To investigate the hypothesis that ocular surface epithelial thickness is correlated with tear osmolarity, conjunctivochalasis (CCh), and dry eye clinical tests.
A case-control study.
A total of 50 patients with different types of dry eye disease (DED) and 15 age- and sex-matched control subjects were enrolled in this study. We performed a detailed diagnostic assessment of tear film and ocular surface parameters, including Ocular Surface Disease Index questionnaire, visual analog scale questionnaire, topographic surface regularity and surface asymmetry indices, tear film break-up time, corneal and conjunctival dye staining, Schirmer I test, and tear osmolarity. The corneal and bulbar conjunctival epithelial thickness (CET) were obtained using the latest version of spectral domain optical coherence tomography (SDOCT).
Patients with aqueous-deficient DED (ADDED) had lower bulbar CET values, particularly in the temporal region, than those of normal subjects and patients with evaporative-type DED (EDED); however, the difference did not quite reach a statistically significant level. Patients with DED and CCh had lower bulbar CET values in temporal (38.52 ± 9.58 µm) and inferior regions (50.79 ± 9.10 µm) compared to those with DED without CCh (47.39 ± 11.71 µm, 60.38 ± 14.36 µm, respectively, P < .02). In the DED group, tear osmolarity was found to be negatively correlated with CET values in temporal bulbar region ( P = .006 and r = −0.403) and central corneal epithelial thickness values ( P = .029 and r = −0.325).
CCh and tear osmolarity are associated with reduced conjunctival epithelial thickness in DED.
D ry eye disease (DED) is mainly classified as reduced aqueous production or volume (ADDE) or as excessive tear evaporation (EDE). Both conditions are associated with tear hyperosmolarity, which contributes to ocular surface disease and pain from dry eye. Studies have demonstrated that tear hyperosmolarity might induce apoptosis of the ocular surface epithelial cells and activate stress signaling pathways that trigger production of inflammatory mediators by the epithelial and resident immune cells. Even though slit-lamp examination with and without vital stains (fluorescein, lissamine green, rose bengal) can give some information on ocular surface health, , high-resolution, rapid imaging techniques are needed to evaluate the effect of DED on the ocular surface cells.
Impression cytology is a widely used technique to study stratified and goblet cell epithelium and inflammatory cells in the conjunctiva. Impression cytology is labor intensive and requires capabilities to process, stain, and image membranes, which has limited its usage in clinical practice. Ultrasound biomicroscopy, in vivo confocal microscopy, and anterior segment optical coherence tomography (AS OCT) are imaging techniques that can evaluate and grade alterations of the ocular surface in DED. OCT technology is capable of high-resolution, cross-sectional imaging of the human corneal and conjunctival layers that can measure corneal, limbal, and conjunctival thickness at the micron level.
The aim of the current study was to use in vivo spectral-domain (SD) AS OCT to compare the corneal and bulbar conjunctival epithelial thickness in normal subjects and patients with different types of DED and also to correlate the results with tear osmolarity and other clinical tests for DED.
The study was conducted according to the tenets of the Declaration of Helsinki and approved by the Baylor College of Medicine Institutional Review Board. The written informed consent form was obtained from all participants after the explanation of the purpose and some possible consequences of the study. A total of 50 patients with different types of DED (41 women and 9 men; mean age: 58.88 ± 12.17 years; range: 31-86 years) were enrolled in this single-center observational study. Fifteen age- and sex-matched control subjects (11 women and 4 men; mean age: 50.67 ± 16.58 years; range: 30-77 years) were also recruited. All control subjects were selected from those who had no complaints of eye irritation, no anterior segment abnormality on biomicroscopic examination, and normal tear and ocular surface tests. Subjects with DED were classified as ADDE or EDE based on the criteria as described in our previous study. These groups were combined for some analyses. The presence of conjunctivochalasis (CCh) based on the presence of lid parallel folds in the temporal and nasal conjunctiva above the lid margin and visualization in OCT images of the inferior tear meniscus were noted.
Exclusion criteria for both groups was as follows: age under 18 years old, a history of any previous ocular surgery or ocular trauma, contact lens wear, active ocular inflammation/infection, or corneal disease or dystrophies that might affect the results.
All patients were asked to complete the Ocular Surface Disease Index (Allergan Inc, Irvine, California, USA) questionnaire (range: 0-100) and SANDE visual analog scale questionnaire (5 grades in 100 mm). Then, topographic surface regularity index and surface asymmetric index were measured using the Tomey TMS-4N computerized videokeratoscopy (Tomey, Phoenix, AZ ph). Finally, an ocular surface examination was performed in the following order: tear film break-up time (TBUT), corneal and conjunctival dye staining, and Schirmer test without anesthesia. TBUT was measured by instilling fluorescein into the lower fornix with a wetted fluorescein strip (BioGlo; HUB, Rancho Cucamonga, California, USA). Corneal fluorescein staining was graded 0-6 in each of 5 zones (central, nasal, temporal, inferior, superior), as previously described. Conjunctival lissamine green staining was graded on a scale of 0-3 exposed in the nasal and temporal bulbar conjunctiva with a total maximum score of 6. Schirmer I test without anesthesia was used to measure tear production by inserting a dry Schirmer test strip over the outer third of the lower eyelid margin. The ocular surface clinical parameters were all measured by the same experienced investigator (S.C.P.).
Tear osmolarity was measured with the TearLab Osmolarity System (OcuSense, San Diego, California, USA). All measurements were performed by a single investigator (K.G.) in the same room at the same temperature prior to any invasive clinical tests. The instrument was calibrated regularly in accordance with manufacturer guidelines.
Anterior Segment Optical Coherence Tomography and Image Analysis
All subjects underwent imaging with Optovue RTVue XR Avanti (Optovue, Inc, Fremont, California, USA) by a single well-trained investigator (K.G.) in order to visualize and measure corneal and conjunctival epithelial thickness. Careful and maximum attention was paid to the standardization of OCT scans. All scans were performed in a standardized manner using the same instrument at the same room conditions. Subjects were asked to blink several times and OCT images were obtained immediately after the last blink (within 2 seconds). Corneal epithelial thickness was automatically processed by using the instrument’s corneal epithelial thickness mapping and pachymetry software, corresponding to a 9-mm-diameter area ( Figure 1 ). While pachymetric and epithelial thickness maps of 9 mm diameter are shown on the right side of test output, other details are shown on the left side, including pachymetric statistics such as SN-IT (superior temporal – inferior temporal) (2-5 mm), S-I (superior – inferior) (2-5 mm), Min (minimum), and Min-Max values, and epithelium statistics such as S (superior) (2-7 mm), I (inferior) (2-7 mm), Min, Max (maximum), and Min-Max ( Figure 1 ).
The temporal (exposed) and inferior (nonexposed) bulbar conjunctiva were imaged to measure epithelial thickness. Temporal corneal-scleral limbus and bulbar conjunctiva were imaged using the same length horizontal scan that was adjusted to pass through the middle of the cornea, as shown in Figure 2 a. Inferior bulbar conjunctiva was imaged using the same horizontal scan length that was taken 1-2 mm below the inferior limbus centered at 6 o’clock, as shown in Figure 2 b. The conjunctival epithelium thickness (CET) was measured with the digital caliper from the surface to the change in brightness of the substantia propria noted in OCT images ( Figure 3 ). Three epithelium thickness measurements were made at 1.5-mm intervals in the central 3 mm of the conjunctival images ( Figure 3 ). The average of the 3 calculated values was used for statistical analysis.
Statistical analyses were performed with SPSS for Mac version 23.0 (SPSS Inc, Chicago, Illinois, USA). For each patient, computer-generated randomization was performed to select 1 eye for statistical analysis. Regarding study variables, comparisons between groups were performed using the nonparametric Mann-Whitney U test, and comparisons between several groups were performed using the Kruskal-Wallis test, because all data were not normally distributed. The Spearman correlation test was performed to evaluate bivariate correlations between epithelial thickness parameters and the dry eye clinical tests. The test was considered to be statistically significant at P < .05.
There were no statistically significant differences in age ( P = .125) and sex ( P = .477) between the DED and control groups. The comparison of ocular surface clinical tests between all DED and control groups is presented in Table 1 . Patients with DED had significantly lower TBUT and Schirmer scores and higher tear osmolarity, Ocular Surface Disease Index, and ocular surface staining scores ( Table 1 ).
|Dry Eye Tests||Control (N = 15)Median / IQR||DED (N = 50)Median / IQR||P Value|
|SAI||0.395 / 0.14||0.390 / 0.27||.548|
|SRI||0.330 / 0.41||0.260 / 0.30||.955|
|Tear osmolarity (mOsm/L)||301 / 11||306 / 20||.048*|
|OSDI||19.79 / 22||45.83 / 30||<.001*|
|VAS-1||2.0 / 2.3||4.0 / 1.0||<.001*|
|VAS-2||/ 3.3||4.0 / 1.0||<.001*|
|Schirmer-1 test||20.0 / 18.0||11.0 / 13||.030*|
|TBUT (seconds)||7.50 / 4.0||4.0 / 2.0||<.001*|
|Corneal staining||0 / 1.0||3.0 / 3.5||<.001*|
|Conjunctival staining – T||1.0 / 1.0||2.0 / 2.0||<.001*|
|Conjunctival staining – N||0.0 / 0.0||0.0 / 1.0||.300|
|Tear meniscus height||324 / 219||275 / 212.5||.117|
In the DED group, means (± standard deviations) of bulbar CET in the temporal and inferior regions were 43.67 ± 11.63 µm and 56.35 ± 13.20 µm, respectively; and in the control group, 42.33 ± 8.62 µm and 52.60 ± 7.86 µm, respectively. Temporal and inferior bulbar CET values were not statistically different between the 2 groups ( P = .827 and P = .396, respectively). Because dry eye can be owing to decreased tear production or increased evaporation, the DED group was subdivided into 2 groups: (1) aqueous-deficient DED (ADDED) and (2) evaporative type DED (EDED). Patients with ADDED had lower bulbar CET values, particularly in the temporal region (exposed area), than those of normal subjects and patients with EDED; however, the difference did not quite reach a statistically significant level ( P > .05, Table 2 ).