Current Eye Research, Early Online, 1–6, 2014 ! Informa Healthcare USA, Inc. ISSN: 0271-3683 print / 1460-2202 online DOI: 10.3109/02713683.2014.898311
RESEARCH REPORT
Interexaminer Reproducibility of Optical Coherence Tomography for Measuring the Tear Film Meniscus Handan Canan, Rana Altan-Yaycioglu, Burak Ulas, Selcuk Sizmaz and Muge Coban-Karatas
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
Department of Ophthalmology, Baskent University Faculty of Medicine, Adana Teaching & Medical Research Center, Adana, Turkey
ABSTRACT Purpose: New optical coherence tomography (OCT) devices with anterior segment module are able to measure the tear meniscus height (TMH) and tear meniscus area (TMA). Since the borders of the area and the height are to be marked by an examiner, the measurements are prone to be subjective. Herein, we aimed to evaluate the interexaminer reproducibility of the tear meniscus measurements with OCT. Materials and Methods: In this prospective cross-sectional study, both eyes of 300 consecutive patients with and without dry eye were studied. Following routine ophthalmologic examination, tear-film OCT images were obtained. Two examiners, who were unaware of the other examiner’s results, measured the TMH (i.e. the line connecting the intersection of the meniscus with the cornea/sclera and eyelid) and cross-sectional TMA. The reliability and correlation of the two examiners’ results were assessed. Results: Four right and six left eyes were excluded from the final analysis, so a total of 590 eyes were evaluated. The mean difference of the two examiners’ measurements of both eyes was 0.001 ± 0.027 mm2 in TMA, and 21.29 ± 39.95 mm in TMH. An agreement between the two examiners was found regarding TMA and TMH measurements for right and left, as well as both eyes (Cronbach’s alpha40.900, for all). Also, the correlation between both variables was high (inter item correlation matrix40.840, for all). Conclusions: We showed a strong statistical agreement for both TMA and TMH measurements. According to our results, we believe that FD-OCT device is dependable in measuring the TMA and TMH values, given that its results are reproducible. Keywords: Dry eye syndromes, optical coherence tomography, optical imaging, tears, tear meniscus
INTRODUCTION
hand, tear-film break up time evaluates the quality of tear film with some limitations, such as the need for fluorescein instillation, problems with standardization of concentration and induction of reflex tearing.4,5 So, researchers focus on non-invasive methods to evaluate tear volume. The tear menisci are formed by the tears lying at the junctions of the bulbar conjunctiva with the margins of both the upper and lower eyelids. In evaluation of tear meniscus, methods that make the measurement in its natural state are preferable. For
Tear film is important in comfort of the eye and also plays a major role in refraction.1 There are many different methods to evaluate the status of the dry eye, mainly focusing on secretion. For more than 100 years Schirmer’s tests are being used to assess tear secretion. However, Schirmer’s test is open to bias since it may cause reflex tear secretion resulting in poor reproducibility,2 low sensitivity, and specificity; and lack of control over reflex lacrimation.3 On the other
Received 16 September 2013; revised 13 February 2014; accepted 16 February 2014; published online 21 April 2014 Correspondence: Rana Altan-Yaycioglu, M.D., F.E.B.O., Professor in Ophthalmology, Department of Ophthalmology, Baskent University Adana Teaching & Medical Research Center, Dadaloglu Mah. 39 Sok. No: 6, 01250 Yuregir, Adana, Turkey. Tel: +90-532-373-5585. Fax: +90-322327-1274. E-mail: [email protected]
1
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
2
H. Canan et al.
this purpose, several methods are suggested including micrometry, photography, videography, reflective meniscometry, Tearscope measurements, strip meniscometry, and dacryomeniscometry.6–11 Photographic digital images demonstrated that tear meniscus height (TMH), along with tear meniscus cross-sectional area (TMA), are significantly reduced in dry-eye patients compared to non dry-eye patients.12 Optical coherence tomography (OCT) is a lowcoherence interferometric technique to produce detailed sectional images of biological tissues. Recently, OCT has been used to evaluate the anterior segment and tear meniscus.13 Fourier-domain (FD) OCT allows for faster, higher resolution sampling in comparison with previous versions of this technology, and leads to results superior to time-domain (TD)-OCT.14 RTVue (Optovue) is an FD-OCT device with a scan time of 0.16 s and resolution of 5 microns, making the analysis of the tear film easier because of less difficulty in determining the tear film boundaries.15 RTVue-CAM has a telecentric geometry, thereby reducing the image distortion that results from differences in air-tissue indices.16 In a previous study, we found tear menisci measurements with FD-OCT promising as being rapid and noninvasive, and suggested it in the diagnosis and follow-up of patients with dry eye.17 In current technology, the tear film measurements with OCT require marking the borders of the tear meniscus, and this might lead to possible errors. In the present study, we aimed to evaluate the interexaminer reproducibility of the tear meniscus measurements with FD-OCT.
MATERIALS AND METHODS Patients In this prospective cross-sectional study, patients with and without dry eye were included. Patients with eyelid problems (e.g. ectropion, entropion, lagophthalmos, ptosis), conjunctival disorders (e.g. symblepharon and conjunctivochalasis), contact lens wearers, those with prosthetic eyes, or with a history of eye surgery, and those who were amaurotic, were excluded from the study. The study was designed according to the tenets of the Declaration of Helsinki. The purposes, design, and possible complications of the study were explained to all patients, and written consents obtained. Threehundred eligible patients – 111 (37%) men and 189 (63%) women – all of whom were examined by the same ophthalmologist (RAY) between February and September, 2011, were included in the study. All patients underwent routine ophthalmologic examination, including visual acuity testing, slit-lamp biomicroscopy, non-contact intraocular pressure (IOP)
measurement, and posterior segment evaluation. In addition, all patients underwent anterior segment OCT imaging. Both eyes (OU) of each patient were included in the study. All subjects were instructed not to use any eye drops in less than 2 h prior to testing.
Imaging OCT The measurements were explained in the previous study.17 OCT measurements were undertaken by a certified technician, who was masked to the clinical parameters of the patient. An FD-OCT system (RTVue-100, OptovueÕ , Fremont, CA), equipped with a low-magnification corneal adaptor module(CAM-L), was used. The CAM produced telecentric scanning for anterior segment imaging using either a wide-angle or high-magnification adaptor lens.16 Two cross-sectional images of tear meniscus were recorded in rapid succession 3 s following each blink. One of the two with a better image was selected for investigation. Two examiners examined the same selected image at different days in a randomized manner. A built-in caliper was used to measure TMH and TMA. Before the initiation of this study, both examiners were educated by an experienced examiner, on how the measurements should be taken. Following that, they agreed on the common rules for marking (Figure 1). The distance between the points at which the meniscus intersected with the cornea/sclera (superiorly) and eyelid (inferiorly) was accepted as the TMH in mm. The borders of the tear meniscus were marked with the caliper. Integrated analysis software was used to calculate the TMA in mm2. First, examiner (E#1, MMK) measured the TMH and TMA. Following, at a separate day a second examiner (E#2, BU), who was unaware of the first results, measured the TMH and TMA in a randomized order of patients.
Statistical Analysis Data were initially entered into an Excel spreadsheet, and were then transferred to SPSS software (Statistical Package for the Social Sciences, version 17.0, SPSS Inc, Chicago, IL). The measurements of the two examiners were evaluated using the reliability analysis. The results were interpreted according to the Cronbach’s alpha score. If the score was higher than 0.80, the correlation was accepted as reliable.18 Also the correlation among the examiners was given as the inter item correlation matrix. The data were also entered to MedCalc statistical software (MedCalc version 12, Ostend, Belgium) to evaluate the results in Bland– Altman plots. Current Eye Research
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
OCT-Tear-Interexaminer Reproducibility
3
FIGURE 1 The image obtained with OCT. (a) The borders of the tear meniscus were marked with the caliper to measure the TMA. Integrated analysis software calculated the area in mm2. (b) The distance between the points at which the meniscus intersected with the cornea/sclera and eyelid was accepted as the TMH (mm).
TABLE 1 The tear parameters measured with anterior segment module of OCT, including the mean ± SD values of TMA (mm2) and TMH (mm). Examiner #1
TMA-OD TMA-OS TMA-OU TMH-OD TMH-OS TMH-OU
Examiner #2
Differences in measurements
Mean ± SD
Range
Mean ± SD
Range
0.043 ± 0.049 0.046 ± 0.055 0.044 ± 0.052 287.9 ± 140.9 294.9 ± 151.6 291.3 ± 146.2
0.001–0.334 0.001–0.406 0.001–0.406 58.0–902.0 47.0–1350.0 47.0–1350.0
0045 ± 0.048 0.046 ± 0.050 0.045 ± 0.049 308.1 ± 140.7 317.4 ± 151.8 312.7 ± 146.2
0.002–0.354 0.001–0.508 0.001–0.508 69.0–984.0 58.0–1450.0 58.0–1450.0
Mean ± SD
Range
Cronbach’s alpha
0.002 ± 0.027 0.000 ± 0.028 0.001 ± 0.027 20.19 ± 44.68 22.40 ± 34.54 21.29 ± 39.95
0–0.3 0–0.269 0–0.3 0–464 0–305 0–464
0.917 0.924 0.920 0.974 0.987 0.981
Inter item correlation matrix 0.846 0.861 0.854 0.949 0.974 0.962
SD, standard deviation; TMA, tear meniscus area; OD, right eye; OS, left eye; OU, both eyes; TMH, tear meniscus height. The differences between the measurements of the two examiners were also given. The results were evaluated using the reliability analysis, and were interpreted according to the Cronbach’s alpha score. A score, higher than 0.80, was accepted as reliable.
RESULTS There were 111 male and 189 female subjects. The mean age was 53.1 ± 18.3 years (ranged from 15 to 92 years). Four right eyes (OD) and six left eyes (OS) had conjunctivochalasis and were excluded from the final analysis. Thus, a total of 590 eyes were evaluated. The mean ± SD values of TMA and TMH measurements, as well as the differences between the two examiners on each image are shown on Table 1. Regarding the TMA measurements, the statistical agreement between examiners was high for OD and OS, as well as both eyes (OU) (Cronbach’s alpha = 0.917, 0.924, and 0.920, respectively). Also, the correlation between the variables was high (inter item correlation matrix = 0.846, 0.861, and 0.854, respectively). Additionally, Bland–Altman plots showed that the measurements were within the limits of agreement (Figure 2). Also, for the TMH measurements, the Bland– Altman plots showed an agreement between the examiners. Reliability analysis revealed a high statistical agreement between examiners !
2014 Informa Healthcare USA, Inc.
(Cronbach’s alpha = 0.974, 0.987, and 0.981, respectively for OD, OS, and OU). The inter-item correlation matrix showed significant correlation between the examiners (0.948, 0.974, and 0.962 for OD, OS, and OU, respectively).
DISCUSSION In dry eye, there is no universally accepted method to measure tear-volume objectively. Technical advancements in anterior segment OCT enabled better measurement of tear meniscus, and thus OCT has been suggested as a reliable alternative in the evaluation of tear meniscus.19–24 In our previous study, we have found FD-OCT as a valuable tool in the evaluation of dry eye by measuring the TMH and TMA.17 In the present study, we investigated the interexaminer reproducibility of the measurements in 300 patients, and showed a significant correlation between the examiners in both TMH and TMA measurements. The tear meniscus has a concave air–tear interface resulting from the hydrophobic surfaces of the cornea
H. Canan et al.
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
4
FIGURE 2 The Bland–Altman plots, showing that the results of two examiners’ measurements were within the limits of agreement. (a, b, c) Results of TMA measurements in all patients, right eyes, and left eyes, respectively. (d, e, f) Results of TMH measurements in all patients, right eyes, and left eyes, respectively. [TMA: Tear meniscus area; TMH: Tear meniscus height; od: Right eye, os: Left eye; _1: Examiner #1; _2: Examiner #2].
and lower eyelid. Tear meniscus measurements such as TMH, TMA and depth have been suggested to be useful in the diagnosis of dry eye.16,20,23,25 Among the three, TMA was proposed as the best indicator of aqueous tear deficiency. Qiu et al. related that to its ability to eliminate some mixed factors such as removing the interference factor of individual
palpebral aperture.26 Several methods have been suggested to measure TMA with OCT. For instance, Zhou et al. divided the area into two triangles,16 Wang et al. fitted the interface with a circular arc.13 In FD-OCT, the borders of the tear meniscus must first be marked, after which the area is computed by the software. The other tear parameter, TMH is measured Current Eye Research
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
OCT-Tear-Interexaminer Reproducibility using a straight line connecting the points at which the meniscus makes contact with cornea and lower eyelid, as suggested by Zhou and Tittler and performed in the present study.16,27. This measurement is different from Ibrahim et al., in that they formed a perpendicular line between the touch points of eyelid, cornea, and tear meniscus element.23 Marking of the borders of the tear meniscus crosssectional area possibly will give more accurate measurements, yet might be dependent on the examiner. Several reports showed the repeatability of tear meniscus measurements with OCT. In three major types of dry eye, namely aqueous tear deficiency, lipid tear deficiency and Sjo¨gren’s syndrome, FD-OCT has shown precise measurement of tear meniscus with favorable repeatability.26 Ibrahim et al. also observed no significant difference between the two measurements taken 2 weeks apart.23 Similarly, Zhou et al. found between visits repeatability with RTVue FD OCT, and this was higher compared to previously used OCT instruments.16 Using swept-source OCT, Fukuda et al. found higher than 95% intergrader intraclass correlations for all measurements.24 On the other hand, the reproducibility was investigated in only few reports.23,27 Tittler et al. examined the reproducibility of OCT tear meniscus measurements performed by two graders in 16 right eyes of 16 patients, and showed that RTVue OCT was able to produce precise measurements of the lower eyelid tear meniscus with high intergrader reproducibility.27 Ibrahim et al. also have found that even by different graders the system was able to produce precise measurements of the tear film.23 These results are in agreement with ours. Compared to previous reports, we included a greater number of patients. We examined the images of tear menisci taken by FDOCT in 590 eyes by two examiners, and have shown interexaminer reproducibility for both TMA and TMH values. Even, when we evaluated right or left eyes separately, the results were similar, and showed the reproducibility of these measurements. According to the Bland–Altman plots, the TMA measurements were in better agreement compared to TMH measurements, and might be the preferred parameter in tear evaluation with OCT. Nevertheless, studies on this topic are ongoing, and there is still a need for larger patient numbers to determine the cut-off values for normal and dry eye. Thus, we believe that both parameters ought to be used in tear meniscus evaluation. As suggested by Tittler et al., a program recognizing the borders and area within the meniscus would provide the best objectivity and accuracy in precision of quantifying tear volume.27 We agree with the authors; though, in current technology, FD-OCT seems like a promising tool in evaluating the tear meniscus. In conclusion, we found the tear meniscus measurements obtained with FD-OCT to be reproducible. !
2014 Informa Healthcare USA, Inc.
5
There was a significant agreement for both TMH and TMA values when two examiners performed the measurements. Though, the measurements are dependent on the examiners’ marking, both TMA and TMH values are reliable and reproducible. Further development of anterior segment OCT with computerized measuring of tear meniscus is necessary to exclude the subjective factors.
ACKNOWLEDGEMENTS We would like to thank M. Mustafa Koc for capturing the anterior segment OCT images (as examiner #1). We also would like to thank Cagla Sariturk for statistical evaluation of the results.
DECLARATION OF INTEREST The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. This study did not receive any financial support.
REFERENCES 1. Dry Eye Workshop. The definition and classification of dry eye disease: report of the definition and classification subcommittee of the International Dry Eye Workshop. Ocul Surf 2007;5:75–92. 2. Lambert DW, Foster CS, Perry HD. Schirmer test after topical anesthesia and tear meniscus heights in normal eyes. Arch Ophthalmol 1979;97:1082–1085. 3. Cho P, Yap M. Schirmer test. I. A review. Optom Vis Sci 1993;70:152–156. 4. Pflugfelder SC, Tseng SC, Sanabria O, Kell H, Garcia CG, Felix C, et al. Evaluation of subjective assessment and objective diagnostic tests for diagnosing tear-film disorders known to cause ocular irritation. Cornea 1998;17: 38–56. 5. Kojima T, Ishida R, Dogru M, Goto E, Takano Y, Matsumoto Y, et al. A new noninvasive tear stability analysis system fort the assessment of dry eyes. Invest Ophthalmol Vis Sci 2004;45:1369–1374. 6. Santodomingo-Rubido J, Wolffsohn JS, Gilmartin B. Comparison between graticule and image capture assessment of lower tear film meniscus height. Contact Lens Ant Eye 2006;29:169–173. 7. Doughty MJ, Laiquazzaman M, Button E. Video-assessment of tear meniscus height in elderly Caucasians and its relationship to the exposed ocular surface. Curr Eye Res 2001;22:420–426. 8. Yokoi N, Bron A, Tiffany JM, Kinoshita S. Reflective meniscometry: a new field of dry eye assessment. Cornea 2000;19:537–543. 9. Uchida A, Uchino M, Goto E, Hosaka E, Kasuya Y, Fukagawa K, et al. Noninvasive interference tear meniscometry in dry eye patients with Sjo¨gren syndrome. Am J Ophthalmol 2007;144:232–237. 10. Dogru M, Ishida K, Matsumoto Y, Goto E, Ishioka M, Kojima T, et al. Strip meniscometry: a new and simple method for tear meniscus evaluation. Invest Ophthalmol Vis Sci 2006;47:1895–1901.
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
6
H. Canan et al.
11. Ho A, Papas EB, Keshavarzi K, Lum E, Ehrmann K. Dacryomeniscometry: an early attempt. Proc SPIE, Ophthalmic Technologies X 2000;3908:23–31. 12. Mainstone JC, Bruce AS, Golding TR. Tear meniscus measurement in the diagnosis of dry eye. Curr Eye Res 1996;15:653–661. 13. Wang J, Aquavella J, Palakuru J, Chung S, Feng C. Relationships between central tear film thickness and tear menisci of the upper and lower eyelids. Invest Ophthalmol Vis Sci 2006;47:4349–4355. 14. Savini G, Carbonelli M, Barboni P. Spectral-domain optical coherence tomography for the diagnosis and follow-up of glaucoma. Curr Opin Ophthalmol 2011;22: 115–123. 15. Bitton E, Keech A, Simpson T, Jones L. Variability of the analysis of the tear meniscus height by the optical coherence tomography. Optom Vis Sci 2007;84: 903–908. 16. Zhou S, Li Y, Lu ATZ, Tang M, Yiu SC, Huang D. Reproducibility of tear meniscus measurement by Fourier-domain optical coherence tomography: a pilot study. Ophthalmic Surg Lasers Imaging 2009;40: 442–447. 17. Altan-Yaycioglu R, Sizmaz S, Canan H, Coban-Karatas M. Optical coherence tomography for measuring the tear film meniscus: correlation with schirmer test and tear-film breakup time. Curr Eye Res 2013; 38: 736–742. 18. Bland JM, Altman DG. Statistics notes: Cronbach’s alpha. Br Med J 1997;314:512. 19. Palakuru JR, Wang J, Aquavella J. Effect of blinking on tear volume after instillation of mid-viscosity artificial tears. Am J Ophthalmol 2008;146:920–924.
20. Shen M, Li J, Wang J, Ma H, Cai C, Tao A, et al. Upper and lower tear menisci in the diagnosis of dry eye. Invest Ophthalmol Vis Sci 2009;50:2722–2726. 21. Qiu X, Gong L, Sun X, Jin H. Age-related variations of human tear meniscus and diagnosis of dry eye with Fourier-domain anterior segment optical coherence tomography. Cornea 2011;30:543–549. 22. Bujak MC, Yiu S, Zhang X, Huang D. Serial measurement of tear meniscus by FD-OCT after instillation of artificial tears in patients with dry eyes. Ophthalmic Surg Lasers Imaging 2011;42:308–313. 23. Ibrahim OM, Dogru M, Takano Y, Satake Y, Wakamatsu TH, Fukugawa K, et al. Application of Visante optical coherence tomography tear meniscus height measurement in the diagnosis of dry eye disease. Ophthalmology 2010; 117:1923–1929. 24. Fukuda R, Usui T, Miyai T, Yamagami S, Amano S. Tear meniscus evaluation by anterior segment swept-source optical coherence tomography. Am J Ophthalmol 2013;155: 620–624. 25. Wang J, Palakuru JR, Aquavella JV. Correlations among upper and lower tear menisci, noninvasive tear break-up time, and the Schirmer test. Am J Ophthalmol 2008;145: 795–800. 26. Qiu X, Gong L, Lu Y, Jin H, Robitaille M. The diagnostic significance of Fourier-domain optical coherence tomography in Sjo¨gren syndrome, aqueous tear deficiency and lipid tear deficiency patients. Acta Ophthalmol 2012;90: e359–e366. 27. Tittler EH, Bujak MC, Nguyen P, Zhang X, Li Y, Yiu SC, Huang D. Between-grader repeatibility of tear meniscus measurements using Fourier-domain OCT in patients with dry eye. Ophthalmic Surg Lasers Imaging 2011;42:423–427.
Current Eye Research
RESEARCH REPORT
Interexaminer Reproducibility of Optical Coherence Tomography for Measuring the Tear Film Meniscus Handan Canan, Rana Altan-Yaycioglu, Burak Ulas, Selcuk Sizmaz and Muge Coban-Karatas
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
Department of Ophthalmology, Baskent University Faculty of Medicine, Adana Teaching & Medical Research Center, Adana, Turkey
ABSTRACT Purpose: New optical coherence tomography (OCT) devices with anterior segment module are able to measure the tear meniscus height (TMH) and tear meniscus area (TMA). Since the borders of the area and the height are to be marked by an examiner, the measurements are prone to be subjective. Herein, we aimed to evaluate the interexaminer reproducibility of the tear meniscus measurements with OCT. Materials and Methods: In this prospective cross-sectional study, both eyes of 300 consecutive patients with and without dry eye were studied. Following routine ophthalmologic examination, tear-film OCT images were obtained. Two examiners, who were unaware of the other examiner’s results, measured the TMH (i.e. the line connecting the intersection of the meniscus with the cornea/sclera and eyelid) and cross-sectional TMA. The reliability and correlation of the two examiners’ results were assessed. Results: Four right and six left eyes were excluded from the final analysis, so a total of 590 eyes were evaluated. The mean difference of the two examiners’ measurements of both eyes was 0.001 ± 0.027 mm2 in TMA, and 21.29 ± 39.95 mm in TMH. An agreement between the two examiners was found regarding TMA and TMH measurements for right and left, as well as both eyes (Cronbach’s alpha40.900, for all). Also, the correlation between both variables was high (inter item correlation matrix40.840, for all). Conclusions: We showed a strong statistical agreement for both TMA and TMH measurements. According to our results, we believe that FD-OCT device is dependable in measuring the TMA and TMH values, given that its results are reproducible. Keywords: Dry eye syndromes, optical coherence tomography, optical imaging, tears, tear meniscus
INTRODUCTION
hand, tear-film break up time evaluates the quality of tear film with some limitations, such as the need for fluorescein instillation, problems with standardization of concentration and induction of reflex tearing.4,5 So, researchers focus on non-invasive methods to evaluate tear volume. The tear menisci are formed by the tears lying at the junctions of the bulbar conjunctiva with the margins of both the upper and lower eyelids. In evaluation of tear meniscus, methods that make the measurement in its natural state are preferable. For
Tear film is important in comfort of the eye and also plays a major role in refraction.1 There are many different methods to evaluate the status of the dry eye, mainly focusing on secretion. For more than 100 years Schirmer’s tests are being used to assess tear secretion. However, Schirmer’s test is open to bias since it may cause reflex tear secretion resulting in poor reproducibility,2 low sensitivity, and specificity; and lack of control over reflex lacrimation.3 On the other
Received 16 September 2013; revised 13 February 2014; accepted 16 February 2014; published online 21 April 2014 Correspondence: Rana Altan-Yaycioglu, M.D., F.E.B.O., Professor in Ophthalmology, Department of Ophthalmology, Baskent University Adana Teaching & Medical Research Center, Dadaloglu Mah. 39 Sok. No: 6, 01250 Yuregir, Adana, Turkey. Tel: +90-532-373-5585. Fax: +90-322327-1274. E-mail: [email protected]
1
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
2
H. Canan et al.
this purpose, several methods are suggested including micrometry, photography, videography, reflective meniscometry, Tearscope measurements, strip meniscometry, and dacryomeniscometry.6–11 Photographic digital images demonstrated that tear meniscus height (TMH), along with tear meniscus cross-sectional area (TMA), are significantly reduced in dry-eye patients compared to non dry-eye patients.12 Optical coherence tomography (OCT) is a lowcoherence interferometric technique to produce detailed sectional images of biological tissues. Recently, OCT has been used to evaluate the anterior segment and tear meniscus.13 Fourier-domain (FD) OCT allows for faster, higher resolution sampling in comparison with previous versions of this technology, and leads to results superior to time-domain (TD)-OCT.14 RTVue (Optovue) is an FD-OCT device with a scan time of 0.16 s and resolution of 5 microns, making the analysis of the tear film easier because of less difficulty in determining the tear film boundaries.15 RTVue-CAM has a telecentric geometry, thereby reducing the image distortion that results from differences in air-tissue indices.16 In a previous study, we found tear menisci measurements with FD-OCT promising as being rapid and noninvasive, and suggested it in the diagnosis and follow-up of patients with dry eye.17 In current technology, the tear film measurements with OCT require marking the borders of the tear meniscus, and this might lead to possible errors. In the present study, we aimed to evaluate the interexaminer reproducibility of the tear meniscus measurements with FD-OCT.
MATERIALS AND METHODS Patients In this prospective cross-sectional study, patients with and without dry eye were included. Patients with eyelid problems (e.g. ectropion, entropion, lagophthalmos, ptosis), conjunctival disorders (e.g. symblepharon and conjunctivochalasis), contact lens wearers, those with prosthetic eyes, or with a history of eye surgery, and those who were amaurotic, were excluded from the study. The study was designed according to the tenets of the Declaration of Helsinki. The purposes, design, and possible complications of the study were explained to all patients, and written consents obtained. Threehundred eligible patients – 111 (37%) men and 189 (63%) women – all of whom were examined by the same ophthalmologist (RAY) between February and September, 2011, were included in the study. All patients underwent routine ophthalmologic examination, including visual acuity testing, slit-lamp biomicroscopy, non-contact intraocular pressure (IOP)
measurement, and posterior segment evaluation. In addition, all patients underwent anterior segment OCT imaging. Both eyes (OU) of each patient were included in the study. All subjects were instructed not to use any eye drops in less than 2 h prior to testing.
Imaging OCT The measurements were explained in the previous study.17 OCT measurements were undertaken by a certified technician, who was masked to the clinical parameters of the patient. An FD-OCT system (RTVue-100, OptovueÕ , Fremont, CA), equipped with a low-magnification corneal adaptor module(CAM-L), was used. The CAM produced telecentric scanning for anterior segment imaging using either a wide-angle or high-magnification adaptor lens.16 Two cross-sectional images of tear meniscus were recorded in rapid succession 3 s following each blink. One of the two with a better image was selected for investigation. Two examiners examined the same selected image at different days in a randomized manner. A built-in caliper was used to measure TMH and TMA. Before the initiation of this study, both examiners were educated by an experienced examiner, on how the measurements should be taken. Following that, they agreed on the common rules for marking (Figure 1). The distance between the points at which the meniscus intersected with the cornea/sclera (superiorly) and eyelid (inferiorly) was accepted as the TMH in mm. The borders of the tear meniscus were marked with the caliper. Integrated analysis software was used to calculate the TMA in mm2. First, examiner (E#1, MMK) measured the TMH and TMA. Following, at a separate day a second examiner (E#2, BU), who was unaware of the first results, measured the TMH and TMA in a randomized order of patients.
Statistical Analysis Data were initially entered into an Excel spreadsheet, and were then transferred to SPSS software (Statistical Package for the Social Sciences, version 17.0, SPSS Inc, Chicago, IL). The measurements of the two examiners were evaluated using the reliability analysis. The results were interpreted according to the Cronbach’s alpha score. If the score was higher than 0.80, the correlation was accepted as reliable.18 Also the correlation among the examiners was given as the inter item correlation matrix. The data were also entered to MedCalc statistical software (MedCalc version 12, Ostend, Belgium) to evaluate the results in Bland– Altman plots. Current Eye Research
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
OCT-Tear-Interexaminer Reproducibility
3
FIGURE 1 The image obtained with OCT. (a) The borders of the tear meniscus were marked with the caliper to measure the TMA. Integrated analysis software calculated the area in mm2. (b) The distance between the points at which the meniscus intersected with the cornea/sclera and eyelid was accepted as the TMH (mm).
TABLE 1 The tear parameters measured with anterior segment module of OCT, including the mean ± SD values of TMA (mm2) and TMH (mm). Examiner #1
TMA-OD TMA-OS TMA-OU TMH-OD TMH-OS TMH-OU
Examiner #2
Differences in measurements
Mean ± SD
Range
Mean ± SD
Range
0.043 ± 0.049 0.046 ± 0.055 0.044 ± 0.052 287.9 ± 140.9 294.9 ± 151.6 291.3 ± 146.2
0.001–0.334 0.001–0.406 0.001–0.406 58.0–902.0 47.0–1350.0 47.0–1350.0
0045 ± 0.048 0.046 ± 0.050 0.045 ± 0.049 308.1 ± 140.7 317.4 ± 151.8 312.7 ± 146.2
0.002–0.354 0.001–0.508 0.001–0.508 69.0–984.0 58.0–1450.0 58.0–1450.0
Mean ± SD
Range
Cronbach’s alpha
0.002 ± 0.027 0.000 ± 0.028 0.001 ± 0.027 20.19 ± 44.68 22.40 ± 34.54 21.29 ± 39.95
0–0.3 0–0.269 0–0.3 0–464 0–305 0–464
0.917 0.924 0.920 0.974 0.987 0.981
Inter item correlation matrix 0.846 0.861 0.854 0.949 0.974 0.962
SD, standard deviation; TMA, tear meniscus area; OD, right eye; OS, left eye; OU, both eyes; TMH, tear meniscus height. The differences between the measurements of the two examiners were also given. The results were evaluated using the reliability analysis, and were interpreted according to the Cronbach’s alpha score. A score, higher than 0.80, was accepted as reliable.
RESULTS There were 111 male and 189 female subjects. The mean age was 53.1 ± 18.3 years (ranged from 15 to 92 years). Four right eyes (OD) and six left eyes (OS) had conjunctivochalasis and were excluded from the final analysis. Thus, a total of 590 eyes were evaluated. The mean ± SD values of TMA and TMH measurements, as well as the differences between the two examiners on each image are shown on Table 1. Regarding the TMA measurements, the statistical agreement between examiners was high for OD and OS, as well as both eyes (OU) (Cronbach’s alpha = 0.917, 0.924, and 0.920, respectively). Also, the correlation between the variables was high (inter item correlation matrix = 0.846, 0.861, and 0.854, respectively). Additionally, Bland–Altman plots showed that the measurements were within the limits of agreement (Figure 2). Also, for the TMH measurements, the Bland– Altman plots showed an agreement between the examiners. Reliability analysis revealed a high statistical agreement between examiners !
2014 Informa Healthcare USA, Inc.
(Cronbach’s alpha = 0.974, 0.987, and 0.981, respectively for OD, OS, and OU). The inter-item correlation matrix showed significant correlation between the examiners (0.948, 0.974, and 0.962 for OD, OS, and OU, respectively).
DISCUSSION In dry eye, there is no universally accepted method to measure tear-volume objectively. Technical advancements in anterior segment OCT enabled better measurement of tear meniscus, and thus OCT has been suggested as a reliable alternative in the evaluation of tear meniscus.19–24 In our previous study, we have found FD-OCT as a valuable tool in the evaluation of dry eye by measuring the TMH and TMA.17 In the present study, we investigated the interexaminer reproducibility of the measurements in 300 patients, and showed a significant correlation between the examiners in both TMH and TMA measurements. The tear meniscus has a concave air–tear interface resulting from the hydrophobic surfaces of the cornea
H. Canan et al.
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
4
FIGURE 2 The Bland–Altman plots, showing that the results of two examiners’ measurements were within the limits of agreement. (a, b, c) Results of TMA measurements in all patients, right eyes, and left eyes, respectively. (d, e, f) Results of TMH measurements in all patients, right eyes, and left eyes, respectively. [TMA: Tear meniscus area; TMH: Tear meniscus height; od: Right eye, os: Left eye; _1: Examiner #1; _2: Examiner #2].
and lower eyelid. Tear meniscus measurements such as TMH, TMA and depth have been suggested to be useful in the diagnosis of dry eye.16,20,23,25 Among the three, TMA was proposed as the best indicator of aqueous tear deficiency. Qiu et al. related that to its ability to eliminate some mixed factors such as removing the interference factor of individual
palpebral aperture.26 Several methods have been suggested to measure TMA with OCT. For instance, Zhou et al. divided the area into two triangles,16 Wang et al. fitted the interface with a circular arc.13 In FD-OCT, the borders of the tear meniscus must first be marked, after which the area is computed by the software. The other tear parameter, TMH is measured Current Eye Research
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
OCT-Tear-Interexaminer Reproducibility using a straight line connecting the points at which the meniscus makes contact with cornea and lower eyelid, as suggested by Zhou and Tittler and performed in the present study.16,27. This measurement is different from Ibrahim et al., in that they formed a perpendicular line between the touch points of eyelid, cornea, and tear meniscus element.23 Marking of the borders of the tear meniscus crosssectional area possibly will give more accurate measurements, yet might be dependent on the examiner. Several reports showed the repeatability of tear meniscus measurements with OCT. In three major types of dry eye, namely aqueous tear deficiency, lipid tear deficiency and Sjo¨gren’s syndrome, FD-OCT has shown precise measurement of tear meniscus with favorable repeatability.26 Ibrahim et al. also observed no significant difference between the two measurements taken 2 weeks apart.23 Similarly, Zhou et al. found between visits repeatability with RTVue FD OCT, and this was higher compared to previously used OCT instruments.16 Using swept-source OCT, Fukuda et al. found higher than 95% intergrader intraclass correlations for all measurements.24 On the other hand, the reproducibility was investigated in only few reports.23,27 Tittler et al. examined the reproducibility of OCT tear meniscus measurements performed by two graders in 16 right eyes of 16 patients, and showed that RTVue OCT was able to produce precise measurements of the lower eyelid tear meniscus with high intergrader reproducibility.27 Ibrahim et al. also have found that even by different graders the system was able to produce precise measurements of the tear film.23 These results are in agreement with ours. Compared to previous reports, we included a greater number of patients. We examined the images of tear menisci taken by FDOCT in 590 eyes by two examiners, and have shown interexaminer reproducibility for both TMA and TMH values. Even, when we evaluated right or left eyes separately, the results were similar, and showed the reproducibility of these measurements. According to the Bland–Altman plots, the TMA measurements were in better agreement compared to TMH measurements, and might be the preferred parameter in tear evaluation with OCT. Nevertheless, studies on this topic are ongoing, and there is still a need for larger patient numbers to determine the cut-off values for normal and dry eye. Thus, we believe that both parameters ought to be used in tear meniscus evaluation. As suggested by Tittler et al., a program recognizing the borders and area within the meniscus would provide the best objectivity and accuracy in precision of quantifying tear volume.27 We agree with the authors; though, in current technology, FD-OCT seems like a promising tool in evaluating the tear meniscus. In conclusion, we found the tear meniscus measurements obtained with FD-OCT to be reproducible. !
2014 Informa Healthcare USA, Inc.
5
There was a significant agreement for both TMH and TMA values when two examiners performed the measurements. Though, the measurements are dependent on the examiners’ marking, both TMA and TMH values are reliable and reproducible. Further development of anterior segment OCT with computerized measuring of tear meniscus is necessary to exclude the subjective factors.
ACKNOWLEDGEMENTS We would like to thank M. Mustafa Koc for capturing the anterior segment OCT images (as examiner #1). We also would like to thank Cagla Sariturk for statistical evaluation of the results.
DECLARATION OF INTEREST The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. This study did not receive any financial support.
REFERENCES 1. Dry Eye Workshop. The definition and classification of dry eye disease: report of the definition and classification subcommittee of the International Dry Eye Workshop. Ocul Surf 2007;5:75–92. 2. Lambert DW, Foster CS, Perry HD. Schirmer test after topical anesthesia and tear meniscus heights in normal eyes. Arch Ophthalmol 1979;97:1082–1085. 3. Cho P, Yap M. Schirmer test. I. A review. Optom Vis Sci 1993;70:152–156. 4. Pflugfelder SC, Tseng SC, Sanabria O, Kell H, Garcia CG, Felix C, et al. Evaluation of subjective assessment and objective diagnostic tests for diagnosing tear-film disorders known to cause ocular irritation. Cornea 1998;17: 38–56. 5. Kojima T, Ishida R, Dogru M, Goto E, Takano Y, Matsumoto Y, et al. A new noninvasive tear stability analysis system fort the assessment of dry eyes. Invest Ophthalmol Vis Sci 2004;45:1369–1374. 6. Santodomingo-Rubido J, Wolffsohn JS, Gilmartin B. Comparison between graticule and image capture assessment of lower tear film meniscus height. Contact Lens Ant Eye 2006;29:169–173. 7. Doughty MJ, Laiquazzaman M, Button E. Video-assessment of tear meniscus height in elderly Caucasians and its relationship to the exposed ocular surface. Curr Eye Res 2001;22:420–426. 8. Yokoi N, Bron A, Tiffany JM, Kinoshita S. Reflective meniscometry: a new field of dry eye assessment. Cornea 2000;19:537–543. 9. Uchida A, Uchino M, Goto E, Hosaka E, Kasuya Y, Fukagawa K, et al. Noninvasive interference tear meniscometry in dry eye patients with Sjo¨gren syndrome. Am J Ophthalmol 2007;144:232–237. 10. Dogru M, Ishida K, Matsumoto Y, Goto E, Ishioka M, Kojima T, et al. Strip meniscometry: a new and simple method for tear meniscus evaluation. Invest Ophthalmol Vis Sci 2006;47:1895–1901.
Curr Eye Res Downloaded from informahealthcare.com by Michigan University on 10/16/14 For personal use only.
6
H. Canan et al.
11. Ho A, Papas EB, Keshavarzi K, Lum E, Ehrmann K. Dacryomeniscometry: an early attempt. Proc SPIE, Ophthalmic Technologies X 2000;3908:23–31. 12. Mainstone JC, Bruce AS, Golding TR. Tear meniscus measurement in the diagnosis of dry eye. Curr Eye Res 1996;15:653–661. 13. Wang J, Aquavella J, Palakuru J, Chung S, Feng C. Relationships between central tear film thickness and tear menisci of the upper and lower eyelids. Invest Ophthalmol Vis Sci 2006;47:4349–4355. 14. Savini G, Carbonelli M, Barboni P. Spectral-domain optical coherence tomography for the diagnosis and follow-up of glaucoma. Curr Opin Ophthalmol 2011;22: 115–123. 15. Bitton E, Keech A, Simpson T, Jones L. Variability of the analysis of the tear meniscus height by the optical coherence tomography. Optom Vis Sci 2007;84: 903–908. 16. Zhou S, Li Y, Lu ATZ, Tang M, Yiu SC, Huang D. Reproducibility of tear meniscus measurement by Fourier-domain optical coherence tomography: a pilot study. Ophthalmic Surg Lasers Imaging 2009;40: 442–447. 17. Altan-Yaycioglu R, Sizmaz S, Canan H, Coban-Karatas M. Optical coherence tomography for measuring the tear film meniscus: correlation with schirmer test and tear-film breakup time. Curr Eye Res 2013; 38: 736–742. 18. Bland JM, Altman DG. Statistics notes: Cronbach’s alpha. Br Med J 1997;314:512. 19. Palakuru JR, Wang J, Aquavella J. Effect of blinking on tear volume after instillation of mid-viscosity artificial tears. Am J Ophthalmol 2008;146:920–924.
20. Shen M, Li J, Wang J, Ma H, Cai C, Tao A, et al. Upper and lower tear menisci in the diagnosis of dry eye. Invest Ophthalmol Vis Sci 2009;50:2722–2726. 21. Qiu X, Gong L, Sun X, Jin H. Age-related variations of human tear meniscus and diagnosis of dry eye with Fourier-domain anterior segment optical coherence tomography. Cornea 2011;30:543–549. 22. Bujak MC, Yiu S, Zhang X, Huang D. Serial measurement of tear meniscus by FD-OCT after instillation of artificial tears in patients with dry eyes. Ophthalmic Surg Lasers Imaging 2011;42:308–313. 23. Ibrahim OM, Dogru M, Takano Y, Satake Y, Wakamatsu TH, Fukugawa K, et al. Application of Visante optical coherence tomography tear meniscus height measurement in the diagnosis of dry eye disease. Ophthalmology 2010; 117:1923–1929. 24. Fukuda R, Usui T, Miyai T, Yamagami S, Amano S. Tear meniscus evaluation by anterior segment swept-source optical coherence tomography. Am J Ophthalmol 2013;155: 620–624. 25. Wang J, Palakuru JR, Aquavella JV. Correlations among upper and lower tear menisci, noninvasive tear break-up time, and the Schirmer test. Am J Ophthalmol 2008;145: 795–800. 26. Qiu X, Gong L, Lu Y, Jin H, Robitaille M. The diagnostic significance of Fourier-domain optical coherence tomography in Sjo¨gren syndrome, aqueous tear deficiency and lipid tear deficiency patients. Acta Ophthalmol 2012;90: e359–e366. 27. Tittler EH, Bujak MC, Nguyen P, Zhang X, Li Y, Yiu SC, Huang D. Between-grader repeatibility of tear meniscus measurements using Fourier-domain OCT in patients with dry eye. Ophthalmic Surg Lasers Imaging 2011;42:423–427.
Current Eye Research
