0.39 ± 2.65 | 0.21 |
| [−1.45 to −0.19] | [−1.86 to −0.35] | [−1.54 to −0.76] | ||
| Best corrected VA (Snellen decimal) | 1.23 ± 0.21 | 1.34 ± 0.17 | 1.08 ± 0.14 | <0.0005 |
| [1.18–1.29] | [1.28–1.40] | [1.02–1.14] | ||
| Mean of three TBUT (seconds) | 8.1 ± 6.9 | 7.8 ± 7.2 | 8.6 ± 6.5 | 0.43 |
| [6.3–10.0] | [5.3–10.3] | [5.8–11.4] |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
