Main

Central corneal thickness (CCT) measurements have been reported in several races, ethnic groups, and countries: Whites,1, 2, 3, 4, 5 African-Americans,2, 3, 4 Blacks,5 Latinos,4, 6 Greenland Eskimos,7 Asians4 (Hong-Kong,8, 9, 10 Chinese,11 Korean,12 Japanese,13 Singapore,14 and Mongolian,15), in Switzerland,16 Germany,17, 18 Spain,19 Denmark,20, 21 Holland,22 Turkey,23 Israel,24 Canada,25 New Zealand,26 India,27, 28 and Rwanda.29 According to Doughty and Zaman,30 who reviewed and performed a meta-analysis of more than 300 articles, if we consider that the normal cornea averages 536±0.031 μm, CCT was found to be significantly thinner (mean CCT values of less than 530 μm) in African-American, Mongolian, and Indian patients than in Caucasian, Asian, Hispanic, or European patients. To the best of our knowledge, CCT measurements have not been reported in African patients other than African-Americans living in the United States or Canada.

In recent years, importance has been attributed to CCT measurement in the screening, diagnosis, and management of ocular hypertension and glaucoma patients.31, 32 Similarly, with the development of refractive surgery procedures, CCT values are of enormous importance during the preoperative evaluation of the patients, the decision whether or not to perform surgery, the type of recommended procedure, and rate of postoperative complications.33, 34

Mixed populations are ideal for comparing CCT measurements in different ethnic groups under the supervision of the same investigators and the use of the same pachymeters. With people immigrating to Israel from almost all countries around the world, Israel demography provides an excellent opportunity to evaluate CCT in populations of different origin, including North Africa.

For this reason, people requesting refractive surgery in Israel constitute an ideal population for examining CCT measurements and for examining the relationship between CCT and patient's origin and preoperative data. It also allows describing CCT measurements in people from North African origin.

The present study was performed to report CCT for patients examined in a refractive surgery centre, and to search for possible associations between CCT and patient origin, age, sex, refractive status, keratometry, and intraocular pressure (IOP).

Methods

Patients included in this report were examined during 2003 in the Refractive Clinic at Soroka University Medical Center, in Beer-Sheva, Israel. The study included consecutive new patients presenting for refractive surgery. All data in this study are limited to the preoperative examinations, and were collected retrospectively. All patients were seen in year 2003.

The following data were collected from the preoperative examination: personal details, age, sex, manifest and cycloplegic refractions, automated keratometry, corneal topography (EyeSys® Vision, Houston, TX, USA), Goldmann applanation tonometry and CCT from both eyes. CCT was measured with an ultrasound pachymeter (Nidek US-1800). After applying a drop of benoxinate hydrochloride 0.4% for local anaesthesia, the pachymeter tip was placed perpendicular to the central cornea. A minimum of nine measurements were taken and the average reading was recorded. All measurements were performed by two of us (JL, TL) between 0400 and 0700. IOP was measured before the pachymetry measurement.

Patients were asked for their place of birth and that of their parents. Seven geographical zones were defined: (1) America, (2) Western and Central Europe, (3) Eastern Europe and Russia, (4) Asia, (5) North-Africa (ie Morocco, Algeria, Tunisia, Libya, and Egypt), (6) Israel, and (7) mixed origin, which included patients whose father and mother were born in different geographical zones. When both parents were born in the same country, patient's origin was recorded in one of the first six groups.

Inclusion criteria for this study were adults 18 years of age or older, no previous ocular surgery, and no contact lens use for 2 weeks before preoperative examination in the case of soft lenses or 4 weeks in the case of gas-permeable and hard lenses. Exclusion criteria included eyes with keratoconus or forme fruste keratoconus as demonstrated by corneal topography and/or Orbscan® (Baush & Lomb, Rochester, NY, USA). Eyes with suspected corneal dystrophy by slit-lamp examination, known ocular pathology, or previous ocular surgery were also excluded.

Data were analysed using SPSS for Windows (version 12.0, Chicago, IL, USA). The CCT measurements were averaged after collection and are presented as mean±SD. The two-tailed Student's t-test was used to compare mean values of parametric data. The categorical variables analysed were sex, and parents' and patient's place of birth. The continuous variables analysed were age, sphere, cylinder, IOP, keratometry, and CCT. Linear regression analysis and analysis of variance (ANOVA) were used to analyse the possible relationship between CCT and the other variables. Statistical analysis was performed for the right eye and the left eye separately. The results were considered significant for P-value of ≤0.05.

Results

A total of 204 patients were included in this study. The demographic characteristics of the patients are given in Table 1. There were more females than males (58.8 vs 41.2%). The preoperative CCT, sphere, cylinder, keratometry, and IOP data of right and left eye are given in Table 2. In the total group, mean CCT was 529.91±33.66 μm in the right eye and 530.04±33.98 μm in the left eye. Patient's place of birth and origin (ie parents' place of birth) are given in Table 3. The majority of patients were born in Israel (n=144, 70.6%), and the majority of parents were born in North-Africa (n=121, 59.3%). Comparisons of preoperative examination data between patients of North-African origin and patients of other origins are presented in Table 4. CCT was found to be significantly thinner in patients of North-African origin (518.87±31.53 vs 545.36±30.44 μm in the right eye, and 518.43±32.07 vs 546.31±29.76 μm in the left eye; P≤0.0001). The distribution of the corneal thickness in the populations is presented in Figure 1. The majority of thinner corneas were present in people of North-African origin. People of North-African origin were also significantly older (35.9 vs 32.4 years, P=0.028). There were no differences between groups when comparing for sex (P=0.387), sphere (P=0.758 and 0.823), cylinder (P=0.60 and 0.619), keratometry (P=0.192 and 0.228), and IOP (P=0.070 and 0.073) for the right eye and the left eye, respectively.

Table 1 Patient demographics
Full size table
Table 2 Preoperative examination data
Full size table
Table 3 Place of birth of patients and their parents
Full size table
Table 4 Preoperative data by origin
Full size table
Figure 1
figure 1

Distribution of central corneal thickness by origin.

Full size image

The multivariate linear model of mean corneal thickness included sex, age, origin, preoperative sphere, cylinder, keratometry, and IOP. Only patient's origin was found to be independently associated with CCT (P<0.0001). The other tested variables were not independently associated with CCT (age, P=0.977; sex, P=0.987; sphere, P=0.998; cylinder, P=0.989; keratometry, P=0.978; and IOP, P=0.968).

Discussion

Doughty and Zaman30 recently performed a meta-analysis of the corneal thickness literature based on 300 articles. They concluded that the mean CCT of eyes reported as ‘normal’ was 534 μm. Overall, studies using slit-lamp-based pachymetry reported marginally lower CCT values (530±0.029 μm) compared to ultrasound-based studies (544±0.034 μm), but the authors attributed this difference to the type of individual studied (nonsurgical vs presurgical patients) rather than the technique itself.

CCT measurements have been reported in several ethnic groups and countries.1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29 Overall, CCT was found to be significantly thinner in African-Americans with an average CCT value in the low 530's μm. Only in two studies from Mongolia15 and India,28 mean reported CCT was below 520 μm (495±32 and 519.92±33.36 μm, respectively). In the present study, we found significantly thinner corneas in people of North-African origin with an average of 518.87±31.53 μm (right eye) than people of other origins (545.36±30.44 μm, right eye) (Table 5).

Table 5 Summary of CCT reported in the literature by patient's race and country in which the study was performed
Full size table

CCT measurement can be affected by several factors: technique of measurement, age, diurnal effects, contact lens wearer, ocular disease, and previous ophthalmic surgery.30 Given the methodology used in this study, it is unlikely that methodological bias could be the explanation for the thinner corneas found in people from North-Africa. Specifically, all the measurements were performed with the same ultrasound pachymeter, by the same two examiners, at the same hours in the afternoon. All people coming to preoperative examination were instructed not to wear soft contact lenses 2 weeks before, or 4 weeks before in the case of gas-permeable and hard lenses. Finally, people with ocular disease (ie keratoconus, Fuchs dystrophy) or previous ocular surgery were excluded from this study. Regarding the age of the patients, we found that North-African people were significantly older (35.9 vs 32.4 years). A substantial age-related increase in CCT has been reported in ultrasound pachymetry studies (approximately 15 μm/decade).30 Thus, even accepting this possible age-related effect on CCT, we could expect thicker and not thinner corneas in people from North-Africa.

Preoperative parameters such as refractive status, keratometry, and IOP can theoretically affect CCT measurement, but no differences were found between the two groups. When we performed multivariate analysis, only patient's origin was found to be independently statistically associated with CCT. The other tested variables (age, sex, refractive status, keratometry, and IOP) were not associated with CCT.

To the best of our knowledge, there is only one report29 of CCT in people from Africa, other than African-Americans actually living in the United States and Canada. The authors reported that CCT measured by slit-lamp device in 38 males and 38 females in Rwanda was 524 and 521 μm, respectively, which is similar to the reported average CCT in a meta-analysis of 125 studies using slit lamp30 (525 vs 544 μm using ultrasound pachymetry). So these Rwanda people had ‘normal’ CCT values.

An important methodological issue is the decision to use the parents' place of birth and not the patient's place of birth as the indicator of ‘origin’ for the statistical analysis. Previous CCT measurements in different ethnic groups were based on self-definition by the patient and sometimes also confirmed by demographic information found in the patient's medical record. As Israel's present population originates from almost all around the world, a self-definition of ‘race’ would be problematic and racial backgrounds difficult to assess, so we preferred to gather ‘objective’ information on place of birth, leaving the determination of ‘origin’ to the algorithm created by the investigators. As a result of this variety, the Israeli population offers an excellent opportunity to compare people coming from different geographic regions.

One potential limitation of this report is that the study population was taken from people examined at a refractive clinic, and it is not a representative population-based sample. We can only conclude that people examined at a refractive clinic whose parents were born in North-Africa have significant thinner corneas than people of other origins. A population-based study is necessary to confirm our findings.

Possible reasons for different CCT values in different ethnic groups have not been reported. Genetic, environmental, climatic and/or multifactorial reasons could be responsible for these differences in CCT. What seems to be clear from our findings is the fact that people from North-Africa coming for refractive surgery had thinner corneas than the reported average values. With the advent of refractive laser procedures during the recent years, leaving a residual stromal bed of more than 250 μm in order to avoid postoperative keratectasia has been suggested.34 At this moment it is unknown whether or not North-African people are more prone to postoperative complications, certainly an issue that merits further study.

Also recently the importance of measuring CCT in all patients with glaucoma or ocular hypertension has been highlighted. In African-American people with glaucoma, Goldmann applanation tonometry uncorrected for CCT can lead to a higher morbidity probably secondary to delay in diagnosis and setting of inadequate treatment targets.4 In the mean time, until a larger study with different design can confirm our findings, it seems to be prudent to pay special attention to CCT in North-African people with glaucoma or at glaucoma risk.