Highlights
- •
This study presents new data on refractive error components: corneal, refractive and internal astigmatism, in a group of young UK based adults.
- •
A weak, but significant correlation is noted for both axial length and MSE, with both internal and refractive (total) astigmatic components of J45, i.e. oblique astigmatism.
Abstract
Purpose
There is a clear benefit in defining internal (IA) and corneal astigmatic error (CA) prior to surgical and other refractive interventions, such as orthokeratology, to minimise risk of unsatisfactory refractive outcomes. Such data would also be of relevance to broader areas of ophthalmic care such as spectacle dispensing and other types of rigid lens fitting. This study offers a detailed characterisation of astigmatic error in a group of university students and specifically investigates compensation of corneal astigmatism by the eye’s internal optics.
Methods
For 176 young-adult participants, objective measurements of refractive error were obtained using the open-view Grand Seiko WAM-5500 autorefractor; corneal curvature and axial length were measured using the Aladdin biometer. Clinical measurements of corneal and refractive astigmatism were converted into vector components J0 and J45; followed by an assessment of corneal astigmatism compensation.
Results
Mean total refractive astigmatism (RA), CA, and IA were 0.24 ± 0.32D, 0.46 ± 0.27D and -0.21 ± 0.25D respectively for J0 and -0.05 ± 0.20D, 0.01 ± 0.16D, and -0.06 ± 0.18D for J45. Significant linear correlations were noted between RA, CA, and IA for both J0 and J45 (P < 0.01). A significant linear regression was also noted between axial length and J45 RA and IA, but not CA. Levels of full compensation were low, 7% and 9% for J0 and J45 respectively, however, a complete absence of compensation was also uncommon particularly for J45 (2%).
Conclusions
In general, partial compensation for corneal astigmatism by the eye’s internal optics is noted, but it is unclear whether this is an active compensatory mechanism. Further, larger scale, studies would be required to characterise differences in corneal astigmatic compensation with respect to ethnicity.
1
Introduction
Corneal astigmatism (CA) often exceeds the total astigmatic error, but a counterbalance between the eye’s internal and corneal optics helps to minimise the total refractive astigmatism (RA) [ , ]. Corneal altering procedures such as refractive surgery and orthokeratology, or fitting refractive solutions such as rigid lenses, may, therefore, disrupt this attenuation of corneal astigmatism leading to treatment induced residual astigmatic error. To minimise the risk of such unsatisfactory treatment outcomes, there is a clear benefit in defining and understanding internal and corneal astigmatic error prior to surgical or other refractive interventions. Detailed study of astigmatism is timely given the renewed and growing clinical interest in approaches such as orthokeratology which, while possibly fuelled by an interest in myopia management, are also used to manage manifest refractive error [ ].
The principal origin of internal astigmatism (IA) is thought to be the crystalline lens; however, smaller contributions may arise from other refractive media such as the vitreous or aqueous humour. Since internal astigmatism is difficult to measure, it is instead commonly accepted as the difference between refractive and corneal astigmatic error [ ]. However, unless the refractive and corneal astigmatic axes coincide, internal astigmatism cannot be derived directly by subtracting corneal astigmatism from refractive astigmatism. Instead, Thibos et al. [ ] advocate the application of Fourier analysis to convert refractive clinical data to vector notation. The vectorial approach permits statistical analysis of vector power components J0 and J45 which respectively represent the orthogonal and oblique astigmatic powers; hence both magnitude of power and axis orientation can be evaluated.
Further characterisation of astigmatic error can be achieved through determining the amount of corneal astigmatism compensation [ , , , ]. Based on work by Muftuoglu et al. [ ] and others, Park et al. [ ] proposed a detailed system of calculating and classifying corneal astigmatism ‘compensation factor’, (CF). For a perfect optical system, full compensation would be achieved with zero refractive (total) astigmatism. However, Park et al. [ ] reported ‘under compensation’ of corneal astigmatism as the most common form of CF in young adults (aged 19–46 years old) (see Methods, Table 1 ).
| CF value | Compensation type in relation to corneal astigmatism | Meaning |
|---|---|---|
| <-0.1 | Same axis augmentation | Total astigmatism increases to values greater than CA, CA axis maintained. |
| −0.1 to 0.1 | No compensation | |
| 0.1 to 0.9 | Under compensation | Total astigmatism decreases to values less than CA, but CA axis remains the same |
| 0.9 to 1.1 | Full compensation | |
| 1.1 to 2 | Over compensation | Amount of total astigmatism decreases to values smaller than CA, axis is also changed to opposite angle |
| >2 | Opposite axis augmentation | Total astigmatism greater than CA and axis opposite angle |
While the primary focus for many studies investigating astigmatism in children is to garner clues about refractive developmental processes, particularly predictions of myopia development; in adults, characterising astigmatism may have a clinical relevance in relation to predicting visual outcomes for refractive and other corneal altering therapies such as orthokeratology.
Previous reports have provided valuable datasets which characterise astigmatism for various age groups and ethnicities (e.g. [ ], ,[ , , , ]). The principal aim of this study was to characterise and understand the interrelationships between corneal, refractive, and internal astigmatism, in a cohort of young UK based students i.e. individuals of an age where refractive surgery or alternatives such as orthokeratology may be a consideration.
2
Materials and methods
Ethical approval was provided by the internal university departmental ethics committee; all aspects of the research were carried out in accordance with the tenets of the Declaration of Helsinki.
2.1
Refractive error, axial length, and keratometry measurements
Non-cycloplegic objective measurements of refractive error were obtained using the infra-red open-view Grand Seiko WAM-5500 autorefractor (Ryusyo Industrial Co. Ltd, Osaka,Japan). Vertex distance was set to 12 mm and autorefractor output increments to 0.12D. Negative cylindrical clinical notation of refractive error was then converted into individual dioptric power vectors using the vectorial method proposed by Thibos et al. [ ] and described by many others (e.g. [ ]): MSE, which represents the spherical equivalent; J0 and J45 which respectively indicate the orthogonal (90 and 180 °) and oblique (45 and 135 °) axes of the Jackson Cross Cylinder. The following formulas were used to generate vectorial components:
Internal astigmatism (IA) was calculated for each of the power vector components (J0 and J45) by subtracting corneal astigmatism (CA) from (total) refractive astigmatism (RA).
2.2
Statistical analysis
All statistical analyses were undertaken using commercially available software (SPSS, IBM, UK). Linear regression analyses and scatterplots were used to investigate the relationship between IA, CA, RA, M and axial length
Student’s paired t -tests were used to check for differences between IA, CA, and RA for vector components J0 and J45.
Using a method proposed by Muftuoglu et al. [ ] ‘compensation factor’ (CF) was calculated; this refers to the ratio describing amount of compensation of corneal astigmatism by refractive astigmatism. To help provide a detailed characterisation of the astigmatic error, CF type was assigned according to the classification system proposed by Park et al. [ ] (see Table 1 for details).
3
Results
3.1
Participants
Data from all participants found to have undergone refractive surgery were excluded from the analysis.
One-hundred and seventy-six young-adult participants were eligible for inclusion in the study (mean age 21.1 ± 2.3 yrs; range 18–36 years, age data available for n = 176) from the university student population. Ethnicity data were available for the majority of subjects (n = 172); ethnicity groupings were provided by participants via questionnaire; the ethnicity options reflected those provided by the Office of National Statistics UK, ( https://www.ons.gov.uk/methodology/classificationsandstandards/measuringequality/ethnicgroupnationalidentityandreligion ). The cohort was predominantly comprised of individuals who identified as being of Indian or Pakistani ethnicities (please see Table 2 for full details).
