Slit-scan tomography evaluation of the anterior chamber and corneal configurations at different ages

Authors


Anders Behndig MD, PhD
Department of Clinical Science/Ophthalmology
Umeå University Hospital
SE-901 85 Umeå
Sweden
Tel: + 46 90 785 37 31
Fax: + 46 90 13 34 99
Email: anders.behndig@ophthal.umu.se

Abstract.

Purpose: To evaluate the aqueous humour and corneal volumes, their correlations to age, sex and refractive status, and their changes with age.

Methods: A total of 153 eyes of 153 healthy volunteers and 58 eyes of 58 patients planned for cataract surgery were examined with Orbscan II slit-scan tomography and the autorefractometer-keratometer. In 16 eyes of 16 volunteers, the same examinations were performed twice with a 4-year interval. Anterior chamber volumes were calculated with a 3-dimensional mapping method, corneal volumes were calculated, and multiple refraction and corneal/anterior chamber configuration variables were registered.

Results: The aqueous humour volume is inversely correlated to the age of the individual (r = − 0.22, p = 0.001), with an average decrease of 1.4 ± 2.6 µl per year on longitudinal follow-up (p = 0.042). Specifically, the posterior part of the anterior chamber undergoes a pronounced reduction in volume with time, whereas the volume of the anterior part increases slightly with time. Increasing steepness and peripheral thinning of the cornea (p = 0.034), and a reduction in corneal volume (p = 0.037) were also seen with increasing age. Males had less steeply curved corneas and higher aqueous humour volumes than females.

Conclusion: The anterior segment of the eye undergoes continuous alterations with age, which differ significantly between the genders. These normal differences and alterations may be of importance in the planning of refractive procedures, and in the evaluation of disease processes.

Introduction

The anatomy of the eye's anterior segment is altered continuously by the process of ageing. These changes often have clinical relevance and are sometimes associated with the development of various ocular diseases. A well known example is the continuous growth of the lens throughout the lifespan of an individual, which may eventually be closely related to presbyopia and nuclear cataract (Bron et al. 2000), and which may also indirectly contribute to increased intraocular pressure (Altan et al. 2004), or even, in extreme cases, to narrow-angle glaucoma (Markowitz & Morin 1984). Age-related changes in the cornea can be less clinically obvious, but they do occur, and can be reflected in, for example, the response to some surgical procedures (Feltham & Wolfe 2000). A more detailed knowledge of the normal configuration of the eye's anterior segment and its natural change over time can be of importance in the planning of refractive procedures such as phakic intraocular lens (IOL) implantation (Hosny et al. 2000; Vetrugno et al. 2000; Cosar & Sener 2003), but also in the understanding and quantification of disease processes such as keratoconus (Auffarth et al. 2000), cataract and glaucoma (Markowitz & Morin 1984; Altan et al. 2004).

The aim of the present study was to map the age-related changes of the anterior chamber and cornea, and to clarify the influence of gender and refractive status on these changes.

Materials and Methods

The investigation was performed at the Department of Clinical Sciences/Ophthalmology, Umeå University, Umeå, Sweden and was approved by the ethical research committee of Umeå University, Umeå, Sweden. Orbscan II (Bausch & Lomb, Inc., San Dimas, California, USA) slit-scan tomography was performed in the right eyes of 153 volunteers (mainly medical students and hospital personnel). The acoustic equivalent correction factor (0.92) was used to achieve equivalence with ultrasonic pachymetry as recommended by the Orbscan II manufacturer. The autokerato-refractor KR 8100 (Topcon, Inc., Paramus, New Jersey, USA) was used to determine the refractive status of all eyes. The persons examined were all healthy, had no known ocular disease and had not undergone previous ocular surgery, and slit-lamp examination revealed no signs of anterior segment disease. The male : female ratio was 51 : 102. The mean age was 32.1 ± 6.2 years (range 22–47 years). The spherical equivalent (SE) ranged from + 3.25 D to −11.25 dioptres (D) and the cylinder from 0 D to 3.00 D. In addition, to increase the age span in the material for the corneal measurements, Orbscan II examinations were performed on 58 eyes of 58 individuals (age 72.2 ± 7.8 years, range 51–85 years) due for cataract surgery, who showed no signs of corneal disease on slit-lamp examination. As the subjects of this group all had cataract, no data for the posterior part of the anterior chamber or for anterior chamber depth were extracted from these measurements (see below). In 16 eyes of 16 individuals, the above-mentioned measurements were performed twice with a 4-year interval. These individuals were aged 39.0 ± 4.4 years at the first examination (range 26–47 years) and the male : female ratio was 6 : 10. All participants provided informed consent.

The following data were extracted from the Orbscan II measurements: anterior and posterior best fit sphere (BFSANT and BFSPOST, respectively), central anterior chamber depth (ACD), horizontal corneal diameter (‘white-to-white’, WW), corneal thickness at the fixation point (CTFIX), and mean corneal thickness (for an 11-mm central zone, CTMEAN). A matrix of depth measurements of the anterior chamber (from the lens/iris to the inner surface of the cornea) and of the anterior part of the anterior chamber (from the limbus plane to the inner surface of the cornea), was extracted from Orbscan II raw data using the ‘anterior lens endo’ and ‘true elevation posterior’ algorithms, respectively. The aqueous humour volume (VAH) was calculated from the ACD matrices, with 3-dimensional mapping as previously described (Behndig & Markstrom 2004). Briefly, a matrix of polar co-ordinates of anterior chamber depth was created, and was processed using the Mathematica software (Wolfram Research, Inc., Champaign, Illinois, USA). In the peripheral part of the anterior chamber, the depth was estimated by polynomial extrapolation. The VAH was calculated using linear regression between the data points and a rotation integral of the anterior chamber depth for volume calculation. The volume of the anterior part of the anterior chamber (VAC ANT) was calculated accordingly, and the volume of the posterior part of the anterior chamber (VAC POST) was calculated as (VAH − VAC ANT). The corneal volume (VC) was calculated using the formula for the volume of a cylinder with either CTFIX (Brubaker 1989) or CTMEAN as the height of the cylinder. From the autorefractometer data, the sphere and cylinder were registered, and the spherical equivalent (Sph) was calculated as (sphere + cylinder/2).

Student's 2-sample t-test was used for interindividual, and Student's paired t-test for longitudinal comparisons. Correlations were tested with Pearson correlations. Backward stepwise multiple regression analyses were performed to identify the variables independently associated with age, sex and Sph, involving the variables showing a significant correlation on univariate analysis, or a significant difference with the t-test. A level of p ≤ 0.05 was considered statistically significant.

Results

Table 1 presents the various anterior segment variables, the differences between males and females, and their correlation with age and spherical equivalent. In brief, the corneas were less steeply curved in males than in females on the anterior corneal surface, whereas the opposite applied to the posterior corneal surface. The corneal volume, when calculated using CTMEAN, was also larger in males than in females. The CTFIX and CTMEAN, however, did not differ significantly between the genders. The ACD and VAH were larger in males, compared with females, although the refractive status did not differ between males and females in this material (sphere: −1.40 ± 2.44 D versus −1.50 ± 2.45 D, p = 0.82; cylinder: −0.82 ± 0.74 D versus −0.84 ± 0.56 D, p = 0.88, and spherical equivalent: −1.88 ± 2.40 D versus −1.81 ± 2.57 D, p = 0.87, respectively).

Table 1.  Various anterior segment variables and their correlations with sex (males : females), age and spherical equivalent. A positive r-value indicates an increase in the variable with age/sph; a negative value indicates a decrease. The ACD and VAH measurements, as well as all the comparisons with the Sph, involved 153 eyes from 153 individuals. All other measurements involved 211 eyes from 211 individuals. Multivariate analysis was performed when significant correlations were found on univariate analysis (age, Sph) or significant differences were found on the t-test (sex).
  BFSANTBFSPOSTWWVC (CTFIX)VC (CTMEAN)CTFIXCTMEANACDVAH
  1. M = males; F = females.

  2. Sph = spherical equivalent.

  3. BFSANT = anterior best fit sphere in mm; BFSPOST = posterior best fit sphere in mm; WW = corneal diameter in mm; VC = corneal volume, with two alternate measurement methods (see text), in µl; CTFIX = central corneal thickness in µm; CTMEAN = mean corneal thickness in µm; ACD = anterior chamber depth in mm; VAH = aqueous humour volume in µl.

 Mean ± SD7.90 ± 0.286.62 ± 0.3211.69 ± 0.3758.64 ± 5.963.31 ± 5.8590 ± 43590 ± 432.97 ± 0.38183 ± 41
MMean ± SD7.96 ± 0.316.61 ± 0.3111.75 ± 0.3759.60 ± 6.564.42 ± 5.9593 ± 39593 ± 393.09 ± 0.31194 ± 40
FMean ± SD7.86 ± 0.256.64 ± 0.3211.66 ± 0.3658.13 ± 5.562.73 ± 5.8545 ± 43588 ± 442.91 ± 0.39174 ± 40
Univariate analysis
Sexp =<0.019 0.025 0.0770.10 0.048 0.530.32<0.0020 0.0019
Ager =−0.31 0.31 0.0580.060−0.16 0.0560.025−0.67−0.22
p =<0.001<0.001 0.400.38 0.047 0.460.072<0.001 0.001
Sphr =0.051−0.125−0.290.049 0.071−0.0090.071−0.32−0.23
 p =0.54 0.12<0.0010.55 0.39 0.910.39<0.001 0.012
Multivariate analysis
Sexp =0.034<0.001NDND0.021NDND 0.0110.002
Agep =0.0140.037NDND0.14NDND 0.0260.013
Sphp =NDND0.004NDNDNDND<0.0010.006

A gradual steepening of the anterior corneal surface, combined with a gradual flattening of the posterior surface was seen with age (0.0046 mm/year and 0.0051 mm/year, respectively; Fig. 1). The steepening of the anterior corneal surface was also seen on the longitudinal follow-up (0.011 mm/year; Fig. 2, Table 2). The CTMEAN was reduced with age (0.38 µm/year), whereas the CTFIX was unaltered. Pronounced reductions of the ACD and the VAH with age were also seen (0.013 mm/year and 1.4 µl/year, respectively; not determined for the older group of patients, see above; Table 1). The latter two variables, together with the corneal diameter, were also significantly correlated to the spherical equivalent.

Figure 1.

The BFSPOST (○), and the BFSANT (•) plotted against age. Note the gradual increase in the BFSPOST (r = 0.31, p < 0.001) and the corresponding decrease in the BFSANT (r = −0.31, p < 0.001).

Figure 2.

The BFSANT measured twice with a 4-year interval in 16 eyes of 16 individuals (individuals: inline image, grey lines; mean: °, black line). A slight, but significant decrease is seen (p = 0.034).

Table 2.  Alterations in various anterior segment variables with time (16 eyes, 16 individuals). The interval between measurements 1 and 2 is 4 years. A slight myopic shift is seen in this material. Note the steepening of the cornea and the reduction in mean corneal thickness and volume. There is also a reduction in the anterior chamber depth and the aqueous humour volume, but the anterior part of the anterior chamber shows an increase in volume.
 BFSANTBFSPOSTVC (CTFIX)VC (CTMEAN)CTFIXCTMEANACDVAHVAC ANTVAC POSTSph
  1. Me 1 = measurement 1; Me 2 = measurement 2.

  2. BFSANT = anterior best fit sphere in mm; BFSPOST = posterior best fit sphere in mm; WW = corneal diameter in mm; VC = corneal volume, with two alternate measurement methods (see text), in µl; CTFIX = central corneal thickness in µm; CTMEAN = mean corneal thickness in µm; ACD = anterior chamber depth in mm; VAH = aqueous humour volume in µl; Sph = spherical equivalent.

Me 17.96 ± 0.286.53 ± 0.2958.3 ± 4.663.7 ± 5.0551 ± 41602 ± 352.87 ± 0.33174 ± 3683.6 ± 16.083.7 ± 36.11.85 ± 2.02
Me 27.91 ± 0.266.55 ± 0.2957.7 ± 5.063.0 ± 5.0548 ± 41597 ± 342.85 ± 0.33168 ± 3692.1 ± 12.974.8 ± 32.01.97 ± 2.25
p =0.0340.290.0900.0370.170.0130.00180.0420.0450.00830.013

Generally, where differences were found on the t-test, or correlations were found on univariate analysis, an independent correlation with age, sex, or Sph was also found on multivariate analysis, with the exception of the corneal volume, which was not independently correlated with age.

Longitudinal comparisons over a period of 4 years confirmed much of the above-mentioned findings (Table 2). A slight myopization was noted in this material, as seen in a reduced spherical equivalent. The BFSANT was reduced, but in this subgroup, the increase in the BFSPOST did not reach statistical significance. The VC was reduced, due to a reduction in the CTMEAN, as was the total VAH. An average decrease in the VAH of 1.4 ± 2.6 µl/year could be calculated. Interestingly, however, the VAC ANT was increased, but overwhelmed by a more pronounced reduction in the VAC POST. The latter finding, however, was not confirmed in the larger material (data not shown).

Discussion

The present study demonstrates that significant changes occur with time in the configurations of the eye's anterior segment. With increasing age, a thinning of the peripheral cornea and a steepening of the anterior surface were noted, findings which confirm those of previous investigations (Hayashi et al. 1995; Pardhan & Beesley 1999). In addition, we noted a flattening of the posterior corneal surface with age. Using the CTFIX to calculate the corneal volume (Brubaker 1989) is likely to result in an underestimation of this volume, in the present study by approximately 8.6%. When the CTMEAN is used to calculate the corneal volume, a slight reduction of this volume is seen with age on longitudinal follow-up, which is likely to be due to age-related thinning of the peripheral cornea.

Using longitudinal comparisons, a time-related increase in the volume of the part of the anterior chamber underlying the corneal dome was found. This finding could not be confirmed in the larger material, possibly due to inaccurate determinations of the limbal plane with the Orbscan II. Due to the high repeatability of the Orbscan II measurements, however, small differences that occurred over a 4-year period could still be detected. Despite the increase in the VAH ANT, the total VAH was reduced with time, due to an even more pronounced decrease in the VAH POST. This decrease is in accordance with results of previous studies of the ACD and VAH (Johnson et al. 1978; Markowitz & Morin 1984; Cosar & Sener 2003), and is most likely due to the growth of the lens (Bron et al. 2000). From the present findings, however, it appears that the posterior part of the anterior chamber diminishes more rapidly than simple ACD or VAH measurements would indicate, which may be of importance when predicting longterm results after implantation of phakic IOLs. Contrary to our findings, the results published by Kinge et al. (1999) showed no decrease in anterior chamber depth over a 3-year period in a study involving 149 individuals. Their study, however, had a shorter follow-up time than ours, and was based on A-scan measurements, which may be less reliable than optical methods of determining anterior chamber depth (Koranyi et al. 2002).

Differences were also found between male and female subjects, but the depth and volume of the anterior chamber and the corneal diameter were the only variables significantly correlated with refractive status in this material. An association between a steep cornea and myopia, as has been suggested (Grosvenor & Goss 1998), could thus not be confirmed in the present study (p = 0.54; Table 1).

Slit-scan tomography using the Orbscan II system has been compared to other methods of measuring corneal thickness and ACD, and is often − although not always (Iskander et al. 2001; Suzuki et al. 2003) − reported to overestimate the corneal thickness slightly (Yaylali et al. 1997; Marsich & Bullimore 2000), which agrees with our own experience in previous investigations. The anterior chamber depth, on the other hand, may be slightly underestimated with the Orbscan II system (Vetrugno et al. 2000), although a previous study carried out by our group showed a good correlation between Orbscan II-based and invasive methods of determining aqueous humour volume (Behndig & Markstrom 2004), and a study by Koranyi et al. (2002) showed good agreement between different optical methods of estimating anterior chamber depth. The repeatability of Orbscan II pachymetry and ACD measurements has been shown to be superior to that of corresponding ultrasonic measurements (Marsich & Bullimore 2000; Vetrugno et al. 2000).

The present study further emphasizes that the anterior segment of the human eye consists of dynamic structures, which change over time, and whose configurations also differ significantly between the genders. Such normal changes and differences should be taken into account when refractive surgical procedures involving these structures are planned, or when anterior segment disease processes are being evaluated.

Ancillary