Study design
We performed a cross-sectional study of patients with glaucoma, including 132 randomly selected eyes (right or left) of 132 patients using glaucoma medications for at least 1 year. Patients receiving certain medications such as anti-inflammatory or immunosuppressant drugs, or patients with previous eye surgery, were excluded from the study. The control group included 87 eyes from 87 asymptomatic people, who were consecutively recruited among hospital staff, nurses, patient relatives, and patients referred for a routine visual acuity examination without ocular diseases who had an intraocular pressure (IOP) of 20 mmHg or lower, normal visual fields, and no familial glaucoma history. Normal participants with dry eye symptoms or who were using any eyedrop, were excluded. All participants were European Caucasians and had no corneal or retinal pathology, no substantial media opacity that obscured the eye fundus, no history of amblyopia, no contraindication to dilation or intolerance to topical anaesthetic or mydriatic agents or fluorescein dye.
Procedures
At the beginning of the visit, an ocular surface analysis using the Keratograph 5M software version 2.5r18 (Oculus Optikgerate GmbH, Wetzlar, Germany) was performed without fluorescein dye. Two experienced examiners (AMB and EC) performed all non-invasive evaluations. All the measurements were taken between 10:00 a.m. and 6:00 p.m. in one single visit in a dimly lit room with controlled temperature (21–24ºC) and humidity (30–60%). The software determined the NITBUT after alignment of the instrument head with the pupillary centre and after the subjects were asked to blink twice, using an infrared light (Fig. 1). Two measurements were performed and if the values were similar, the first measure was recorded; if the difference between both measures of the same eye exceeded 2 seconds, a third measure was performed; then the first of the two measures closer to the third one was recorded. After acquisition, two parameters were recorded; the First NITBUT (F-NITBUT) defined as the time, in seconds, between the last complete blink and the first perturbation or irregularity of the 22 concentric rings of the Placido disc reflected on the corneal surface, and the Average NITBUT (A-NITBUT) defined as the average of all tear film breakups occurring in the measured period of up to 24.98 seconds (time limit set by the device’s software). The device has a minimal measure time of about 4 seconds to record NITBUT values. If the patient blinked after a brief period, before a value can be recorded, the examination was repeated. If the patient still was not able to keep the eye open for the minimal time of 4 seconds, that eye was not considered for further data analysis and excluded from the study. While recording the time of NITBUT values, the location of each rupture is also recorded by 168 small areas, or breakup areas (BUA). A graph with both measures and the areas of tear breakup are displayed (Fig. 1).
The number of simultaneous BUA at the F-NITBUT is referred as first breakup areas (F-BUA), and the total number of breakup areas at the end of the NITBUT measure, is referred as total breakup areas (T-BUA). The T-BUA is expressed as the percentage of the total exposed area measured by the device in each eye. The steepness [expressed in grades] of the increase in BUA after the F-NITBUT; that is, progression from F-BUA to T-BUA; allows analysis of tear film breakup rates and is referred as Dry Area Growth Rate (DAGR), as previously described [11]. A custom program written in MATLAB (MathWorks, Natick, MA, USA) was used to analyse the data recorded by the Keratograph 5M to further evaluate these previous parameters. We generate a matrix in which the columns represented the time in fractions of one hundredth of a second (therefore 2498 columns), and the rows represented the randomly selected eye from each patient. We traced all the breaking points of each patient in an orderly manner looking for how many areas were break-up in each instant of time. Thus, we obtained a vector per patient of each group. In each group, we calculated the average of break-up zones at different times and the confidence interval, obtaining two vectors, one for normal cases and another one for cases with glaucoma.
The Keratograph 5M was also used to quantify conjunctival hyperaemia (CH), evaluated after the NITBUT measure using a white light and automatically classified by the device software with the Jenvis grading scale from 0 to 4 using one decimal number (Fig. 2). The infrared camera of the device was also used to view the anatomy of the meibomian glands of the upper and lower eyelids. The status of the meibomian glands was evaluated with the Meiboscore [12], using a 0 to 3 scale (grade 0, no gland loss; grade 1, area of gland loss up to 33% of the total gland area; grade 2, area of gland loss between 33%–66%; and grade 3, area of gland loss of 67% or more), and determined by hand tracing every picture of each eye with the ImageJ software (public domain, National Institutes of Health, USA).
Examiners (2 ophthalmologists), masked to the Keratograph 5M results, performed ocular surface staining with 5 uL of 2% sodium fluorescein solution for TFBUT and for Oxford grading scale [13]. Schirmer test type 1 without anaesthesia was measured at the end of the visit.
Statistical Analysis
The Shapiro-Wilk test assessed the normality of the quantitative variables. Normally distributed variables were expressed as the mean and standard deviation (SD), and non-normally distributed variables were summarized using box plots. Differences between both groups were analysed using Student t-test for parametric data and Mann-Whitney U test for non-parametric data. Pearson's correlation coefficient was used to assess the relationship between parametric variables and Spearman in the case of non-parametric variables. Data was evaluated using the SPSS program version 20.0.1 (SPSS Inc., Chicago, IL, USA).