Comparisonbetweengraticuleandimagecaptureassessmentoflower.docx

Comparison between graticule and image capture assessment of lower tear film meniscus height J. Santodomingo-Rubido, a, , J.S. Wolffsohna and B. GilmartinaaOphthalmic Research Group, School of Life and Health Sciences, Aston University, Birmingham, UK AbstractPurposeTo compare graticule and image capture assessment of the lower tear film meniscus height (TMH).MethodsLower tear film meniscus height measures were taken in the right eyes of 55 healthy subjects at two study visits separated by 6 months. Two images of the TMH were captured in each subject with a digital camera attached to a slit-lamp biomicroscope and stored in a computer for future analysis. Using the best of two images, the TMH was quantified by manually drawing a line across the tear meniscus profile, following which the TMH was measured in pixels and converted into millimetres, where one pixel corresponded to 0.0018 mm. Additionally, graticule measures were carried out by direct observation using a calibrated graticule inserted into the same slit-lamp eyepiece. The graticule was calibrated so that actual readings, in 0.03 mm increments, could be made with a 40× ocular.ResultsSmaller values of TMH were found in this study compared to previous studies. TMH, as measured with the image capture technique (0.13 ± 0.04 mm), was significantly greater (by approximately 0.01 ± 0.05 mm, p = 0.03) than that measured with the graticule technique (0.12 ± 0.05 mm). No bias was found across the range sampled. Repeatability of the TMH measurements taken at two study visits showed that graticule measures were significantly different (0.02 ± 0.05 mm, p = 0.01) and highly correlated (r = 0.52, p < 0.0001), whereas image capture measures were similar (0.01 ± 0.03 mm, p = 0.16), and also highly correlated (r = 0.56, p < 0.0001).ConclusionsAlthough graticule and image analysis techniques showed similar mean values for TMH, the image capture technique was more repeatable than the graticule technique and this can be attributed to the higher measurement resolution of the image capture (i.e. 0.0018 mm) compared to the graticule technique (i.e. 0.03 mm).Keywords: Tear meniscus height; Tear prism; Tear film; RepeatabilityArticle Outline1. Introduction 2. Methods 2.1. Statistical analysis3. Results 4. Discussion Acknowledgements References1. IntroductionThe tear meniscus, which is formed between the lid surface and the bulbar conjunctiva, is present along the superior and inferior lid margins. It has been estimated that the tear meniscus holds 7590% of the total volume of the tear film 1. Therefore, careful examination of the lower lid tear meniscus height (TMH) is likely to provide a simple but clinically useful indication of tear volume. Assessment of the TMH is commonly used by eye care practitioners universally as part of routine ocular assessments 2 and 3. However, mean absolute values of TMH in normal eyes have been reported to vary between 0.14 and 0.46 mm depending on the technique employed 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 and 20. Studies using calibrated graticules inserted into the slit-lamp eyepiece have reported mean values which vary between 0.15 and 0.35 mm in normal healthy eyes (Table 1) 4, 5, 6, 7, 8 and 9. Another study has used the calibrated variable slit beam height of a slit-lamp to measure the TMH and reported a mean value of 0.32 mm 10. Video capture techniques measure the TMH by taking an image of the meniscus in which the upper and lower extent of the tear meniscus is located subjectively (often by drawing a line) and then, using computer software, convert the number of pixels between these points to a physical distance using a calibration factor. Mean values of the TMH between 0.17 and 0.43 in normal healthy eyes have been reported using image capture techniques (Table 1) 11, 12, 13, 14, 15, 16 and 17. Measures of the TMH can also be obtained by means of an optical pachymeter attached to a slit-lamp and mean values of the TMH in normal subjects have, in this case, been found to vary between 0.16 and 0.38 mm 17, 18 and 19. Finally, optical coherence tomography, which uses low-coherence interferometry to produce two-dimensional images of optical scattering from tissues, can allow cross-sectional imaging of the TMH, giving mean reported values of 0.14 20 and 0.27 mm 17 in normal healthy eyes. Larger values of TMH have been reported in studies in which fluorescein was instilled into the eye to enhance visualization of the tear meniscus 14 and 21. Fluorescein is known to disrupt the tear film by reducing stability 22, 23 and 24 and this might initiate reflex lacrimation leading to higher TMH values. It is also possible that increased tear meniscus height with the use of fluorescein occurs as a result of ocular irritation and/or adding fluorescein containing-dye solution to the tear meniscus 25. Although differences in the mean value of the TMH have been reported in the literature, the measurement of TMH has been found to be useful in differentiating between normal, dry eye, and nasolacrimal duct obstruction subjects 26 and 27. Furthermore, TMH measurements have shown promising results in evaluating the efficacy of punctual occlusion in subjects with different forms of dry eye 28. Significant differences in TMH values have also been found between tolerant and intolerant contact lens wearers 16. Whereas calibrated graticules inserted into the slit-lamp eyepiece might be more commonly used in clinical settings, the use of image capture techniques are predominantly used in research settings. Since the relationship between these two methods of TMH assessment has not been compared in the same subjects, the purpose of this study is to compare a graticule versus an image capture technique in the measurement of the TMH.Table 1. Mean ± standard deviation (S.D.) lower tear menicus height (TMH) values reported in the literature using graticule and image capture techniquesGraticule Image capture Author Mean ± S.D. TMH Author Mean ± S.D. TMH Lamberts et al. 40.23 ± 0.09Zaman et al. 110.18 ± 0.11Tomlinson et al. 50.35 ± 0.11Doughty et al. 120.17 ± 0.05Miller et al. 60.22 ± 0.08Doughty et al. 130.19 ± 0.09Jordan and Baum 70.30 ± 0.06Kwong and Cho 150.24 ± 0.07Papas and Vajdic 80.15 ± 0.04Glasson et al. 160.43 ± 0.11Lim and Lee 90.19 ± 0.05Johnson and Murphy 170.34 ± 0.05Mean ± S.D. TMH0.24 ± 0.07Mean ± S.D. TMH0.26 ± 0.10Full-size tableResults for normal subjects without instillation of fluorescein are shown.View Within Article2. MethodsFifty-five healthy subjects (24 males and 31 females), with a mean age of 20.4 (standard deviation S.D. 1.6) years (range 18.424.7 years, median 20.1 years), with a mean ± S.D. spherical equivalent of 2.46 ± 2.34 D, and astigmatism <1.00 D, had measurements of TMH performed by one of the authors (J.S.-R.) at two scheduled visits separated by 6 months. Ethical approval for this study was obtained from the Aston University Ethical Committee. All subjects were given written information about the study before they signed a written statement of consent to participate. The study complied with the Declaration of Helsinki. TMH was measured as the distance between the darker edge of the lower eyelid and the tear strip. Subjects were positioned on the slit-lamp chin rest and instructed to look at a target located to maintain primary eye gaze. In order to avoid reflex tearing, a relatively short light beam of low intensity was used to prevent direct shinning of the light into the pupil during TMH measurement. Two images (800 × 600 pixels) of the lower fornix and a portion of the inferior sclera and lower eyelid were brought into focus, captured using a digital camera (CKY-F58 3-CCD, JVC Americas Corp., Wayne, USA) attached to a slit-lamp (Topcon SL-7F, Tokyo, Japan) by WinTV software (v4.6, Hauppauge!, NY, USA) and stored in a computer for future analysis (Fig. 1). Using the best of two images, the TMH was quantified by manually drawing a line across the tear meniscus profile, following which the TMH was measured in pixels and converted into millimetres, where one pixel corresponded to 0.0018 mm. Following image capture, the TMH was subsequently measured subjectively with a graticule (WF10Xmicro, Olympus, Tokyo) inserted into the eyepiece of the slit-lamp. The graticule was calibrated so that actual readings, in 0.03 mm increments, could be made with a 40× ocular. With both techniques, three readings were recorded and a mean obtained.Fig. 1. Image of the tear meniscus height captured by a camera attached to the slit-lamp. Fig. 1. Image of the tear meniscus height captured by a camera attached to the slit-lamp.View Within Article2.1. Statistical analysisThe distribution of values and repeatability of three consecutive measures of the TMH at the first study visit were assessed with the coefficient of skewness and the mean standard deviation (S.D.) of repeated measures, respectively. The bias between the two measuring techniques and repeatability between study visits for each technique (i.e. mean difference, S.D., and 95% confidence limits) were calculated and presented graphically 29. Comparisons and correlations between measures were performed using paired two-tailed t-tests and the Pearson's product moment correlation coefficient, respectively. Data for the right eye only were used to avoid the confounding effect of using non-independent data from both eyes 30. The level of statistical significance was taken as 5%.3. ResultsA mean ± S.D. value of TMH of 0.13 ± 0.04 (range 0.040.23 mm, median 0.13 mm) and 0.12 ± 0.05 mm (range 0.040.30 mm, median 0.11 mm) was found with the image capture and graticule techniques, respectively, at the first study visit (Fig. 2). A lower coefficient of skewness was found with the image capture measures (0.44) compared to the graticule measures (1.04) at the first visit. The repeatability of three consecutive measures at the first visit was 0.01 and 0.004 mm for the graticule and image capture measures, respectively. TMH, as measured with the image capture technique, was significantly longer (by approximately 0.01 ± 0.05 mm, p = 0.03) than that measured with the graticule technique (Fig. 3). The image capture technique could be expected to read approximately as much as 0.09 above or 0.07 below the subjective technique. There was no significant bias between the two methods of measurement for the whole range of TMH values evident in this study. Both measuring techniques were highly correlated (r = 0.61, p < 0.0001). Repeatability of the TMH measurements taken at the two study visits showed that graticule measures were significantly different (0.02 ± 0.05 mm, p = 0.01) and highly correlated (r = 0.52, p < 0.0001), whereas image capture measures were similar (0.01 ± 0.03 mm, p = 0.16), but also highly correlated (r = 0.56, p < 0.0001) (Fig. 4).Fig. 2. Histogram to show the distribution of the graticule and image capture measures of the lower tear meniscus height from the 59 right eyes evaluated. Fig. 2. Histogram to show the distribution of the graticule and image capture measures of the lower tear meniscus height from the 59 right eyes evaluated.View Within ArticleFig. 3. Differences between image capture and graticule measures of the lower tear meniscus height (TMH). Fig. 3. Differences between image capture and graticule measures of the lower tear meniscus height (TMH).Fig. 4. Repeatability of the graticule and image capture measures of the lower tear meniscus height (TMH) between the two study visits. Fig. 4. Repeatability of the graticule and image capture measures of the lower tear meniscus height (TMH) between the two study visits.View Within Article4. DiscussionConsistent with previous studies, the values of TMH found in this study were positively skewed for both the graticule and image capture techniques 4 and 12. In addition, lower mean values of the TMH were found compared to previous studies (Table 1). However, there are a number of reports which, although slightly higher on average, have shown similar values to the TMH values reported here 8, 11, 12, 18 and 19. Most subjects employed in this study were normal silicone hydrogel contact lens wearers (76%) and a previous study has shown lower values of TMH in silicone hydrogel lens wearers compared to normal non-contact lens wearers 6. Special care was taken to ensure that subjects maintained primary eye gaze during the measurement of the TMH as it was noted prior to the study that when subjects were asked to look upwards the height of the tear meniscus approximately doubled, and when subjects were looking nasally, about two-thirds of the subjects showed higher values of the TMH than when they were maintaining primary eye gaze. Although some studies have measured TMH while subjects were maintaining primary eye gaze 12, 13, 15 and 17, the extent to which others studies have controlled eye gaze is unclear 4, 5, 6, 7, 8, 9, 10, 11, 14, 16, 18, 19 and 20. In taking TMH measurements, special care was taken in the present study to avoid reflex tearing by direct shinning of the slit-lamp beam light into the pupil. Whereas some studies have taken steps to avoid this as a source of error 12, 13, 15 and 17, this may not have been the case in others 4, 5, 6, 7, 8, 9, 10, 11, 14, 16, 18, 19 and 20. A statistically significant difference between the two techniques of TMH measurement was found. Nevertheless, the relatively close match in the mean TMH values between the two techniques is in agreement with previous studies (Table 1). A better repeatability for three consecutive readings and over time was found with the image capture compared to the graticule technique, a finding likely to be attributable to the higher measurement resolution of the former (i.e. 0.0018 mm) compared to the latter (i.e. 0.03 mm). Two earlier studies using a similar image capture technique have reported comparable levels of repeatability to those found in this study 15 and 16. In addition to the assessment of the lower TMH, a study of both the upper and lower tear meniscus height, radius of curvatures, width and cross-sectional areas is likely to provide valuable information for the diagnosis and treatment of subjects with dry eyes, nasolacrimal duct obstruction and contact lens complications 14, 16, 26, 28, 31 and 32. Clearly, the validity and repeatability of these tear meniscus parameters need to be assessed. In summary, although graticule and image analysis techniques showed similar mean values for TMH, the image capt