Clinical evaluation of the Oculus Keratograph.docx

《Clinical evaluation of the Oculus Keratograph.docx》由会员分享，可在线阅读，更多相关《Clinical evaluation of the Oculus Keratograph.docx（4页珍藏版）》请在得力文库 - 分享文档赚钱的网站上搜索。

1、Contact Lens & Anterior Eye 35 (2012) 171174 Contents lists available at SciVerse ScienceDirect Contact Lens & Anterior Eye jou rn al h om epa ge: Clinical evaluation of the Oculus Keratograph N. Best a,b , L. Drury a , J.S. Wolffsohn b, a Specsavers, 41 High Row, Darlington, Co Durham DL3 7QW, UK

2、b Life and Health Science, Aston University, Ophthalmic Research Group, Birmingham, UK a r t i c l e i n f o a b s t r a c t Keywords: Non-invasive tear break-up time Tear lm stability Keratometry Objective Corneal topography Aim: To determine the validity and reliability of the measurement of corne

3、al curvature and non-invasive tear break-up time (NITBUT) measures using the Oculus Keratograph. Method: One hundred eyes of 100 patients had their corneal curvature assessed with the Keratograph and the Nidek ARKT TonorefII. NITBUT was then measured objectively with the Keratograph with Tear Film S

4、can software and subjectively with the Keeler Tearscope. The Keratograph measurements of corneal curvature and NITBUT were repeated to test reliability. The ocular sensitivity disease index questionnaire was completed to quantify ocular comfort. Results: The Keratograph consistently measured signica

5、ntly atter corneal curvatures than the ARKT (MSE difference: +1.83 0.44D), but was repeatable (p 0.05). Keratograph NITBUT measurements were signicantly lower than observation using the Tearscope (by 12.35 7.45 s; p 0.001) and decreased on subsequent measurement (by 1.64 6.03 s; p 0.01). The Keratog

6、raph measures the rst time the tears break up anywhere on the cornea with 63% of subjects having NITBUTs 5 s and a further 22% having readings between 5 and 10 s. The Tearscope results were found to correlate better with the patients symptoms (r = 0.32) compared to the Keratograph (r = 0.19). Conclu

7、sions: The Keratograph requires a calibration off-set to be comparable to other keratometry devices. Its current software detects very early tear lm changes, recording signicantly lower NITBUT values than conventional subjective assessment. Adjustments to instrumentation software have the potential

8、to enhance the value of Keratograph objective measures in clinical practice. 2012 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved. 1. Introduction In optometric practice corneal curvature is routinely measured with a keratometer prior to rigid lens tting. A keratomet

9、er is an instrument used to examine the central 3.03.5 mm of the cornea providing information on the radii of curvature, the directions of the principle meridians, the degree of corneal astigmatism and the presence of any corneal distortion. Keratometers only assesses the central corneal curvature,

10、but most corneas atten towards the periphery as prolate ellipses 1. Videokeratoscopes, generally known as topographers, typically assess corneal curvature over a wider (up to 10 mm diameter) region of the cornea by reecting an illuminated placido disc of known proportions off the tear lm and compari

11、ng this to the imaged reection. Image processing software detects the location of the rings objectively in multiple meridians and displays the data in the form of contour maps along with simulated keratometry readings in the principal axes. As well as providing generally more Corresponding author at

12、: Life and Health Science, Aston University, Aston Tri- angle, Birmingham B4 7ET, UK. Tel.: +44 0121 2044140; fax: +44 0121 2044048. E-mail address: j.s.w.wolffsohnaston.ac.uk (J.S. Wolffsohn). reliable information on corneal topography over a wider corneal area the reection quality of the placido m

13、ires indicates the quality of the tear lm over time. Whilst this has been utilised in a research setting 2, until now no commercial devices have been available to objectively assess non-invasive tear break-up time. Objectively analysing the placido reections from the tear lm over time after a blink

14、has been shown to have higher sensitivity, but similar speci- city in predicting symptomatic dry eye than uorescein break-up time. Tear stability is routinely assessed in clinical practice to aid in the diagnosis of dry eye disease and to help predict the likelihood of contact lens induced dry eye i

15、n neophyte contact lens wearers. This is most commonly done by assessing the tear lm break up time (BUT), a measurement of the time which elapses between a patient blinking and their tear lm beginning to break up or a subsequent uncontrollable blink. It is often assessed following the instillation o

16、f sodium uorescein dye into the tears and observation with a slit lamp microscope using blue light and a yellow enhancement lter 3. There is concern that the presence of uorescein in the tear lm will destabilise the tears and for this reason it is preferable to measure tear lm non-invasively without

17、 rst instilling uorescein 46 This type of tear lm measurement is referred to as non- invasive tear break-up time (NITBUT) although it should be noted 1367-0484/$ see front matter 2012 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved. http:/dx.doi.org/10.1016/j.clae.20

18、12.04.002 172 N. Best et al. / Contact Lens & Anterior Eye 35 (2012) 171174 that changes in meniscus curvature have been observed even with this minimally invasive technique suggesting it is easy to induce minor degrees of reex tearing 6. The repeatability of measurements with one of the main subjec

19、- tive devices for assessing NIBUT, the Tearscope (Keeler, Windsor, UK) appear to be more reliable that other techniques such as obser- vations through a slit lamp or of video keratoscope mires, although Tearscope measures are still quite variable 7 and there is consid- erable interexaminer variabil

20、ity 8. The Diagnostic Methodology Subcommittee of the International Dry Eye Workshop stated it was important to develop objective analysis methods of NIBUT to help standardise tear lm examination methods and improve compara- bility of measurements 6. The Keratograph (OculusOptikgerate GmbH, Wetzlar,

21、 German) is the rst commercially available device with software (“Tear Film Scan”) which permits an automated, examiner independent tech- nique for measuring NITBUT. The aim of this study was to determine the validity and reliability of the measurement of corneal curvature and NITBUT measures using

22、the Keratograph. 2. Methods One hundred consecutive patients with no known anterior eye disease (average age 37 13 years, range 1967 years; 65 female) were recruited from the staff and patients of a community opto- metric practice in the North East of England. Consent was obtained after explanation

23、of the study and possible consequences of taking part. The study was approved by the ethical committee of Aston University and conformed to the Declaration of Helsinki. Due to the similar nature of the two eyes, data from only right eyes were analysed to avoid statistical bias. A single keratometry

24、reading was captured with a validated Tonoref II (Software version 1.05; Nidek, Nagoya, Japan) 9 following alignment of the instrument head with the centre of the pupil and after the patient had been asked to blink. Two further topography images of the patients right eye were sub- sequently captured

25、 with the Oculus Keratograph (software version 2.73r19). All measurements were taken by a trained optometrist or contact lens optician and took approximately 30 s. Both instru- ments were calibrated by their manufacturers immediately prior to the study. NITBUT was measured on the same patients with

26、the Keeler Tearscope (average of 3 reading) by one researcher and then within 5 minutes twice with the objective Keratograph (average of 3 read- ings) by another masked researcher, in random order to prevent bias. Once the Keratograph assessment drops below an unspeci- ed level, the instrument stops

27、 measuring and this time was also recorded. Tearscope measures were made using a grid to create lines of non-reection so that tear lm break-up could be observed at any point over the corneal surface. The Tearscope was hand-held and the tear lm observed through the magnifying lens attachment. The Ocu

28、lar Surface Disease Index questionnaire (OSDI) was then completed to relate the tear lm stability to the subjective comfort of the eye. 2.1. Statistical analysis Validity was assessed by applying BlandAltman analysis to the comparison between the instruments with the average read- ing plotted agains

29、t the difference for each subject 10. Reliability was determined from the 95% condence interval of the differ- ence between the repeated Keratograph measurements. Normally distributed components were compared by t-test. Assessing variance in cylindrical components can be problem- atical 11 so the cy

30、linder and axis component were converted into a vector representation 12: Fig. 1. Difference in mean spherical equivalent (MSE) between the Oculus Ker- atograph and Nidek ARKT Tonoref II (black symbols) and repeated Keratograph measures (grey symbols) compared to the mean. n = 100 eyes. a spherical

31、lens of power mean spherical equivalent (MSE = sphere + (cylinder/2) Jackson cross-cylinder power at axis 0 (J0 = cylinder/2cos2 axis) Jackson cross-cylinder power at axis 45 (J45 = cylinder/2sin2 axis) The Ocular Surface Disease Index (OSDI) is a 12 item, 5 category Likert scale that investigates s

32、ymptoms, triggers and consequences of dry eye. OSDI scores were converted to a 100 point scale 13 and correlated with NITBUT to assess the discrimination of the devices. 3. Results 3.1. Topography The average corneal curvature was 7.74 0.29 mm with an aver- age difference between the orthogonal meri

33、dians of 0.14 0.15 mm (ARKT measures). On average the MSE as measured by the Ker- atograph was found to be more positive than the ARKT (MSE difference: +1.83 0.44D, p 0.001; Fig. 1). However, there was no signicant difference in the astigmatic components (differences, J0 = +0.01 0.27D, p = 0.61; J45

34、 = 0.03 0.18D, p = 0.13; Fig. 2). The Keratograph topography repeated measures were similar for MSE (difference: +0.11 0.97D, p = 0.35), J0 (difference: 0.10 1.12D, p = 0.29) and J45 (difference: 0.10 0.60, p = 0.37). 3.2. NITBUT NITBUT measured with the Keratograph ranged from 0.36 s to 29.00 s, wi

35、th 63% of readings being 5 s and 85% 10 s. This com- pared to the TearScope NITBUT range of 5.0 s to 30.8 s with none 5 s and 15% 10 s. On average the NITBUT measured by the Ker- atograph was 12.35 shorter than measured with the Tearscope (SD 7.45 s, p 0.001; Fig. 3). The duration over which the Ker

36、ato- graph measured for each subject was more similar to the NITBUT of the Tearscope (1.7 3.6 s longer, correlation r = 0.88), although the difference was still signicant (p 0.001). The second Kerato- graph NITBUT was on average 1.64 s less than the rst (SD 6.03, p 0.01). OSDI correlated more strong

37、ly with NITBUT measured with the Tearscope (r = 0.32) compared to the keratograph (NIT- BUT: r = 0.19; total measurement time: r = 0.19). N. Best et al. / Contact Lens & Anterior Eye 35 (2012) 171174 173 Fig. 4. Output of a Keratograph NITBUT measurement: left live video of the Placido mires; top ri

38、ght detected distortion of Placido mires overlaid on the image demonstrating progress of tear lm break-up over time: bottom right Tear-Map displaying rst break up time in each location. Fig. 2. Difference in J0 (red symbols) and J45 (blue symbols) astigmatic compo- nents between the Oculus Keratogra

39、ph and Nidek ARKT Tonoref II (dark colours) and repeated Keratograph measures (light colours) compared to the mean. n = 100 eyes. 4. Discussion Approximately half of current contact lens wearers suffer from dryness and discomfort, particularly towards the end of the day 14. This inevitably leads to

40、dissatisfaction and possible discon- tinuation of lens wear. Prior to tting their patients with contact lenses there are a number of tests available to the practitioner to assess the quality and quantity of tears, to allow advice to be given on an individuals suitability for contact lenses and to re

41、commend the most appropriate modality and lens type. These tests include lid parallel-conjunctival folds, NITBUT, invasive break up time, corneal and conjunctival staining, lid wiper epitheliopathy, limbal hyper- aemia, tear prism height measurement, phenol red test and various questionnaires 6,15.

42、NITBUT has been shown to be the clinical test with the highest sensitivity and specicity for dry eye 6. NITBUT involves projecting a light source onto the cornea and imaging the reected Purkinje image or grid. Traditionally this reection has been observed by a practitioner for any disruption Fig. 3.

43、 Difference in NITBUT as measured with the Keratograph when compared to the Tearscope (black symbols) and on repeated measurement with the Keratograph (grey symbols) compared to the mean. n = 100. which would indicate tear lm break up. The Tearscope uses a cold cathode light to reect off the tear lm

44、, combined with a heat sink which draws the heat away from the light via the handle to reduce any drying effect. The area of the cornea covered is larger than a Purkinje image and a grid can be inserted into the Tearscope so that the tear break in the projected image can be more easily detected and

45、localised to a region of the corneal surface. In comparison, the Keratograph illumination system consists of 200 red LEDs (wavelength 653 nm), that emit little heat, reducing thermally induced alterations to the tear lm. An illuminated ring pattern is projected onto the cornea in the form of a Placi

46、do disk consisting of 22 rings. Once the patient is correctly aligned the soft- ware prompts the practitioner to ask the patient to blink twice. The second blink triggers the video recording and measurement. The measurement nishes when one of two events occurs; either sig- nicant distortion of the r

47、eected image of the Placido rings occurs; or the subject blinks. The computer displays a video of the reected mires, an image with the detected mire distortion overlaid and a corneal map with the time of the rst detected tear lm break- up detected at each location colour coded (Fig. 4). Time to rst

48、break-up and total measuring time is also recorded (to 1/100th of a second). NITBUT as measured with the Keratograph was consistently shorter than measurements recorded with the Tearscope, much more than would be expected from the subjective observer response time. This is because the Keratograph records the rst incident of break-up anywhere in the tear lm irrespective of how small or transient the area of break-up. Such small or tran- sient regions of break up would probably not be detected by an observer viewing the Tearscope mires. Alternatively the soft- ware could be detecting