Introduction
The tear film is vital for ocular health and provides many functions for the eyes, including maintenance of a smooth optical surface, lubrication of the cornea and conjunctiva, defense against pathogenic organisms, and removal of debris from the ocular surface. The tear film is comprised of the aqueous, lipid, and mucin components.1,2 The lipid and mucin components provide tear film stability by preventing evaporation and promoting tear film adherence to the cornea.1,3
The tear film breakup time (TFBUT) test is a noninvasive assessment of precorneal tear film stability that is commonly performed when conjunctival hyperemia, chemosis, ocular discharge, or keratitis are noted.4 In general, this test is performed by a veterinary ophthalmologist and requires fluorescein stain, a cobalt filter, and slit-lamp biomicroscopy.2 The TFBUT is the most commonly used test for clinical evaluation of tear film stability in veterinary medicine.5–10 Results of the TFBUT test are highly variable within and among species.5,6,11–15 Expected TFBUTs in healthy dogs are approximately 20 seconds, with a range of approximately 15 to 40 seconds reported in the literature.5,8,11,16–19 The variations noted in TFBUT readings are postulated to be attributable to the amount of fluorescein instilled, interobserver variation (subjective interpretation), and environmental factors.14 Because of these variations, other tear film tests such as automated tear film surface–quality breakup times, meibometry, and tear osmolarity readings20,21 are often used to confirm qualitative tear abnormalities in human patients; however, these tests are not commonly performed in veterinary practice.11,14,22,23
The TFBUT has been used as the primary assessment of tear quality for dogs and cats in several investigations. Descriptions of the TFBUT technique vary slightly among sources,2,7,8,11,12 and to our knowledge, a detailed, standardized test has not been outlined for dogs. For example, fluorescein has been administered by use of standard fluorescein-impregnated strips and eyewash,5,9 modified fluorescein-impregnated strips,14 and 2% fluorescein dye solution15 in various species. This variation in methods could affect test results and interpretation.
Additionally, topical medications, such as anesthetic solutions or fluorescein stain, have been shown to decrease TFBUT in people and horses, suggesting a destabilizing effect on the tear film.15,24,25 Fluorescein stain and topical anesthetics are frequently used as diagnostic tools in veterinary ophthalmic practice. In the US, proparacaine has been reported to be the preferred topical anesthetic agent for use in veterinary patients15,26 because it is readily available and has few documented adverse effects when used for diagnostic purposes.27 To the authors’ knowledge, the effects of topical anesthetics and fluorescein stain on tear film stability in dogs have not been reported.
The objective of the study reported here was to determine the intra- and interobserver reliability of the TFBUT test as it would be performed in dogs in clinical practice. A secondary objective was to assess reliability of the TFBUT test after application of proparacaine hydrochloride ophthalmic solution. We hypothesized that the described TFBUT test would have moderate reliability within and between observers when performed according to 1 described protocol. We also hypothesized that proparacaine application and repeated TFBUT testing would decrease TFBUT measurements for dogs.
Materials and Methods
Dogs
A convenience sample of 21 dogs that belonged to staff members of the Mississippi State University College of Veterinary Medicine and had no clinical history of ophthalmic disease was recruited for study participation. All dogs that met these criteria were eligible for the study, and detection of ocular abnormalities after enrollment was not a cause for exclusion. The study protocol was approved by the Mississippi State University Institutional Animal Care and Use Committee and was performed in accordance with the Association for Research in Vision and Ophthalmology’s statement on the use of animals in ophthalmic and vision research. Each staff member provided written informed consent for their dog’s participation.
Age, sex, breed, and reproductive status were recorded. Current medical history, including administration of systemic or ocular medications, was recorded. One to 3 days prior to the study, all dogs underwent complete physical and ophthalmic examinations. Ophthalmic examination techniques included Schirmer tear test I (standardized sterile strips; Merck Animal Health) fluorescein staining (Fluoro-I Strip AT; Ayerst Laboratories), rebound tonometry (Tono-Vet; iCare Finland), slit-lamp biomicroscopy (Kowa SL-17; Kowa Co Ltd), and indirect ophthalmoscopy (Vantage Plus Digital LED; Keeler Ophthalmic Instruments Inc) without pupil dilation. All results were recorded. All dogs were discharged from the hospital after the examinations were completed.
Study design
The dogs were randomly assigned to undergo TFBUT testing on 2 days (n = 10 [day 1] and 11 [day 2]) with a cup draw method. Two clinicians participated in the study; observer 1 was a veterinary ophthalmology specialty intern, and observer 2 was a board-certified veterinary ophthalmologist. Prior to testing on days 1 and 2, each observer read a detailed description of the TFBUT test technique to be used in the study.
On day 1, the first group of dogs was brought to the study location and each was randomly assigned a unique number (1 to 10) with the cup draw method. On day 2, the remaining dogs were brought to the same location and randomly assigned numbers (1 to 11) in the same manner. Two examination rooms were used for the TFBUT tests. The dogs were housed for the day in a kennel facility in the same building as the examination rooms.
Each dog underwent 4 fluorescein stain–based TFBUT tests for each eye (Appendix 1). On the basis of the results of a coin flip, the right eye of each dog was examined first throughout the study. Two TFBUT tests (1 by each observer) were performed without proparacaine administration in testing periods 1 and 2, and 2 TFBUT tests were performed after administration of 0.5% proparacaine hydrochloride ophthalmic solution (Bausch and Lomb Inc) in testing periods 3 and 4, with a 1-hour interval between testing periods. At least 1 hour, not exceeding 1 hour and 20 minutes, lapsed between examinations for all dogs to allow sufficient time for tear turnover,28 fluorescein stain clearance,27 and proparacaine degradation,29 when applicable. On day 1, observer 1 performed TFBUT tests for dogs 1 through 5 in numerical order, while observer 2 performed TFBUT tests for the remaining dogs in numerical order, during testing period 1. Each subset of dogs was then tested in the same order by the other observer during testing period 2. The testing order used for period 1 was followed after proparacaine administration in period 3, and the testing order used for period 2 was followed after proparacaine administration in period 4. After the fourth testing period, residual fluorescein stain was rinsed from the eyes with eyewash solution (Advanced Eye Relief; Bausch and Lomb Inc), and an artificial tears solution with 0.3% viscoadaptive hyaluronan (I-Drop Vet Gel; I-MED Animal Health) was instilled in each eye. The same protocol was followed for testing periods on day 2.
TFBUT tests
For the TFBUT test, a 0.3-mL aliquot of eyewash was applied to 1 standard fluorescein-impregnated strip. Excess eyewash was not shaken off the strip. The strip was lightly touched to the dorsal bulbar conjunctiva of the right eye and then the left eye. The dog was allowed to blink, the lights were turned off, and the eyelids of the right eye were manually closed and then held open, leaving the eyelid margin in contact with the globe. The slit-lamp biomicroscope was set on the cobalt blue filter with the diffuse beam at a midlevel light intensity to observe the dorsolateral quadrant. To determine the TFBUT, each observer verbally instructed an assistant to start a timer at the opening of the eyelids and to stop the timer once dark spots or lines were noted in the fluorescein stain.2 The assistant recorded the TFBUT test in seconds for the right eye. The timer was reset, and the left eye underwent testing in the same manner. TFBUTs were recorded 3 times/eye, and the mean value for each eye was calculated. The observers were masked to the elapsed times for each TFBUT test throughout the study. When proparacaine was administered for testing periods 3 and 4, a drop was placed on the right eye and then the left eye, 1 minute elapsed, and the TFBUT test was performed as described.
Statistical analysis
To assess the inter- and intraobserver test reliability of the TFBUT test, intraclass correlation coefficients (ICCs) were calculated from intercept-only linear mixed models with commercially available software (MIXED procedure version 9.4; SAS Institute Inc).30 The mean TFBUT was used as the dependent variable and sample identity (eye within dog and proparacaine status), and observer identity were included as random effects to assess interobserver reliability (agreement) in separate models for TFBUTs without topical anesthetic and after proparacaine administration. The TFBUT, including all 3 replicates taken at each measurement, was used as the dependent variable and sample identity was included as a random effect in separate models for each observer to assess intraobserver reliability. On the basis of previously established guidelines, ICC values < 0.5 were equivalent to poor reliability, values between 0.5 and 0.75 were equivalent to moderate reliability, values between 0.75 and 0.9 were equivalent to good reliability, and values > 0.90 were equivalent to excellent reliability.31 Linear mixed models were used to assess associations between testing period, observer, eye, and ophthalmic disorder (present vs absent) and the mean TFBUT. The initial model included the 4 main effects and their pairwise interactions as fixed effects. Dog identity was included as a random effect. The distribution of the conditional residuals was assessed for all models to ensure the assumptions of normality and homoskedasticity for the statistical models had been met. If an interaction term was not significant, it was removed and the model was rerun. For each fixed effect that was significant in the model, pairwise comparisons of least-squares means for each level of the fixed effect were made. Statistical significance of differences in the least-squares means was determined after adjusting for multiple comparisons by use of a Tukey-Kramer post hoc test. Values of P < 0.05 were considered significant.
Results
The study sample primarily comprised mixed-breed dogs (n = 9); there were 3 Dachshunds, 2 French Bulldogs, 2 Labrador Retrievers, and 1 each of the following breeds: Beagle, Whippet, Golden Retriever, Rottweiler, and Blue Heeler. Ages of the 21 dogs ranged from 5 months to 17 years, with a mean ± SD of 5.5 ± 4.12 years. All dogs were systemically healthy, and none were receiving medications other than routine flea, tick, and heartworm preventatives.
On ophthalmic examination at study enrollment, 2 dogs had mild bilateral mucoid ocular discharge. One of the 2 dogs had a low Schirmer tear test I result for the left eye (14 mm/min; expected values,2 18.64 ± 4.47 mm/min to 23.90 ± 5.12 mm/min), with a measurement of 17 mm/min for the right eye. The other dog in which ocular discharge was noted had Schirmer tear test values of 18 and 21 mm/min in the left and right eyes, respectively. Minor ocular abnormalities detected in 5 other dogs included distichiasis (n = 3), an eyelid mass (1), and medial entropion (1). No blepharospasm or excessive lacrimation was noted in these dogs.
The overall adjusted mean ± SD TFBUTs recorded by both observers over both days were as follows: 7.69 ± 0.54 seconds for period 1, 7.75 ± 0.54 seconds for period 2, 8.12 ± 0.54 seconds for period 3, and 6.25 ± 0.54 seconds for period 4. The mean ± SD TFBUTs recorded by each observer during each testing period over the 2 study days were summarized (Table 1). On statistical analysis, none of the interaction terms were found to be significant (P ≥ 0.091); consequently, the final mixed model included the 4 main effects (testing period, observer, eye, and ophthalmic disorder). No significant difference in TFBUTs was found between the right and left eyes (P = 0.300) or between the presence or absence of an ophthalmic disorder (P = 0.467). Observer (P < 0.001) and order of testing (P = 0.001) were significantly associated with TFBUTs. Overall, the mean TFBUTs assessed by observers 1 and 2 were 5.9 and 8.6 seconds, respectively, when adjusted for other effects in the model. TFBUTs in periods 1 (P = 0.019), 2 (P = 0.008), and 3 (P = 0.001) were significantly different from those in period 4 for both study days combined (Table 2). Specifically, TFBUTs were significantly lower in period 4 than those in all other periods, with an overall decrease of 1.4 seconds from the period 1 results when adjusted for the other effects in the model. No significant difference was found between TFBUTs for periods 1 versus 2, 1 versus 3, or 2 versus 3.
Mean ± SD tear film breakup time (TFBUT) in seconds measured for 21 dogs by 2 observers (a veterinary ophthalmology specialty intern [observer 1] and a board-certified veterinary ophthalmologist [observer 2]) during 4 testing periods over 2 days in a prospective, randomized study to determine intra- and interobserver reliability of the TFBUT test as it would be performed in a clinical environment with and without administration of a topical anesthetic.
| Observer | Testing period | |||
|---|---|---|---|---|
| 1 | 2 | 3 | 4 | |
| 1 | 6.38 ± 3.30 | 6.59 ± 3.60 | 6.63 ± 3.52 | 4.83 ± 1.77 |
| 2 | 9.00 ± 3.69 | 8.91 ± 4.16 | 9.51 ± 4.46 | 7.70 ± 3.20 |
On each study day, a randomly selected group of dogs (10 on day 1 and 11 on day 2) underwent fluorescein-based TFBUT testing according to a predetermined protocol. Each dog was evaluated twice by both observers; dogs tested in periods 1 and 3 by observer 1 were tested in periods 2 and 4 by observer 2 and vice versa. Results for each test were read in triplicate. No topical anesthetic was used in periods 1 and 2; a single drop of 0.5% proparacaine solution was instilled in each eye in periods 3 and 4, and 1 min elapsed between topical administration of the anesthetic and TFBUT testing. There was a 1-h interval between testing periods.
See Appendix for details on testing periods.
Differences of least-squares means used to determine whether there were statistical differences in TFBUTs (pairwise comparisons) between testing periods as determined by 2 observers for the 21 dogs in Table 1, after adjustment for other effects in the model and Tukey-Kramer post hoc analysis.
| Test period comparison | TFBUT difference | Adjusted P value (after post hoc analysis) |
|---|---|---|
| 1 vs 2 | –0.1284 | 0.999 |
| 1 vs 3 | –0.5050 | 0.696 |
| 1 vs 4 | 1.3663 | 0.019 |
| 2 vs 3 | –0.3766 | 0.848 |
| 2 vs 4 | 1.4946 | 0.008 |
| 3 vs 4 | 1.8713 | 0.001 |
Values of P < 0.05 were considered significant.
Evaluation of intraobserver reliability for all testing periods revealed that TFBUT tests performed by observer 1 had poor reliability (ICC, 0.47) and those performed by observer 2 had moderate reliability (ICC, 0.53). Interobserver reliability was poor (ICC, 0.30) when dogs did not receive a topical anesthetic and decreased further when proparacaine was instilled in the eyes before testing (ICC, 0.19). The TFBUTs measured by observer 1 were a mean of 2.7 seconds shorter than those measured by observer 2 (P < 0.001) when adjusted for the other effects in the model.
Discussion
The present study revealed that results of the fluorescein stain–based TFBUT test in dogs had poor reliability when performed by one observer and moderate reliability when performed by the other. Intraobserver reliability of the TFBUT test was poor with and without the use of a topical anesthetic. Administration of 0.5% proparacaine ophthalmic solution and repeated testing decreased TFBUTs and decreased reliability of the TFBUT test between observers. Other factors that may have affected test reliability in this study included the experience levels of the observers and the manner in which the test was performed (manual restraint of dogs without sedation and the fluorescein instillation technique).
The TFBUT test is inherently subjective. Both observers read a description of a common method for performing the fluorescein stain–based TFBUT test in practice and were meticulous in performing the test as described. However, the values obtained in the TFBUT test are dependent on the observer’s interpretation of tear film evaporation from the ocular surface, which can vary owing to factors that are not objectively measurable. Experience levels may contribute to reliability in TFBUT readings. With practice, clinicians may become more proficient at performing a TFBUT test with slit-lamp biomicroscopy.17 There were consistencies in the TFBUTs identified in the present study despite poor statistical agreement between observers. Observer 2 was a more-experienced clinician than observer 1 thus, the differences in TFBUT test reliability may have reflected experience level. Observer 1 consistently recorded shorter TFBUT measurements than observer 2, with a mean difference of approximately 3 seconds throughout all 4 testing periods. This fairly uniform difference could have accounted for the overall low interobserver reliability. In clinical practice, if these consistencies are known between observers, poor interobserver reliability may not be clinically important when interpreting TFBUT test results.
The goal of the study reported here was to determine inter- and intraobserver reliability of the TFBUT test as it is currently performed in clinical practice, and this may also have contributed to poor interobserver reliability. The dogs were not sedated, and the eyelids were manually held open for visibility of the dorsolateral quadrant. Previous investigations have found substantial variation in TFBUT measurements in healthy unsedated5,11 and sedated dogs.11 In people, the eyelids are not manipulated during fluorescein stain–based TFBUT testing, as manual pressure can cause some degree of expression of the meibomian glands.32 However, a 2019 study33 found no significant difference in TFBUT measurements for client-owned dogs when blinking was controlled manually by the investigators, compared with the measurements when dogs blinked spontaneously. In addition, elevation of the nictitating membrane could not be prevented in the present study, and this may have affected TFBUT interpretation by inhibiting visual access to the dorsolateral quadrant or disrupting the tear film as the membrane swept across the cornea.33 Although these factors could have contributed to the overall low reliability of TFBUT test results for dogs of the present study, these are factors that are present during routine performance of the test in clinical practice. Even though each TFBUT was measured 3 times/test and the values were mathematically averaged, these factors likely influenced TFBUT interpretation.
It has been reported that the volume of fluorescein instilled into the eye is a major source of variability in TFBUT testing, suggesting that a larger volume of fluorescein leads to a longer TFBUT reading.34–38 In the study reported here, a standard fluorescein-impregnated strip was used with a controlled amount of eyewash (0.3 mL) applied to each strip 1 time, without rewetting before application of the stain to the contralateral eye. The same fluorescein strip was used for both eyes and was always applied to the right eye first. There was no significant difference in TFBUTs between the right and left eyes, but the amounts of fluorescein instilled could have varied with this method. The described method was comparable to other methods used for measurement of TFBUTs. In people, the TFBUT test has been performed by the use of 1% or 2% fluorescein solution instilled into the tear film with a micropipette, standard fluorescein-impregnated strips wet with eyewash solution, or dry-eye test modified fluorescein-impregnated strips (designed to decrease ocular irritation and variability of standard fluorescein-impregnated strips) wet with eyewash solution.37,39–41 No clinically relevant difference in human TFBUTs was found between the use of a micropipette and a standard fluorescein strip for application of the stain.37,39 Additionally, in normal human patients (ie, those without dry-eye conditions), TFBUT results obtained by use of a modified fluorescein-impregnated strip did not differ from those obtained with a standard fluorescein strip or micropipette in a clinical setting.39 When the modified fluorescein-impregnated strip was used for TFBUT testing in healthy cats, the TFBUT test had poor test-retest reliability with an ICC of 0.20.14 One drop of eyewash was used for each strip, suggesting that the strip was wet with 50 μL of eyewash each time.2 It was not stated whether the same fluorescein strip was used for both eyes.14 The use of a new standard fluorescein strip and 0.3 mL eyewash solution for each eye may improve reliability in future TFBUT studies and in clinical practice.
The testing periods were significantly associated with TFBUT values in the study reported here. TFBUTs in period 4 were significantly shorter than those obtained in testing periods 1, 2, and 3 for both observers, which indicated a destabilization of the tear film. This destabilization may have been due to the topical application of 0.5% proparacaine solution in periods 3 and 4. Topical anesthetics have been shown to destabilize the tear film in other species, including horses and people.15,24 It should be noted that by the time of TFBUT measurement in period 4, dogs had also received 4 applications of fluorescein stain. Fluorescein stain has been shown to cause irritation, reflex tearing,34 and shorter TFBUTs in people.35,42 These results in the present study suggested that the cumulative effect of 4 applications of fluorescein and 2 applications of proparacaine destabilized the tear film. We suggest that the TFBUT of a quickly evaporating tear film was more difficult to measure consistently, and that this decreased the reliability of the test.
A washout period of 1 hour was provided between testing periods in the present study on the basis of studies23,27,43 that investigated fluorescein clearance from the tear film and conjunctival sac in dogs. Gelatt et al23 found a minimal fluorescein stain retention of 0.14% at 45 minutes after the topical application of 6 μL of fluorescein stain. Oriá et al27 found that fluorescein stain retention was < 5% 1 hour after topical application of one drop of the solution. A 2008 review36 of human medical literature indicated that a fluorescein clearance time of 20 minutes is deemed normal. Regarding the tear turnover rate, it has been established that 10 minutes should allow for complete tear turnover in dogs.44 One hour may have been long enough for fluorescein clearance in the present study, but fluorescein application may have caused mild irritation and reflex tearing and thus altered the basal tear film quality.34 Future studies may need to include a longer washout period between TFBUTs to allow for the tear film to become reestablished if multiple applications of fluorescein are used.
An important limitation of the study reported here was the lack of exclusion criteria other than a known history of ophthalmologic disease. This approach was elected because the aim was to determine the reliability of the TFBUT test in a diverse sample of dogs, similar to testing performed in dogs in clinical practice. The ages, breeds, and sexes of dogs varied, and these factors did not have a significant effect on TFBUT reliability; however, these variables may have contributed to the generally low mean TFBUTs of these dogs.12,16,44 Overall, the TFBUTs found in the present study were lower than those previously published for healthy dogs (≥ 20 seconds).5,8,11,16–19 Three of 21 dogs in the present study were brachycephalic. Previous publications have suggested TFBUTs of brachycephalic dogs can be decreased to approximately 10 seconds.12,16 In the present study, eyelid and eyelash disorders were identified in 5 of 21 (24%) dogs, and these factors may have decreased the mean TFBUT measurements. In a recent study,16 TFBUTs were documented to range from 10 to > 20 seconds in dogs with distichiasis, suggesting the presence of destabilized tear film. Eyelash and eyelid disorders have also been linked to exposure-related dysfunctional tear syndrome in human patients, leading to excessive drying of the cornea.45 In the authors’ opinion, dogs in the present study may have had an overall lower mean TFBUT, compared with previously reported values for healthy dogs, owing to the breeds represented or presence of ocular disorders known to destabilize the tear film. Presence of ophthalmic disorders was not significantly associated with inter- and intraobserver reliability of TFBUT tests in the present study. However, the TFBUT may be more difficult to interpret with a destabilized tear film, and this could have contributed to the low to moderate reliability between and within observers.
Other limitations of the study included a low number of observers. Inclusion of additional observers might increase interobserver reliability of the test by mitigating subjective observer bias.
Our findings indicated that the use of topical anesthesia and multiple applications of fluorescein stain decreases TFBUT in dogs, suggesting a negative effect on tear film quality and stability. The results also suggested that reliability of the TFBUT test within and between observers may decrease as tear film stability decreases. Clinical TFBUT testing as performed in this study had poor interobserver reliability and had moderate intraobserver reliability for the more experienced of the 2 observers. However, the mean difference between the observers was 2 to 3 seconds and thus not likely to change clinical interpretation of the TFBUT test. Small changes to method of fluorescein instillation and testing protocol may improve reliability of the TFBUT test. The authors suggest that, similar to reports of clinical TFBUT testing in people, veterinary ophthalmologists working within a single practice should create a standard fluorescein instillation protocol and a unique set of expected values for TFBUT tests in their practice to help improve overall reliability.37
Acknowledgments
Supported by the Mississippi State University College of Veterinary Medicine Office of Research and Graduate Studies and the Department of Clinical Sciences.
The authors declare that there were no conflicts of interest.
The authors thank Dr. Robin Fontenot for thoughtful review of the manuscript.
References
- 1. ↑
Mishima S, Maurice DM. The oily layer of the tear film and evaporation from the corneal surface. Exp Eye Res. 1961;1:39–45.
- 2. ↑
Gelatt KN, Ben-Shlomo G, Gilger BC, Hendrix DVH, Kern TJ, Plummer CE. In: Featherstone HJ, Heinrich CL, eds. The eye examination and diagnostic procedures. Veterinary Ophthalmology. 6th ed. John Wiley and Sons Inc; 2021:601–618.
- 3. ↑
Millar TJ, Schuett BS. The real reason for having a meibomian lipid layer covering the outer surface of the tear film – a review. Exp Eye Res. 2015;137:125–138.
- 4. ↑
Moore CP. Qualitative tear film disease. Vet Clin North Am Small Anim Pract. 1990;20(3):565–581.
- 5. ↑
Cullen CL, Ihle SL, Webb AA, McCarville C. Keratoconjunctival effects of diabetes mellitus in dogs. Vet Ophthalmol. 2005;8(4):215–224.
- 6. ↑
Grahn BH, Sisler S, Storey E. Qualitative tear film and conjunctival goblet cell assessment of cats with corneal sequestra. Vet Ophthalmol. 2005;8(3):167–170.
- 7. ↑
Saito A, Watanabe Y, Kotani T. Morphologic changes of the anterior corneal epithelium caused by third eyelid removal in dogs. Vet Ophthalmol. 2004;7(2):113–119.
- 8. ↑
Biondi F, Dornbusch PT, Sampaio M, Montiani-Ferreira F. Infrared ocular thermography in dogs with and without keratoconjunctivitis sicca. Vet Ophthalmol. 2015;18(1):28–34.
- 9. ↑
Cullen CL, Lim C, Sykes J. Tear film breakup times in young healthy cats before and after anesthesia. Vet Ophthalmol. 2005;8(3):159–165.
- 10. ↑
Lim CC, Cullen CL. Schirmer tear test values and tear film break-up times in cats with conjunctivitis. Vet Ophthalmol. 2005;8(5):305–310.
- 11. ↑
Moore CP, Wilsman NJ, Nordheim EV, Majors LJ, Collier LL. Density and distribution of canine conjunctival goblet cells. Invest Ophthalmol Vis Sci. 1987;28(12):1925–1932.
- 12. ↑
Arnold TS, Wittenburg LA, Powell CC. Effect of topical naltrexone 0.3% on corneal sensitivity and tear parameters in normal brachycephalic dogs. Vet Ophthalmol. 2014;17(5):328–333.
- 13.
Carrington SD, Bedford PG, Guillon JP, Woodward EG. Biomicroscopy of the tear film: the tear film of the Pekingese dog. Vet Rec. 1989;124(13):323–328.
- 14. ↑
Sebbag L, Kass PH, Maggs DJ. Reference values, intertest correlations, and test-retest repeatability of selected tear film tests in healthy cats. J Am Vet Med Assoc. 2015;246(4):426–435.
- 15. ↑
Monclin SJ, Farnir F, Grauwels M. Determination of tear break-up time reference values and ocular tolerance of tetracaine hydrochloride eyedops in healthy horses. Equine Vet J. 2011;43(1):74–77.
- 16. ↑
Palella Gómez A, Mazzucchelli S, Scurrell E, Smith K, Pinheiro de Lacerda R. Evaluation of partial tarsal plate excision using a transconjunctival approach for the treatment of distichiasis in dogs. Vet Ophthalmol. 2020;23(3):506–514.
- 17. ↑
Shukla AK, Pinard CL, Flynn BL, Bauman CA. Effects of orally administered gabapentin, tramadol, and meloxicam on ocular variables in healthy dogs. Am J Vet Res. 2020;81(12):973–984.
- 18.
Evans PM, Lynch GL, Labelle P. Effects of oral administration of diphenhydramine on pupil diameter, intraocular pressure, tear production, tear film quality, conjunctival goblet cell density, and corneal sensitivity of clinically normal adult dogs. Am J Vet Res. 2012;73(12):1983–1986.
- 19. ↑
Saito A, Kotani T. Estimation of lacrimal level and testing methods on normal Beagles. Vet Ophthalmol. 2001;4(1):7–11.
- 20. ↑
Perry HD. Dry eye disease: pathophysiology, classification, and diagnosis. Am J Manag Care. 2008;14(suppl 3);S79–S87.
- 21. ↑
Downie LE. Automated tear film surface quality breakup time as a novel clinical marker for tear hyperosmolarity in dry eye disease. Invest Ophthalmol Vis Sci. 2015;56(12):7260–7268.
- 23. ↑
Gelatt KN, MacKay EO, Widenhouse C, Widenhouse TC, Stopek JB. Effect of lacrimal punctal occlusion on tear production and tear fluorescein dilution in normal dogs. Vet Ophthalmol. 2006;9(1):23–27.
- 24. ↑
George O, Omokhua P. Comparative analysis of the effects of topical anaesthetic agents on tear quantity and tear quality. J Niger Optom Assoc. 2010;16(1):30–33.
- 25. ↑
Lemp MA, Hamill JR Jr. Factors affecting tear film breakup in normal eyes. Arch Ophthalmol. 1973;89(2):103–105.
- 26. ↑
Venturi F, Blocker T, Dees DD, Madsen R, Brinkis J. Corneal anesthetic effect and ocular tolerance of 3.5% lidocaine gel in comparison with 0.5% aqueous proparacaine and 0.5% viscous tetracaine in normal canines. Vet Ophthalmol. 2017;20(5):405–410.
- 27. ↑
Oriá AP, Rebouças MF, Martins Filho E, Dórea Neto FA, Raposo AC, Sebbag L. Photography-based method for assessing fluorescein clearance test in dogs. BMC Vet Res. 2018;14(1):269.
- 28. ↑
Webber WR, Jones DP, Wright P. Fluorophotometric measurements of tear turnover rate in normal healthy persons: Evidence for a circadian rhythm. Eye (Lond). 1987;1(5):615–620.
- 29. ↑
Herring IP, Bobofchak MA, Landry MP, Ward DL. Duration of effect and effect of multiple doses of topical ophthalmic 0.5% proparacaine hydrochloride in clinically normal dogs. Am J Vet Res. 2005;66(1):77–80.
- 30. ↑
Maki E. Interclass reliability in healthcare studies: calculating the intraclass correlation coefficient (ICC) in SAS. Accessed April 23, 2020. https://www.sas.com/content/dam/SAS/en_ca/User%20Group%20Presentations/Health-User-Groups/Maki-InterraterReliability-Apr2014.pdf
- 31. ↑
Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med. 2016;15(2):155–163.
- 32. ↑
Nagymihályi A, Dikstein S, Tiffany JM. The influence of eyelid temperature on the delivery of meibomian oil. Exp Eye Res. 2004;78(3):367–370.
- 33. ↑
Iwashita H, Saito A. Comparing two different methods of tear film break up time testing in dogs. Vet Ophthalmol. 2019;22(3):E17. Abstract.
- 34. ↑
Yokoi N, Komuro A. Non-invasive methods of assessing the tear film. Exp Eye Res. 2004;78(3):399–407.
- 35. ↑
Mengher LS, Bron AJ, Tonge SR, Gilbert DJ. Effect of fluorescein instillation on the pre-corneal tear film stability. Curr Eye Res. 1985;4(1):9–12.
- 36. ↑
Savini G, Prabhawasat P, Kojima T, Grueterich M, Espana E, Goto E. The challenge of dry eye diagnosis. Clin Ophthalmol. 2008;2(1):31–55.
- 37. ↑
Johnson ME, Murphy PJ. The effect of instilled fluorescein solution volume on the values and repeatability of TBUT measurements. Cornea. 2005;24(7):811–817.
- 38. ↑
Kim KT, Kim J-H, Kong YT, Chae JB, Hyung S. Reliability of a new modified tear breakup time method: dry tear breakup time. Graefes Arch Clin Exp Ophthalmol. 2015;253(8):1355–1361.
- 39. ↑
Paugh JR, Tse J, Nguyen T, et al. Efficacy of the fluorescein tear breakup time test in dry eye. Cornea. 2020;39(1):92–98.
- 40.
Korb DR, Greiner JV, Herman J. Comparison of fluorescein break-up time measurement reproducibility using standard fluorescein strips versus the dry eye test (DET) method. Cornea. 2001;20(8):811–815.
- 41. ↑
Abdul-Fattah AM, Bhargava HN, Korb DR, Glonek T, Finnemore VM, Greiner JV. Quantitative in vitro comparison of fluorescein delivery to the eye via impregnated paper strip and volumetric techniques. Optom Vis Sci. 2002;79(7):435–438.
- 42. ↑
Wang MTM, Craig JP. Comparative evaluation of clinical methods of tear film stability assessment: a randomized crossover trial. JAMA Ophthalmol. 2018;136(3):291–294.
- 43. ↑
Sebbag L, Kirner NS, Allbaugh RA, Reis A, Mochel JP. Kinetics of fluorescein in tear film after eye drop instillation in Beagle dogs: does size really matter? Front Vet Sci. 2019;6:457.
- 44. ↑
Sebbag L, Allbaugh RA, Wehrman RF, et al. Fluorophotometric assessment of tear volume and turnover rate in healthy dogs and cats. J Ocul Pharmacol Ther. 2019;35(9):497–502.
- 45. ↑
Behrens A, Doyle JJ, Stern L, et al. Dysfunctional tear syndrome: a Delphi approach to treatment recommendations. Cornea. 2006;25(8):900–907.
Appendix 1
Sequence of examination for 21 dogs (10 on day 1 and 11 on day 2) that underwent TFBUT testing in a prospective, randomized study to determine intra- and interobserver reliability of the TFBUT test as it would be performed in a clinical environment with and without administration of a topical anesthetic.
| Study day and observer | No topical anesthetic | Proparacaine | ||
|---|---|---|---|---|
| Period 1 | Period 2 | Period 3 | Period 4 | |
| Day 1 | ||||
| Observer 1 | Dogs 1–5 | Dogs 6–10 | Dogs 1–5 | Dogs 6–10 |
| Observer 2 | Dogs 6–10 | Dogs 1–5 | Dogs 6–10 | Dogs 1–5 |
| Day 2 | ||||
| Observer 1 | Dogs 1–5 | Dogs 6–11 | Dogs 1–5 | Dogs 6–11 |
| Observer 2 | Dogs 6–11 | Dogs 1–5 | Dogs 6–11 | Dogs 1–5 |
On each study day, a randomly selected group of dogs (10 on day 1 and 11 on day 2) underwent fluorescein-based TFBUT testing according to a predetermined protocol. At the beginning of the day, each dog was numbered by random assignment, and each subset of dogs underwent TFBUT testing in numerical order by the observers during testing periods as shown. In periods 3 and 4, a single drop of 0.5% proparacaine solution was instilled in each eye, and 1 min elapsed before TFBUT testing. There was a ≥ 1-hour interval between testing periods.
