Central corneal thickness in Iranian congenital glaucoma patients.

PURPOSE: To compare central corneal thickness (CCT) in subjects with controlled primary congenital glaucoma (PCG) and nonglaucomatous subjects and to investigate the correlation between CCT and intraocular pressure (IOP) in the study population.
MATERIALS AND METHODS: Twenty-three consecutive PCG cases with controlled IOP and no clinical evidence of corneal edema comprised the Study Group. There was an interval of at least 2 months between last intraocular surgery and inclusion in the study. Twenty-one subjects with strabismus or lacrimal drainage insufficiency who did not have glaucoma or any history of intraocular surgery or ocular trauma comprised the control group. The Control Group was age and sex-matched. Data from ultrasonic pachymetry and applanation tonometry were analyzed for differences between groups. Correlation of the study parameters was investigated. A P-value less than 0.05 was statistically significant.
RESULTS: Data from both eyes of subjects in the Study Group and Control Group were included in the original analysis. Mean CCT was statistically significantly higher in the Study Group compared to the Control Group (589.42 ± 53.44 μm vs. 556.14 ± 30.51 μm, respectively; P=0.001). There was a significant correlation between CCT and IOP (r=0.63; P<0.0001). Similar statistically significant outcomes were observed when only one eye per subject was used in a reanalysis of the data for the Study and Control Groups.
CONCLUSION: Patients with PCG who had controlled IOP have statistically significantly thicker corneas than nonglaucomatous age and sex-matched subjects The thicker cornea could significantly alter IOP measurement with applanation tonometry. Pachymetry should be considered an essential part of the evaluation for PCG.

INTRODUCTION

Blindness in children is one of the five major priorities in The Right to Sight: Vision 2020 global initiative.1 In 1999, there were an estimated 1.4 million blind children globally and twice as many with low vision.2 Blindness was expected to increase by 500,000 blind children yearly.2 Glaucoma remains a major sight-threatening condition in children, accounting for 2.5-10% of the cases of blindness.3 The incidence of primary infantile glaucoma is 1 in every 10,000-20,000 live births in Western countries and is reported to be as high as 1 in every 2500 live births in the Middle East.45 Timely diagnosis and proper management of congenital glaucoma is crucial to reduce the inherent morbidity.

Corneal enlargement and increased axial length in primary congenital glaucoma (PCG) result in a buphthalmic appearance. Some have proposed that corneal stretching could result in increased corneal diameter and reduced central corneal thickness (CCT).6 Alternatively, corneal edema due to disruption in corneal endothelial integrity could result in increased thickness. The importance of CCT for the diagnosis, management, and prognosis of adult glaucoma is well documented.7 It seems to be equally important in children. CCT has a marked effect on intraocular pressure (IOP) readings with most available tonometers including the Goldmann applanation tonometer8–10 and the Tonopen (Reichert Technologies, Depew, NY).1011 In addition, CCT may be an independent risk factor for the development of glaucoma.12

To date, very few studies have investigated corneal thickness in pediatric glaucoma.6910 Current evidence on CCT in PCG patients is inconclusive and even contradictory. Additionally, there are ethnic differences in the prevalence and nature of PCG. Literature is sparse regarding PCG in Middle Eastern children. The goal of the current study was to evaluate CCT in PCG patients in Iran, and compare it with a matched cohort of control subjects. A secondary goal was to evaluate any correlation between IOP and CCT.

MATERIALS AND METHODS

This was a descriptive, cross-sectional, comparative study on all consecutive PCG cases presenting to the glaucoma clinic at Farabi Eye Hospital, a University-based hospital serving as a primary and major tertiary eye hospital in Tehran, Iran. All subjects presented between January 2008 and January 2009 (Study Group). The study protocol was approved by the Ethical Committee of Tehran University of Medical Sciences and the study adhered to the tenets of Declaration of Helsinki. All patients or their legal guardians provided an oral or written informed consent prior to enrollment in the study.

All PCG cases were included in the study, regardless of age. PCG was diagnosed by a glaucoma specialist based on typical glaucomatous optic disc damage, IOP, corneal diameter, and biomicroscopy. All subjects had controlled disease at the time of examination with an IOP below 24 mmHg. Prior to enrollment, there was an interval of at least 2 months since last intraocular surgery. Patients with systemic disease, corneal scar, or clinical corneal edema were excluded from the study. An age and sex-matched cohort of nonglaucomatous subjects was selected among patients presenting to the lacrimal/ strabismus service of the hospital (Control Group). In the Control Group, subjects without glaucoma, younger than 2-year old were selected among patients with insufficient lacrimal drainage and underwent an examination under anesthesia (EUA) before lacrimal probing. Older children without glaucoma were selected among patients with nontraumatic strabismus without a history of ocular surgery.

A thorough ophthalmic examination was performed on all subjects including biomicroscopy, applanation tonometry, funduscopy, and pachymetry. EUA was performed after general inhalational anesthesia using halothane, and IOP was measured in the first 5 minutes after administration of the anesthetic. Pachymetry was performed with Sonomed 200P ultrasonic pachymeter (Sonomed, Lake Success, NY). The horizontal corneal diameter was measured using a single surgical caliper. IOP was measured using a TonoPen (TonoPen XL; Mentor, Norwell, MA) in patients undergoing EUA and a Goldmann applanation tonometer (Model AT 900 C/M; Haag-Streit, Bern, Switzerland) in the out-patient clinic. A single surgeon performed all measurements and the mean of at least three acceptable measurements was recorded.

Data are presented as mean ± standard deviation (SD). Comparison of variables between glaucomatous (Study Group) and nonglaucomatous (Control Group) eyes was performed using the independent sample t test and the Levene test was used to check the preassumption of equality of variances. Nonparametric equivalent was used as appropriate based on the results of the Kolmogorov-Smirnov test. All statistical tests were two-tailed and a statistical significance level of 0.05 was adopted. The study had a predefined sample size of 40 eyes in each study arm to reach a statistical power of 80% to detect a significant difference in CCT between groups. All statistical analyses were performed using SPSS version 13 software (SPSS Inc., Chicago, IL).

RESULTS

There were 23 subjects in the Study Group and 21 subjects in the Control Group. The mean age of the Study Group was 11.1 ± 7.53 years (mode 8 years; range 1-31 years). The mean age of the Control Group was 10.7 ± 8.45 years (mode 8 years; range 1-31 years).

Eight (35%) subjects in the Study Group were female. There were 21 right glaucomatous eyes (50%). Data were analyzed by randomly selecting eye per subject and both eyes per subject in the Study and Control Groups.

Entering both eyes of eligible glaucomatous subjects (Study Group) and nonglaucomatous subjects (Control Group) into the analysis pool (total 42 eyes in each Group): The mean CCT of the Study Group was 589.42 ± 53.44 μm (range 469-693 μm). The mean CCT of the Control Group (42 eyes) was 556.14 ± 30.51 μm (range 483-620 μm). The Study Group had statistically significantly thicker cornea than the Control Group (P=0.001). The mean IOP in the Study Group was 16.52 ± 4.61 mmHg (range 4-24 mmHg) and the mean horizontal corneal diameter was 14.27 ± 1.44 mm (range 12-20 mm). In the Study group, the CCT was positively correlated with IOP (r= 0.623, P<0.0001). Multiple regression analysis indicated that the number of previous surgeries, number of medications and corneal diameter had no effect on correlation between CCT and IOP.

Reanalysis of the data by randomly selecting one eye per subject indicated the mean CCT in the Study Group was still statistically significantly thicker than the Control Group (587.86 ± 56.03 μm vs.560.17 ± 27.17 μm, respectively; P=0.04). There was a strong correlation between CCT and IOP (r=0.630; P=0.001).

DISCUSSION

In this study, we found statistically significantly thicker corneas in subjects with PCG compared to a Control Group of nonglaucomatous subjects. CCT was positively correlated with IOP in subjects with glaucoma. Corneal thickening in congenital glaucoma could be an inherent component of the pathophysiology and related to the racial and genetic background. Alternately, the thicker corneas could be due to the presence of subclinical corneal edema as a result of endothelial dysfunction and long-standing IOP elevation. Additionally, corneal edema could be associated with corneal scarring that permanently increases corneal thickness.

Corneal thickness has a known effect on IOP measurement with applanation tonometry.8 This may in part be explained by the effect of corneal thickness on ocular biomechanics. Indeed, CCT may be a surrogate measure for ocular rigidity.13–15 Corneal rigidity affects the technique of tonometry which is dependent on corneal deformation. Hence, IOP is underestimated in structurally thinner corneas, whereas IOP is overestimated in structurally thicker corneas. A major exception is corneal edema, which causes an underestimation of IOP despite thickening.81116

Published data are ambigious regarding CCT in PCG cases with both thicker10 and thinner17 corneas being reported. While some congenital glaucoma cases have thinner corneas, others have thicker corneas. Thicker corneas could be due to structural changes, scarring or edema, which have differential effects on tonometry. Thinner corneas due to corneal stretching in PCG cases could behave differently. Hence, in congenital glaucoma cases, corneal thickness could have complex effects. Generally, it has been reported that CCT is significantly correlated with IOP in congenital glaucoma cases and could increase in patients without clinically visible corneal edema.17

Mastropasqua and associates18 performed an in vivo confocal microscopic evaluation of two patients with glaucomatous megalocornea. In both patients there was a mild reduction of keratocyte density in the mid- and posterior stroma, a specific abnormal “clew-shaped” morphology of stromal nerves, and the presence of discontinuous hyper-reflective structures overhanging the endothelial layer at the level of the Descemet's membrane.18 The endothelium showed severe polymegathism, pleomorphism, markedly decreased cell density, and focal cellular lesions.18 Mastropasqua and associates18 attributed reduced corneal endothelial density to corneal distension. This is in contrast to nonglaucomatous megalocornea which has normal endothelial cell density and morphology.19

Corneal and endothelial thinning caused by progressive stretching could result in corneal scar and decompensation later in life. The thinned endothelium is more susceptible to the endothelial cell loss as one ages and to the loss of function in adulthood.

Some authors reported increased corneal thickness in PCG cases and attributed this to corneal edema.20 Sampaolesi and Caruso reported that glaucomatous eyes had thicker corneas than nonglaucomatous eyes (0.64 ± 0.24 mm vs. 0.54 mm).20 However, Sampaolesi and Caruso measured glaucomatous patients preoperatively and immediate postoperatively and the effect of possible corneal edema due to postoperative factors could not be ruled out.20 Nonetheless, the thicker cornea in their PCG cases concurs with our findings in PCG patients without clinically visible corneal edema.

Muir and colleagues10 reported significantly higher mean CCT in children with ocular hypertension (595 ± 39 μm) compared to normal children (555 ± 37 μm). We found thicker corneas in definite PCG cases.

Henriques and colleagues21 reported a significant thinning in central cornea of 30 congenital glaucoma cases (525.4 ± 53.3 μm compared to 556.7 ± 26.7 μm in control subjects; P=0.01). In addition, the central corneal thickness was inversely correlated with corneal diameter and axial length.21 Our results contradicts those of Henriques and colleagues.21 The differences between studies may be due to racial differences between Brazilian and Iranian PCG patients. In addition, the PCG cases included by Henriques and colleagues21 had lower IOP than our series which could explain the differences in outcomes. Lastly, the subjects in the Henriques and colleagues21 series were much younger than the subjects in our study.

Wygnanski-Jaff and Barequet22 reported nine congenital glaucoma cases with thinner CCT in the eye with the more severe glaucomatous damage in bilateral cases or in the glaucomatous eyes compared with the normal fellow eye in unilateral glaucoma cases. This was likely due to corneal stretching and/or scarring although it contradicts the results of the current study.

Lopes and associates23 reported thinner corneas in primary congenital glaucoma cases compared to controls (543.3 ± 66.9 μm vs. 555.6 ± 38.4 μm, respectively). However, considering all the different types of diagnosed glaucoma, the average CCT was thicker in pediatric glaucoma.23 Lopes and associates23 stated that CCT could have a clinically significant effect on IOP measurements in more than half of the patients with pediatric glaucoma and suggested pachymetry results be considered in the management of pediatric glaucoma cases.23 Notably, eyes with glaucoma that had prior glaucoma surgery had a significantly lower CCT than eyes that had not undergone surgery (559.1 ± 76.4 μm vs. 650.6 ± 137.1 μm, respectively; P=0.027).23 A decrease in CCT may therefore also be helpful in tracking improvement over time.

Henriques and associates21 suggested that corneal thinning in congenital glaucoma caused by stretching pressure, along with clinical and subclinical edema caused by ruptures in Descemet's membrane and endothelium could result in inaccurate IOP measurement - generally an underestimation.21

Interestingly, Lopes and associates23 reported that the CCT of eyes with Haab's striae was significantly lower (495.1 ± 70.1 μm) than eyes without Haab's striae (575.5 ± 45.2 μm) (P=0.015). This observation may lead to greater insight of the role of corneal stretching on corneal thinning in PCG cases.

In addition to the known effect of CCT on applanation tonometry, the thin cornea due to stretching in PCG could be completely different from a “normal” thin cornea.24 Tong and associates24 demonstrated that a change in CCT in children was associated with a greater difference in measured IOP than in adults. This is particularly important in marginally controlled cases with apparent fair IOP control and smoldering progression of glaucomatous damage.24

There are several practical limitations in the follow-up of congenital glaucoma cases. IOP measurement could be inaccurate because of possible IOP reduction by general anesthesia,25 corneal edema, corneal scar, and questionable accuracy of tonometers used during EUA, especially in patients with unusual corneal thickness.78101116 In addition, the frequency of IOP reading and investigation of diurnal variation are limited by the EUA time frame. Hence, several surrogate measures such as axial length,202627 corneal diameter626 and refraction28 have been used to monitor long-term IOP control of congenital glaucoma before 3 years of age. As the patient ages, biometric parameters such as corneal diameter and axial length become less important in disease monitoring and IOP, visual fields, cup-to-disc ratio and optic disc appearance become more clinically relevant, as in open-angle glaucoma, for PCG.21

We agree with other investigators41022232930 that incorporating CCT measurement in the routine examinations of pediatric glaucoma could improve disease management. We found a strong correlation between CCT and IOP in glaucomatous children. This is in accordance with previous studies on both normal and glaucomatous children.1024

Our study was limited by wide variation in age in the study cohort. However, this is likely not a significant drawback as CCT is not associated with age in pediatric glaucoma or normal children.23 Indeed, the CCT in newborns has been found to be greater than 580 mm31 and decreases to adult values by 3 years of age.32 Another drawback was the lack of data on axial length in the current study. We suggest incorporating axial length in analyzing corneal thickness changes of PCG cases in future studies. Furthermore, halothane anesthesia could result in imprecise IOP measurement, mainly underestimation. We performed IOP measurement within the initial 5 minutes of the anesthesia to mitigate this effect. Although using different types of tonometers could affect the precision of data, good agreement between Goldmann/Perkins and TonoPen tonometers has been previously reported.33

In conclusion, we found that the CCT is higher in Iranian PCG cases than nonglaucomatous controls and recommend pachymetry as an essential component of the evaluation for diagnosis and management of pediatric glaucoma. Larger scale studies with more stringent patient selection criteria, and assessment of other biometric variables such as axial length in a longitudinal study would be valuable in determining the magnitude and importance of this effect.