Corneal Thickness in Highlanders

Introduction

Corneal thickness is an important parameter in Ophthalmology with both diagnostic and therapeutic implications. It influences intraocular pressure (IOP) readings when measured by applanation tonometry, thicker corneas leading to overestimation of IOP and thinner, an underestimation of IOP thus impacting the treatment of glaucoma (Rand et al., 2005). Thinner corneas are an independent risk factor for the development of glaucoma in individuals with ocular hypertension (Medeiros and Weinreb, 2012). Corneal thickness measurement is of paramount importance in refractive surgeries especially in prevention of ectasia after LASIK surgery (Zhao et al., 2015) and evaluation of various corneal pathologies like keratoconus, keratoglobus, and pellucid marginal degeneration.

There are many factors, which can affect corneal thickness, one of which is increasing altitude (Morris et al., 2007; Jha, 2012). This has gained immense importance with increasing numbers of people travelling to higher altitudes for mountaineering, leisure, or work. Corneal thickness increases as an individual ascends to higher altitudes, that is, above 10,000 feet. The increase in corneal thickness affects visual acuity in patients who have undergone radial keratotomy and LASIK (Jha, 2012). Various studies and case reports have highlighted increase in corneal thickness in lowlanders on ascending to high altitude (Morris et al., 2007; Jha, 2012). However, there are no studies available in the published literature pertaining to corneal thickness of the highlanders who are inhabitants of such altitudes.

Information of the health status of highlanders is very sparse due to difficulty in collecting data from inaccessible areas and people living in distant areas in small pockets. Extrapolation of values of normalcy of people who inhabit lower altitudes is used in people living at high altitude without knowing if the same is applicable to them. Therefore, to learn about one of the most important parameters affecting measurement of IOP, that is, corneal thickness, we measured the corneal thickness in highlanders who have lived in altitudes more than 11,000 feet without having come to lower altitudes. Extensive literature search was done by various search engines, including Google, PubMed, Embase, and Medline, but we could not find studies pertaining to corneal thickness measurement in highlanders. In view of the above, we carried out the present study to determine the corneal thickness of highlanders who lived at heights of more than 11,000 feet and compared it with corneal thickness of lowlanders.

Materials and Methods

The study population consisted of persons living in high altitude (more than 11,000 feet) who had never visited the lowland (below 11,000 feet). A convenience sampling from the highlander population accessible to researcher was taken. The highlander participants of the study consisted of inhabitants at an altitude of more than 11,000 feet above sea level in Leh District of Ladakh region of India. Inclusion criteria included adult highlanders, who were not suffering from any ocular diseases, had not used contact lens, were also not on any topical or systemic medication, and had never visited the lowlands. The lowlander participants consisted of inhabitants at an altitude of 1500 feet above sea level in Delhi, India and not suffering from ocular diseases, had not used contact lens, and were not on systemic or topical medication. An informed consent was taken from each participant.

The sample size was calculated assuming alpha error to be 0.05, power of study to be 90%, estimated difference between central corneal thickness (CCT) to be 10 μm, and standard deviation to be 30 and assuming that to be equal in both the populations. The calculated sample size was 382, that is, 191 in each group. However, a total of 466 (254 highlanders; 212 lowlanders) participants were finally enrolled, which gives power of 95% to the study.

CCT was measured by ultrasound pachymetry (DGH-555B, Pachette 4; DGH Technology, Inc., Exton, PA) after instillation of one drop of Proparacaine hydrochloride 0.5% eye drops. All measurements were taken between 10 a.m. and 5 p.m. to minimize the effects of diurnal variation on CCT. The method was standardized and used in both populations. All the measurements were taken by researchers themselves. A mean of 25 measurements of CCT of every participant was obtained for each eye using ultrasonic pachymeter. The study was approved by local institutional ethics committee at Ladakh and Delhi.

Data were analyzed using Stata version 13. “t”-Test was used to compare quantitative variables between two groups. Chi square test was used for categorical variable. Regression analysis was used to adjust the effect of age and sex on CCT. Regression modeling was done using random effect with one step cluster using maximum likelihood estimation. The goodness of fit of the model was checked.

We used data from both eyes so that no data were wasted and used the correlation coefficient between readings of both eyes in a single subject (hierarchal model) to obtain more power to the study.

Results

A total of 212 lowlanders and 254 highlanders participated in the study. There were 12.7% and 46.4% of females in lowlanders and highlanders, respectively (Table 1). The mean (standard deviation) age of lowlanders and highlanders was 41.8 (15.9) and 47.7 (17.7) years, respectively.

The mean CCT readings are shown in Table 2 and Figure 1. There was no statistically significant difference in mean CCT readings of right eye and left eye in either lowlanders or highlanders [532.7 (34.7) μm in right eye versus 534.4 (35.9) μm in left eye in lowlanders and 520.7 (32.5) μm in right eye versus 521.3 (33.8) μm left eye in highlanders]. Distribution of CCT reading in right eye and left eye is shown in Figure 2. The CCT reading of both eyes was highly correlated (rho = 0.94).

OPEN IN VIEWER

FIG. 1.

Box and whisker plot of CCT reading in lowlanders (0) and highlanders (1) in right (blue color) and left (red color) eye. CCT, central corneal thickness.

OPEN IN VIEWER

FIG. 2.

Distribution of CCT reading in right eye (blue circle) and Left eye (red circle).

Since the reading within two eyes is highly correlated or clustered together, random effect cluster modeling with maximum likelihood estimation was used to adjust the effect of covariates (age and gender) on lowlanders and highlanders. The goodness of fit of the model was statistically significant (p < 0.001). The model shows that highlanders have 11.95 μm lower mean CCT reading compared to lowlanders after adjusting for age and sex (p < 0.001). Age also had a significant effect after adjustment for location and sex (p = 0.001). A year increase in age leads to 0.31 μm decrease in CCT reading, that is, a decadal increase in age will decrease the corneal thickness by 3.1 μm. Sex does not have statistically significant effect after adjusting for age and location. The results are shown in Table 3.

Discussion

Corneal pachymetry is the method of measuring the thickness of the cornea. It can be done by various contact (ultrasound pachymetry and confocal microscopy) and noncontact (Pentacam, Sirius, Galilei, anterior segment optical coherence tomography, Orbscan, and so on) methods (Mashige, 2013). The gold standard for measuring corneal thickness is by ultrasound pachymetry (Sadoughi et al., 2015). Many factors affect CCT like race, age, sex, anthropometric parameters, drugs, time of day, blink rate, and type of equipment used. Studies on CCT among various ethnic groups have shown a wide range of CCT, varying between 508 and 563 μm (Mashige, 2013). It is known that Japanese have thinner corneas than Chinese and Filipinos. The African Americans have thinner corneas than normal, the average CCT of nonglaucomatous subjects being 512.4–550 μm (Soatiana et al., 2014).

Our study has shown a mean CCT of 532.7 (34.71) μm of right eye and 534.4 (35.9) μm in left eye in lowlanders, compared to mean CCT of 520.7 (32.5) μm in the right eye and CCT of 521.3 (33.8) μm in the left eye of highlanders. There is a statistically significant difference between the CCT of the two groups studied, thinner corneas being found in highlanders. However, it should be noted that the participants in highlander group belonging to the ethnic Ladakh region and the lowlanders to people belonging to North India, staying at low altitudes, are different in terms of their diet, habits, and environmental factors (like oxygen concentration and barometric pressure). And there is a need to further study the effect of these factors on CCT. It is difficult to precisely point out the reason for thinner corneas in highlanders, but it may be conjectured that it is an epigenetic change to allow better diffusion of oxygen through the cornea due to low atmospheric pressure at high altitude.

On comparing the CCT of highlander group of our study with other ethnic groups in the published literature who have thinner corneas, we found that the average CCT was 531.7 μm in Japanese population and varied between 512.4 and 550 μm in nonglaucomatous subjects in the sub-Saharan region (Soatiana et al., 2014). CCT is different in different races. The finding of lesser CCT in highlanders may be a racial difference, like the differences found among Chinese CCT-555.6 μm, Caucasians 550.4 μm, Hispanics 548.1 μm, Filipinos 550.6 μm, African Americans 521.0 μm, Japanese 531.7 μm (Aghaian et al., 2004), Chinese in Hong Kong 575 μm, Spanish population 548.21 μm, and two Australian Aborigine populations 514 and 511 μm (Gros-Otero et al., 2011).

Pan et al., in a multiethnic study of three varied ethnicities residing in the same geographical location in China, found a difference in CCT among various ethnic groups: adults of Han ethnicity were found to have thinner corneas in comparison with ethnic minorities like Bai or Yi. They have also stated that people of different ethnicities have different thickness of corneas, even if they have been exposed to the same environment (Pan et al., 2011). However, Godar et al. (2012) found no significant correlation of gender and ethnicity with CCT at 5% level in a hospital based study conducted in Nepal. Similarly, native highlanders may also have a lower CCT in comparison to lowlanders, most probably due to an ethnic difference although the effect of environment cannot be ruled out.

Our study also found CCT of the two eyes correlating with each other and there was no correlation of CCT with gender. Studies published in the literature have shown varied findings of difference of CCT between sexes. Shimmyo et al., Hahn et al., and Garcia Medina et al. found males to have thicker corneas than females (Mashige, 2013). A study by Suzuki et al. (2005) found CCT to be greater in men than women in Japanese population, although they found it to correlate negatively with age. In contrast, Gros-Otero et al. (2011) found no relation of CCT with sex. Another confounding factor in females is the relationship of CCT with menstrual cycle. Ghahfarokhi et al. (2015) found that the cornea is thickest during ovulation time and thinnest at the end of the cycle; therefore, CCT of women will differ with their menstrual cycle.

In this study, we also found that there was a decrease in corneal thickness with age, an increase in age showed a decrease of 0.31 μm/year in CCT. In studies done by Iyamu et al. (2013) and Hawker et al. (2009), CCT correlated negatively very weakly with age (Pearson's r = −0.063, p = 0.047). In the Gutenberg Health study, the younger age group was found to have thicker corneas than the older (Hoffmann et al., 2013). The Barbados Eye study also found thinner corneas with increasing age (Nemesure et al., 2003). Aghaian et al. (2004) in another multiracial cross-sectional study found a 3 μm decrease in corneal thickness per decade increase of age. In contrast, Foster et al. (1998) found a larger (5–6 μm) decrease in CCT per decade of age in a cross-sectional study of Mongolian patients. The landmark Ocular hypertension treatment study too revealed thinner corneas, 6.3 μm decrease in corneal thickness per decade of age on cross-sectional analysis (Brandt et al., 2001). In another hospital based study on hilly tribes of Nepal by Godar et al. (2012), they reported mean CCT of 538 (32) μm in right eye and 540 (30) μm in left eye. The authors also noted decreasing corneal thickness with increasing age (Godar et al., 2012).

There are certain limitations of the study. First, the convenience sampling limits the generalizability of the findings. Second, the study was conducted in participants with different ethnic backgrounds as people with different ethnic background may have different CCT, the variation due to ethnicity or high altitude cannot be discerned from the study.

However, this is the first study showing lesser corneal thickness values in highlanders who live in altitudes of more than 11,000 feet. The importance of finding lies in the fact that these two groups belong to two different populations having different mean CCT, hence the values which are normal (or abnormal) for lowlanders as a group may not apply to highlanders (as a group) or vice versa. Unlike the lowlanders whose CCT increases temporarily on ascent to these heights, the inhabitants of such heights have thinner corneas. Ascent to higher altitudes does result in corneal edema. Greater corneal thickness is associated with shorter acclimatization, and the amount of reduction in the partial pressure of oxygen parallels the increase in CCT (Bosch et al., 2010).

The thinner corneas of highlanders may have a bearing on diagnosis and treatment of glaucoma, refractive surgery, and contact lens fitting implantation of Intacs, and astigmatic keratectomy done on such patients although thinner corneas increasing their susceptibility to diseases is not known. The study also opens the scope of further research in this area.