Purpose
To assess the associations of central corneal thickness (CCT) with ocular and systemic factors in a West African population.
Design
Population-based cross-sectional study.
Methods
Participants aged ≥40 years in the Tema Eye Survey who had clinically normal corneas were included in this study. CCT was determined bilaterally using handheld ultrasound pachymetry. The association between CCT and ocular or systemic factors was analyzed with univariable linear regression. Multivariable linear regression analysis was performed for variables significantly associated with CCT in the univariable analysis. Main outcome measures were CCT (μm) and its associations with age, sex, IOP, cup-to-disc ratio (CDR), glaucoma, hypertension, and diabetes.
Results
A total of 6806 eligible subjects were identified, of whom 5603 (82.3%) participated in the Tema Eye Survey. A total of 4737 participants (84.5% of participants) of West African descent and clinically normal corneas were included. The mean age ± standard deviation (SD) of participants was 51.2 ± 9.7 years and 38.7% were male. The mean ± SD CCT in the population was 533.9 ± 34.0 μm. In the multivariable linear regression analysis, increased CCT was significantly associated with younger age, male sex, and higher IOP (all P < .001). There were no significant associations between CCT and CDR, glaucoma, hypertension, and diabetes.
Conclusions
This population-based cross-sectional survey of West African adults found a significant association between increased CCT and younger age, male sex, and higher IOP but not glaucoma or CDR. Variations in corneal thickness may influence the accuracy of IOP measurements in African persons.
Central corneal thickness (CCT) is an important parameter in refractive surgery, in the assessment of corneal disease, and for risk profiling in ocular hypertension and glaucoma. Not only does CCT influence the accuracy of Goldmann applanation and most other forms of tonometry, but CCT was also reported to be a strong predictor of the development of primary open-angle glaucoma (POAG) in the Ocular Hypertension Treatment Study (OHTS) and the European Glaucoma Prevention Study (EGPS). In the analysis of the pooled data of these 2 studies, the risk of developing POAG was reported to double for every 40-μm decrease in CCT. CCT was reported to be an independent predictive factor for longer-term progression of POAG in the Early Manifest Glaucoma Trial (EMGT) and the Barbados Eye Studies, and the prevalence of glaucoma was higher in individuals with thin CCT in the Los Angeles Latino Eye Study (LALES).
African-American subjects have thinner CCT measurements than white subjects in ophthalmology clinic–based studies in Texas and North Carolina, averaging approximately 530 μm. This is consistent with the population-based CCT measurements of Afro-Caribbeans in the Barbados Eye Studies. Interestingly, in the OHTS, race was not a predictor for the development of POAG when adjusted for factors such as CCT and baseline vertical cup-to-disc ratio (CDR). Currently, there are no population-based data on CCT measurements in African people within the African continent itself, though a clinic-based study conducted in West Nigeria reported that the mean CCT of 70 eyes was 532 μm and was similar to the reported CCT values of African Americans and Afro-Caribbeans. However, clinic-based studies contain inherent bias, since subjects presenting to clinics are likely patients with eye diseases that could affect corneal thickness.
The Tema Eye Survey is a population-based eye survey of 5603 inhabitants of Tema, Ghana. This survey offers the opportunity to investigate the distribution of CCT in a large West African population-based cohort of largely unmedicated subjects, and to evaluate the relationship between corneal thickness and other systemic and ocular factors.
Methods
The Tema Eye Survey
The Tema Eye Survey is a cross-sectional population-based prevalence study of vision loss and ocular disease in 5603 residents of Tema, Ghana, West Africa. It was prospectively approved by the Ethics Committee of the Ministry of Health of Ghana and the home institutions of all investigators. Written informed consent was provided by each subject before any study-related activity was carried out, using a consent form approved by the Ghana Ethics Committee.
All participants were of West African ethnic origin, though 7.7% were from outside Ghana, including Mali (40), Togo (37), Nigeria (14), and Cote d’Ivoire (13). After a census in randomly selected clusters of 5 communities, subjects aged 40 years and above were invited for a field examination consisting of visual acuity measurement using the Early Treatment Diabetic Retinopathy Study (ETDRS) method, autorefraction, intraocular pressure (IOP) measurement, ultrasonic pachymetry, visual field screening, and fundus photography. Subjects failing to meet predefined criteria in each of these investigations were invited for a clinic-based comprehensive ophthalmologic examination including Humphrey visual field test, refraction, gonioscopy, anterior segment optical coherence tomography (AS-OCT) examination, and dilated fundus examination. Data were collected in the field and clinic by direct computer data entry into customized databases (Microsoft Access 2003; Microsoft Corporation, Redmond, Washington, USA). Demographic data were recorded and correlated with the information in the house-to-house census in each randomly selected cluster. All participants included in this study were interviewed regarding their medical history. Details of the study population and study design have been previously reported.
Eligibility
Of 6806 subjects who were eligible for the Tema Eye Survey, 5603 (82.3%) participated in an examination in the field, where corneal examination was performed with a flashlight. In subjects who passed the field examinations (n = 3734), the cornea was assumed to be sufficiently healthy to be enrolled in the study without further corneal examination. Eligible patients who underwent a field examination were excluded if 2 valid ultrasound pachymetry measurements were not obtained from each eye, or if CCT was less than 400 μm in either eye. Of the 1869 subjects (33.3%) who failed 1 or more screening examinations, 1538 returned to the clinic for a complete examination by an ophthalmologist. Subjects were excluded if pathologic conditions involving the cornea were subsequently diagnosed using the slit lamp, including corneal scar, corneal edema, corneal dystrophy, pterygium, and guttata. We also excluded participants who were using glaucoma medications (22 participants), as this has been shown to affect CCT measurements. Previous studies have reported that latanoprost reduced CCT measurements by 4–20 μm after more than 2 years of treatment, while timolol administration for 28 days was associated with a reversible increase in CCT of 12 μm. After the above criteria were applied, 4737 subjects (84.5% of total participating subjects) were included in the final analysis.
Ultrasound Pachymetry
A handheld ultrasound pachymetry device, Pachmate 55 (DGH, Exton, Pennsylvania, USA), was used to measure CCT in both eyes, after the cornea was anesthetized with 1 drop of proparacaine hydrochloride 0.5% (Alcaine; Alcon Labs, Fort Worth, Texas, USA). The probe tip was placed at the very center of the cornea and the pachymeter was held horizontally perpendicular to the corneal surface. If the patient moved and the probe decentered, then the measurements were repeated.
Ultrasound pachymetry was performed in each case by 1 of 2 trained technicians. Twenty-five sequential measurements were obtained from the center of the cornea and averaged. Values with standard deviation (SD) of 5 μm or less were considered suitable for inclusion. Two sets of readings were taken for each eye and the data were recorded in the Tema Eye Survey database. The 2 CCT readings taken from the right and left eyes were averaged together to provide a single CCT value for each subject.
Measurement and Definitions of Other Variables
IOP was measured bilaterally with Tonopen XL (Reichert Ophthalmic Instruments, Depew, New York, USA) after instillation of 1 drop of proparacaine hydrochloride 0.5% (Alcaine; Alcon Labs). Two IOP measurements were taken for each eye, and the averages of both measurements from both eyes were calculated for each subject. The vertical CDR was assessed by the Moorfields Eye Hospital Reading Center from dilated optic disc photographs taken with a handheld digital fundus camera (Kowa Genesis D; Kowa Company, Ltd, Tokyo, Japan) at the field examination, or from stereoscopic optic disc photographs taken with the Nidek 3Dx camera (Nidek Co, Ltd, Gamagori, Japan) if the participants were subsequently referred to the clinic. Glaucoma was diagnosed based on the definition by Foster and associates, and is consistent with that used in our previous study. Systolic and diastolic blood pressure (BP) were measured with a digital automatic BP monitor (Omron HEM-907; Hoffman Estates, Illinois, USA). Hypertension was defined as systolic BP ≥140 mm Hg, diastolic BP ≥90 mm Hg, or self-reported physician-diagnosed hypertension. Diabetes mellitus was defined as self-reported physician-diagnosed diabetes.
Statistics
All statistical analyses were performed using Stata Software version 12 (StataCorp, LP, College Station, Texas, USA). Pearson correlation was applied to calculate the correlation statistics between age grouping and CCT readings, tabulated on sexes. For all the variables (age, IOP, CDR, sex, hypertension, diabetes mellitus, and glaucoma), univariable linear regression was used to assess the association between these variables and CCT. These analyses were repeated with the inclusion of the 22 participants on glaucoma medications, and with the use of generalized estimating equations. Further assessment of the association between the variables of interest (glaucoma, IOP, CDR, hypertension, and diabetes mellitus) were all further adjusted with age and sex. The t test was used to compare CCT measurements between participants who were referred to the clinic for a slit-lamp examination and those who only underwent a field examination. Multivariable linear regression analysis was performed for variables that were significantly associated with CCT in the univariable analysis. P < .05 was considered statistically significant.
Results
Field examinations were conducted in 5603 subjects. Two adequate CCT measurements were taken from each eye in 5426 subjects, after exclusion of 177 subjects in whom 2 CCT measurements with SD of 5 μm or less could not be taken from both eyes. Another 667 subjects were excluded after corneal pathologies were noted in a clinic examination, and 22 subjects were excluded because they were using glaucoma medications. A total of 4737 subjects (84.5%) were included in the final analyses. The mean ± SD age of the study cohort was 51.2 ± 9.7 years, and 38.7% were male.
The age–sex distribution of CCT among 4737 participants is shown in Table 1 . The mean ± SD CCT in the study population was 533.9 ± 34.0 μm. A significant inverse relationship between age and CCT was found (R = −0.140; P < .001) in both male and female participants, and CCT was greater in male participants compared with female participants (537.7 ± 34.8 vs 531.6 ± 33.3 μm; P < .001) ( Figure ).
| Total | Female | Male | P Value a | ||||
|---|---|---|---|---|---|---|---|
| N | Mean (SD)/μm | N | Mean (SD)/μm | N | Mean (SD)/μm | ||
| Age (y) | |||||||
| 40–49 | 2431 | 537.5 (33.0) | 1559 | 534.8 (32.2) | 872 | 542.5 (33.8) | <.001 |
| 50–59 | 1364 | 533.4 (33.9) | 825 | 530.7 (33.1) | 539 | 537.7 (34.6) | |
| 60–69 | 643 | 527.7 (35.7) | 350 | 526.8 (35.2) | 293 | 528.9 (36.3) | |
| 70–79 | 232 | 522.9 (34.0) | 120 | 519.0 (35.9) | 112 | 527.0 (31.6) | |
| ≥80 | 67 | 511.8 (30.4) | 48 | 510.1 (29.3) | 19 | 516.1 (33.5) | |
| All ages | 4737 | 533.9 (34.0) | 2902 | 531.6 (33.3) | 1835 | 537.7 (34.8) | |
| R | −0.140 | −0.142 | −0.158 | ||||
| P value b | <.001 | <.001 | <.001 | ||||
a t test comparing male and female subjects.
b Pearson correlation coefficient between age and central corneal thickness.
There were no significant differences in CCT measurements between participants with and without glaucoma (534.2, 95% confidence interval [CI] 529.3–539.0 vs 533.9, 95% CI 532.9–534.9; P = .93), hypertension (535.0, 95% CI 533.2–536.8 vs 537.3, 95% CI 535.5–539.1; P = .079), and diabetes mellitus (531.6, 95% CI 527.3–535.9 vs 534.6, 533.0–536.2; P = .25). Including the 22 participants on glaucoma medications in the analysis did not alter the results and the CCT measurements between participants with and without glaucoma were still not significantly different. There was no significant difference in the mean CCT measurement between participants who were referred to the clinic for a slit-lamp examination and those who only underwent a field examination.
In the univariable linear regression analysis with CCT as a dependent variable and other ocular and systemic factors as independent variables, higher IOP was significantly associated with increased CCT (coefficient 1.57; P < .001). Similar results were obtained when the analysis was repeated with generalized linear equations, with IOP being significantly associated with CCT ( P < .001). This association remained significant after adjustment for age and sex ( Table 2 ). In the multivariable linear regression analysis for variables significant on the univariable analysis with CCT as the dependent variable, increased CCT was significantly associated with younger age, male sex, and higher IOP (all P < .001) ( Table 3 ).
| Univariable | Adjusted for Age and Sex | |||
|---|---|---|---|---|
| β Coefficient | P Value | β Coefficient | P Value | |
| Glaucoma | 0.22 | .929 | 2.03 | .412 |
| Cup-to-disc ratio | −0.61 | .872 | −0.79 | .835 |
| Intraocular pressure | 1.57 | <.001 | 1.73 | <.001 |
| Hypertension | −2.28 | .079 | 0.13 | .921 |
| Diabetes | −3.02 | .246 | −0.70 | .787 |
| Variables | β Coefficient | 95% CI for β Coefficient | P Value | |
|---|---|---|---|---|
| Lower | Upper | |||
| Age | −0.58 | −0.68 | −0.48 | <.001 |
| Sex (reference = female) | 7.01 | 5.08 | 8.94 | <.001 |
| Intraocular pressure | 1.73 | 1.47 | 2.00 | <.001 |
Results
Field examinations were conducted in 5603 subjects. Two adequate CCT measurements were taken from each eye in 5426 subjects, after exclusion of 177 subjects in whom 2 CCT measurements with SD of 5 μm or less could not be taken from both eyes. Another 667 subjects were excluded after corneal pathologies were noted in a clinic examination, and 22 subjects were excluded because they were using glaucoma medications. A total of 4737 subjects (84.5%) were included in the final analyses. The mean ± SD age of the study cohort was 51.2 ± 9.7 years, and 38.7% were male.
The age–sex distribution of CCT among 4737 participants is shown in Table 1 . The mean ± SD CCT in the study population was 533.9 ± 34.0 μm. A significant inverse relationship between age and CCT was found (R = −0.140; P < .001) in both male and female participants, and CCT was greater in male participants compared with female participants (537.7 ± 34.8 vs 531.6 ± 33.3 μm; P < .001) ( Figure ).
